research article propagation analysis for automated...

16
Research Article Propagation Analysis for Automated Switching of Embedded RFID with GPS in Wireless Sensor Network Platform F. A. Poad and W. Ismail Auto-ID Laboratory, School of Electrical and Electronic Engineering, Universiti Sains Malaysia, Seri Ampangan, 14300 Seberang Perai Selatan, Pulau Pinang, Malaysia Correspondence should be addressed to W. Ismail; [email protected] Received 23 July 2014; Revised 1 October 2014; Accepted 6 October 2014 Academic Editor: Gregorio Martinez Perez Copyright © 2015 F. A. Poad and W. Ismail. is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. A 2.45GHz active integrated ZigBee RFID system embedded with GPS is developed with an automated switching protocol algorithm utilizing wireless sensor network (WSN) platform to track the physical belongings or people in indoor and outdoor environments within a control area. In addition, to make the system contactless when the tag is not joining the network, the function of global system for mobile (GSM) communication is added to the integration. e communication between RFID reader and RFID tags is via WSN platform that supports mesh network and self-healing mechanism. Energy efficiency, robustness, and reliability are key factors in the design algorithm of the system. Comparisons are made between the existing active RFID system and the proposed active RFID system to study the performance of the proposed RFID system with an automated switching algorithm implemented on it. Based on the experimental study and statistical analysis done for all situations, a conclusion has been made where the proposed embedded RFID tag provided better signal propagation compared to the existing RFID tag and the battery lifetime for embedded RFID tag with switching mechanism is better than without switching mechanism. 1. Introduction RFID and WSN are two important wireless technologies that have a wide variety of applications in current and future systems. ere are a number of advantages to merging these two technologies. RFID tags are much cheaper than sensor nodes. It is economical to use RFID tags to replace some of the sensor nodes in WSN. Moreover, because an object that is embedded with an RFID tag is trackable, RFID technology provides a reasonable addition to WSN in tracking objects that otherwise are difficult to sense especially in an indoor environment. On the other hand, WSN offers a number of advantages over traditional RFID systems [1]. Sensors can provide various sensing capabilities to RFID tags, push logic into nodes to enable RFID readers and tags to have intelligence, and provide the RFID system with the capability of operating in multihop communication, which potentially can extend the applications of RFIDs. In addition, GPS is the best solution to track the outdoor location since it is unable to work completely indoors or in an environment with obstacles and GSM is for global communication. By adding GPS and GSM into RFID and WSN, a comprehensive system is developed which can provide both positioning technologies within either a control area or a global area. Recently, most of the researchers focused on the track- ing such as buses, containers, logistics, vehicles, humans, and assets only for indoor or outdoor environment [210]. However, little has been done in the tracking of both Indoor and outdoor location especially using WSN platform [6]. Due to this reason, the embedment between RFID, GPS, GSM, and WSN is the main contribution part and the automated switching algorithm is implemented to facilitate indoor and outdoor location tracking simultaneously as presented in previous work [11]; thus this new paper is a continued work from the previous publication in [11]. e algorithm developed includes multimode configuration; Hindawi Publishing Corporation International Journal of Distributed Sensor Networks Volume 2015, Article ID 392385, 15 pages http://dx.doi.org/10.1155/2015/392385

Upload: others

Post on 21-Sep-2020

3 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Research Article Propagation Analysis for Automated ...downloads.hindawi.com/journals/ijdsn/2015/392385.pdf · Hybrid tag includes active RFID, GPS, satellite, and sensors [ ] MHz

Research ArticlePropagation Analysis for Automated Switching of EmbeddedRFID with GPS in Wireless Sensor Network Platform

F A Poad and W Ismail

Auto-ID Laboratory School of Electrical and Electronic Engineering Universiti Sains Malaysia Seri Ampangan14300 Seberang Perai Selatan Pulau Pinang Malaysia

Correspondence should be addressed to W Ismail eewidadusmmy

Received 23 July 2014 Revised 1 October 2014 Accepted 6 October 2014

Academic Editor Gregorio Martinez Perez

Copyright copy 2015 F A Poad and W Ismail This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

A 245GHz active integrated ZigBee RFID system embedded with GPS is developed with an automated switching protocolalgorithm utilizing wireless sensor network (WSN) platform to track the physical belongings or people in indoor and outdoorenvironments within a control area In addition tomake the system contactless when the tag is not joining the network the functionof global system formobile (GSM) communication is added to the integrationThe communication between RFID reader and RFIDtags is viaWSN platform that supports mesh network and self-healing mechanism Energy efficiency robustness and reliability arekey factors in the design algorithm of the system Comparisons aremade between the existing active RFID system and the proposedactive RFID system to study the performance of the proposed RFID systemwith an automated switching algorithm implemented onit Based on the experimental study and statistical analysis done for all situations a conclusion has been made where the proposedembedded RFID tag provided better signal propagation compared to the existing RFID tag and the battery lifetime for embeddedRFID tag with switching mechanism is better than without switching mechanism

1 Introduction

RFID andWSN are two important wireless technologies thathave a wide variety of applications in current and futuresystems There are a number of advantages to merging thesetwo technologies RFID tags are much cheaper than sensornodes It is economical to use RFID tags to replace some ofthe sensor nodes inWSNMoreover because an object that isembedded with an RFID tag is trackable RFID technologyprovides a reasonable addition to WSN in tracking objectsthat otherwise are difficult to sense especially in an indoorenvironment On the other hand WSN offers a numberof advantages over traditional RFID systems [1] Sensorscan provide various sensing capabilities to RFID tags pushlogic into nodes to enable RFID readers and tags to haveintelligence and provide the RFID system with the capabilityof operating in multihop communication which potentiallycan extend the applications of RFIDs In addition GPS is

the best solution to track the outdoor location since it isunable to work completely indoors or in an environmentwith obstacles and GSM is for global communication Byadding GPS and GSM into RFID andWSN a comprehensivesystem is developed which can provide both positioningtechnologies within either a control area or a global area

Recently most of the researchers focused on the track-ing such as buses containers logistics vehicles humansand assets only for indoor or outdoor environment [2ndash10]However little has been done in the tracking of both Indoorand outdoor location especially using WSN platform [6]Due to this reason the embedment between RFID GPSGSM and WSN is the main contribution part and theautomated switching algorithm is implemented to facilitateindoor and outdoor location tracking simultaneously aspresented in previous work [11] thus this new paper isa continued work from the previous publication in [11]The algorithm developed includes multimode configuration

Hindawi Publishing CorporationInternational Journal of Distributed Sensor NetworksVolume 2015 Article ID 392385 15 pageshttpdxdoiorg1011552015392385

2 International Journal of Distributed Sensor Networks

sleep and continuous mode multiprotocol reader talk first(RTF) and tag talk first (TTF) and multimode communica-tion and application programming interface (API)mode andapplication transparent (AT) mode

2 Literature Review

US Patent 20080030306 A1 [12] describes an RFID devicethat is an active transponder which includes a transmitter areceiver and a microprocessor The invention relates RFIDdevices to many applications for example for determiningthe location of objects or person that is tagged with RFIDtag However the document does not describe any additionalsolutions on how to monitor or locate the objects or personsespecially for an indoor and outdoor location environmentrespectively US Patent 20090085745 A1 [13] describes anRFID tracking device that includes aGPS capable of detectingthe unique identification (ID) from the RFID tag of theobjects and obtaining the location coordinate provided byGPS However GPS is unable to work completely indoorsor in an environment with obstacles which thus contributesto the weaknesses of the design The WO 2003050960 A2[2] is another patented system that combines the RFIDand GPS functionality on intelligent label The system usedactive RFID reader and GPS to track the location and sendthe location data information to the RFID reader in shortdistances only without involving WSN platformThis systemis however unable to work completely indoors or in anenvironment with obstacles The EP 1752908 A22006 [3]is another patented system that integrates the GPS func-tionalities in a portable RFID system to track the locationof an object or pallet inside or outside the structures suchas warehouses and the factory automation environmentHowever this system integrates the GPS receiver on theRFID reader portion to track the location and send theinformation to themonitoring station via wireless transceiverin short distances without introducingWSN platformThis isin contrast with the proposed embedded RFID system thatutilized the GPS function on the RFID tag portion to trackthe location

Real time location system (RLTS) with ZigBee Technol-ogy (EP 2196816 A12008) [4] is another work using a similarZigBee Technology IEEE 802154ZigBee standard as theproposed embedded RFID tag that operates in unlicensedfrequency bands (868MHz in Europe 915MHz in USAand 24GHz almost worldwide) The system involved WSNin the implementation however the system only provideslocation tracking for indoor environmentwithout integrationwith GPS technology US 20090043504 A1 [5] develops asystem and method for locating tracking and monitoringthe status of personnel andor asserts However the systemdid not involve RFID or GPS technology and only focuses ona computer-implemented method for generating a positionestimation of a tracker in an indoor location A most similarsystem developed by Numerex and Savi Technology [6] ahybrid tag that includes active RFID GPS satellite commu-nication and sensors introduces an automated switching

mechanism however the system has different approach fromthe proposed embedded RFID tag The system combinedsatellite communication and GPS tracking with 433MHzactive RFID into a single device controlled by a micro-processor without introducing WSN in their platform Thesatellite system hardware is expensive and each time the tagcommunicateswith the satellite system a fee is chargedwhichcontributes to the system drawback

Therefore a better solution is needed to improve theexistingRFID systemby the development of a 245GHz activeRFID tagwith automated switchingmechanismbetweenGPSand RFID tracking which utilized GSM and WSN platformwith the capabilities of M2M communication between RFIDreader and RFID tags The GSM communication is availableonly when the RFID tag is not joining the network thedesigned method reduces the cost of the embedded RFIDtag compared to previous work [6] which used satellitecommunication as a medium of transmission when the tag isnot in the reader read range Table 1 shows a brief comparisonbetween previous system and proposed system that is relatedto indoor and outdoor tracking

3 System Design

In this work the embedded RFID tag communicates withthe RFID reader via WSN platform The WSN consists ofrouters that are utilized to route the data from RFID readerto RFID tag or otherwise The major factors considered inthe design are energy efficiency robustness and reliabilityThe embedded RFID tags are battery powered and thereforeto increase network lifetime energy must be saved in everyhardware and software solution composing the networkarchitecture Data communication is responsible for thegreatest weight in the energy budget when compared withdata sensing and processing [14] In previous work doneby [14] they use short range communication using optimalspacing between nodes to reduce the energy consumptionby about 15 to 38 depending on the network densityTherefore it is desirable to use short range communicationinstead of long range communication between nodes inWSNplatformThus in this work the embedded RFID tag is presetwith minimum power transmission to reduce the energyconsumption as suggested by [14] Another way to minimizethe energy consumption is by optimizing the data volumeto be transmitted signal processing in the RFID tag andthe percentage of time the embedded RFID tag is on byusing sleepawake protocol algorithm as suggested by [7]In addition the energy harvesting technology also can beused to increase theWSN network lifetime [15ndash17]The RFIDreader and routers on the other hand do not necessarily havestrict restrictions on energy and processing power since theyare powered by fixed supply

The GPS receiver embedded with the RFID tag utilizedin this work used National Marine Electronic Association(NMEA 0183) protocol that is capable of estimating the loca-tion information with 33m accuracy without aid of satellitebased augmentation system (SBAS) The active antenna onboard helps the system integrators to do the system design

International Journal of Distributed Sensor Networks 3

Table 1 Comparison of existing and proposed system

Title Wireless technology GPS GSM M2M WSN Indoors andoutdoors Real time Long range

Combination RFID and GPSfunctionality on intelligent label(WO 2003050960 A2) [2]

Active RFID system and GPStechnology Yes No No No No Yes No

Portable RFID reader havinglocation determination(EP 1752908 A22006) [3]

Active RFID system utilizingwireless local area network

(WLAN)Yes No No Yes No Yes No

Real time location system (RLTS)with ZigBee Technology(EP 2196816 A12008) [4]

Did not involve RFID technology No No No Yes No Yes Yes

System and method for locatingtracking and monitoring the statusof personnel andor assets(US 20090043504 A1) [5]

Did not involve RFID technology No No No No No No No

Hybrid tag includes active RFIDGPS satellite and sensors [6]

433MHz active RFID systemGPS and satellitecommunication

Yes No No No Yes Yes No

Proposed RFID system245GHz active RFID WSN

based on IEEE 802154ZigBeeTechnology GPS technology

Yes Yes Yes Yes Yes Yes Yes

easily The standard update rate is 1 Hz and can be up to5Hz as well The current consumption during acquisition is63mA however it will become lower while tracking by about59mA for first 5 minutes 42mA after 5 minutes and 33mAafter 20 minutes The standard output sentences are GGAGLL GSA GSV RMC VTG and ZDA [18]

The embedded RFID tag is periodically sent out locationdata from GPS receiver to RFID reader via WSN platformat monitoring station to track and trace the movement andsequences location of the tagged person or item only whenhaving valid GPS data from satellites However if there areno valid data from satellites the localization will be doneby RFID tag using RSS values retrieved from the proposedembedded RFID tag The RSS value can be requested man-ually or automatically from the RFID reader which later isused to calculate the distance from the RFID routers whichhave fixed data location However this approach may delayaccessing the required information A compromise betweendelay and power efficiency can be achieved by optimallyselecting the interval for periodic transmission of requireddata The location information of the embedded RFID tagalong with the time stamp and identification number istransmitted to theRFID reader periodically to themonitoringstation Periodic transmission of location information facil-itates continuous monitoring but leads to increased powerconsumption and bandwidth occupancy

The subsections to follow elaborate the embedded RFIDtag RFID reader protocols and algorithms involved in theGPS RFIDWSN based tracking system

31 Embedded Active RFID Tag with GPS Functionalities Themotivation of this embedded RFID system is to provide userswith a technology that can track and trace their physical

belonging assets vehicles humans or animal remotely froma monitoring station using WSN platform by providingthe identification and location data as well as contributingto the M2M communication without human interventionThe system is also suitable to be used in supply chainmanagement medical emergencies and crowd control appli-cation and the most popular application demand recentlyis to support information and communication technologiesin collaboration during emergency response and disastermanagement With the aid of this embedded RFID systemall the information about the conditions can be obtained andproblems regarding to the application can be solved remotelyThus the embedded RFID system can reduce the manpoweruse in each management system as well as reducing the costof labour

The novelty of the proposed embedded RFID tag devel-oped in this work is an automated switching protocol ofindoor and outdoor location tracking andmonitoring systemwhich includes a microcontroller embedded with a GPSreceiver together with a wireless communicationmodule andGSM modem to become a single RFID tag which can beused to transmit and receive data fromto the RFID readerThe embedded RFID tag with the cooperation of the RFIDrouters will form a WSN network that can cover a large areaof data transmission and reception The embedded RFID tagis an active RFID that communicates with the RFID readersystem at a frequency of 245GHz ISM band that has anability to transfer the location data to the active RFID readerin a range of 60m indoors and 1500m outdoors and can beextended to more than 1500m via RFID router in multihopcommunication using WSN platform [1] The embedment ofGSMcommunication is to support the global communicationespecially when the embedded RFID tag is not in the wirelessnetwork coverage area However as long as the embedded

4 International Journal of Distributed Sensor Networks

Power management

Input 30V button Cell battery

Reset button

LED TxRx indicator

Wireless communication

module(WSN protocol)

Antenna

Figure 1 Block diagram of the existing RFID tag [19]

Power management

GPS module RTC

38 Kbytes RAM 64 Kbytes flash memory

Microcontroller

Reset button

LED TxRx indicator

Wireless communication

module(WSN protocol)

Input 84VDC output 33VDC and 25VDC

GSM modem

Antenna

Figure 2 Block diagram of the proposed embedded RFID tag

RFID tag is covered by thewireless network theGSMmodemwill not be activated The embedded RFID tag is batterypowered and therefore requires power efficient hardware andminimum amount of signal processing Due to these reasonsan algorithm based on AT mode and cyclic sleep mechanismis implemented to minimize the amount of signal processingand reduce the power consumption Figures 1 and 2 show theblock diagram of the existing RFID tag [19] and proposedembedded RFID tag that is implemented in this work

32 Active RFID Reader The application of RFID is a goodchoice in M2M communication as proposed in [19] and [20]and as such it is appropriate to be used in this proposed RFIDsystem Figure 3 shows the block diagram of the existingactive RFID reader [19] while Figure 4 shows the block dia-gram of the proposed active RFID reader that is implementedin this work The proposed active RFID reader consists ofa wireless communication module two control buttons andreal time clock connected to a microcontroller and RS232driver board for interfacing with a personal computer locatedat the monitoring station The location is periodically sentfrom the embedded RFID tag to RFID reader along with itsID longitude latitude and timestamp

Whenever there is a need to locate items or persons thatcannot be detected by the GPS receiver the RFID reader shall

broadcast a command Each embedded RFID tag received acommand and checks whether satellites detect their loca-tion or not If not the embedded RFID tag will send anacknowledgment and the ID along with RSS value to theRFID reader to indicate their location and distance and turnto sleep mode to reduce energy consumption If there is asatellites reception the embedded RFID tagwill remain silent(no acknowledgment) and continue with data reception fromGPS receiver and periodically send location information tothe RFID reader The RFID RS 232 driver board is added toRFID reader for serial data transmission and reception to orfrom the computer at the monitoring station and is poweredup by a fixed power supply of 9 VDC All communications inRFID reader system and RFID router are configured in APImode

33 Communication Protocols and Algorithms The mini-mization of energy consumptions highly relies on the pro-cessing and communication requirements of the protocolsand algorithms at various layers of the RFID system Thecommunication protocol defines how to exchange instruc-tion and data between reader and tag in both directionsThe ISO 18000-4 has suggested the usage of RTF protocolin 245GHz active RFID system [21] The concept of this

International Journal of Distributed Sensor Networks 5

Power management

Input 5-12VDCoutput 33VDC

Reset button

LED TxRx indicator

FDTI mm232r USB driver

Wireless communication

module(WSN protocol)

Antenna

Figure 3 Block diagram of the existing active RFID reader [19]

Power management

Indoor button RTC Reset tag ID button

38 Kbytes RAM 64 Kbytes flash

memory Microcontroller

MAX3232

RS232 serial communication

Reset button

LED TxRx indicator

Wireless communication

module(WSN protocol)

Input 84VDC output 33VDC and 25VDC

Antenna

Figure 4 Block diagram of the proposed active RFID reader

protocol is that any tag shall not perform data transmissionunless it has received and properly decoded the instructionsent by the reader RTF protocol is identified as one of thesolutions for anticollision in multitag communication [22]and is an effective way to reduce energy consumption of theRFID tag [23]

The proposed active RFID system employed multipro-tocol communication RTF and TTF The TTF protocol isimplemented when the tag is having valid location data fromGPS receiver The method of switching from one protocolto another can be many ways For example active RFID taglistens to RFID reader broadcast command before switchingfrom RTF to TTF mode [24] Meanwhile in [25] the RFIDtag is switchable between TTF and RTF mode within aswitching time frame based on RFID tagrsquos transceiver con-figuration In this work the automated switching algorithmis based on the data location provided by the GPS receiverIf the location is detected by the GPS receiver the embeddedRFID tag will activate the TTF protocol and send the locationdata to the RFID reader If the location is not detected bythe GPS receiver the embedded RFID tag will turn to sleepmode and wake up periodically within certain time frame tosense the incoming signal from RFID reader At this pointthe RFID tag will automatically switch from TTF to RTFmode and wait for interrogation signal from RFID reader

asking for the indoor location data Instead of indoor locationmonitoring the RFID reader also used RTFmode to resettingthe tag ID from the host computer to a specific embeddedRFID tag and thus this function contributes to the M2Mcommunication to change RFID tagrsquos information from theRFID reader All communications described in this work areassumed in WSN platform however if the embedded RFIDtag is not in the WSN coverage the GSM communicationwill be activated Figures 5(a) and 5(b) show the algorithmflow chart of the active RFID reader and tag with the abilityof switching mechanism for indoor and outdoor application

4 Experimental Study forPropagation Analysis

There are four types of experiments conducted in an indoorenvironment with line of sight (LOS) direction The firstexperiment is to measure the radiation pattern for bothexisting and proposed RFID system in order to evaluate theperformance of the developed system based on automatedswitching algorithm presented The second experiment isdesigned based on DOE method to identify the relationshipinteractions and effects of range polarization and angle ofpropagation based on RSS for both types of RFID systemThe third experiment is to find the most significant power

6 International Journal of Distributed Sensor Networks

Start

Initialization

Send collection command

Received ACK

Receive packet(1) Receive GPS data (2) Indoor location data (3) ACK packet

Packet received

Transmit ACK signal

Time out

Time out

Transmit mode

Transmitcommand to

RFID tag

Send to host computer via

RS-232

Receive mode

No

Yes

No

Yes

No

No

Received interrupt from

RS232

Receive interrupt from RTC alarm

Yes

Yes

Receive ACK

Disable interrupt

Yes

No

(a)

Collection command

Start

Initialization

Cyclic sleepmode

Receive GPSsignal

(UART)

Receive UARTinterrupt

GPS data valid

Transmit mode

Valid

Receive mode

Transmit ACK signal

Send ID data

Receive ACK

Time out

Time out

RetrieveRSSI

Retrievetag ID

Transmitmode

Send ID andlocation data

ReceiveACK

Receive RTCalarm interrupt

Networkcoverage

Disableinterruption

Send ID andlocation data

via GSM

Yes

No

No

Yes

Yes

No

No

Yes

No

Yes

NoNo

Yes

Yes

(b)

Figure 5 (a) The flow chart of algorithm implemented in the proposed RFID reader (b) The flow chart of algorithm implemented in theproposed RFID tag

Hostterminal

(PC) RS232 cable

RFIDreader

Antenna

RFID tag

RFID reader

Antenna

180∘

180∘

180∘

180∘

270∘

270∘

0∘360

0∘360

Distance d

H

E-field polarization

H-field polarization

Figure 6 Illustration of polarization measurements

level that provides the best variation of RSS for the proposedRFID system to operate in certain range of indoor locationThe fourth experiment is designed based on DOE methodto investigate the interactions and effects between powerlevel range and RSS for both existing and proposed RFIDsystem All experiments conducted in this work used the

same experimental setup as shown in Figure 6 There aretwo test beds involved for all experiments First test bed isfor nonembedded RFID system (stand-alone system withoutGPS) which consists of existing RFID reader and RFID tag[19] while the second test bed is for the embedded RFIDsystem which consists of a proposed active RFID reader and

International Journal of Distributed Sensor Networks 7

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(a)

NonembeddedEmbedded

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

(b)

Figure 7 H-plane pattern at (a) 2m and (b) 10m distance

RFID tag embedded with GPS The reader and tag are facingeach other at a specific distance which later communicates ina single hop ZigBee network

41 Radiation Pattern Measurement The radiation patternmeasurements are done for both existing RFID system [19]and proposed RFID system to evaluate and compare theperformance of the developed system based on automatedswitching algorithm presented at two different distanceswhich are 2m and 10m The far-field distance is obtained byusing (1) [26] based on characteristic of the antenna used inthis research work Consider

119877119891119891 gt21198712

120582 (1)

where 119877119891119891 is the distance between transmit and receiveantenna 119871 is an antenna aperture under test (825 cm) and120582 is the wavelength of transmission (1205824) indicating as120582 = 119862119891 where 119862 is the speed of light 3 times 108msminus2 andthe operating frequency 119891 = 245GHz Therefore 119877119891119891 isgreater than 042m when the antenna aperture value is notmore than 01m Hence 2m and 10mmeasurement distancesare in the far-field region The algorithm requires a readerto send information to tag and the tag will acknowledgethe packet received from reader and gives information thatis requested by the reader The measurements of RSS areperformed using whip antenna for both RFID reader and tagFor existing RFID system the RSS value is obtained from theX-CTU terminal by using ATDB command However for theproposed RFID system the RSS are retrieved from the algo-rithm implemented in the RFID system The power level of

the wireless communication module has been programmedto maximum value with the output power +10 dBm Theorientation of the antenna for H-field polarization is verticalwhile for E-field polarization it is horizontal as shown inFigure 6 The reader antenna is rotated in 119911-direction from0∘ to 360∘ degrees with 15∘ degree increment steps

Figures 7(a) and 7(b) show the comparison of radiationpatterns of existing RFID system and proposed RFID systemfor H-plane polarization at 2m and 10m distances respec-tively The result for H-field radiation pattern at 2m distanceis an omnidirectional It can be seen that the normalizedradiation pattern for proposed RFID system is slightly betterthan the existing RFID system in between 90∘ and 195∘degrees However the radiation pattern results for existingRFID system at 10m distance are better than proposedRFID system by about 09 dBm to 64 dBm differences Thedifferences in antenna pattern may be due to the immediatesurroundings since actual performance depends on manyfactors in the environment Obstructions in the propagationpath or other wireless networks or systems will also affectthe performance Figures 8(a) and 8(b) show the comparisonof radiation patterns for E-plane polarization at 2m and10mdistances respectivelyThenormalized radiation patternfor proposed RFID system is better than the existing RFIDsystem at 2m and 10m distances Based on these initialfindings it can be concluded that the radiation for shorterdistance is stronger than longer distance in both H-planeand E-plane and the embedded mechanism for RFID systemdoes affect the radiation pattern notmore than 15 differencefrom existing RFID system It shows that the proposed RFIDsystem contributes to stronger radiation compared to existingRFID system This may be due to the embedded power

8 International Journal of Distributed Sensor Networks

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(a)

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(b)

Figure 8 E-plane pattern at (a) 2m and (b) 10m distance

management that is implemented in the embedded activeRFID tag that provided power stability as mentioned by [27]thus increasing the performance of the system in the network

The second experiment is the continuity of the radiationpattern measurement which involved four factors with twolevels of treatment 24 The factors are referred to as type ofRFID (nonembedded embedded) power level (2 dBm10 dBm) range (5m 10m) polarization (E H) and angleof polarization (0∘ 90∘) The response of the experiment isin RSS and 2 replicates are used in this work to reduce themeasurement errors and noiseThe order in which the obser-vations are taken is selected randomly so that the design is acompletely randomized design To identify the interactionsand effects between type of RFID range polarization andangle of propagation with RSS a statistical method calledanalysis of variance (ANOVA) is used as presented in [28]and the effect estimate summaries for the proposed systemare shown in Table 2

In this work the main effect of A really dominates theprocess accounting for 63635 percent of the total variabilitywhereas themain effect of C accounts for about 7328 percentwhile the AB and BC interactions account for about 8808percent and 6616 percent respectively thus it is proved thatthe RSS value can be used as a prediction method to measuredistance and location of tags from reader in WSN platformsince the range contributes highly significantly compared toother factors The ANOVA in Table 3 is used to confirm themagnitude of these effects From Table 3 it can be seen thatthe main effects of A and C are highly significant since bothof the factors have very small 119875 values (lt0001) For 4-wayinteraction the ABCD interactions are also highly significant(low 119875 values) thus it can be concluded that there is a stronginteraction between four factors with RSS values

Figures 9 and 10 show the interactions and main effectsplot for radiation measurement that is used to interpret theresults from Tables 1 and 2

The graph shows that the RSS for existing RFID system isbetter than proposed RFID system at 10m distance howeverthe RSS for proposed RFID system is better than the existingRFID system at 2m distance In terms of angle the RSS for0∘ and 90∘ degree proposed RFID system are better thanexisting RFID system however the RSS value for 0∘ degreesis better than 90∘ degrees for both RFID system In terms ofpolarization the H-field polarization gives more significantvalues of RSS compared to E-field polarization for both RFIDsystems The findings show that the RFID system will havebetter RSS value if it is located nearer to the reader at 0∘degrees direction of propagation with H-field polarizationof antenna This statement is supported by Friis free spaceequation as shown in [26]

119875119903 (119889) =119875119905119866119905119866119903120582

2

(4120587)21198892119871 (2)

where119875119905 is the transmitted power119875119903(119889) is the received power119866119905 is the transmitter antenna gain 119866119903 is the receiver antennagain 119889 is the transmit and receive separation and 119871 is the sys-tem loss factor dependent upon line attenuation filter lossesand antenna losses and not related to propagation Accordingto Friis equation the electromagnetic (EM) signals whentraveling through wireless channels experience fading effectsdue to various effects but for ideal case the transmission thatis considered with a direct line of sight will give the bestvalue of received power The effects of impedance mismatchmisalignment of the antenna pointing and polarization and

International Journal of Distributed Sensor Networks 9

Table 2 Effect estimate summaries

Term constant Effect Sum of squares Percentage contributionRange (119860) minus1109 983460 63635Angle (119861) minus079 4960 0321Polarization (119862) 376 113250 7328Type of RFID (119863) 160 20480 1325Range lowast angle (119860119861) minus358 102250 6616Range lowast polarization (119860119862) minus073 4210 0272Range lowast type of RFID (119860119863) minus226 40950 2650Angle lowast polarization (119861119862) minus412 136120 8808Angle lowast type of RFID (119861119863) 034 0910 0058Polarization lowast type of RFID (119862119863) minus186 27750 1796Range lowast angle lowast polarization (119860119861119862) minus124 12250 0793Range lowast angle lowast type of RFID (119860119861119863) 140 15680 1015Range lowast polarization lowast type of RFID (119860119862119863) minus075 4500 0291Angle lowast polarization lowast type of RFID (119861119862119863) minus100 8000 0518Range lowast angle lowast polarization lowast type of RFID (119860119861119862119863) minus206 34030 2202

Table 3 ANOVA for the radiation pattern measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Range (119860) 1 983460 983461 429340 0000Angle (119861) 1 4960 4961 2170 0160Polarization (119862) 1 113250 113251 49440 0000Type of RFID (119863) 1 20480 20480 8940 0009Range lowast angle (119860119861) 1 102250 102245 44640 0000Range lowast polarization (119860119862) 1 4210 4205 1840 0194Range lowast type of RFID (119860119863) 1 40950 40951 17880 0001Angle lowast polarization (119861119862) 1 136120 136125 59430 0000Angle lowast type of RFID (119861119863) 1 0910 0911 0400 0537Polarization lowast type of RFID (119862119863) 1 27750 27751 12120 0003Range lowast angle lowast polarization (119860119861119862) 1 12250 12251 5350 0034Range lowast angle lowast type of RFID (119860119861119863) 1 15680 15680 6850 0019Range lowast polarization lowast type RFID (119860119862119863) 1 4500 4500 1960 0180Angle lowast polarization lowast type of RFID (119861119862119863) 1 8000 8000 3490 0080Range lowast angle lowast polarization lowast type of RFID (119860119861119862119863) 1 34030 34031 14860 0001Residual error 16 36650 2291Pure error 16 36650 2291Total 31 1545450

absorption can be included by adding additional factors asgiven in [26]

119875119903

119875119905

= 119866119905 (120579119905 120601119905) 119866119903 (120579119903 120601119903) (120582

4120587119877)

2

times (1 minus1003816100381610038161003816Γ1199051003816100381610038161003816

2) (1 minus1003816100381610038161003816Γ1199031003816100381610038161003816

2)1003816100381610038161003816119886119905 sdot 1198861199031003816100381610038161003816

2119890minus120572119877

(3)

where 119866119905(120579119905 120601119905) is the gain of the transmit antenna in thedirection 119866119903(120579119903 120601119903) is the gain of the receive antenna in thedirection (120579119903 120601119903) Γ119905 and Γ119903 are the reflection coefficients ofthe transmit and receive antennas respectively and 119886119905 and119886119903 are the polarization vectors of the transmit and receiveantennas respectively taken in the appropriate directions

120572 is the absorption coefficient of the intervening mediumHowever these additional factors will contribute to reductionof antenna received power Thus in the line of sight directionthe received power will become higher compared to non-lineof sight direction as experimented in this section

42 Power Level versus Range Measurement The power levelversus range measurement is done for distance between 5mand 100m indoorswith 5m incrementThedistance is limiteddue to the buildingrsquos built-up area which is limited to 100-meter lengthThe response is RSS values which are measuredfor existing and proposed RFID system at two levels of powerwhich are 10 dBm (max) and 2 dBm (min) At each point10 single values of receiving power are collected to calculate

10 International Journal of Distributed Sensor Networks

210

Range

090

Angle

Polarization

Type of RFID

Interactions plot

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

2 10

0 90

Range

Angle

Polarization

Type of RFID

E

E

H

H

N

P

N P

Figure 9 Interaction plot for radiation measurement

Range

Mea

nM

ean

Angle

Polarization Type of RFID

Main effects plot for RSSIData means

minus36

minus39

minus42

minus45

minus48

minus36

minus39

minus42

minus45

minus48

2 10 0 90

E H N P

Figure 10 Main effects plot for radiation measurement

the mean Figures 11 and 12 show the comparison betweenRSS values of existing and proposed RFID system based ondistance of 5 meters to 100 meters indoors at 10 dBm and2 dBm power levels The graph shows that the RSS valuesare decreased when the distance is increased at both powerlevels The existing RFID system gives better RSS valuesthan proposed RFID system at 10 dBm power level while the

proposed RFID system gives better RSS values than existingRFID system at 2 dBm power level To compare the perfor-mance of two types of RFID system based on power leveland range the third experiment is conducted based on DOEmethod with three factors having two levels of treatment 23The factors are referred to as type of RFID (nonembeddedembedded) power level (2 dBm 10 dBm) and range (5m

International Journal of Distributed Sensor Networks 11

00 20 40 60 80 100 120

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

RSSI

(dBm

)

Figure 11 Received signal strength values at 10 dBm in an indoorenvironment

00 20 40 60 80 100 120

RSSI

(dBm

)

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

minus80

Figure 12 Received signal strength values at 2 dBm in an indoorenvironment

10m) The response of the experiment is RSS and 2 replicatesare used in this work to reduce the measurement errorsand noise The order in which the observations are takenis selected randomly so that the design is a completelyrandomized design Later the statistical method ANOVA isperformed using Minitab 162 and the results are displayedin Tables 4 and 5 for effect estimate summary and ANOVAfor power level versus range measurement From the effectestimate summary presented in Table 4 the main effect of Creally dominates the process accounting for 91381 percentof the total variability whereas the main effect of A and theAC interaction account for about 6144 and 1727 percentrespectively thus the DOE for this experiment also provedthat the RSS value can be used as a prediction method tomeasure distance and location of tags from reader in wireless

sensor network platform since the range contributions arehighly significant compared to other factors The ANOVAin Table 4 is used to confirm the magnitude of these effectsFrom Table 5 the main effects of A B and C are highlysignificant since the factors have very small119875 values (lt0001)For 2-way interaction the AB and AC interaction are alsohighly significant (low 119875 values) thus it can be concludedthat there is a strong interaction between two factors withRSS values in this experimental study Figures 13 and 14 showthe interaction and effects plot for power level versus rangemeasurement that is used to interpret the results from Tables4 and 5 It can be seen that theRSS for embeddedRFID systemis nearly equal to nonembedded RFID system at 10 dBmpower level However at 2 dBm power level the RSS forembedded RFID system is better than nonembedded RFIDsystem In terms of range the RSS for embeddedRFID systemis better compared to nonembedded RFID system at both 5-meter and 100-meter range Based on the results obtainedit can be summarized that for indoor tracking purposesthe low power level gives better RSS values for embeddedRFID system compared to nonembedded RFID system Thisfinding can be supported by [29] which stated that the lowtransmit power is appropriate and sufficient for area withlimited range however if the high transmit power is use inthe area with limited range it will resulting in disturbancethat will cause degradation in radiation performanceThus todesign an efficient indoor localization system these findingsmust be taken into account as a guideline implementing thefull setup for real time location monitoring system in WSNplatform

5 Battery Life Testing

A single RFID tag requires a minimum of 27 V to 33 Vvoltage to operate as a transmitter and receiver in thenetwork Howver the embedded RFID tag proposed in thiswork requires minimum of 50V to power up the embeddeddevices Thus the embedded RFID tag can be powered upusing 9 VDC power supply if the tag is located at fixedlocation or Li-Po rechargeable batteries can be used if thetag is mobile The tag has been programmed to consumethe lowest amount of energy as much as possible sinceit introduced an automated switching mechanism betweenindoor and outdoor location tracking and operating in sleepmode at certain timeTheRFID tag only sends data if the validoutdoor location is detected by the GPS receiver however ifthe embedded RFID tag is located inside the building or theplace where the signal from satellite cannot be detected bythe GPS receiver it will turn to sleep mode and wait for theRFID reader interrogation every hourTheGSMmodemonlyworks when the embedded RFID tag is out of the wirelessnetwork coverage to consume the lowest amount of energy

The battery lifetime measurement is performed for threesystems which are stand-alone RFID tag (A) [20] embeddedRFID tag with switching mechanism (B) and embeddedRFID tag without switching mechanism (C) in an indoorenvironment (inside building) where the GPS signals are notavailable most of the time The results show that the battery

12 International Journal of Distributed Sensor Networks

Table 4 Effect estimate summary

Term constant Effect Sum of squares Percent contributionPower (119860) 724 209530 6144Type of RFID (119861) 179 12780 0375Range (119862) minus2791 3116430 91381Power lowast type of RFID (119860119861) minus141 7980 0234Power lowast range (119860119862) 384 58910 1727Type RFID lowast range (119861119862) 074 2180 0064Power lowast type of RFID lowast range (119860119861119862) minus056 1270 00372

Table 5 ANOVA for power level versus range measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Power (119860) 1 209530 209530 1294370 0000Type of RFID (119861) 1 12780 12780 78950 0000Range (119862) 1 3116430 3116430 19252080 0000Power lowast type of RFID (119860119861) 1 7980 7980 49300 0000Power lowast range (119860119862) 1 58910 58910 363900 0000Type RFID lowast range (119861119862) 1 2180 2180 13440 0006Power lowast type of RFID lowast range (119860119861119862) 1 1270 1270 7820 0023Residual error 8 1290 0160Pure error 8 1300 0160Total 15 3410360

for the tag drops from 824V to 711 V in the duration of 85hours for the proposed embedded RFID tag with switchingmechanism compared to other systems as shown in Figure 15

The results for system B are much better than systemC due to the switching introduced However the result forsystem A is better than system B due to the nonembeddedsystem According to these real implementation results thetag will be expected to stop operating approximately in24 hours when the battery reaches the minimum voltagerequired which is 50 V However this result does not includethe situation where the proposed embedded RFID tag islocated outside building which might increase the energyusage The estimation results also did not include otherexternal factors that might reduce the battery life Thus thetag is estimated to cover the operation for a day withouthaving to recharge the battery until the next day To reducethe energy consumption and maximize the time of usage anenergy harvesting technologywill be implemented later in theembedded RFID tag as our next target

6 Conclusion

This paper describes the implementation of an automatedswitching algorithm in the design of a 245GHz active inte-grated RFID system for indoor and outdoor location trackingutilizing WSN platform Four experimental measurementsbased on DOE method are performed in order to validatethe performance of the proposed RFID system as well as tostudy the relationship interaction and effects of each factor

to RSS values extracted from existing and proposed RFIDsystem The results obtained from first experiment show thatembedded RFID system does affect the radiation patternnot more than 15 differences from existing RFID system Itshows that the proposed RFID system contributes to strongerradiation compared to existing RFID system This may bedue to the embedded design and improvement of algorithmimplemented in the proposed RFID system which increasedthe power stability of the system which thus provided highersignal strength [28]

In the second experiment the results show that the pro-posed RFID tag has better RSS value when it is located nearerto the reader at 0∘ degrees direction of propagation with H-field polarization of antenna In the third experiment theexisting RFID system gives better RSS values than proposedRFID system at 10 dBm power level while the proposedRFID system gives better RSS values than existing RFIDsystem at 2 dBm power level In the fourth experiment it canbe summarized that for indoor tracking purposes the lowpower level gives better RSS values for the proposed RFIDtag compared to the existing RFID system The statisticalanalysis ANOVA proved that the RSS values are significantwith range thus it can be concluded that the RSS can beused as a prediction method to evaluate distance betweenRFID reader and RFID tag in WSN platform Based onthe experimental study and statistical analysis done for allsituations a conclusion has been made where the proposedRFID system with automated switching algorithm is betterthan the existing RFID system in terms of performance andvariability of the factors The battery life testing also gives

International Journal of Distributed Sensor Networks 13

Power

Range

Interactions plot

2 10 5 100

minus30

minus45

minus60

minus30

minus45

minus60

minus30

minus45

minus60

Range

Power

Type of RFID

210

5100

Type of RFID

E

N

EN

Figure 13 Interaction plot for power level versus range measurement

Power

Mea

nM

ean

Type RFID

Range

Main effects plot for RSSI

minus40

minus50

minus60

minus40

minus50

minus60

2 10

5 100

EN

Figure 14 Main effects plot for power level versus range measurement

significant findings where the proposed embedded RFIDtag with switching mechanism introduced is much betterthan embedded RFID tag without switching mechanismTo increase the performance of the embedded RFID tag interms of time of usage an energy harvesting technology willbe introduced later as a part of our next contribution Interms of hardware costing the proposed approach is moreacceptable compared to the system developed by [6] since thetechnology used in this work especially GSM communicationis much cheaper than satellite communication used in the

existing system [6] In addition this proposed system doesnot totally depend on GSM since the communication willbe done by RFID which is free as long as the embeddedRFID tag is working in the control area of WSN platformHowever the existing work [6] did not support long rangecommunication and thus when the tag is moving far fromthe reader the satellite communication needs to be used andthis will lead to high cost of communication As a conclusionthe proposed embedded RFID tag provided better signalpropagation compared to the existing RFID tag [19] and

14 International Journal of Distributed Sensor Networks

00 1 2 3 4 5 6 7 8 9

Batte

ry re

duct

ion

()

Time (hours)

Battery lifetime testing

Embedded RFID tag with switchingEmbedded RFID tag without switchingStandalone RFID tag

minus5

minus10

minus15

minus20

minus25

minus30

Figure 15 Battery life testing based on three types of RFIDdesigned

the battery lifetime for embedded RFID tag with switchingmechanism is better than that without switchingmechanism

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank Malaysia Ministry of HigherEducation (LRGS Fund) for sponsoring the research anddeveloping this project Also special appreciation is due toProfessor DrMohamadKamal Abdul Rahim fromUniversitiTeknologiMalaysia Head of the LRGS Project and AssociateProfessor Dr Alyani Ismail Universiti PutraMalaysia for thesupport

References

[1] DIGI XBeeXBee PRO ZB RF Module Datasheet 2013[2] J B Howard ldquoCombination RFID and GPS functionality on

intelligent labelrdquo WO 2003050960 A2 2003[3] S Chand V R Bapat K H Hall et al ldquoPortable RFID reader

having a location determination systemrdquo EP 1752908 A220062006

[4] G O Garsia S I M Nunez L A Saavedra and M D I TapialdquoReal time location system (RTLS) with ZigBee technologyrdquo EP2196816 A12008 2008

[5] A Bandyopadhyay D Hakim B E Funk E A Kohn C AToelis and G Blankenship ldquoSystem and method for locatingtracking andor monitoring the status of personnel andorassets both indoorsrdquo US 20090043504 A1 2009

[6] RFID Journal ldquoHybrid tags includes active RFID GPS satelliteand sensorrdquo RFID Journal 2009 httpwwwrfidjournalcomarticlesview4635

[7] M Mohandes M Haleem M Deriche and K BalakrishnanldquoWireless sensor networks for pilgrims trackingrdquo IEEE Embed-ded Systems Letters vol 4 no 4 pp 106ndash109 2012

[8] N D Simoes J L Goncalves M L Caeiro M J Boavida andF D Cardoso ldquoZigBeeGPS tracking system for rowing racesrdquoin Proceedings of the International Conference on Computer as aTool (EUROCON rsquo11) Lisbon Portugal April 2011

[9] Y Xu R Jiang S Yan and D Xiong ldquoThe research of safetymonitoring system applied in school bus based on the internetof thingsrdquo Procedia Engineering vol 15 pp 2464ndash2468 2011

[10] L Zheng M Li C Wu et al ldquoDevelopment of a smartmobile farming service systemrdquo Mathematical and ComputerModelling vol 54 no 3-4 pp 1194ndash1203 2011

[11] F A Poad and W Ismail ldquoAutomated switching mechanismfor indoor and outdoor propagation with embedded RFID andGPS in wireless sensor network platformrdquo in Proceedings ofthe International Conference on Wireless Network pp 710ndash714London UK 2014

[12] J OrsquoToole J Tuttle M Tuttle et al ldquoRadio frequency identifi-cation sensor tag apparatusesrdquo US 20080030306 A1 2008

[13] V M Gupta and S Annambhotla ldquoRFID tracker and locatorrdquoUS 20090085745 A1 2009

[14] F Shebli I Dayoub and J M Rouvaen ldquoMinimizing energyconsumptionwithinwireless sensor networkrdquoUbiquitous Com-puting and Communication Journal vol 87 2007

[15] R Ding J Hou and B Xing ldquoResearch of wireless sensornetwork nodes based on ambient energy harvestingrdquo in Pro-ceedings of the 6th IEEE International Conference on IntelligentNetworks and Intelligent Systems (ICINIS rsquo13) pp 286ndash288Shenyang China November 2013

[16] A Cammarano C Petrioli andD Spenza ldquoPro-Energy a novelenergy prediction model for solar and wind energy-harvestingwireless sensor networksrdquo in Proceedings of the 9th IEEEInternational Conference on Mobile Ad-Hoc and Sensor Systems(MASS rsquo12) pp 75ndash83 Las Vegas Nev USA October 2012

[17] J Song and Y K Tan ldquoEnergy consumption analysis of ZigBee-based energy harvesting wireless sensor networksrdquo in Proceed-ings of the IEEE International Conference on CommunicationSystems (ICCS rsquo12) pp 468ndash472 November 2012

[18] San Jose Technology Inc Datasheet GPS Engine Board ModelFV M8 2014 httpswwwsparkfuncomdatasheetsGPSFV-M8 Specpdf

[19] W Ismail and R Abdulla ldquoPortable RFID Reader for RTLS(Real Time Location Systems)rdquo PCT International Filing NoPCTMY2011000167 Country Australia Pub No WO2012102600 2012

[20] W Ismail C Z Zulkifli and M G Rahman ldquoWireless pro-duction monitoring systemrdquo PCT International Filing PCTMY2013000037 2013

[21] ISOIEC 18000-4 Information Technology-Radio FrequencyIdentification for ItemManagement-Part7 Parameters for ActiveAir Interface Communications at 245 GHz ISO 18000-42008International Organization for Standardization GenevaSwitzerland 2008

[22] D-H Shih P-L Sun D C Yen and S-M Huang ldquoTaxonomyand survey of RFID anti-collision protocolsrdquo Computer Com-munications vol 29 no 11 pp 2150ndash2166 2006

[23] B Nilsson L Bengtsson P-A Wiberg and B SvenssonldquoProtocols for active RFIDmdashthe energy consumption aspectrdquo inProceedings of the IEEE 2nd International Symposium on Indus-trial Embedded Systems (SIES rsquo07) pp 41ndash48 July 2007

[24] H L van Eeden ldquoWireless Communication Systemrdquo US Patent0045923 2009

International Journal of Distributed Sensor Networks 15

[25] H Haar K Bischof and H Scharke ldquoRFID transponderrdquo WOPatent 109212 2006

[26] C A Balanis AntennaTheory Analysis and Design JohnWileyamp Sons New York NY USA 3rd edition 2005

[27] P Sorrells Optimizing Read Range in RFID System MicrochipTechnology Inc 2000 httpedncomdesignwireless-network-ing4341980Optimizing-read-range-in-RFID-systems

[28] A Ammu L Mapa and A H Jayatissa ldquoEffect of factors onRFID tag readability-statistical analysisrdquo in Proceedings of theIEEE International Conference on ElectroInformation Technol-ogy (EIT rsquo09) pp 355ndash358 June 2009

[29] R Zurawski Embedded System Handbook Second EditionNetwork Embedded Systems CRC Press 2009

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

Page 2: Research Article Propagation Analysis for Automated ...downloads.hindawi.com/journals/ijdsn/2015/392385.pdf · Hybrid tag includes active RFID, GPS, satellite, and sensors [ ] MHz

2 International Journal of Distributed Sensor Networks

sleep and continuous mode multiprotocol reader talk first(RTF) and tag talk first (TTF) and multimode communica-tion and application programming interface (API)mode andapplication transparent (AT) mode

2 Literature Review

US Patent 20080030306 A1 [12] describes an RFID devicethat is an active transponder which includes a transmitter areceiver and a microprocessor The invention relates RFIDdevices to many applications for example for determiningthe location of objects or person that is tagged with RFIDtag However the document does not describe any additionalsolutions on how to monitor or locate the objects or personsespecially for an indoor and outdoor location environmentrespectively US Patent 20090085745 A1 [13] describes anRFID tracking device that includes aGPS capable of detectingthe unique identification (ID) from the RFID tag of theobjects and obtaining the location coordinate provided byGPS However GPS is unable to work completely indoorsor in an environment with obstacles which thus contributesto the weaknesses of the design The WO 2003050960 A2[2] is another patented system that combines the RFIDand GPS functionality on intelligent label The system usedactive RFID reader and GPS to track the location and sendthe location data information to the RFID reader in shortdistances only without involving WSN platformThis systemis however unable to work completely indoors or in anenvironment with obstacles The EP 1752908 A22006 [3]is another patented system that integrates the GPS func-tionalities in a portable RFID system to track the locationof an object or pallet inside or outside the structures suchas warehouses and the factory automation environmentHowever this system integrates the GPS receiver on theRFID reader portion to track the location and send theinformation to themonitoring station via wireless transceiverin short distances without introducingWSN platformThis isin contrast with the proposed embedded RFID system thatutilized the GPS function on the RFID tag portion to trackthe location

Real time location system (RLTS) with ZigBee Technol-ogy (EP 2196816 A12008) [4] is another work using a similarZigBee Technology IEEE 802154ZigBee standard as theproposed embedded RFID tag that operates in unlicensedfrequency bands (868MHz in Europe 915MHz in USAand 24GHz almost worldwide) The system involved WSNin the implementation however the system only provideslocation tracking for indoor environmentwithout integrationwith GPS technology US 20090043504 A1 [5] develops asystem and method for locating tracking and monitoringthe status of personnel andor asserts However the systemdid not involve RFID or GPS technology and only focuses ona computer-implemented method for generating a positionestimation of a tracker in an indoor location A most similarsystem developed by Numerex and Savi Technology [6] ahybrid tag that includes active RFID GPS satellite commu-nication and sensors introduces an automated switching

mechanism however the system has different approach fromthe proposed embedded RFID tag The system combinedsatellite communication and GPS tracking with 433MHzactive RFID into a single device controlled by a micro-processor without introducing WSN in their platform Thesatellite system hardware is expensive and each time the tagcommunicateswith the satellite system a fee is chargedwhichcontributes to the system drawback

Therefore a better solution is needed to improve theexistingRFID systemby the development of a 245GHz activeRFID tagwith automated switchingmechanismbetweenGPSand RFID tracking which utilized GSM and WSN platformwith the capabilities of M2M communication between RFIDreader and RFID tags The GSM communication is availableonly when the RFID tag is not joining the network thedesigned method reduces the cost of the embedded RFIDtag compared to previous work [6] which used satellitecommunication as a medium of transmission when the tag isnot in the reader read range Table 1 shows a brief comparisonbetween previous system and proposed system that is relatedto indoor and outdoor tracking

3 System Design

In this work the embedded RFID tag communicates withthe RFID reader via WSN platform The WSN consists ofrouters that are utilized to route the data from RFID readerto RFID tag or otherwise The major factors considered inthe design are energy efficiency robustness and reliabilityThe embedded RFID tags are battery powered and thereforeto increase network lifetime energy must be saved in everyhardware and software solution composing the networkarchitecture Data communication is responsible for thegreatest weight in the energy budget when compared withdata sensing and processing [14] In previous work doneby [14] they use short range communication using optimalspacing between nodes to reduce the energy consumptionby about 15 to 38 depending on the network densityTherefore it is desirable to use short range communicationinstead of long range communication between nodes inWSNplatformThus in this work the embedded RFID tag is presetwith minimum power transmission to reduce the energyconsumption as suggested by [14] Another way to minimizethe energy consumption is by optimizing the data volumeto be transmitted signal processing in the RFID tag andthe percentage of time the embedded RFID tag is on byusing sleepawake protocol algorithm as suggested by [7]In addition the energy harvesting technology also can beused to increase theWSN network lifetime [15ndash17]The RFIDreader and routers on the other hand do not necessarily havestrict restrictions on energy and processing power since theyare powered by fixed supply

The GPS receiver embedded with the RFID tag utilizedin this work used National Marine Electronic Association(NMEA 0183) protocol that is capable of estimating the loca-tion information with 33m accuracy without aid of satellitebased augmentation system (SBAS) The active antenna onboard helps the system integrators to do the system design

International Journal of Distributed Sensor Networks 3

Table 1 Comparison of existing and proposed system

Title Wireless technology GPS GSM M2M WSN Indoors andoutdoors Real time Long range

Combination RFID and GPSfunctionality on intelligent label(WO 2003050960 A2) [2]

Active RFID system and GPStechnology Yes No No No No Yes No

Portable RFID reader havinglocation determination(EP 1752908 A22006) [3]

Active RFID system utilizingwireless local area network

(WLAN)Yes No No Yes No Yes No

Real time location system (RLTS)with ZigBee Technology(EP 2196816 A12008) [4]

Did not involve RFID technology No No No Yes No Yes Yes

System and method for locatingtracking and monitoring the statusof personnel andor assets(US 20090043504 A1) [5]

Did not involve RFID technology No No No No No No No

Hybrid tag includes active RFIDGPS satellite and sensors [6]

433MHz active RFID systemGPS and satellitecommunication

Yes No No No Yes Yes No

Proposed RFID system245GHz active RFID WSN

based on IEEE 802154ZigBeeTechnology GPS technology

Yes Yes Yes Yes Yes Yes Yes

easily The standard update rate is 1 Hz and can be up to5Hz as well The current consumption during acquisition is63mA however it will become lower while tracking by about59mA for first 5 minutes 42mA after 5 minutes and 33mAafter 20 minutes The standard output sentences are GGAGLL GSA GSV RMC VTG and ZDA [18]

The embedded RFID tag is periodically sent out locationdata from GPS receiver to RFID reader via WSN platformat monitoring station to track and trace the movement andsequences location of the tagged person or item only whenhaving valid GPS data from satellites However if there areno valid data from satellites the localization will be doneby RFID tag using RSS values retrieved from the proposedembedded RFID tag The RSS value can be requested man-ually or automatically from the RFID reader which later isused to calculate the distance from the RFID routers whichhave fixed data location However this approach may delayaccessing the required information A compromise betweendelay and power efficiency can be achieved by optimallyselecting the interval for periodic transmission of requireddata The location information of the embedded RFID tagalong with the time stamp and identification number istransmitted to theRFID reader periodically to themonitoringstation Periodic transmission of location information facil-itates continuous monitoring but leads to increased powerconsumption and bandwidth occupancy

The subsections to follow elaborate the embedded RFIDtag RFID reader protocols and algorithms involved in theGPS RFIDWSN based tracking system

31 Embedded Active RFID Tag with GPS Functionalities Themotivation of this embedded RFID system is to provide userswith a technology that can track and trace their physical

belonging assets vehicles humans or animal remotely froma monitoring station using WSN platform by providingthe identification and location data as well as contributingto the M2M communication without human interventionThe system is also suitable to be used in supply chainmanagement medical emergencies and crowd control appli-cation and the most popular application demand recentlyis to support information and communication technologiesin collaboration during emergency response and disastermanagement With the aid of this embedded RFID systemall the information about the conditions can be obtained andproblems regarding to the application can be solved remotelyThus the embedded RFID system can reduce the manpoweruse in each management system as well as reducing the costof labour

The novelty of the proposed embedded RFID tag devel-oped in this work is an automated switching protocol ofindoor and outdoor location tracking andmonitoring systemwhich includes a microcontroller embedded with a GPSreceiver together with a wireless communicationmodule andGSM modem to become a single RFID tag which can beused to transmit and receive data fromto the RFID readerThe embedded RFID tag with the cooperation of the RFIDrouters will form a WSN network that can cover a large areaof data transmission and reception The embedded RFID tagis an active RFID that communicates with the RFID readersystem at a frequency of 245GHz ISM band that has anability to transfer the location data to the active RFID readerin a range of 60m indoors and 1500m outdoors and can beextended to more than 1500m via RFID router in multihopcommunication using WSN platform [1] The embedment ofGSMcommunication is to support the global communicationespecially when the embedded RFID tag is not in the wirelessnetwork coverage area However as long as the embedded

4 International Journal of Distributed Sensor Networks

Power management

Input 30V button Cell battery

Reset button

LED TxRx indicator

Wireless communication

module(WSN protocol)

Antenna

Figure 1 Block diagram of the existing RFID tag [19]

Power management

GPS module RTC

38 Kbytes RAM 64 Kbytes flash memory

Microcontroller

Reset button

LED TxRx indicator

Wireless communication

module(WSN protocol)

Input 84VDC output 33VDC and 25VDC

GSM modem

Antenna

Figure 2 Block diagram of the proposed embedded RFID tag

RFID tag is covered by thewireless network theGSMmodemwill not be activated The embedded RFID tag is batterypowered and therefore requires power efficient hardware andminimum amount of signal processing Due to these reasonsan algorithm based on AT mode and cyclic sleep mechanismis implemented to minimize the amount of signal processingand reduce the power consumption Figures 1 and 2 show theblock diagram of the existing RFID tag [19] and proposedembedded RFID tag that is implemented in this work

32 Active RFID Reader The application of RFID is a goodchoice in M2M communication as proposed in [19] and [20]and as such it is appropriate to be used in this proposed RFIDsystem Figure 3 shows the block diagram of the existingactive RFID reader [19] while Figure 4 shows the block dia-gram of the proposed active RFID reader that is implementedin this work The proposed active RFID reader consists ofa wireless communication module two control buttons andreal time clock connected to a microcontroller and RS232driver board for interfacing with a personal computer locatedat the monitoring station The location is periodically sentfrom the embedded RFID tag to RFID reader along with itsID longitude latitude and timestamp

Whenever there is a need to locate items or persons thatcannot be detected by the GPS receiver the RFID reader shall

broadcast a command Each embedded RFID tag received acommand and checks whether satellites detect their loca-tion or not If not the embedded RFID tag will send anacknowledgment and the ID along with RSS value to theRFID reader to indicate their location and distance and turnto sleep mode to reduce energy consumption If there is asatellites reception the embedded RFID tagwill remain silent(no acknowledgment) and continue with data reception fromGPS receiver and periodically send location information tothe RFID reader The RFID RS 232 driver board is added toRFID reader for serial data transmission and reception to orfrom the computer at the monitoring station and is poweredup by a fixed power supply of 9 VDC All communications inRFID reader system and RFID router are configured in APImode

33 Communication Protocols and Algorithms The mini-mization of energy consumptions highly relies on the pro-cessing and communication requirements of the protocolsand algorithms at various layers of the RFID system Thecommunication protocol defines how to exchange instruc-tion and data between reader and tag in both directionsThe ISO 18000-4 has suggested the usage of RTF protocolin 245GHz active RFID system [21] The concept of this

International Journal of Distributed Sensor Networks 5

Power management

Input 5-12VDCoutput 33VDC

Reset button

LED TxRx indicator

FDTI mm232r USB driver

Wireless communication

module(WSN protocol)

Antenna

Figure 3 Block diagram of the existing active RFID reader [19]

Power management

Indoor button RTC Reset tag ID button

38 Kbytes RAM 64 Kbytes flash

memory Microcontroller

MAX3232

RS232 serial communication

Reset button

LED TxRx indicator

Wireless communication

module(WSN protocol)

Input 84VDC output 33VDC and 25VDC

Antenna

Figure 4 Block diagram of the proposed active RFID reader

protocol is that any tag shall not perform data transmissionunless it has received and properly decoded the instructionsent by the reader RTF protocol is identified as one of thesolutions for anticollision in multitag communication [22]and is an effective way to reduce energy consumption of theRFID tag [23]

The proposed active RFID system employed multipro-tocol communication RTF and TTF The TTF protocol isimplemented when the tag is having valid location data fromGPS receiver The method of switching from one protocolto another can be many ways For example active RFID taglistens to RFID reader broadcast command before switchingfrom RTF to TTF mode [24] Meanwhile in [25] the RFIDtag is switchable between TTF and RTF mode within aswitching time frame based on RFID tagrsquos transceiver con-figuration In this work the automated switching algorithmis based on the data location provided by the GPS receiverIf the location is detected by the GPS receiver the embeddedRFID tag will activate the TTF protocol and send the locationdata to the RFID reader If the location is not detected bythe GPS receiver the embedded RFID tag will turn to sleepmode and wake up periodically within certain time frame tosense the incoming signal from RFID reader At this pointthe RFID tag will automatically switch from TTF to RTFmode and wait for interrogation signal from RFID reader

asking for the indoor location data Instead of indoor locationmonitoring the RFID reader also used RTFmode to resettingthe tag ID from the host computer to a specific embeddedRFID tag and thus this function contributes to the M2Mcommunication to change RFID tagrsquos information from theRFID reader All communications described in this work areassumed in WSN platform however if the embedded RFIDtag is not in the WSN coverage the GSM communicationwill be activated Figures 5(a) and 5(b) show the algorithmflow chart of the active RFID reader and tag with the abilityof switching mechanism for indoor and outdoor application

4 Experimental Study forPropagation Analysis

There are four types of experiments conducted in an indoorenvironment with line of sight (LOS) direction The firstexperiment is to measure the radiation pattern for bothexisting and proposed RFID system in order to evaluate theperformance of the developed system based on automatedswitching algorithm presented The second experiment isdesigned based on DOE method to identify the relationshipinteractions and effects of range polarization and angle ofpropagation based on RSS for both types of RFID systemThe third experiment is to find the most significant power

6 International Journal of Distributed Sensor Networks

Start

Initialization

Send collection command

Received ACK

Receive packet(1) Receive GPS data (2) Indoor location data (3) ACK packet

Packet received

Transmit ACK signal

Time out

Time out

Transmit mode

Transmitcommand to

RFID tag

Send to host computer via

RS-232

Receive mode

No

Yes

No

Yes

No

No

Received interrupt from

RS232

Receive interrupt from RTC alarm

Yes

Yes

Receive ACK

Disable interrupt

Yes

No

(a)

Collection command

Start

Initialization

Cyclic sleepmode

Receive GPSsignal

(UART)

Receive UARTinterrupt

GPS data valid

Transmit mode

Valid

Receive mode

Transmit ACK signal

Send ID data

Receive ACK

Time out

Time out

RetrieveRSSI

Retrievetag ID

Transmitmode

Send ID andlocation data

ReceiveACK

Receive RTCalarm interrupt

Networkcoverage

Disableinterruption

Send ID andlocation data

via GSM

Yes

No

No

Yes

Yes

No

No

Yes

No

Yes

NoNo

Yes

Yes

(b)

Figure 5 (a) The flow chart of algorithm implemented in the proposed RFID reader (b) The flow chart of algorithm implemented in theproposed RFID tag

Hostterminal

(PC) RS232 cable

RFIDreader

Antenna

RFID tag

RFID reader

Antenna

180∘

180∘

180∘

180∘

270∘

270∘

0∘360

0∘360

Distance d

H

E-field polarization

H-field polarization

Figure 6 Illustration of polarization measurements

level that provides the best variation of RSS for the proposedRFID system to operate in certain range of indoor locationThe fourth experiment is designed based on DOE methodto investigate the interactions and effects between powerlevel range and RSS for both existing and proposed RFIDsystem All experiments conducted in this work used the

same experimental setup as shown in Figure 6 There aretwo test beds involved for all experiments First test bed isfor nonembedded RFID system (stand-alone system withoutGPS) which consists of existing RFID reader and RFID tag[19] while the second test bed is for the embedded RFIDsystem which consists of a proposed active RFID reader and

International Journal of Distributed Sensor Networks 7

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(a)

NonembeddedEmbedded

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

(b)

Figure 7 H-plane pattern at (a) 2m and (b) 10m distance

RFID tag embedded with GPS The reader and tag are facingeach other at a specific distance which later communicates ina single hop ZigBee network

41 Radiation Pattern Measurement The radiation patternmeasurements are done for both existing RFID system [19]and proposed RFID system to evaluate and compare theperformance of the developed system based on automatedswitching algorithm presented at two different distanceswhich are 2m and 10m The far-field distance is obtained byusing (1) [26] based on characteristic of the antenna used inthis research work Consider

119877119891119891 gt21198712

120582 (1)

where 119877119891119891 is the distance between transmit and receiveantenna 119871 is an antenna aperture under test (825 cm) and120582 is the wavelength of transmission (1205824) indicating as120582 = 119862119891 where 119862 is the speed of light 3 times 108msminus2 andthe operating frequency 119891 = 245GHz Therefore 119877119891119891 isgreater than 042m when the antenna aperture value is notmore than 01m Hence 2m and 10mmeasurement distancesare in the far-field region The algorithm requires a readerto send information to tag and the tag will acknowledgethe packet received from reader and gives information thatis requested by the reader The measurements of RSS areperformed using whip antenna for both RFID reader and tagFor existing RFID system the RSS value is obtained from theX-CTU terminal by using ATDB command However for theproposed RFID system the RSS are retrieved from the algo-rithm implemented in the RFID system The power level of

the wireless communication module has been programmedto maximum value with the output power +10 dBm Theorientation of the antenna for H-field polarization is verticalwhile for E-field polarization it is horizontal as shown inFigure 6 The reader antenna is rotated in 119911-direction from0∘ to 360∘ degrees with 15∘ degree increment steps

Figures 7(a) and 7(b) show the comparison of radiationpatterns of existing RFID system and proposed RFID systemfor H-plane polarization at 2m and 10m distances respec-tively The result for H-field radiation pattern at 2m distanceis an omnidirectional It can be seen that the normalizedradiation pattern for proposed RFID system is slightly betterthan the existing RFID system in between 90∘ and 195∘degrees However the radiation pattern results for existingRFID system at 10m distance are better than proposedRFID system by about 09 dBm to 64 dBm differences Thedifferences in antenna pattern may be due to the immediatesurroundings since actual performance depends on manyfactors in the environment Obstructions in the propagationpath or other wireless networks or systems will also affectthe performance Figures 8(a) and 8(b) show the comparisonof radiation patterns for E-plane polarization at 2m and10mdistances respectivelyThenormalized radiation patternfor proposed RFID system is better than the existing RFIDsystem at 2m and 10m distances Based on these initialfindings it can be concluded that the radiation for shorterdistance is stronger than longer distance in both H-planeand E-plane and the embedded mechanism for RFID systemdoes affect the radiation pattern notmore than 15 differencefrom existing RFID system It shows that the proposed RFIDsystem contributes to stronger radiation compared to existingRFID system This may be due to the embedded power

8 International Journal of Distributed Sensor Networks

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(a)

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(b)

Figure 8 E-plane pattern at (a) 2m and (b) 10m distance

management that is implemented in the embedded activeRFID tag that provided power stability as mentioned by [27]thus increasing the performance of the system in the network

The second experiment is the continuity of the radiationpattern measurement which involved four factors with twolevels of treatment 24 The factors are referred to as type ofRFID (nonembedded embedded) power level (2 dBm10 dBm) range (5m 10m) polarization (E H) and angleof polarization (0∘ 90∘) The response of the experiment isin RSS and 2 replicates are used in this work to reduce themeasurement errors and noiseThe order in which the obser-vations are taken is selected randomly so that the design is acompletely randomized design To identify the interactionsand effects between type of RFID range polarization andangle of propagation with RSS a statistical method calledanalysis of variance (ANOVA) is used as presented in [28]and the effect estimate summaries for the proposed systemare shown in Table 2

In this work the main effect of A really dominates theprocess accounting for 63635 percent of the total variabilitywhereas themain effect of C accounts for about 7328 percentwhile the AB and BC interactions account for about 8808percent and 6616 percent respectively thus it is proved thatthe RSS value can be used as a prediction method to measuredistance and location of tags from reader in WSN platformsince the range contributes highly significantly compared toother factors The ANOVA in Table 3 is used to confirm themagnitude of these effects From Table 3 it can be seen thatthe main effects of A and C are highly significant since bothof the factors have very small 119875 values (lt0001) For 4-wayinteraction the ABCD interactions are also highly significant(low 119875 values) thus it can be concluded that there is a stronginteraction between four factors with RSS values

Figures 9 and 10 show the interactions and main effectsplot for radiation measurement that is used to interpret theresults from Tables 1 and 2

The graph shows that the RSS for existing RFID system isbetter than proposed RFID system at 10m distance howeverthe RSS for proposed RFID system is better than the existingRFID system at 2m distance In terms of angle the RSS for0∘ and 90∘ degree proposed RFID system are better thanexisting RFID system however the RSS value for 0∘ degreesis better than 90∘ degrees for both RFID system In terms ofpolarization the H-field polarization gives more significantvalues of RSS compared to E-field polarization for both RFIDsystems The findings show that the RFID system will havebetter RSS value if it is located nearer to the reader at 0∘degrees direction of propagation with H-field polarizationof antenna This statement is supported by Friis free spaceequation as shown in [26]

119875119903 (119889) =119875119905119866119905119866119903120582

2

(4120587)21198892119871 (2)

where119875119905 is the transmitted power119875119903(119889) is the received power119866119905 is the transmitter antenna gain 119866119903 is the receiver antennagain 119889 is the transmit and receive separation and 119871 is the sys-tem loss factor dependent upon line attenuation filter lossesand antenna losses and not related to propagation Accordingto Friis equation the electromagnetic (EM) signals whentraveling through wireless channels experience fading effectsdue to various effects but for ideal case the transmission thatis considered with a direct line of sight will give the bestvalue of received power The effects of impedance mismatchmisalignment of the antenna pointing and polarization and

International Journal of Distributed Sensor Networks 9

Table 2 Effect estimate summaries

Term constant Effect Sum of squares Percentage contributionRange (119860) minus1109 983460 63635Angle (119861) minus079 4960 0321Polarization (119862) 376 113250 7328Type of RFID (119863) 160 20480 1325Range lowast angle (119860119861) minus358 102250 6616Range lowast polarization (119860119862) minus073 4210 0272Range lowast type of RFID (119860119863) minus226 40950 2650Angle lowast polarization (119861119862) minus412 136120 8808Angle lowast type of RFID (119861119863) 034 0910 0058Polarization lowast type of RFID (119862119863) minus186 27750 1796Range lowast angle lowast polarization (119860119861119862) minus124 12250 0793Range lowast angle lowast type of RFID (119860119861119863) 140 15680 1015Range lowast polarization lowast type of RFID (119860119862119863) minus075 4500 0291Angle lowast polarization lowast type of RFID (119861119862119863) minus100 8000 0518Range lowast angle lowast polarization lowast type of RFID (119860119861119862119863) minus206 34030 2202

Table 3 ANOVA for the radiation pattern measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Range (119860) 1 983460 983461 429340 0000Angle (119861) 1 4960 4961 2170 0160Polarization (119862) 1 113250 113251 49440 0000Type of RFID (119863) 1 20480 20480 8940 0009Range lowast angle (119860119861) 1 102250 102245 44640 0000Range lowast polarization (119860119862) 1 4210 4205 1840 0194Range lowast type of RFID (119860119863) 1 40950 40951 17880 0001Angle lowast polarization (119861119862) 1 136120 136125 59430 0000Angle lowast type of RFID (119861119863) 1 0910 0911 0400 0537Polarization lowast type of RFID (119862119863) 1 27750 27751 12120 0003Range lowast angle lowast polarization (119860119861119862) 1 12250 12251 5350 0034Range lowast angle lowast type of RFID (119860119861119863) 1 15680 15680 6850 0019Range lowast polarization lowast type RFID (119860119862119863) 1 4500 4500 1960 0180Angle lowast polarization lowast type of RFID (119861119862119863) 1 8000 8000 3490 0080Range lowast angle lowast polarization lowast type of RFID (119860119861119862119863) 1 34030 34031 14860 0001Residual error 16 36650 2291Pure error 16 36650 2291Total 31 1545450

absorption can be included by adding additional factors asgiven in [26]

119875119903

119875119905

= 119866119905 (120579119905 120601119905) 119866119903 (120579119903 120601119903) (120582

4120587119877)

2

times (1 minus1003816100381610038161003816Γ1199051003816100381610038161003816

2) (1 minus1003816100381610038161003816Γ1199031003816100381610038161003816

2)1003816100381610038161003816119886119905 sdot 1198861199031003816100381610038161003816

2119890minus120572119877

(3)

where 119866119905(120579119905 120601119905) is the gain of the transmit antenna in thedirection 119866119903(120579119903 120601119903) is the gain of the receive antenna in thedirection (120579119903 120601119903) Γ119905 and Γ119903 are the reflection coefficients ofthe transmit and receive antennas respectively and 119886119905 and119886119903 are the polarization vectors of the transmit and receiveantennas respectively taken in the appropriate directions

120572 is the absorption coefficient of the intervening mediumHowever these additional factors will contribute to reductionof antenna received power Thus in the line of sight directionthe received power will become higher compared to non-lineof sight direction as experimented in this section

42 Power Level versus Range Measurement The power levelversus range measurement is done for distance between 5mand 100m indoorswith 5m incrementThedistance is limiteddue to the buildingrsquos built-up area which is limited to 100-meter lengthThe response is RSS values which are measuredfor existing and proposed RFID system at two levels of powerwhich are 10 dBm (max) and 2 dBm (min) At each point10 single values of receiving power are collected to calculate

10 International Journal of Distributed Sensor Networks

210

Range

090

Angle

Polarization

Type of RFID

Interactions plot

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

2 10

0 90

Range

Angle

Polarization

Type of RFID

E

E

H

H

N

P

N P

Figure 9 Interaction plot for radiation measurement

Range

Mea

nM

ean

Angle

Polarization Type of RFID

Main effects plot for RSSIData means

minus36

minus39

minus42

minus45

minus48

minus36

minus39

minus42

minus45

minus48

2 10 0 90

E H N P

Figure 10 Main effects plot for radiation measurement

the mean Figures 11 and 12 show the comparison betweenRSS values of existing and proposed RFID system based ondistance of 5 meters to 100 meters indoors at 10 dBm and2 dBm power levels The graph shows that the RSS valuesare decreased when the distance is increased at both powerlevels The existing RFID system gives better RSS valuesthan proposed RFID system at 10 dBm power level while the

proposed RFID system gives better RSS values than existingRFID system at 2 dBm power level To compare the perfor-mance of two types of RFID system based on power leveland range the third experiment is conducted based on DOEmethod with three factors having two levels of treatment 23The factors are referred to as type of RFID (nonembeddedembedded) power level (2 dBm 10 dBm) and range (5m

International Journal of Distributed Sensor Networks 11

00 20 40 60 80 100 120

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

RSSI

(dBm

)

Figure 11 Received signal strength values at 10 dBm in an indoorenvironment

00 20 40 60 80 100 120

RSSI

(dBm

)

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

minus80

Figure 12 Received signal strength values at 2 dBm in an indoorenvironment

10m) The response of the experiment is RSS and 2 replicatesare used in this work to reduce the measurement errorsand noise The order in which the observations are takenis selected randomly so that the design is a completelyrandomized design Later the statistical method ANOVA isperformed using Minitab 162 and the results are displayedin Tables 4 and 5 for effect estimate summary and ANOVAfor power level versus range measurement From the effectestimate summary presented in Table 4 the main effect of Creally dominates the process accounting for 91381 percentof the total variability whereas the main effect of A and theAC interaction account for about 6144 and 1727 percentrespectively thus the DOE for this experiment also provedthat the RSS value can be used as a prediction method tomeasure distance and location of tags from reader in wireless

sensor network platform since the range contributions arehighly significant compared to other factors The ANOVAin Table 4 is used to confirm the magnitude of these effectsFrom Table 5 the main effects of A B and C are highlysignificant since the factors have very small119875 values (lt0001)For 2-way interaction the AB and AC interaction are alsohighly significant (low 119875 values) thus it can be concludedthat there is a strong interaction between two factors withRSS values in this experimental study Figures 13 and 14 showthe interaction and effects plot for power level versus rangemeasurement that is used to interpret the results from Tables4 and 5 It can be seen that theRSS for embeddedRFID systemis nearly equal to nonembedded RFID system at 10 dBmpower level However at 2 dBm power level the RSS forembedded RFID system is better than nonembedded RFIDsystem In terms of range the RSS for embeddedRFID systemis better compared to nonembedded RFID system at both 5-meter and 100-meter range Based on the results obtainedit can be summarized that for indoor tracking purposesthe low power level gives better RSS values for embeddedRFID system compared to nonembedded RFID system Thisfinding can be supported by [29] which stated that the lowtransmit power is appropriate and sufficient for area withlimited range however if the high transmit power is use inthe area with limited range it will resulting in disturbancethat will cause degradation in radiation performanceThus todesign an efficient indoor localization system these findingsmust be taken into account as a guideline implementing thefull setup for real time location monitoring system in WSNplatform

5 Battery Life Testing

A single RFID tag requires a minimum of 27 V to 33 Vvoltage to operate as a transmitter and receiver in thenetwork Howver the embedded RFID tag proposed in thiswork requires minimum of 50V to power up the embeddeddevices Thus the embedded RFID tag can be powered upusing 9 VDC power supply if the tag is located at fixedlocation or Li-Po rechargeable batteries can be used if thetag is mobile The tag has been programmed to consumethe lowest amount of energy as much as possible sinceit introduced an automated switching mechanism betweenindoor and outdoor location tracking and operating in sleepmode at certain timeTheRFID tag only sends data if the validoutdoor location is detected by the GPS receiver however ifthe embedded RFID tag is located inside the building or theplace where the signal from satellite cannot be detected bythe GPS receiver it will turn to sleep mode and wait for theRFID reader interrogation every hourTheGSMmodemonlyworks when the embedded RFID tag is out of the wirelessnetwork coverage to consume the lowest amount of energy

The battery lifetime measurement is performed for threesystems which are stand-alone RFID tag (A) [20] embeddedRFID tag with switching mechanism (B) and embeddedRFID tag without switching mechanism (C) in an indoorenvironment (inside building) where the GPS signals are notavailable most of the time The results show that the battery

12 International Journal of Distributed Sensor Networks

Table 4 Effect estimate summary

Term constant Effect Sum of squares Percent contributionPower (119860) 724 209530 6144Type of RFID (119861) 179 12780 0375Range (119862) minus2791 3116430 91381Power lowast type of RFID (119860119861) minus141 7980 0234Power lowast range (119860119862) 384 58910 1727Type RFID lowast range (119861119862) 074 2180 0064Power lowast type of RFID lowast range (119860119861119862) minus056 1270 00372

Table 5 ANOVA for power level versus range measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Power (119860) 1 209530 209530 1294370 0000Type of RFID (119861) 1 12780 12780 78950 0000Range (119862) 1 3116430 3116430 19252080 0000Power lowast type of RFID (119860119861) 1 7980 7980 49300 0000Power lowast range (119860119862) 1 58910 58910 363900 0000Type RFID lowast range (119861119862) 1 2180 2180 13440 0006Power lowast type of RFID lowast range (119860119861119862) 1 1270 1270 7820 0023Residual error 8 1290 0160Pure error 8 1300 0160Total 15 3410360

for the tag drops from 824V to 711 V in the duration of 85hours for the proposed embedded RFID tag with switchingmechanism compared to other systems as shown in Figure 15

The results for system B are much better than systemC due to the switching introduced However the result forsystem A is better than system B due to the nonembeddedsystem According to these real implementation results thetag will be expected to stop operating approximately in24 hours when the battery reaches the minimum voltagerequired which is 50 V However this result does not includethe situation where the proposed embedded RFID tag islocated outside building which might increase the energyusage The estimation results also did not include otherexternal factors that might reduce the battery life Thus thetag is estimated to cover the operation for a day withouthaving to recharge the battery until the next day To reducethe energy consumption and maximize the time of usage anenergy harvesting technologywill be implemented later in theembedded RFID tag as our next target

6 Conclusion

This paper describes the implementation of an automatedswitching algorithm in the design of a 245GHz active inte-grated RFID system for indoor and outdoor location trackingutilizing WSN platform Four experimental measurementsbased on DOE method are performed in order to validatethe performance of the proposed RFID system as well as tostudy the relationship interaction and effects of each factor

to RSS values extracted from existing and proposed RFIDsystem The results obtained from first experiment show thatembedded RFID system does affect the radiation patternnot more than 15 differences from existing RFID system Itshows that the proposed RFID system contributes to strongerradiation compared to existing RFID system This may bedue to the embedded design and improvement of algorithmimplemented in the proposed RFID system which increasedthe power stability of the system which thus provided highersignal strength [28]

In the second experiment the results show that the pro-posed RFID tag has better RSS value when it is located nearerto the reader at 0∘ degrees direction of propagation with H-field polarization of antenna In the third experiment theexisting RFID system gives better RSS values than proposedRFID system at 10 dBm power level while the proposedRFID system gives better RSS values than existing RFIDsystem at 2 dBm power level In the fourth experiment it canbe summarized that for indoor tracking purposes the lowpower level gives better RSS values for the proposed RFIDtag compared to the existing RFID system The statisticalanalysis ANOVA proved that the RSS values are significantwith range thus it can be concluded that the RSS can beused as a prediction method to evaluate distance betweenRFID reader and RFID tag in WSN platform Based onthe experimental study and statistical analysis done for allsituations a conclusion has been made where the proposedRFID system with automated switching algorithm is betterthan the existing RFID system in terms of performance andvariability of the factors The battery life testing also gives

International Journal of Distributed Sensor Networks 13

Power

Range

Interactions plot

2 10 5 100

minus30

minus45

minus60

minus30

minus45

minus60

minus30

minus45

minus60

Range

Power

Type of RFID

210

5100

Type of RFID

E

N

EN

Figure 13 Interaction plot for power level versus range measurement

Power

Mea

nM

ean

Type RFID

Range

Main effects plot for RSSI

minus40

minus50

minus60

minus40

minus50

minus60

2 10

5 100

EN

Figure 14 Main effects plot for power level versus range measurement

significant findings where the proposed embedded RFIDtag with switching mechanism introduced is much betterthan embedded RFID tag without switching mechanismTo increase the performance of the embedded RFID tag interms of time of usage an energy harvesting technology willbe introduced later as a part of our next contribution Interms of hardware costing the proposed approach is moreacceptable compared to the system developed by [6] since thetechnology used in this work especially GSM communicationis much cheaper than satellite communication used in the

existing system [6] In addition this proposed system doesnot totally depend on GSM since the communication willbe done by RFID which is free as long as the embeddedRFID tag is working in the control area of WSN platformHowever the existing work [6] did not support long rangecommunication and thus when the tag is moving far fromthe reader the satellite communication needs to be used andthis will lead to high cost of communication As a conclusionthe proposed embedded RFID tag provided better signalpropagation compared to the existing RFID tag [19] and

14 International Journal of Distributed Sensor Networks

00 1 2 3 4 5 6 7 8 9

Batte

ry re

duct

ion

()

Time (hours)

Battery lifetime testing

Embedded RFID tag with switchingEmbedded RFID tag without switchingStandalone RFID tag

minus5

minus10

minus15

minus20

minus25

minus30

Figure 15 Battery life testing based on three types of RFIDdesigned

the battery lifetime for embedded RFID tag with switchingmechanism is better than that without switchingmechanism

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank Malaysia Ministry of HigherEducation (LRGS Fund) for sponsoring the research anddeveloping this project Also special appreciation is due toProfessor DrMohamadKamal Abdul Rahim fromUniversitiTeknologiMalaysia Head of the LRGS Project and AssociateProfessor Dr Alyani Ismail Universiti PutraMalaysia for thesupport

References

[1] DIGI XBeeXBee PRO ZB RF Module Datasheet 2013[2] J B Howard ldquoCombination RFID and GPS functionality on

intelligent labelrdquo WO 2003050960 A2 2003[3] S Chand V R Bapat K H Hall et al ldquoPortable RFID reader

having a location determination systemrdquo EP 1752908 A220062006

[4] G O Garsia S I M Nunez L A Saavedra and M D I TapialdquoReal time location system (RTLS) with ZigBee technologyrdquo EP2196816 A12008 2008

[5] A Bandyopadhyay D Hakim B E Funk E A Kohn C AToelis and G Blankenship ldquoSystem and method for locatingtracking andor monitoring the status of personnel andorassets both indoorsrdquo US 20090043504 A1 2009

[6] RFID Journal ldquoHybrid tags includes active RFID GPS satelliteand sensorrdquo RFID Journal 2009 httpwwwrfidjournalcomarticlesview4635

[7] M Mohandes M Haleem M Deriche and K BalakrishnanldquoWireless sensor networks for pilgrims trackingrdquo IEEE Embed-ded Systems Letters vol 4 no 4 pp 106ndash109 2012

[8] N D Simoes J L Goncalves M L Caeiro M J Boavida andF D Cardoso ldquoZigBeeGPS tracking system for rowing racesrdquoin Proceedings of the International Conference on Computer as aTool (EUROCON rsquo11) Lisbon Portugal April 2011

[9] Y Xu R Jiang S Yan and D Xiong ldquoThe research of safetymonitoring system applied in school bus based on the internetof thingsrdquo Procedia Engineering vol 15 pp 2464ndash2468 2011

[10] L Zheng M Li C Wu et al ldquoDevelopment of a smartmobile farming service systemrdquo Mathematical and ComputerModelling vol 54 no 3-4 pp 1194ndash1203 2011

[11] F A Poad and W Ismail ldquoAutomated switching mechanismfor indoor and outdoor propagation with embedded RFID andGPS in wireless sensor network platformrdquo in Proceedings ofthe International Conference on Wireless Network pp 710ndash714London UK 2014

[12] J OrsquoToole J Tuttle M Tuttle et al ldquoRadio frequency identifi-cation sensor tag apparatusesrdquo US 20080030306 A1 2008

[13] V M Gupta and S Annambhotla ldquoRFID tracker and locatorrdquoUS 20090085745 A1 2009

[14] F Shebli I Dayoub and J M Rouvaen ldquoMinimizing energyconsumptionwithinwireless sensor networkrdquoUbiquitous Com-puting and Communication Journal vol 87 2007

[15] R Ding J Hou and B Xing ldquoResearch of wireless sensornetwork nodes based on ambient energy harvestingrdquo in Pro-ceedings of the 6th IEEE International Conference on IntelligentNetworks and Intelligent Systems (ICINIS rsquo13) pp 286ndash288Shenyang China November 2013

[16] A Cammarano C Petrioli andD Spenza ldquoPro-Energy a novelenergy prediction model for solar and wind energy-harvestingwireless sensor networksrdquo in Proceedings of the 9th IEEEInternational Conference on Mobile Ad-Hoc and Sensor Systems(MASS rsquo12) pp 75ndash83 Las Vegas Nev USA October 2012

[17] J Song and Y K Tan ldquoEnergy consumption analysis of ZigBee-based energy harvesting wireless sensor networksrdquo in Proceed-ings of the IEEE International Conference on CommunicationSystems (ICCS rsquo12) pp 468ndash472 November 2012

[18] San Jose Technology Inc Datasheet GPS Engine Board ModelFV M8 2014 httpswwwsparkfuncomdatasheetsGPSFV-M8 Specpdf

[19] W Ismail and R Abdulla ldquoPortable RFID Reader for RTLS(Real Time Location Systems)rdquo PCT International Filing NoPCTMY2011000167 Country Australia Pub No WO2012102600 2012

[20] W Ismail C Z Zulkifli and M G Rahman ldquoWireless pro-duction monitoring systemrdquo PCT International Filing PCTMY2013000037 2013

[21] ISOIEC 18000-4 Information Technology-Radio FrequencyIdentification for ItemManagement-Part7 Parameters for ActiveAir Interface Communications at 245 GHz ISO 18000-42008International Organization for Standardization GenevaSwitzerland 2008

[22] D-H Shih P-L Sun D C Yen and S-M Huang ldquoTaxonomyand survey of RFID anti-collision protocolsrdquo Computer Com-munications vol 29 no 11 pp 2150ndash2166 2006

[23] B Nilsson L Bengtsson P-A Wiberg and B SvenssonldquoProtocols for active RFIDmdashthe energy consumption aspectrdquo inProceedings of the IEEE 2nd International Symposium on Indus-trial Embedded Systems (SIES rsquo07) pp 41ndash48 July 2007

[24] H L van Eeden ldquoWireless Communication Systemrdquo US Patent0045923 2009

International Journal of Distributed Sensor Networks 15

[25] H Haar K Bischof and H Scharke ldquoRFID transponderrdquo WOPatent 109212 2006

[26] C A Balanis AntennaTheory Analysis and Design JohnWileyamp Sons New York NY USA 3rd edition 2005

[27] P Sorrells Optimizing Read Range in RFID System MicrochipTechnology Inc 2000 httpedncomdesignwireless-network-ing4341980Optimizing-read-range-in-RFID-systems

[28] A Ammu L Mapa and A H Jayatissa ldquoEffect of factors onRFID tag readability-statistical analysisrdquo in Proceedings of theIEEE International Conference on ElectroInformation Technol-ogy (EIT rsquo09) pp 355ndash358 June 2009

[29] R Zurawski Embedded System Handbook Second EditionNetwork Embedded Systems CRC Press 2009

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

Page 3: Research Article Propagation Analysis for Automated ...downloads.hindawi.com/journals/ijdsn/2015/392385.pdf · Hybrid tag includes active RFID, GPS, satellite, and sensors [ ] MHz

International Journal of Distributed Sensor Networks 3

Table 1 Comparison of existing and proposed system

Title Wireless technology GPS GSM M2M WSN Indoors andoutdoors Real time Long range

Combination RFID and GPSfunctionality on intelligent label(WO 2003050960 A2) [2]

Active RFID system and GPStechnology Yes No No No No Yes No

Portable RFID reader havinglocation determination(EP 1752908 A22006) [3]

Active RFID system utilizingwireless local area network

(WLAN)Yes No No Yes No Yes No

Real time location system (RLTS)with ZigBee Technology(EP 2196816 A12008) [4]

Did not involve RFID technology No No No Yes No Yes Yes

System and method for locatingtracking and monitoring the statusof personnel andor assets(US 20090043504 A1) [5]

Did not involve RFID technology No No No No No No No

Hybrid tag includes active RFIDGPS satellite and sensors [6]

433MHz active RFID systemGPS and satellitecommunication

Yes No No No Yes Yes No

Proposed RFID system245GHz active RFID WSN

based on IEEE 802154ZigBeeTechnology GPS technology

Yes Yes Yes Yes Yes Yes Yes

easily The standard update rate is 1 Hz and can be up to5Hz as well The current consumption during acquisition is63mA however it will become lower while tracking by about59mA for first 5 minutes 42mA after 5 minutes and 33mAafter 20 minutes The standard output sentences are GGAGLL GSA GSV RMC VTG and ZDA [18]

The embedded RFID tag is periodically sent out locationdata from GPS receiver to RFID reader via WSN platformat monitoring station to track and trace the movement andsequences location of the tagged person or item only whenhaving valid GPS data from satellites However if there areno valid data from satellites the localization will be doneby RFID tag using RSS values retrieved from the proposedembedded RFID tag The RSS value can be requested man-ually or automatically from the RFID reader which later isused to calculate the distance from the RFID routers whichhave fixed data location However this approach may delayaccessing the required information A compromise betweendelay and power efficiency can be achieved by optimallyselecting the interval for periodic transmission of requireddata The location information of the embedded RFID tagalong with the time stamp and identification number istransmitted to theRFID reader periodically to themonitoringstation Periodic transmission of location information facil-itates continuous monitoring but leads to increased powerconsumption and bandwidth occupancy

The subsections to follow elaborate the embedded RFIDtag RFID reader protocols and algorithms involved in theGPS RFIDWSN based tracking system

31 Embedded Active RFID Tag with GPS Functionalities Themotivation of this embedded RFID system is to provide userswith a technology that can track and trace their physical

belonging assets vehicles humans or animal remotely froma monitoring station using WSN platform by providingthe identification and location data as well as contributingto the M2M communication without human interventionThe system is also suitable to be used in supply chainmanagement medical emergencies and crowd control appli-cation and the most popular application demand recentlyis to support information and communication technologiesin collaboration during emergency response and disastermanagement With the aid of this embedded RFID systemall the information about the conditions can be obtained andproblems regarding to the application can be solved remotelyThus the embedded RFID system can reduce the manpoweruse in each management system as well as reducing the costof labour

The novelty of the proposed embedded RFID tag devel-oped in this work is an automated switching protocol ofindoor and outdoor location tracking andmonitoring systemwhich includes a microcontroller embedded with a GPSreceiver together with a wireless communicationmodule andGSM modem to become a single RFID tag which can beused to transmit and receive data fromto the RFID readerThe embedded RFID tag with the cooperation of the RFIDrouters will form a WSN network that can cover a large areaof data transmission and reception The embedded RFID tagis an active RFID that communicates with the RFID readersystem at a frequency of 245GHz ISM band that has anability to transfer the location data to the active RFID readerin a range of 60m indoors and 1500m outdoors and can beextended to more than 1500m via RFID router in multihopcommunication using WSN platform [1] The embedment ofGSMcommunication is to support the global communicationespecially when the embedded RFID tag is not in the wirelessnetwork coverage area However as long as the embedded

4 International Journal of Distributed Sensor Networks

Power management

Input 30V button Cell battery

Reset button

LED TxRx indicator

Wireless communication

module(WSN protocol)

Antenna

Figure 1 Block diagram of the existing RFID tag [19]

Power management

GPS module RTC

38 Kbytes RAM 64 Kbytes flash memory

Microcontroller

Reset button

LED TxRx indicator

Wireless communication

module(WSN protocol)

Input 84VDC output 33VDC and 25VDC

GSM modem

Antenna

Figure 2 Block diagram of the proposed embedded RFID tag

RFID tag is covered by thewireless network theGSMmodemwill not be activated The embedded RFID tag is batterypowered and therefore requires power efficient hardware andminimum amount of signal processing Due to these reasonsan algorithm based on AT mode and cyclic sleep mechanismis implemented to minimize the amount of signal processingand reduce the power consumption Figures 1 and 2 show theblock diagram of the existing RFID tag [19] and proposedembedded RFID tag that is implemented in this work

32 Active RFID Reader The application of RFID is a goodchoice in M2M communication as proposed in [19] and [20]and as such it is appropriate to be used in this proposed RFIDsystem Figure 3 shows the block diagram of the existingactive RFID reader [19] while Figure 4 shows the block dia-gram of the proposed active RFID reader that is implementedin this work The proposed active RFID reader consists ofa wireless communication module two control buttons andreal time clock connected to a microcontroller and RS232driver board for interfacing with a personal computer locatedat the monitoring station The location is periodically sentfrom the embedded RFID tag to RFID reader along with itsID longitude latitude and timestamp

Whenever there is a need to locate items or persons thatcannot be detected by the GPS receiver the RFID reader shall

broadcast a command Each embedded RFID tag received acommand and checks whether satellites detect their loca-tion or not If not the embedded RFID tag will send anacknowledgment and the ID along with RSS value to theRFID reader to indicate their location and distance and turnto sleep mode to reduce energy consumption If there is asatellites reception the embedded RFID tagwill remain silent(no acknowledgment) and continue with data reception fromGPS receiver and periodically send location information tothe RFID reader The RFID RS 232 driver board is added toRFID reader for serial data transmission and reception to orfrom the computer at the monitoring station and is poweredup by a fixed power supply of 9 VDC All communications inRFID reader system and RFID router are configured in APImode

33 Communication Protocols and Algorithms The mini-mization of energy consumptions highly relies on the pro-cessing and communication requirements of the protocolsand algorithms at various layers of the RFID system Thecommunication protocol defines how to exchange instruc-tion and data between reader and tag in both directionsThe ISO 18000-4 has suggested the usage of RTF protocolin 245GHz active RFID system [21] The concept of this

International Journal of Distributed Sensor Networks 5

Power management

Input 5-12VDCoutput 33VDC

Reset button

LED TxRx indicator

FDTI mm232r USB driver

Wireless communication

module(WSN protocol)

Antenna

Figure 3 Block diagram of the existing active RFID reader [19]

Power management

Indoor button RTC Reset tag ID button

38 Kbytes RAM 64 Kbytes flash

memory Microcontroller

MAX3232

RS232 serial communication

Reset button

LED TxRx indicator

Wireless communication

module(WSN protocol)

Input 84VDC output 33VDC and 25VDC

Antenna

Figure 4 Block diagram of the proposed active RFID reader

protocol is that any tag shall not perform data transmissionunless it has received and properly decoded the instructionsent by the reader RTF protocol is identified as one of thesolutions for anticollision in multitag communication [22]and is an effective way to reduce energy consumption of theRFID tag [23]

The proposed active RFID system employed multipro-tocol communication RTF and TTF The TTF protocol isimplemented when the tag is having valid location data fromGPS receiver The method of switching from one protocolto another can be many ways For example active RFID taglistens to RFID reader broadcast command before switchingfrom RTF to TTF mode [24] Meanwhile in [25] the RFIDtag is switchable between TTF and RTF mode within aswitching time frame based on RFID tagrsquos transceiver con-figuration In this work the automated switching algorithmis based on the data location provided by the GPS receiverIf the location is detected by the GPS receiver the embeddedRFID tag will activate the TTF protocol and send the locationdata to the RFID reader If the location is not detected bythe GPS receiver the embedded RFID tag will turn to sleepmode and wake up periodically within certain time frame tosense the incoming signal from RFID reader At this pointthe RFID tag will automatically switch from TTF to RTFmode and wait for interrogation signal from RFID reader

asking for the indoor location data Instead of indoor locationmonitoring the RFID reader also used RTFmode to resettingthe tag ID from the host computer to a specific embeddedRFID tag and thus this function contributes to the M2Mcommunication to change RFID tagrsquos information from theRFID reader All communications described in this work areassumed in WSN platform however if the embedded RFIDtag is not in the WSN coverage the GSM communicationwill be activated Figures 5(a) and 5(b) show the algorithmflow chart of the active RFID reader and tag with the abilityof switching mechanism for indoor and outdoor application

4 Experimental Study forPropagation Analysis

There are four types of experiments conducted in an indoorenvironment with line of sight (LOS) direction The firstexperiment is to measure the radiation pattern for bothexisting and proposed RFID system in order to evaluate theperformance of the developed system based on automatedswitching algorithm presented The second experiment isdesigned based on DOE method to identify the relationshipinteractions and effects of range polarization and angle ofpropagation based on RSS for both types of RFID systemThe third experiment is to find the most significant power

6 International Journal of Distributed Sensor Networks

Start

Initialization

Send collection command

Received ACK

Receive packet(1) Receive GPS data (2) Indoor location data (3) ACK packet

Packet received

Transmit ACK signal

Time out

Time out

Transmit mode

Transmitcommand to

RFID tag

Send to host computer via

RS-232

Receive mode

No

Yes

No

Yes

No

No

Received interrupt from

RS232

Receive interrupt from RTC alarm

Yes

Yes

Receive ACK

Disable interrupt

Yes

No

(a)

Collection command

Start

Initialization

Cyclic sleepmode

Receive GPSsignal

(UART)

Receive UARTinterrupt

GPS data valid

Transmit mode

Valid

Receive mode

Transmit ACK signal

Send ID data

Receive ACK

Time out

Time out

RetrieveRSSI

Retrievetag ID

Transmitmode

Send ID andlocation data

ReceiveACK

Receive RTCalarm interrupt

Networkcoverage

Disableinterruption

Send ID andlocation data

via GSM

Yes

No

No

Yes

Yes

No

No

Yes

No

Yes

NoNo

Yes

Yes

(b)

Figure 5 (a) The flow chart of algorithm implemented in the proposed RFID reader (b) The flow chart of algorithm implemented in theproposed RFID tag

Hostterminal

(PC) RS232 cable

RFIDreader

Antenna

RFID tag

RFID reader

Antenna

180∘

180∘

180∘

180∘

270∘

270∘

0∘360

0∘360

Distance d

H

E-field polarization

H-field polarization

Figure 6 Illustration of polarization measurements

level that provides the best variation of RSS for the proposedRFID system to operate in certain range of indoor locationThe fourth experiment is designed based on DOE methodto investigate the interactions and effects between powerlevel range and RSS for both existing and proposed RFIDsystem All experiments conducted in this work used the

same experimental setup as shown in Figure 6 There aretwo test beds involved for all experiments First test bed isfor nonembedded RFID system (stand-alone system withoutGPS) which consists of existing RFID reader and RFID tag[19] while the second test bed is for the embedded RFIDsystem which consists of a proposed active RFID reader and

International Journal of Distributed Sensor Networks 7

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(a)

NonembeddedEmbedded

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

(b)

Figure 7 H-plane pattern at (a) 2m and (b) 10m distance

RFID tag embedded with GPS The reader and tag are facingeach other at a specific distance which later communicates ina single hop ZigBee network

41 Radiation Pattern Measurement The radiation patternmeasurements are done for both existing RFID system [19]and proposed RFID system to evaluate and compare theperformance of the developed system based on automatedswitching algorithm presented at two different distanceswhich are 2m and 10m The far-field distance is obtained byusing (1) [26] based on characteristic of the antenna used inthis research work Consider

119877119891119891 gt21198712

120582 (1)

where 119877119891119891 is the distance between transmit and receiveantenna 119871 is an antenna aperture under test (825 cm) and120582 is the wavelength of transmission (1205824) indicating as120582 = 119862119891 where 119862 is the speed of light 3 times 108msminus2 andthe operating frequency 119891 = 245GHz Therefore 119877119891119891 isgreater than 042m when the antenna aperture value is notmore than 01m Hence 2m and 10mmeasurement distancesare in the far-field region The algorithm requires a readerto send information to tag and the tag will acknowledgethe packet received from reader and gives information thatis requested by the reader The measurements of RSS areperformed using whip antenna for both RFID reader and tagFor existing RFID system the RSS value is obtained from theX-CTU terminal by using ATDB command However for theproposed RFID system the RSS are retrieved from the algo-rithm implemented in the RFID system The power level of

the wireless communication module has been programmedto maximum value with the output power +10 dBm Theorientation of the antenna for H-field polarization is verticalwhile for E-field polarization it is horizontal as shown inFigure 6 The reader antenna is rotated in 119911-direction from0∘ to 360∘ degrees with 15∘ degree increment steps

Figures 7(a) and 7(b) show the comparison of radiationpatterns of existing RFID system and proposed RFID systemfor H-plane polarization at 2m and 10m distances respec-tively The result for H-field radiation pattern at 2m distanceis an omnidirectional It can be seen that the normalizedradiation pattern for proposed RFID system is slightly betterthan the existing RFID system in between 90∘ and 195∘degrees However the radiation pattern results for existingRFID system at 10m distance are better than proposedRFID system by about 09 dBm to 64 dBm differences Thedifferences in antenna pattern may be due to the immediatesurroundings since actual performance depends on manyfactors in the environment Obstructions in the propagationpath or other wireless networks or systems will also affectthe performance Figures 8(a) and 8(b) show the comparisonof radiation patterns for E-plane polarization at 2m and10mdistances respectivelyThenormalized radiation patternfor proposed RFID system is better than the existing RFIDsystem at 2m and 10m distances Based on these initialfindings it can be concluded that the radiation for shorterdistance is stronger than longer distance in both H-planeand E-plane and the embedded mechanism for RFID systemdoes affect the radiation pattern notmore than 15 differencefrom existing RFID system It shows that the proposed RFIDsystem contributes to stronger radiation compared to existingRFID system This may be due to the embedded power

8 International Journal of Distributed Sensor Networks

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(a)

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(b)

Figure 8 E-plane pattern at (a) 2m and (b) 10m distance

management that is implemented in the embedded activeRFID tag that provided power stability as mentioned by [27]thus increasing the performance of the system in the network

The second experiment is the continuity of the radiationpattern measurement which involved four factors with twolevels of treatment 24 The factors are referred to as type ofRFID (nonembedded embedded) power level (2 dBm10 dBm) range (5m 10m) polarization (E H) and angleof polarization (0∘ 90∘) The response of the experiment isin RSS and 2 replicates are used in this work to reduce themeasurement errors and noiseThe order in which the obser-vations are taken is selected randomly so that the design is acompletely randomized design To identify the interactionsand effects between type of RFID range polarization andangle of propagation with RSS a statistical method calledanalysis of variance (ANOVA) is used as presented in [28]and the effect estimate summaries for the proposed systemare shown in Table 2

In this work the main effect of A really dominates theprocess accounting for 63635 percent of the total variabilitywhereas themain effect of C accounts for about 7328 percentwhile the AB and BC interactions account for about 8808percent and 6616 percent respectively thus it is proved thatthe RSS value can be used as a prediction method to measuredistance and location of tags from reader in WSN platformsince the range contributes highly significantly compared toother factors The ANOVA in Table 3 is used to confirm themagnitude of these effects From Table 3 it can be seen thatthe main effects of A and C are highly significant since bothof the factors have very small 119875 values (lt0001) For 4-wayinteraction the ABCD interactions are also highly significant(low 119875 values) thus it can be concluded that there is a stronginteraction between four factors with RSS values

Figures 9 and 10 show the interactions and main effectsplot for radiation measurement that is used to interpret theresults from Tables 1 and 2

The graph shows that the RSS for existing RFID system isbetter than proposed RFID system at 10m distance howeverthe RSS for proposed RFID system is better than the existingRFID system at 2m distance In terms of angle the RSS for0∘ and 90∘ degree proposed RFID system are better thanexisting RFID system however the RSS value for 0∘ degreesis better than 90∘ degrees for both RFID system In terms ofpolarization the H-field polarization gives more significantvalues of RSS compared to E-field polarization for both RFIDsystems The findings show that the RFID system will havebetter RSS value if it is located nearer to the reader at 0∘degrees direction of propagation with H-field polarizationof antenna This statement is supported by Friis free spaceequation as shown in [26]

119875119903 (119889) =119875119905119866119905119866119903120582

2

(4120587)21198892119871 (2)

where119875119905 is the transmitted power119875119903(119889) is the received power119866119905 is the transmitter antenna gain 119866119903 is the receiver antennagain 119889 is the transmit and receive separation and 119871 is the sys-tem loss factor dependent upon line attenuation filter lossesand antenna losses and not related to propagation Accordingto Friis equation the electromagnetic (EM) signals whentraveling through wireless channels experience fading effectsdue to various effects but for ideal case the transmission thatis considered with a direct line of sight will give the bestvalue of received power The effects of impedance mismatchmisalignment of the antenna pointing and polarization and

International Journal of Distributed Sensor Networks 9

Table 2 Effect estimate summaries

Term constant Effect Sum of squares Percentage contributionRange (119860) minus1109 983460 63635Angle (119861) minus079 4960 0321Polarization (119862) 376 113250 7328Type of RFID (119863) 160 20480 1325Range lowast angle (119860119861) minus358 102250 6616Range lowast polarization (119860119862) minus073 4210 0272Range lowast type of RFID (119860119863) minus226 40950 2650Angle lowast polarization (119861119862) minus412 136120 8808Angle lowast type of RFID (119861119863) 034 0910 0058Polarization lowast type of RFID (119862119863) minus186 27750 1796Range lowast angle lowast polarization (119860119861119862) minus124 12250 0793Range lowast angle lowast type of RFID (119860119861119863) 140 15680 1015Range lowast polarization lowast type of RFID (119860119862119863) minus075 4500 0291Angle lowast polarization lowast type of RFID (119861119862119863) minus100 8000 0518Range lowast angle lowast polarization lowast type of RFID (119860119861119862119863) minus206 34030 2202

Table 3 ANOVA for the radiation pattern measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Range (119860) 1 983460 983461 429340 0000Angle (119861) 1 4960 4961 2170 0160Polarization (119862) 1 113250 113251 49440 0000Type of RFID (119863) 1 20480 20480 8940 0009Range lowast angle (119860119861) 1 102250 102245 44640 0000Range lowast polarization (119860119862) 1 4210 4205 1840 0194Range lowast type of RFID (119860119863) 1 40950 40951 17880 0001Angle lowast polarization (119861119862) 1 136120 136125 59430 0000Angle lowast type of RFID (119861119863) 1 0910 0911 0400 0537Polarization lowast type of RFID (119862119863) 1 27750 27751 12120 0003Range lowast angle lowast polarization (119860119861119862) 1 12250 12251 5350 0034Range lowast angle lowast type of RFID (119860119861119863) 1 15680 15680 6850 0019Range lowast polarization lowast type RFID (119860119862119863) 1 4500 4500 1960 0180Angle lowast polarization lowast type of RFID (119861119862119863) 1 8000 8000 3490 0080Range lowast angle lowast polarization lowast type of RFID (119860119861119862119863) 1 34030 34031 14860 0001Residual error 16 36650 2291Pure error 16 36650 2291Total 31 1545450

absorption can be included by adding additional factors asgiven in [26]

119875119903

119875119905

= 119866119905 (120579119905 120601119905) 119866119903 (120579119903 120601119903) (120582

4120587119877)

2

times (1 minus1003816100381610038161003816Γ1199051003816100381610038161003816

2) (1 minus1003816100381610038161003816Γ1199031003816100381610038161003816

2)1003816100381610038161003816119886119905 sdot 1198861199031003816100381610038161003816

2119890minus120572119877

(3)

where 119866119905(120579119905 120601119905) is the gain of the transmit antenna in thedirection 119866119903(120579119903 120601119903) is the gain of the receive antenna in thedirection (120579119903 120601119903) Γ119905 and Γ119903 are the reflection coefficients ofthe transmit and receive antennas respectively and 119886119905 and119886119903 are the polarization vectors of the transmit and receiveantennas respectively taken in the appropriate directions

120572 is the absorption coefficient of the intervening mediumHowever these additional factors will contribute to reductionof antenna received power Thus in the line of sight directionthe received power will become higher compared to non-lineof sight direction as experimented in this section

42 Power Level versus Range Measurement The power levelversus range measurement is done for distance between 5mand 100m indoorswith 5m incrementThedistance is limiteddue to the buildingrsquos built-up area which is limited to 100-meter lengthThe response is RSS values which are measuredfor existing and proposed RFID system at two levels of powerwhich are 10 dBm (max) and 2 dBm (min) At each point10 single values of receiving power are collected to calculate

10 International Journal of Distributed Sensor Networks

210

Range

090

Angle

Polarization

Type of RFID

Interactions plot

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

2 10

0 90

Range

Angle

Polarization

Type of RFID

E

E

H

H

N

P

N P

Figure 9 Interaction plot for radiation measurement

Range

Mea

nM

ean

Angle

Polarization Type of RFID

Main effects plot for RSSIData means

minus36

minus39

minus42

minus45

minus48

minus36

minus39

minus42

minus45

minus48

2 10 0 90

E H N P

Figure 10 Main effects plot for radiation measurement

the mean Figures 11 and 12 show the comparison betweenRSS values of existing and proposed RFID system based ondistance of 5 meters to 100 meters indoors at 10 dBm and2 dBm power levels The graph shows that the RSS valuesare decreased when the distance is increased at both powerlevels The existing RFID system gives better RSS valuesthan proposed RFID system at 10 dBm power level while the

proposed RFID system gives better RSS values than existingRFID system at 2 dBm power level To compare the perfor-mance of two types of RFID system based on power leveland range the third experiment is conducted based on DOEmethod with three factors having two levels of treatment 23The factors are referred to as type of RFID (nonembeddedembedded) power level (2 dBm 10 dBm) and range (5m

International Journal of Distributed Sensor Networks 11

00 20 40 60 80 100 120

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

RSSI

(dBm

)

Figure 11 Received signal strength values at 10 dBm in an indoorenvironment

00 20 40 60 80 100 120

RSSI

(dBm

)

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

minus80

Figure 12 Received signal strength values at 2 dBm in an indoorenvironment

10m) The response of the experiment is RSS and 2 replicatesare used in this work to reduce the measurement errorsand noise The order in which the observations are takenis selected randomly so that the design is a completelyrandomized design Later the statistical method ANOVA isperformed using Minitab 162 and the results are displayedin Tables 4 and 5 for effect estimate summary and ANOVAfor power level versus range measurement From the effectestimate summary presented in Table 4 the main effect of Creally dominates the process accounting for 91381 percentof the total variability whereas the main effect of A and theAC interaction account for about 6144 and 1727 percentrespectively thus the DOE for this experiment also provedthat the RSS value can be used as a prediction method tomeasure distance and location of tags from reader in wireless

sensor network platform since the range contributions arehighly significant compared to other factors The ANOVAin Table 4 is used to confirm the magnitude of these effectsFrom Table 5 the main effects of A B and C are highlysignificant since the factors have very small119875 values (lt0001)For 2-way interaction the AB and AC interaction are alsohighly significant (low 119875 values) thus it can be concludedthat there is a strong interaction between two factors withRSS values in this experimental study Figures 13 and 14 showthe interaction and effects plot for power level versus rangemeasurement that is used to interpret the results from Tables4 and 5 It can be seen that theRSS for embeddedRFID systemis nearly equal to nonembedded RFID system at 10 dBmpower level However at 2 dBm power level the RSS forembedded RFID system is better than nonembedded RFIDsystem In terms of range the RSS for embeddedRFID systemis better compared to nonembedded RFID system at both 5-meter and 100-meter range Based on the results obtainedit can be summarized that for indoor tracking purposesthe low power level gives better RSS values for embeddedRFID system compared to nonembedded RFID system Thisfinding can be supported by [29] which stated that the lowtransmit power is appropriate and sufficient for area withlimited range however if the high transmit power is use inthe area with limited range it will resulting in disturbancethat will cause degradation in radiation performanceThus todesign an efficient indoor localization system these findingsmust be taken into account as a guideline implementing thefull setup for real time location monitoring system in WSNplatform

5 Battery Life Testing

A single RFID tag requires a minimum of 27 V to 33 Vvoltage to operate as a transmitter and receiver in thenetwork Howver the embedded RFID tag proposed in thiswork requires minimum of 50V to power up the embeddeddevices Thus the embedded RFID tag can be powered upusing 9 VDC power supply if the tag is located at fixedlocation or Li-Po rechargeable batteries can be used if thetag is mobile The tag has been programmed to consumethe lowest amount of energy as much as possible sinceit introduced an automated switching mechanism betweenindoor and outdoor location tracking and operating in sleepmode at certain timeTheRFID tag only sends data if the validoutdoor location is detected by the GPS receiver however ifthe embedded RFID tag is located inside the building or theplace where the signal from satellite cannot be detected bythe GPS receiver it will turn to sleep mode and wait for theRFID reader interrogation every hourTheGSMmodemonlyworks when the embedded RFID tag is out of the wirelessnetwork coverage to consume the lowest amount of energy

The battery lifetime measurement is performed for threesystems which are stand-alone RFID tag (A) [20] embeddedRFID tag with switching mechanism (B) and embeddedRFID tag without switching mechanism (C) in an indoorenvironment (inside building) where the GPS signals are notavailable most of the time The results show that the battery

12 International Journal of Distributed Sensor Networks

Table 4 Effect estimate summary

Term constant Effect Sum of squares Percent contributionPower (119860) 724 209530 6144Type of RFID (119861) 179 12780 0375Range (119862) minus2791 3116430 91381Power lowast type of RFID (119860119861) minus141 7980 0234Power lowast range (119860119862) 384 58910 1727Type RFID lowast range (119861119862) 074 2180 0064Power lowast type of RFID lowast range (119860119861119862) minus056 1270 00372

Table 5 ANOVA for power level versus range measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Power (119860) 1 209530 209530 1294370 0000Type of RFID (119861) 1 12780 12780 78950 0000Range (119862) 1 3116430 3116430 19252080 0000Power lowast type of RFID (119860119861) 1 7980 7980 49300 0000Power lowast range (119860119862) 1 58910 58910 363900 0000Type RFID lowast range (119861119862) 1 2180 2180 13440 0006Power lowast type of RFID lowast range (119860119861119862) 1 1270 1270 7820 0023Residual error 8 1290 0160Pure error 8 1300 0160Total 15 3410360

for the tag drops from 824V to 711 V in the duration of 85hours for the proposed embedded RFID tag with switchingmechanism compared to other systems as shown in Figure 15

The results for system B are much better than systemC due to the switching introduced However the result forsystem A is better than system B due to the nonembeddedsystem According to these real implementation results thetag will be expected to stop operating approximately in24 hours when the battery reaches the minimum voltagerequired which is 50 V However this result does not includethe situation where the proposed embedded RFID tag islocated outside building which might increase the energyusage The estimation results also did not include otherexternal factors that might reduce the battery life Thus thetag is estimated to cover the operation for a day withouthaving to recharge the battery until the next day To reducethe energy consumption and maximize the time of usage anenergy harvesting technologywill be implemented later in theembedded RFID tag as our next target

6 Conclusion

This paper describes the implementation of an automatedswitching algorithm in the design of a 245GHz active inte-grated RFID system for indoor and outdoor location trackingutilizing WSN platform Four experimental measurementsbased on DOE method are performed in order to validatethe performance of the proposed RFID system as well as tostudy the relationship interaction and effects of each factor

to RSS values extracted from existing and proposed RFIDsystem The results obtained from first experiment show thatembedded RFID system does affect the radiation patternnot more than 15 differences from existing RFID system Itshows that the proposed RFID system contributes to strongerradiation compared to existing RFID system This may bedue to the embedded design and improvement of algorithmimplemented in the proposed RFID system which increasedthe power stability of the system which thus provided highersignal strength [28]

In the second experiment the results show that the pro-posed RFID tag has better RSS value when it is located nearerto the reader at 0∘ degrees direction of propagation with H-field polarization of antenna In the third experiment theexisting RFID system gives better RSS values than proposedRFID system at 10 dBm power level while the proposedRFID system gives better RSS values than existing RFIDsystem at 2 dBm power level In the fourth experiment it canbe summarized that for indoor tracking purposes the lowpower level gives better RSS values for the proposed RFIDtag compared to the existing RFID system The statisticalanalysis ANOVA proved that the RSS values are significantwith range thus it can be concluded that the RSS can beused as a prediction method to evaluate distance betweenRFID reader and RFID tag in WSN platform Based onthe experimental study and statistical analysis done for allsituations a conclusion has been made where the proposedRFID system with automated switching algorithm is betterthan the existing RFID system in terms of performance andvariability of the factors The battery life testing also gives

International Journal of Distributed Sensor Networks 13

Power

Range

Interactions plot

2 10 5 100

minus30

minus45

minus60

minus30

minus45

minus60

minus30

minus45

minus60

Range

Power

Type of RFID

210

5100

Type of RFID

E

N

EN

Figure 13 Interaction plot for power level versus range measurement

Power

Mea

nM

ean

Type RFID

Range

Main effects plot for RSSI

minus40

minus50

minus60

minus40

minus50

minus60

2 10

5 100

EN

Figure 14 Main effects plot for power level versus range measurement

significant findings where the proposed embedded RFIDtag with switching mechanism introduced is much betterthan embedded RFID tag without switching mechanismTo increase the performance of the embedded RFID tag interms of time of usage an energy harvesting technology willbe introduced later as a part of our next contribution Interms of hardware costing the proposed approach is moreacceptable compared to the system developed by [6] since thetechnology used in this work especially GSM communicationis much cheaper than satellite communication used in the

existing system [6] In addition this proposed system doesnot totally depend on GSM since the communication willbe done by RFID which is free as long as the embeddedRFID tag is working in the control area of WSN platformHowever the existing work [6] did not support long rangecommunication and thus when the tag is moving far fromthe reader the satellite communication needs to be used andthis will lead to high cost of communication As a conclusionthe proposed embedded RFID tag provided better signalpropagation compared to the existing RFID tag [19] and

14 International Journal of Distributed Sensor Networks

00 1 2 3 4 5 6 7 8 9

Batte

ry re

duct

ion

()

Time (hours)

Battery lifetime testing

Embedded RFID tag with switchingEmbedded RFID tag without switchingStandalone RFID tag

minus5

minus10

minus15

minus20

minus25

minus30

Figure 15 Battery life testing based on three types of RFIDdesigned

the battery lifetime for embedded RFID tag with switchingmechanism is better than that without switchingmechanism

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank Malaysia Ministry of HigherEducation (LRGS Fund) for sponsoring the research anddeveloping this project Also special appreciation is due toProfessor DrMohamadKamal Abdul Rahim fromUniversitiTeknologiMalaysia Head of the LRGS Project and AssociateProfessor Dr Alyani Ismail Universiti PutraMalaysia for thesupport

References

[1] DIGI XBeeXBee PRO ZB RF Module Datasheet 2013[2] J B Howard ldquoCombination RFID and GPS functionality on

intelligent labelrdquo WO 2003050960 A2 2003[3] S Chand V R Bapat K H Hall et al ldquoPortable RFID reader

having a location determination systemrdquo EP 1752908 A220062006

[4] G O Garsia S I M Nunez L A Saavedra and M D I TapialdquoReal time location system (RTLS) with ZigBee technologyrdquo EP2196816 A12008 2008

[5] A Bandyopadhyay D Hakim B E Funk E A Kohn C AToelis and G Blankenship ldquoSystem and method for locatingtracking andor monitoring the status of personnel andorassets both indoorsrdquo US 20090043504 A1 2009

[6] RFID Journal ldquoHybrid tags includes active RFID GPS satelliteand sensorrdquo RFID Journal 2009 httpwwwrfidjournalcomarticlesview4635

[7] M Mohandes M Haleem M Deriche and K BalakrishnanldquoWireless sensor networks for pilgrims trackingrdquo IEEE Embed-ded Systems Letters vol 4 no 4 pp 106ndash109 2012

[8] N D Simoes J L Goncalves M L Caeiro M J Boavida andF D Cardoso ldquoZigBeeGPS tracking system for rowing racesrdquoin Proceedings of the International Conference on Computer as aTool (EUROCON rsquo11) Lisbon Portugal April 2011

[9] Y Xu R Jiang S Yan and D Xiong ldquoThe research of safetymonitoring system applied in school bus based on the internetof thingsrdquo Procedia Engineering vol 15 pp 2464ndash2468 2011

[10] L Zheng M Li C Wu et al ldquoDevelopment of a smartmobile farming service systemrdquo Mathematical and ComputerModelling vol 54 no 3-4 pp 1194ndash1203 2011

[11] F A Poad and W Ismail ldquoAutomated switching mechanismfor indoor and outdoor propagation with embedded RFID andGPS in wireless sensor network platformrdquo in Proceedings ofthe International Conference on Wireless Network pp 710ndash714London UK 2014

[12] J OrsquoToole J Tuttle M Tuttle et al ldquoRadio frequency identifi-cation sensor tag apparatusesrdquo US 20080030306 A1 2008

[13] V M Gupta and S Annambhotla ldquoRFID tracker and locatorrdquoUS 20090085745 A1 2009

[14] F Shebli I Dayoub and J M Rouvaen ldquoMinimizing energyconsumptionwithinwireless sensor networkrdquoUbiquitous Com-puting and Communication Journal vol 87 2007

[15] R Ding J Hou and B Xing ldquoResearch of wireless sensornetwork nodes based on ambient energy harvestingrdquo in Pro-ceedings of the 6th IEEE International Conference on IntelligentNetworks and Intelligent Systems (ICINIS rsquo13) pp 286ndash288Shenyang China November 2013

[16] A Cammarano C Petrioli andD Spenza ldquoPro-Energy a novelenergy prediction model for solar and wind energy-harvestingwireless sensor networksrdquo in Proceedings of the 9th IEEEInternational Conference on Mobile Ad-Hoc and Sensor Systems(MASS rsquo12) pp 75ndash83 Las Vegas Nev USA October 2012

[17] J Song and Y K Tan ldquoEnergy consumption analysis of ZigBee-based energy harvesting wireless sensor networksrdquo in Proceed-ings of the IEEE International Conference on CommunicationSystems (ICCS rsquo12) pp 468ndash472 November 2012

[18] San Jose Technology Inc Datasheet GPS Engine Board ModelFV M8 2014 httpswwwsparkfuncomdatasheetsGPSFV-M8 Specpdf

[19] W Ismail and R Abdulla ldquoPortable RFID Reader for RTLS(Real Time Location Systems)rdquo PCT International Filing NoPCTMY2011000167 Country Australia Pub No WO2012102600 2012

[20] W Ismail C Z Zulkifli and M G Rahman ldquoWireless pro-duction monitoring systemrdquo PCT International Filing PCTMY2013000037 2013

[21] ISOIEC 18000-4 Information Technology-Radio FrequencyIdentification for ItemManagement-Part7 Parameters for ActiveAir Interface Communications at 245 GHz ISO 18000-42008International Organization for Standardization GenevaSwitzerland 2008

[22] D-H Shih P-L Sun D C Yen and S-M Huang ldquoTaxonomyand survey of RFID anti-collision protocolsrdquo Computer Com-munications vol 29 no 11 pp 2150ndash2166 2006

[23] B Nilsson L Bengtsson P-A Wiberg and B SvenssonldquoProtocols for active RFIDmdashthe energy consumption aspectrdquo inProceedings of the IEEE 2nd International Symposium on Indus-trial Embedded Systems (SIES rsquo07) pp 41ndash48 July 2007

[24] H L van Eeden ldquoWireless Communication Systemrdquo US Patent0045923 2009

International Journal of Distributed Sensor Networks 15

[25] H Haar K Bischof and H Scharke ldquoRFID transponderrdquo WOPatent 109212 2006

[26] C A Balanis AntennaTheory Analysis and Design JohnWileyamp Sons New York NY USA 3rd edition 2005

[27] P Sorrells Optimizing Read Range in RFID System MicrochipTechnology Inc 2000 httpedncomdesignwireless-network-ing4341980Optimizing-read-range-in-RFID-systems

[28] A Ammu L Mapa and A H Jayatissa ldquoEffect of factors onRFID tag readability-statistical analysisrdquo in Proceedings of theIEEE International Conference on ElectroInformation Technol-ogy (EIT rsquo09) pp 355ndash358 June 2009

[29] R Zurawski Embedded System Handbook Second EditionNetwork Embedded Systems CRC Press 2009

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

Page 4: Research Article Propagation Analysis for Automated ...downloads.hindawi.com/journals/ijdsn/2015/392385.pdf · Hybrid tag includes active RFID, GPS, satellite, and sensors [ ] MHz

4 International Journal of Distributed Sensor Networks

Power management

Input 30V button Cell battery

Reset button

LED TxRx indicator

Wireless communication

module(WSN protocol)

Antenna

Figure 1 Block diagram of the existing RFID tag [19]

Power management

GPS module RTC

38 Kbytes RAM 64 Kbytes flash memory

Microcontroller

Reset button

LED TxRx indicator

Wireless communication

module(WSN protocol)

Input 84VDC output 33VDC and 25VDC

GSM modem

Antenna

Figure 2 Block diagram of the proposed embedded RFID tag

RFID tag is covered by thewireless network theGSMmodemwill not be activated The embedded RFID tag is batterypowered and therefore requires power efficient hardware andminimum amount of signal processing Due to these reasonsan algorithm based on AT mode and cyclic sleep mechanismis implemented to minimize the amount of signal processingand reduce the power consumption Figures 1 and 2 show theblock diagram of the existing RFID tag [19] and proposedembedded RFID tag that is implemented in this work

32 Active RFID Reader The application of RFID is a goodchoice in M2M communication as proposed in [19] and [20]and as such it is appropriate to be used in this proposed RFIDsystem Figure 3 shows the block diagram of the existingactive RFID reader [19] while Figure 4 shows the block dia-gram of the proposed active RFID reader that is implementedin this work The proposed active RFID reader consists ofa wireless communication module two control buttons andreal time clock connected to a microcontroller and RS232driver board for interfacing with a personal computer locatedat the monitoring station The location is periodically sentfrom the embedded RFID tag to RFID reader along with itsID longitude latitude and timestamp

Whenever there is a need to locate items or persons thatcannot be detected by the GPS receiver the RFID reader shall

broadcast a command Each embedded RFID tag received acommand and checks whether satellites detect their loca-tion or not If not the embedded RFID tag will send anacknowledgment and the ID along with RSS value to theRFID reader to indicate their location and distance and turnto sleep mode to reduce energy consumption If there is asatellites reception the embedded RFID tagwill remain silent(no acknowledgment) and continue with data reception fromGPS receiver and periodically send location information tothe RFID reader The RFID RS 232 driver board is added toRFID reader for serial data transmission and reception to orfrom the computer at the monitoring station and is poweredup by a fixed power supply of 9 VDC All communications inRFID reader system and RFID router are configured in APImode

33 Communication Protocols and Algorithms The mini-mization of energy consumptions highly relies on the pro-cessing and communication requirements of the protocolsand algorithms at various layers of the RFID system Thecommunication protocol defines how to exchange instruc-tion and data between reader and tag in both directionsThe ISO 18000-4 has suggested the usage of RTF protocolin 245GHz active RFID system [21] The concept of this

International Journal of Distributed Sensor Networks 5

Power management

Input 5-12VDCoutput 33VDC

Reset button

LED TxRx indicator

FDTI mm232r USB driver

Wireless communication

module(WSN protocol)

Antenna

Figure 3 Block diagram of the existing active RFID reader [19]

Power management

Indoor button RTC Reset tag ID button

38 Kbytes RAM 64 Kbytes flash

memory Microcontroller

MAX3232

RS232 serial communication

Reset button

LED TxRx indicator

Wireless communication

module(WSN protocol)

Input 84VDC output 33VDC and 25VDC

Antenna

Figure 4 Block diagram of the proposed active RFID reader

protocol is that any tag shall not perform data transmissionunless it has received and properly decoded the instructionsent by the reader RTF protocol is identified as one of thesolutions for anticollision in multitag communication [22]and is an effective way to reduce energy consumption of theRFID tag [23]

The proposed active RFID system employed multipro-tocol communication RTF and TTF The TTF protocol isimplemented when the tag is having valid location data fromGPS receiver The method of switching from one protocolto another can be many ways For example active RFID taglistens to RFID reader broadcast command before switchingfrom RTF to TTF mode [24] Meanwhile in [25] the RFIDtag is switchable between TTF and RTF mode within aswitching time frame based on RFID tagrsquos transceiver con-figuration In this work the automated switching algorithmis based on the data location provided by the GPS receiverIf the location is detected by the GPS receiver the embeddedRFID tag will activate the TTF protocol and send the locationdata to the RFID reader If the location is not detected bythe GPS receiver the embedded RFID tag will turn to sleepmode and wake up periodically within certain time frame tosense the incoming signal from RFID reader At this pointthe RFID tag will automatically switch from TTF to RTFmode and wait for interrogation signal from RFID reader

asking for the indoor location data Instead of indoor locationmonitoring the RFID reader also used RTFmode to resettingthe tag ID from the host computer to a specific embeddedRFID tag and thus this function contributes to the M2Mcommunication to change RFID tagrsquos information from theRFID reader All communications described in this work areassumed in WSN platform however if the embedded RFIDtag is not in the WSN coverage the GSM communicationwill be activated Figures 5(a) and 5(b) show the algorithmflow chart of the active RFID reader and tag with the abilityof switching mechanism for indoor and outdoor application

4 Experimental Study forPropagation Analysis

There are four types of experiments conducted in an indoorenvironment with line of sight (LOS) direction The firstexperiment is to measure the radiation pattern for bothexisting and proposed RFID system in order to evaluate theperformance of the developed system based on automatedswitching algorithm presented The second experiment isdesigned based on DOE method to identify the relationshipinteractions and effects of range polarization and angle ofpropagation based on RSS for both types of RFID systemThe third experiment is to find the most significant power

6 International Journal of Distributed Sensor Networks

Start

Initialization

Send collection command

Received ACK

Receive packet(1) Receive GPS data (2) Indoor location data (3) ACK packet

Packet received

Transmit ACK signal

Time out

Time out

Transmit mode

Transmitcommand to

RFID tag

Send to host computer via

RS-232

Receive mode

No

Yes

No

Yes

No

No

Received interrupt from

RS232

Receive interrupt from RTC alarm

Yes

Yes

Receive ACK

Disable interrupt

Yes

No

(a)

Collection command

Start

Initialization

Cyclic sleepmode

Receive GPSsignal

(UART)

Receive UARTinterrupt

GPS data valid

Transmit mode

Valid

Receive mode

Transmit ACK signal

Send ID data

Receive ACK

Time out

Time out

RetrieveRSSI

Retrievetag ID

Transmitmode

Send ID andlocation data

ReceiveACK

Receive RTCalarm interrupt

Networkcoverage

Disableinterruption

Send ID andlocation data

via GSM

Yes

No

No

Yes

Yes

No

No

Yes

No

Yes

NoNo

Yes

Yes

(b)

Figure 5 (a) The flow chart of algorithm implemented in the proposed RFID reader (b) The flow chart of algorithm implemented in theproposed RFID tag

Hostterminal

(PC) RS232 cable

RFIDreader

Antenna

RFID tag

RFID reader

Antenna

180∘

180∘

180∘

180∘

270∘

270∘

0∘360

0∘360

Distance d

H

E-field polarization

H-field polarization

Figure 6 Illustration of polarization measurements

level that provides the best variation of RSS for the proposedRFID system to operate in certain range of indoor locationThe fourth experiment is designed based on DOE methodto investigate the interactions and effects between powerlevel range and RSS for both existing and proposed RFIDsystem All experiments conducted in this work used the

same experimental setup as shown in Figure 6 There aretwo test beds involved for all experiments First test bed isfor nonembedded RFID system (stand-alone system withoutGPS) which consists of existing RFID reader and RFID tag[19] while the second test bed is for the embedded RFIDsystem which consists of a proposed active RFID reader and

International Journal of Distributed Sensor Networks 7

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(a)

NonembeddedEmbedded

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

(b)

Figure 7 H-plane pattern at (a) 2m and (b) 10m distance

RFID tag embedded with GPS The reader and tag are facingeach other at a specific distance which later communicates ina single hop ZigBee network

41 Radiation Pattern Measurement The radiation patternmeasurements are done for both existing RFID system [19]and proposed RFID system to evaluate and compare theperformance of the developed system based on automatedswitching algorithm presented at two different distanceswhich are 2m and 10m The far-field distance is obtained byusing (1) [26] based on characteristic of the antenna used inthis research work Consider

119877119891119891 gt21198712

120582 (1)

where 119877119891119891 is the distance between transmit and receiveantenna 119871 is an antenna aperture under test (825 cm) and120582 is the wavelength of transmission (1205824) indicating as120582 = 119862119891 where 119862 is the speed of light 3 times 108msminus2 andthe operating frequency 119891 = 245GHz Therefore 119877119891119891 isgreater than 042m when the antenna aperture value is notmore than 01m Hence 2m and 10mmeasurement distancesare in the far-field region The algorithm requires a readerto send information to tag and the tag will acknowledgethe packet received from reader and gives information thatis requested by the reader The measurements of RSS areperformed using whip antenna for both RFID reader and tagFor existing RFID system the RSS value is obtained from theX-CTU terminal by using ATDB command However for theproposed RFID system the RSS are retrieved from the algo-rithm implemented in the RFID system The power level of

the wireless communication module has been programmedto maximum value with the output power +10 dBm Theorientation of the antenna for H-field polarization is verticalwhile for E-field polarization it is horizontal as shown inFigure 6 The reader antenna is rotated in 119911-direction from0∘ to 360∘ degrees with 15∘ degree increment steps

Figures 7(a) and 7(b) show the comparison of radiationpatterns of existing RFID system and proposed RFID systemfor H-plane polarization at 2m and 10m distances respec-tively The result for H-field radiation pattern at 2m distanceis an omnidirectional It can be seen that the normalizedradiation pattern for proposed RFID system is slightly betterthan the existing RFID system in between 90∘ and 195∘degrees However the radiation pattern results for existingRFID system at 10m distance are better than proposedRFID system by about 09 dBm to 64 dBm differences Thedifferences in antenna pattern may be due to the immediatesurroundings since actual performance depends on manyfactors in the environment Obstructions in the propagationpath or other wireless networks or systems will also affectthe performance Figures 8(a) and 8(b) show the comparisonof radiation patterns for E-plane polarization at 2m and10mdistances respectivelyThenormalized radiation patternfor proposed RFID system is better than the existing RFIDsystem at 2m and 10m distances Based on these initialfindings it can be concluded that the radiation for shorterdistance is stronger than longer distance in both H-planeand E-plane and the embedded mechanism for RFID systemdoes affect the radiation pattern notmore than 15 differencefrom existing RFID system It shows that the proposed RFIDsystem contributes to stronger radiation compared to existingRFID system This may be due to the embedded power

8 International Journal of Distributed Sensor Networks

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(a)

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(b)

Figure 8 E-plane pattern at (a) 2m and (b) 10m distance

management that is implemented in the embedded activeRFID tag that provided power stability as mentioned by [27]thus increasing the performance of the system in the network

The second experiment is the continuity of the radiationpattern measurement which involved four factors with twolevels of treatment 24 The factors are referred to as type ofRFID (nonembedded embedded) power level (2 dBm10 dBm) range (5m 10m) polarization (E H) and angleof polarization (0∘ 90∘) The response of the experiment isin RSS and 2 replicates are used in this work to reduce themeasurement errors and noiseThe order in which the obser-vations are taken is selected randomly so that the design is acompletely randomized design To identify the interactionsand effects between type of RFID range polarization andangle of propagation with RSS a statistical method calledanalysis of variance (ANOVA) is used as presented in [28]and the effect estimate summaries for the proposed systemare shown in Table 2

In this work the main effect of A really dominates theprocess accounting for 63635 percent of the total variabilitywhereas themain effect of C accounts for about 7328 percentwhile the AB and BC interactions account for about 8808percent and 6616 percent respectively thus it is proved thatthe RSS value can be used as a prediction method to measuredistance and location of tags from reader in WSN platformsince the range contributes highly significantly compared toother factors The ANOVA in Table 3 is used to confirm themagnitude of these effects From Table 3 it can be seen thatthe main effects of A and C are highly significant since bothof the factors have very small 119875 values (lt0001) For 4-wayinteraction the ABCD interactions are also highly significant(low 119875 values) thus it can be concluded that there is a stronginteraction between four factors with RSS values

Figures 9 and 10 show the interactions and main effectsplot for radiation measurement that is used to interpret theresults from Tables 1 and 2

The graph shows that the RSS for existing RFID system isbetter than proposed RFID system at 10m distance howeverthe RSS for proposed RFID system is better than the existingRFID system at 2m distance In terms of angle the RSS for0∘ and 90∘ degree proposed RFID system are better thanexisting RFID system however the RSS value for 0∘ degreesis better than 90∘ degrees for both RFID system In terms ofpolarization the H-field polarization gives more significantvalues of RSS compared to E-field polarization for both RFIDsystems The findings show that the RFID system will havebetter RSS value if it is located nearer to the reader at 0∘degrees direction of propagation with H-field polarizationof antenna This statement is supported by Friis free spaceequation as shown in [26]

119875119903 (119889) =119875119905119866119905119866119903120582

2

(4120587)21198892119871 (2)

where119875119905 is the transmitted power119875119903(119889) is the received power119866119905 is the transmitter antenna gain 119866119903 is the receiver antennagain 119889 is the transmit and receive separation and 119871 is the sys-tem loss factor dependent upon line attenuation filter lossesand antenna losses and not related to propagation Accordingto Friis equation the electromagnetic (EM) signals whentraveling through wireless channels experience fading effectsdue to various effects but for ideal case the transmission thatis considered with a direct line of sight will give the bestvalue of received power The effects of impedance mismatchmisalignment of the antenna pointing and polarization and

International Journal of Distributed Sensor Networks 9

Table 2 Effect estimate summaries

Term constant Effect Sum of squares Percentage contributionRange (119860) minus1109 983460 63635Angle (119861) minus079 4960 0321Polarization (119862) 376 113250 7328Type of RFID (119863) 160 20480 1325Range lowast angle (119860119861) minus358 102250 6616Range lowast polarization (119860119862) minus073 4210 0272Range lowast type of RFID (119860119863) minus226 40950 2650Angle lowast polarization (119861119862) minus412 136120 8808Angle lowast type of RFID (119861119863) 034 0910 0058Polarization lowast type of RFID (119862119863) minus186 27750 1796Range lowast angle lowast polarization (119860119861119862) minus124 12250 0793Range lowast angle lowast type of RFID (119860119861119863) 140 15680 1015Range lowast polarization lowast type of RFID (119860119862119863) minus075 4500 0291Angle lowast polarization lowast type of RFID (119861119862119863) minus100 8000 0518Range lowast angle lowast polarization lowast type of RFID (119860119861119862119863) minus206 34030 2202

Table 3 ANOVA for the radiation pattern measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Range (119860) 1 983460 983461 429340 0000Angle (119861) 1 4960 4961 2170 0160Polarization (119862) 1 113250 113251 49440 0000Type of RFID (119863) 1 20480 20480 8940 0009Range lowast angle (119860119861) 1 102250 102245 44640 0000Range lowast polarization (119860119862) 1 4210 4205 1840 0194Range lowast type of RFID (119860119863) 1 40950 40951 17880 0001Angle lowast polarization (119861119862) 1 136120 136125 59430 0000Angle lowast type of RFID (119861119863) 1 0910 0911 0400 0537Polarization lowast type of RFID (119862119863) 1 27750 27751 12120 0003Range lowast angle lowast polarization (119860119861119862) 1 12250 12251 5350 0034Range lowast angle lowast type of RFID (119860119861119863) 1 15680 15680 6850 0019Range lowast polarization lowast type RFID (119860119862119863) 1 4500 4500 1960 0180Angle lowast polarization lowast type of RFID (119861119862119863) 1 8000 8000 3490 0080Range lowast angle lowast polarization lowast type of RFID (119860119861119862119863) 1 34030 34031 14860 0001Residual error 16 36650 2291Pure error 16 36650 2291Total 31 1545450

absorption can be included by adding additional factors asgiven in [26]

119875119903

119875119905

= 119866119905 (120579119905 120601119905) 119866119903 (120579119903 120601119903) (120582

4120587119877)

2

times (1 minus1003816100381610038161003816Γ1199051003816100381610038161003816

2) (1 minus1003816100381610038161003816Γ1199031003816100381610038161003816

2)1003816100381610038161003816119886119905 sdot 1198861199031003816100381610038161003816

2119890minus120572119877

(3)

where 119866119905(120579119905 120601119905) is the gain of the transmit antenna in thedirection 119866119903(120579119903 120601119903) is the gain of the receive antenna in thedirection (120579119903 120601119903) Γ119905 and Γ119903 are the reflection coefficients ofthe transmit and receive antennas respectively and 119886119905 and119886119903 are the polarization vectors of the transmit and receiveantennas respectively taken in the appropriate directions

120572 is the absorption coefficient of the intervening mediumHowever these additional factors will contribute to reductionof antenna received power Thus in the line of sight directionthe received power will become higher compared to non-lineof sight direction as experimented in this section

42 Power Level versus Range Measurement The power levelversus range measurement is done for distance between 5mand 100m indoorswith 5m incrementThedistance is limiteddue to the buildingrsquos built-up area which is limited to 100-meter lengthThe response is RSS values which are measuredfor existing and proposed RFID system at two levels of powerwhich are 10 dBm (max) and 2 dBm (min) At each point10 single values of receiving power are collected to calculate

10 International Journal of Distributed Sensor Networks

210

Range

090

Angle

Polarization

Type of RFID

Interactions plot

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

2 10

0 90

Range

Angle

Polarization

Type of RFID

E

E

H

H

N

P

N P

Figure 9 Interaction plot for radiation measurement

Range

Mea

nM

ean

Angle

Polarization Type of RFID

Main effects plot for RSSIData means

minus36

minus39

minus42

minus45

minus48

minus36

minus39

minus42

minus45

minus48

2 10 0 90

E H N P

Figure 10 Main effects plot for radiation measurement

the mean Figures 11 and 12 show the comparison betweenRSS values of existing and proposed RFID system based ondistance of 5 meters to 100 meters indoors at 10 dBm and2 dBm power levels The graph shows that the RSS valuesare decreased when the distance is increased at both powerlevels The existing RFID system gives better RSS valuesthan proposed RFID system at 10 dBm power level while the

proposed RFID system gives better RSS values than existingRFID system at 2 dBm power level To compare the perfor-mance of two types of RFID system based on power leveland range the third experiment is conducted based on DOEmethod with three factors having two levels of treatment 23The factors are referred to as type of RFID (nonembeddedembedded) power level (2 dBm 10 dBm) and range (5m

International Journal of Distributed Sensor Networks 11

00 20 40 60 80 100 120

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

RSSI

(dBm

)

Figure 11 Received signal strength values at 10 dBm in an indoorenvironment

00 20 40 60 80 100 120

RSSI

(dBm

)

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

minus80

Figure 12 Received signal strength values at 2 dBm in an indoorenvironment

10m) The response of the experiment is RSS and 2 replicatesare used in this work to reduce the measurement errorsand noise The order in which the observations are takenis selected randomly so that the design is a completelyrandomized design Later the statistical method ANOVA isperformed using Minitab 162 and the results are displayedin Tables 4 and 5 for effect estimate summary and ANOVAfor power level versus range measurement From the effectestimate summary presented in Table 4 the main effect of Creally dominates the process accounting for 91381 percentof the total variability whereas the main effect of A and theAC interaction account for about 6144 and 1727 percentrespectively thus the DOE for this experiment also provedthat the RSS value can be used as a prediction method tomeasure distance and location of tags from reader in wireless

sensor network platform since the range contributions arehighly significant compared to other factors The ANOVAin Table 4 is used to confirm the magnitude of these effectsFrom Table 5 the main effects of A B and C are highlysignificant since the factors have very small119875 values (lt0001)For 2-way interaction the AB and AC interaction are alsohighly significant (low 119875 values) thus it can be concludedthat there is a strong interaction between two factors withRSS values in this experimental study Figures 13 and 14 showthe interaction and effects plot for power level versus rangemeasurement that is used to interpret the results from Tables4 and 5 It can be seen that theRSS for embeddedRFID systemis nearly equal to nonembedded RFID system at 10 dBmpower level However at 2 dBm power level the RSS forembedded RFID system is better than nonembedded RFIDsystem In terms of range the RSS for embeddedRFID systemis better compared to nonembedded RFID system at both 5-meter and 100-meter range Based on the results obtainedit can be summarized that for indoor tracking purposesthe low power level gives better RSS values for embeddedRFID system compared to nonembedded RFID system Thisfinding can be supported by [29] which stated that the lowtransmit power is appropriate and sufficient for area withlimited range however if the high transmit power is use inthe area with limited range it will resulting in disturbancethat will cause degradation in radiation performanceThus todesign an efficient indoor localization system these findingsmust be taken into account as a guideline implementing thefull setup for real time location monitoring system in WSNplatform

5 Battery Life Testing

A single RFID tag requires a minimum of 27 V to 33 Vvoltage to operate as a transmitter and receiver in thenetwork Howver the embedded RFID tag proposed in thiswork requires minimum of 50V to power up the embeddeddevices Thus the embedded RFID tag can be powered upusing 9 VDC power supply if the tag is located at fixedlocation or Li-Po rechargeable batteries can be used if thetag is mobile The tag has been programmed to consumethe lowest amount of energy as much as possible sinceit introduced an automated switching mechanism betweenindoor and outdoor location tracking and operating in sleepmode at certain timeTheRFID tag only sends data if the validoutdoor location is detected by the GPS receiver however ifthe embedded RFID tag is located inside the building or theplace where the signal from satellite cannot be detected bythe GPS receiver it will turn to sleep mode and wait for theRFID reader interrogation every hourTheGSMmodemonlyworks when the embedded RFID tag is out of the wirelessnetwork coverage to consume the lowest amount of energy

The battery lifetime measurement is performed for threesystems which are stand-alone RFID tag (A) [20] embeddedRFID tag with switching mechanism (B) and embeddedRFID tag without switching mechanism (C) in an indoorenvironment (inside building) where the GPS signals are notavailable most of the time The results show that the battery

12 International Journal of Distributed Sensor Networks

Table 4 Effect estimate summary

Term constant Effect Sum of squares Percent contributionPower (119860) 724 209530 6144Type of RFID (119861) 179 12780 0375Range (119862) minus2791 3116430 91381Power lowast type of RFID (119860119861) minus141 7980 0234Power lowast range (119860119862) 384 58910 1727Type RFID lowast range (119861119862) 074 2180 0064Power lowast type of RFID lowast range (119860119861119862) minus056 1270 00372

Table 5 ANOVA for power level versus range measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Power (119860) 1 209530 209530 1294370 0000Type of RFID (119861) 1 12780 12780 78950 0000Range (119862) 1 3116430 3116430 19252080 0000Power lowast type of RFID (119860119861) 1 7980 7980 49300 0000Power lowast range (119860119862) 1 58910 58910 363900 0000Type RFID lowast range (119861119862) 1 2180 2180 13440 0006Power lowast type of RFID lowast range (119860119861119862) 1 1270 1270 7820 0023Residual error 8 1290 0160Pure error 8 1300 0160Total 15 3410360

for the tag drops from 824V to 711 V in the duration of 85hours for the proposed embedded RFID tag with switchingmechanism compared to other systems as shown in Figure 15

The results for system B are much better than systemC due to the switching introduced However the result forsystem A is better than system B due to the nonembeddedsystem According to these real implementation results thetag will be expected to stop operating approximately in24 hours when the battery reaches the minimum voltagerequired which is 50 V However this result does not includethe situation where the proposed embedded RFID tag islocated outside building which might increase the energyusage The estimation results also did not include otherexternal factors that might reduce the battery life Thus thetag is estimated to cover the operation for a day withouthaving to recharge the battery until the next day To reducethe energy consumption and maximize the time of usage anenergy harvesting technologywill be implemented later in theembedded RFID tag as our next target

6 Conclusion

This paper describes the implementation of an automatedswitching algorithm in the design of a 245GHz active inte-grated RFID system for indoor and outdoor location trackingutilizing WSN platform Four experimental measurementsbased on DOE method are performed in order to validatethe performance of the proposed RFID system as well as tostudy the relationship interaction and effects of each factor

to RSS values extracted from existing and proposed RFIDsystem The results obtained from first experiment show thatembedded RFID system does affect the radiation patternnot more than 15 differences from existing RFID system Itshows that the proposed RFID system contributes to strongerradiation compared to existing RFID system This may bedue to the embedded design and improvement of algorithmimplemented in the proposed RFID system which increasedthe power stability of the system which thus provided highersignal strength [28]

In the second experiment the results show that the pro-posed RFID tag has better RSS value when it is located nearerto the reader at 0∘ degrees direction of propagation with H-field polarization of antenna In the third experiment theexisting RFID system gives better RSS values than proposedRFID system at 10 dBm power level while the proposedRFID system gives better RSS values than existing RFIDsystem at 2 dBm power level In the fourth experiment it canbe summarized that for indoor tracking purposes the lowpower level gives better RSS values for the proposed RFIDtag compared to the existing RFID system The statisticalanalysis ANOVA proved that the RSS values are significantwith range thus it can be concluded that the RSS can beused as a prediction method to evaluate distance betweenRFID reader and RFID tag in WSN platform Based onthe experimental study and statistical analysis done for allsituations a conclusion has been made where the proposedRFID system with automated switching algorithm is betterthan the existing RFID system in terms of performance andvariability of the factors The battery life testing also gives

International Journal of Distributed Sensor Networks 13

Power

Range

Interactions plot

2 10 5 100

minus30

minus45

minus60

minus30

minus45

minus60

minus30

minus45

minus60

Range

Power

Type of RFID

210

5100

Type of RFID

E

N

EN

Figure 13 Interaction plot for power level versus range measurement

Power

Mea

nM

ean

Type RFID

Range

Main effects plot for RSSI

minus40

minus50

minus60

minus40

minus50

minus60

2 10

5 100

EN

Figure 14 Main effects plot for power level versus range measurement

significant findings where the proposed embedded RFIDtag with switching mechanism introduced is much betterthan embedded RFID tag without switching mechanismTo increase the performance of the embedded RFID tag interms of time of usage an energy harvesting technology willbe introduced later as a part of our next contribution Interms of hardware costing the proposed approach is moreacceptable compared to the system developed by [6] since thetechnology used in this work especially GSM communicationis much cheaper than satellite communication used in the

existing system [6] In addition this proposed system doesnot totally depend on GSM since the communication willbe done by RFID which is free as long as the embeddedRFID tag is working in the control area of WSN platformHowever the existing work [6] did not support long rangecommunication and thus when the tag is moving far fromthe reader the satellite communication needs to be used andthis will lead to high cost of communication As a conclusionthe proposed embedded RFID tag provided better signalpropagation compared to the existing RFID tag [19] and

14 International Journal of Distributed Sensor Networks

00 1 2 3 4 5 6 7 8 9

Batte

ry re

duct

ion

()

Time (hours)

Battery lifetime testing

Embedded RFID tag with switchingEmbedded RFID tag without switchingStandalone RFID tag

minus5

minus10

minus15

minus20

minus25

minus30

Figure 15 Battery life testing based on three types of RFIDdesigned

the battery lifetime for embedded RFID tag with switchingmechanism is better than that without switchingmechanism

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank Malaysia Ministry of HigherEducation (LRGS Fund) for sponsoring the research anddeveloping this project Also special appreciation is due toProfessor DrMohamadKamal Abdul Rahim fromUniversitiTeknologiMalaysia Head of the LRGS Project and AssociateProfessor Dr Alyani Ismail Universiti PutraMalaysia for thesupport

References

[1] DIGI XBeeXBee PRO ZB RF Module Datasheet 2013[2] J B Howard ldquoCombination RFID and GPS functionality on

intelligent labelrdquo WO 2003050960 A2 2003[3] S Chand V R Bapat K H Hall et al ldquoPortable RFID reader

having a location determination systemrdquo EP 1752908 A220062006

[4] G O Garsia S I M Nunez L A Saavedra and M D I TapialdquoReal time location system (RTLS) with ZigBee technologyrdquo EP2196816 A12008 2008

[5] A Bandyopadhyay D Hakim B E Funk E A Kohn C AToelis and G Blankenship ldquoSystem and method for locatingtracking andor monitoring the status of personnel andorassets both indoorsrdquo US 20090043504 A1 2009

[6] RFID Journal ldquoHybrid tags includes active RFID GPS satelliteand sensorrdquo RFID Journal 2009 httpwwwrfidjournalcomarticlesview4635

[7] M Mohandes M Haleem M Deriche and K BalakrishnanldquoWireless sensor networks for pilgrims trackingrdquo IEEE Embed-ded Systems Letters vol 4 no 4 pp 106ndash109 2012

[8] N D Simoes J L Goncalves M L Caeiro M J Boavida andF D Cardoso ldquoZigBeeGPS tracking system for rowing racesrdquoin Proceedings of the International Conference on Computer as aTool (EUROCON rsquo11) Lisbon Portugal April 2011

[9] Y Xu R Jiang S Yan and D Xiong ldquoThe research of safetymonitoring system applied in school bus based on the internetof thingsrdquo Procedia Engineering vol 15 pp 2464ndash2468 2011

[10] L Zheng M Li C Wu et al ldquoDevelopment of a smartmobile farming service systemrdquo Mathematical and ComputerModelling vol 54 no 3-4 pp 1194ndash1203 2011

[11] F A Poad and W Ismail ldquoAutomated switching mechanismfor indoor and outdoor propagation with embedded RFID andGPS in wireless sensor network platformrdquo in Proceedings ofthe International Conference on Wireless Network pp 710ndash714London UK 2014

[12] J OrsquoToole J Tuttle M Tuttle et al ldquoRadio frequency identifi-cation sensor tag apparatusesrdquo US 20080030306 A1 2008

[13] V M Gupta and S Annambhotla ldquoRFID tracker and locatorrdquoUS 20090085745 A1 2009

[14] F Shebli I Dayoub and J M Rouvaen ldquoMinimizing energyconsumptionwithinwireless sensor networkrdquoUbiquitous Com-puting and Communication Journal vol 87 2007

[15] R Ding J Hou and B Xing ldquoResearch of wireless sensornetwork nodes based on ambient energy harvestingrdquo in Pro-ceedings of the 6th IEEE International Conference on IntelligentNetworks and Intelligent Systems (ICINIS rsquo13) pp 286ndash288Shenyang China November 2013

[16] A Cammarano C Petrioli andD Spenza ldquoPro-Energy a novelenergy prediction model for solar and wind energy-harvestingwireless sensor networksrdquo in Proceedings of the 9th IEEEInternational Conference on Mobile Ad-Hoc and Sensor Systems(MASS rsquo12) pp 75ndash83 Las Vegas Nev USA October 2012

[17] J Song and Y K Tan ldquoEnergy consumption analysis of ZigBee-based energy harvesting wireless sensor networksrdquo in Proceed-ings of the IEEE International Conference on CommunicationSystems (ICCS rsquo12) pp 468ndash472 November 2012

[18] San Jose Technology Inc Datasheet GPS Engine Board ModelFV M8 2014 httpswwwsparkfuncomdatasheetsGPSFV-M8 Specpdf

[19] W Ismail and R Abdulla ldquoPortable RFID Reader for RTLS(Real Time Location Systems)rdquo PCT International Filing NoPCTMY2011000167 Country Australia Pub No WO2012102600 2012

[20] W Ismail C Z Zulkifli and M G Rahman ldquoWireless pro-duction monitoring systemrdquo PCT International Filing PCTMY2013000037 2013

[21] ISOIEC 18000-4 Information Technology-Radio FrequencyIdentification for ItemManagement-Part7 Parameters for ActiveAir Interface Communications at 245 GHz ISO 18000-42008International Organization for Standardization GenevaSwitzerland 2008

[22] D-H Shih P-L Sun D C Yen and S-M Huang ldquoTaxonomyand survey of RFID anti-collision protocolsrdquo Computer Com-munications vol 29 no 11 pp 2150ndash2166 2006

[23] B Nilsson L Bengtsson P-A Wiberg and B SvenssonldquoProtocols for active RFIDmdashthe energy consumption aspectrdquo inProceedings of the IEEE 2nd International Symposium on Indus-trial Embedded Systems (SIES rsquo07) pp 41ndash48 July 2007

[24] H L van Eeden ldquoWireless Communication Systemrdquo US Patent0045923 2009

International Journal of Distributed Sensor Networks 15

[25] H Haar K Bischof and H Scharke ldquoRFID transponderrdquo WOPatent 109212 2006

[26] C A Balanis AntennaTheory Analysis and Design JohnWileyamp Sons New York NY USA 3rd edition 2005

[27] P Sorrells Optimizing Read Range in RFID System MicrochipTechnology Inc 2000 httpedncomdesignwireless-network-ing4341980Optimizing-read-range-in-RFID-systems

[28] A Ammu L Mapa and A H Jayatissa ldquoEffect of factors onRFID tag readability-statistical analysisrdquo in Proceedings of theIEEE International Conference on ElectroInformation Technol-ogy (EIT rsquo09) pp 355ndash358 June 2009

[29] R Zurawski Embedded System Handbook Second EditionNetwork Embedded Systems CRC Press 2009

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

Page 5: Research Article Propagation Analysis for Automated ...downloads.hindawi.com/journals/ijdsn/2015/392385.pdf · Hybrid tag includes active RFID, GPS, satellite, and sensors [ ] MHz

International Journal of Distributed Sensor Networks 5

Power management

Input 5-12VDCoutput 33VDC

Reset button

LED TxRx indicator

FDTI mm232r USB driver

Wireless communication

module(WSN protocol)

Antenna

Figure 3 Block diagram of the existing active RFID reader [19]

Power management

Indoor button RTC Reset tag ID button

38 Kbytes RAM 64 Kbytes flash

memory Microcontroller

MAX3232

RS232 serial communication

Reset button

LED TxRx indicator

Wireless communication

module(WSN protocol)

Input 84VDC output 33VDC and 25VDC

Antenna

Figure 4 Block diagram of the proposed active RFID reader

protocol is that any tag shall not perform data transmissionunless it has received and properly decoded the instructionsent by the reader RTF protocol is identified as one of thesolutions for anticollision in multitag communication [22]and is an effective way to reduce energy consumption of theRFID tag [23]

The proposed active RFID system employed multipro-tocol communication RTF and TTF The TTF protocol isimplemented when the tag is having valid location data fromGPS receiver The method of switching from one protocolto another can be many ways For example active RFID taglistens to RFID reader broadcast command before switchingfrom RTF to TTF mode [24] Meanwhile in [25] the RFIDtag is switchable between TTF and RTF mode within aswitching time frame based on RFID tagrsquos transceiver con-figuration In this work the automated switching algorithmis based on the data location provided by the GPS receiverIf the location is detected by the GPS receiver the embeddedRFID tag will activate the TTF protocol and send the locationdata to the RFID reader If the location is not detected bythe GPS receiver the embedded RFID tag will turn to sleepmode and wake up periodically within certain time frame tosense the incoming signal from RFID reader At this pointthe RFID tag will automatically switch from TTF to RTFmode and wait for interrogation signal from RFID reader

asking for the indoor location data Instead of indoor locationmonitoring the RFID reader also used RTFmode to resettingthe tag ID from the host computer to a specific embeddedRFID tag and thus this function contributes to the M2Mcommunication to change RFID tagrsquos information from theRFID reader All communications described in this work areassumed in WSN platform however if the embedded RFIDtag is not in the WSN coverage the GSM communicationwill be activated Figures 5(a) and 5(b) show the algorithmflow chart of the active RFID reader and tag with the abilityof switching mechanism for indoor and outdoor application

4 Experimental Study forPropagation Analysis

There are four types of experiments conducted in an indoorenvironment with line of sight (LOS) direction The firstexperiment is to measure the radiation pattern for bothexisting and proposed RFID system in order to evaluate theperformance of the developed system based on automatedswitching algorithm presented The second experiment isdesigned based on DOE method to identify the relationshipinteractions and effects of range polarization and angle ofpropagation based on RSS for both types of RFID systemThe third experiment is to find the most significant power

6 International Journal of Distributed Sensor Networks

Start

Initialization

Send collection command

Received ACK

Receive packet(1) Receive GPS data (2) Indoor location data (3) ACK packet

Packet received

Transmit ACK signal

Time out

Time out

Transmit mode

Transmitcommand to

RFID tag

Send to host computer via

RS-232

Receive mode

No

Yes

No

Yes

No

No

Received interrupt from

RS232

Receive interrupt from RTC alarm

Yes

Yes

Receive ACK

Disable interrupt

Yes

No

(a)

Collection command

Start

Initialization

Cyclic sleepmode

Receive GPSsignal

(UART)

Receive UARTinterrupt

GPS data valid

Transmit mode

Valid

Receive mode

Transmit ACK signal

Send ID data

Receive ACK

Time out

Time out

RetrieveRSSI

Retrievetag ID

Transmitmode

Send ID andlocation data

ReceiveACK

Receive RTCalarm interrupt

Networkcoverage

Disableinterruption

Send ID andlocation data

via GSM

Yes

No

No

Yes

Yes

No

No

Yes

No

Yes

NoNo

Yes

Yes

(b)

Figure 5 (a) The flow chart of algorithm implemented in the proposed RFID reader (b) The flow chart of algorithm implemented in theproposed RFID tag

Hostterminal

(PC) RS232 cable

RFIDreader

Antenna

RFID tag

RFID reader

Antenna

180∘

180∘

180∘

180∘

270∘

270∘

0∘360

0∘360

Distance d

H

E-field polarization

H-field polarization

Figure 6 Illustration of polarization measurements

level that provides the best variation of RSS for the proposedRFID system to operate in certain range of indoor locationThe fourth experiment is designed based on DOE methodto investigate the interactions and effects between powerlevel range and RSS for both existing and proposed RFIDsystem All experiments conducted in this work used the

same experimental setup as shown in Figure 6 There aretwo test beds involved for all experiments First test bed isfor nonembedded RFID system (stand-alone system withoutGPS) which consists of existing RFID reader and RFID tag[19] while the second test bed is for the embedded RFIDsystem which consists of a proposed active RFID reader and

International Journal of Distributed Sensor Networks 7

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(a)

NonembeddedEmbedded

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

(b)

Figure 7 H-plane pattern at (a) 2m and (b) 10m distance

RFID tag embedded with GPS The reader and tag are facingeach other at a specific distance which later communicates ina single hop ZigBee network

41 Radiation Pattern Measurement The radiation patternmeasurements are done for both existing RFID system [19]and proposed RFID system to evaluate and compare theperformance of the developed system based on automatedswitching algorithm presented at two different distanceswhich are 2m and 10m The far-field distance is obtained byusing (1) [26] based on characteristic of the antenna used inthis research work Consider

119877119891119891 gt21198712

120582 (1)

where 119877119891119891 is the distance between transmit and receiveantenna 119871 is an antenna aperture under test (825 cm) and120582 is the wavelength of transmission (1205824) indicating as120582 = 119862119891 where 119862 is the speed of light 3 times 108msminus2 andthe operating frequency 119891 = 245GHz Therefore 119877119891119891 isgreater than 042m when the antenna aperture value is notmore than 01m Hence 2m and 10mmeasurement distancesare in the far-field region The algorithm requires a readerto send information to tag and the tag will acknowledgethe packet received from reader and gives information thatis requested by the reader The measurements of RSS areperformed using whip antenna for both RFID reader and tagFor existing RFID system the RSS value is obtained from theX-CTU terminal by using ATDB command However for theproposed RFID system the RSS are retrieved from the algo-rithm implemented in the RFID system The power level of

the wireless communication module has been programmedto maximum value with the output power +10 dBm Theorientation of the antenna for H-field polarization is verticalwhile for E-field polarization it is horizontal as shown inFigure 6 The reader antenna is rotated in 119911-direction from0∘ to 360∘ degrees with 15∘ degree increment steps

Figures 7(a) and 7(b) show the comparison of radiationpatterns of existing RFID system and proposed RFID systemfor H-plane polarization at 2m and 10m distances respec-tively The result for H-field radiation pattern at 2m distanceis an omnidirectional It can be seen that the normalizedradiation pattern for proposed RFID system is slightly betterthan the existing RFID system in between 90∘ and 195∘degrees However the radiation pattern results for existingRFID system at 10m distance are better than proposedRFID system by about 09 dBm to 64 dBm differences Thedifferences in antenna pattern may be due to the immediatesurroundings since actual performance depends on manyfactors in the environment Obstructions in the propagationpath or other wireless networks or systems will also affectthe performance Figures 8(a) and 8(b) show the comparisonof radiation patterns for E-plane polarization at 2m and10mdistances respectivelyThenormalized radiation patternfor proposed RFID system is better than the existing RFIDsystem at 2m and 10m distances Based on these initialfindings it can be concluded that the radiation for shorterdistance is stronger than longer distance in both H-planeand E-plane and the embedded mechanism for RFID systemdoes affect the radiation pattern notmore than 15 differencefrom existing RFID system It shows that the proposed RFIDsystem contributes to stronger radiation compared to existingRFID system This may be due to the embedded power

8 International Journal of Distributed Sensor Networks

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(a)

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(b)

Figure 8 E-plane pattern at (a) 2m and (b) 10m distance

management that is implemented in the embedded activeRFID tag that provided power stability as mentioned by [27]thus increasing the performance of the system in the network

The second experiment is the continuity of the radiationpattern measurement which involved four factors with twolevels of treatment 24 The factors are referred to as type ofRFID (nonembedded embedded) power level (2 dBm10 dBm) range (5m 10m) polarization (E H) and angleof polarization (0∘ 90∘) The response of the experiment isin RSS and 2 replicates are used in this work to reduce themeasurement errors and noiseThe order in which the obser-vations are taken is selected randomly so that the design is acompletely randomized design To identify the interactionsand effects between type of RFID range polarization andangle of propagation with RSS a statistical method calledanalysis of variance (ANOVA) is used as presented in [28]and the effect estimate summaries for the proposed systemare shown in Table 2

In this work the main effect of A really dominates theprocess accounting for 63635 percent of the total variabilitywhereas themain effect of C accounts for about 7328 percentwhile the AB and BC interactions account for about 8808percent and 6616 percent respectively thus it is proved thatthe RSS value can be used as a prediction method to measuredistance and location of tags from reader in WSN platformsince the range contributes highly significantly compared toother factors The ANOVA in Table 3 is used to confirm themagnitude of these effects From Table 3 it can be seen thatthe main effects of A and C are highly significant since bothof the factors have very small 119875 values (lt0001) For 4-wayinteraction the ABCD interactions are also highly significant(low 119875 values) thus it can be concluded that there is a stronginteraction between four factors with RSS values

Figures 9 and 10 show the interactions and main effectsplot for radiation measurement that is used to interpret theresults from Tables 1 and 2

The graph shows that the RSS for existing RFID system isbetter than proposed RFID system at 10m distance howeverthe RSS for proposed RFID system is better than the existingRFID system at 2m distance In terms of angle the RSS for0∘ and 90∘ degree proposed RFID system are better thanexisting RFID system however the RSS value for 0∘ degreesis better than 90∘ degrees for both RFID system In terms ofpolarization the H-field polarization gives more significantvalues of RSS compared to E-field polarization for both RFIDsystems The findings show that the RFID system will havebetter RSS value if it is located nearer to the reader at 0∘degrees direction of propagation with H-field polarizationof antenna This statement is supported by Friis free spaceequation as shown in [26]

119875119903 (119889) =119875119905119866119905119866119903120582

2

(4120587)21198892119871 (2)

where119875119905 is the transmitted power119875119903(119889) is the received power119866119905 is the transmitter antenna gain 119866119903 is the receiver antennagain 119889 is the transmit and receive separation and 119871 is the sys-tem loss factor dependent upon line attenuation filter lossesand antenna losses and not related to propagation Accordingto Friis equation the electromagnetic (EM) signals whentraveling through wireless channels experience fading effectsdue to various effects but for ideal case the transmission thatis considered with a direct line of sight will give the bestvalue of received power The effects of impedance mismatchmisalignment of the antenna pointing and polarization and

International Journal of Distributed Sensor Networks 9

Table 2 Effect estimate summaries

Term constant Effect Sum of squares Percentage contributionRange (119860) minus1109 983460 63635Angle (119861) minus079 4960 0321Polarization (119862) 376 113250 7328Type of RFID (119863) 160 20480 1325Range lowast angle (119860119861) minus358 102250 6616Range lowast polarization (119860119862) minus073 4210 0272Range lowast type of RFID (119860119863) minus226 40950 2650Angle lowast polarization (119861119862) minus412 136120 8808Angle lowast type of RFID (119861119863) 034 0910 0058Polarization lowast type of RFID (119862119863) minus186 27750 1796Range lowast angle lowast polarization (119860119861119862) minus124 12250 0793Range lowast angle lowast type of RFID (119860119861119863) 140 15680 1015Range lowast polarization lowast type of RFID (119860119862119863) minus075 4500 0291Angle lowast polarization lowast type of RFID (119861119862119863) minus100 8000 0518Range lowast angle lowast polarization lowast type of RFID (119860119861119862119863) minus206 34030 2202

Table 3 ANOVA for the radiation pattern measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Range (119860) 1 983460 983461 429340 0000Angle (119861) 1 4960 4961 2170 0160Polarization (119862) 1 113250 113251 49440 0000Type of RFID (119863) 1 20480 20480 8940 0009Range lowast angle (119860119861) 1 102250 102245 44640 0000Range lowast polarization (119860119862) 1 4210 4205 1840 0194Range lowast type of RFID (119860119863) 1 40950 40951 17880 0001Angle lowast polarization (119861119862) 1 136120 136125 59430 0000Angle lowast type of RFID (119861119863) 1 0910 0911 0400 0537Polarization lowast type of RFID (119862119863) 1 27750 27751 12120 0003Range lowast angle lowast polarization (119860119861119862) 1 12250 12251 5350 0034Range lowast angle lowast type of RFID (119860119861119863) 1 15680 15680 6850 0019Range lowast polarization lowast type RFID (119860119862119863) 1 4500 4500 1960 0180Angle lowast polarization lowast type of RFID (119861119862119863) 1 8000 8000 3490 0080Range lowast angle lowast polarization lowast type of RFID (119860119861119862119863) 1 34030 34031 14860 0001Residual error 16 36650 2291Pure error 16 36650 2291Total 31 1545450

absorption can be included by adding additional factors asgiven in [26]

119875119903

119875119905

= 119866119905 (120579119905 120601119905) 119866119903 (120579119903 120601119903) (120582

4120587119877)

2

times (1 minus1003816100381610038161003816Γ1199051003816100381610038161003816

2) (1 minus1003816100381610038161003816Γ1199031003816100381610038161003816

2)1003816100381610038161003816119886119905 sdot 1198861199031003816100381610038161003816

2119890minus120572119877

(3)

where 119866119905(120579119905 120601119905) is the gain of the transmit antenna in thedirection 119866119903(120579119903 120601119903) is the gain of the receive antenna in thedirection (120579119903 120601119903) Γ119905 and Γ119903 are the reflection coefficients ofthe transmit and receive antennas respectively and 119886119905 and119886119903 are the polarization vectors of the transmit and receiveantennas respectively taken in the appropriate directions

120572 is the absorption coefficient of the intervening mediumHowever these additional factors will contribute to reductionof antenna received power Thus in the line of sight directionthe received power will become higher compared to non-lineof sight direction as experimented in this section

42 Power Level versus Range Measurement The power levelversus range measurement is done for distance between 5mand 100m indoorswith 5m incrementThedistance is limiteddue to the buildingrsquos built-up area which is limited to 100-meter lengthThe response is RSS values which are measuredfor existing and proposed RFID system at two levels of powerwhich are 10 dBm (max) and 2 dBm (min) At each point10 single values of receiving power are collected to calculate

10 International Journal of Distributed Sensor Networks

210

Range

090

Angle

Polarization

Type of RFID

Interactions plot

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

2 10

0 90

Range

Angle

Polarization

Type of RFID

E

E

H

H

N

P

N P

Figure 9 Interaction plot for radiation measurement

Range

Mea

nM

ean

Angle

Polarization Type of RFID

Main effects plot for RSSIData means

minus36

minus39

minus42

minus45

minus48

minus36

minus39

minus42

minus45

minus48

2 10 0 90

E H N P

Figure 10 Main effects plot for radiation measurement

the mean Figures 11 and 12 show the comparison betweenRSS values of existing and proposed RFID system based ondistance of 5 meters to 100 meters indoors at 10 dBm and2 dBm power levels The graph shows that the RSS valuesare decreased when the distance is increased at both powerlevels The existing RFID system gives better RSS valuesthan proposed RFID system at 10 dBm power level while the

proposed RFID system gives better RSS values than existingRFID system at 2 dBm power level To compare the perfor-mance of two types of RFID system based on power leveland range the third experiment is conducted based on DOEmethod with three factors having two levels of treatment 23The factors are referred to as type of RFID (nonembeddedembedded) power level (2 dBm 10 dBm) and range (5m

International Journal of Distributed Sensor Networks 11

00 20 40 60 80 100 120

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

RSSI

(dBm

)

Figure 11 Received signal strength values at 10 dBm in an indoorenvironment

00 20 40 60 80 100 120

RSSI

(dBm

)

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

minus80

Figure 12 Received signal strength values at 2 dBm in an indoorenvironment

10m) The response of the experiment is RSS and 2 replicatesare used in this work to reduce the measurement errorsand noise The order in which the observations are takenis selected randomly so that the design is a completelyrandomized design Later the statistical method ANOVA isperformed using Minitab 162 and the results are displayedin Tables 4 and 5 for effect estimate summary and ANOVAfor power level versus range measurement From the effectestimate summary presented in Table 4 the main effect of Creally dominates the process accounting for 91381 percentof the total variability whereas the main effect of A and theAC interaction account for about 6144 and 1727 percentrespectively thus the DOE for this experiment also provedthat the RSS value can be used as a prediction method tomeasure distance and location of tags from reader in wireless

sensor network platform since the range contributions arehighly significant compared to other factors The ANOVAin Table 4 is used to confirm the magnitude of these effectsFrom Table 5 the main effects of A B and C are highlysignificant since the factors have very small119875 values (lt0001)For 2-way interaction the AB and AC interaction are alsohighly significant (low 119875 values) thus it can be concludedthat there is a strong interaction between two factors withRSS values in this experimental study Figures 13 and 14 showthe interaction and effects plot for power level versus rangemeasurement that is used to interpret the results from Tables4 and 5 It can be seen that theRSS for embeddedRFID systemis nearly equal to nonembedded RFID system at 10 dBmpower level However at 2 dBm power level the RSS forembedded RFID system is better than nonembedded RFIDsystem In terms of range the RSS for embeddedRFID systemis better compared to nonembedded RFID system at both 5-meter and 100-meter range Based on the results obtainedit can be summarized that for indoor tracking purposesthe low power level gives better RSS values for embeddedRFID system compared to nonembedded RFID system Thisfinding can be supported by [29] which stated that the lowtransmit power is appropriate and sufficient for area withlimited range however if the high transmit power is use inthe area with limited range it will resulting in disturbancethat will cause degradation in radiation performanceThus todesign an efficient indoor localization system these findingsmust be taken into account as a guideline implementing thefull setup for real time location monitoring system in WSNplatform

5 Battery Life Testing

A single RFID tag requires a minimum of 27 V to 33 Vvoltage to operate as a transmitter and receiver in thenetwork Howver the embedded RFID tag proposed in thiswork requires minimum of 50V to power up the embeddeddevices Thus the embedded RFID tag can be powered upusing 9 VDC power supply if the tag is located at fixedlocation or Li-Po rechargeable batteries can be used if thetag is mobile The tag has been programmed to consumethe lowest amount of energy as much as possible sinceit introduced an automated switching mechanism betweenindoor and outdoor location tracking and operating in sleepmode at certain timeTheRFID tag only sends data if the validoutdoor location is detected by the GPS receiver however ifthe embedded RFID tag is located inside the building or theplace where the signal from satellite cannot be detected bythe GPS receiver it will turn to sleep mode and wait for theRFID reader interrogation every hourTheGSMmodemonlyworks when the embedded RFID tag is out of the wirelessnetwork coverage to consume the lowest amount of energy

The battery lifetime measurement is performed for threesystems which are stand-alone RFID tag (A) [20] embeddedRFID tag with switching mechanism (B) and embeddedRFID tag without switching mechanism (C) in an indoorenvironment (inside building) where the GPS signals are notavailable most of the time The results show that the battery

12 International Journal of Distributed Sensor Networks

Table 4 Effect estimate summary

Term constant Effect Sum of squares Percent contributionPower (119860) 724 209530 6144Type of RFID (119861) 179 12780 0375Range (119862) minus2791 3116430 91381Power lowast type of RFID (119860119861) minus141 7980 0234Power lowast range (119860119862) 384 58910 1727Type RFID lowast range (119861119862) 074 2180 0064Power lowast type of RFID lowast range (119860119861119862) minus056 1270 00372

Table 5 ANOVA for power level versus range measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Power (119860) 1 209530 209530 1294370 0000Type of RFID (119861) 1 12780 12780 78950 0000Range (119862) 1 3116430 3116430 19252080 0000Power lowast type of RFID (119860119861) 1 7980 7980 49300 0000Power lowast range (119860119862) 1 58910 58910 363900 0000Type RFID lowast range (119861119862) 1 2180 2180 13440 0006Power lowast type of RFID lowast range (119860119861119862) 1 1270 1270 7820 0023Residual error 8 1290 0160Pure error 8 1300 0160Total 15 3410360

for the tag drops from 824V to 711 V in the duration of 85hours for the proposed embedded RFID tag with switchingmechanism compared to other systems as shown in Figure 15

The results for system B are much better than systemC due to the switching introduced However the result forsystem A is better than system B due to the nonembeddedsystem According to these real implementation results thetag will be expected to stop operating approximately in24 hours when the battery reaches the minimum voltagerequired which is 50 V However this result does not includethe situation where the proposed embedded RFID tag islocated outside building which might increase the energyusage The estimation results also did not include otherexternal factors that might reduce the battery life Thus thetag is estimated to cover the operation for a day withouthaving to recharge the battery until the next day To reducethe energy consumption and maximize the time of usage anenergy harvesting technologywill be implemented later in theembedded RFID tag as our next target

6 Conclusion

This paper describes the implementation of an automatedswitching algorithm in the design of a 245GHz active inte-grated RFID system for indoor and outdoor location trackingutilizing WSN platform Four experimental measurementsbased on DOE method are performed in order to validatethe performance of the proposed RFID system as well as tostudy the relationship interaction and effects of each factor

to RSS values extracted from existing and proposed RFIDsystem The results obtained from first experiment show thatembedded RFID system does affect the radiation patternnot more than 15 differences from existing RFID system Itshows that the proposed RFID system contributes to strongerradiation compared to existing RFID system This may bedue to the embedded design and improvement of algorithmimplemented in the proposed RFID system which increasedthe power stability of the system which thus provided highersignal strength [28]

In the second experiment the results show that the pro-posed RFID tag has better RSS value when it is located nearerto the reader at 0∘ degrees direction of propagation with H-field polarization of antenna In the third experiment theexisting RFID system gives better RSS values than proposedRFID system at 10 dBm power level while the proposedRFID system gives better RSS values than existing RFIDsystem at 2 dBm power level In the fourth experiment it canbe summarized that for indoor tracking purposes the lowpower level gives better RSS values for the proposed RFIDtag compared to the existing RFID system The statisticalanalysis ANOVA proved that the RSS values are significantwith range thus it can be concluded that the RSS can beused as a prediction method to evaluate distance betweenRFID reader and RFID tag in WSN platform Based onthe experimental study and statistical analysis done for allsituations a conclusion has been made where the proposedRFID system with automated switching algorithm is betterthan the existing RFID system in terms of performance andvariability of the factors The battery life testing also gives

International Journal of Distributed Sensor Networks 13

Power

Range

Interactions plot

2 10 5 100

minus30

minus45

minus60

minus30

minus45

minus60

minus30

minus45

minus60

Range

Power

Type of RFID

210

5100

Type of RFID

E

N

EN

Figure 13 Interaction plot for power level versus range measurement

Power

Mea

nM

ean

Type RFID

Range

Main effects plot for RSSI

minus40

minus50

minus60

minus40

minus50

minus60

2 10

5 100

EN

Figure 14 Main effects plot for power level versus range measurement

significant findings where the proposed embedded RFIDtag with switching mechanism introduced is much betterthan embedded RFID tag without switching mechanismTo increase the performance of the embedded RFID tag interms of time of usage an energy harvesting technology willbe introduced later as a part of our next contribution Interms of hardware costing the proposed approach is moreacceptable compared to the system developed by [6] since thetechnology used in this work especially GSM communicationis much cheaper than satellite communication used in the

existing system [6] In addition this proposed system doesnot totally depend on GSM since the communication willbe done by RFID which is free as long as the embeddedRFID tag is working in the control area of WSN platformHowever the existing work [6] did not support long rangecommunication and thus when the tag is moving far fromthe reader the satellite communication needs to be used andthis will lead to high cost of communication As a conclusionthe proposed embedded RFID tag provided better signalpropagation compared to the existing RFID tag [19] and

14 International Journal of Distributed Sensor Networks

00 1 2 3 4 5 6 7 8 9

Batte

ry re

duct

ion

()

Time (hours)

Battery lifetime testing

Embedded RFID tag with switchingEmbedded RFID tag without switchingStandalone RFID tag

minus5

minus10

minus15

minus20

minus25

minus30

Figure 15 Battery life testing based on three types of RFIDdesigned

the battery lifetime for embedded RFID tag with switchingmechanism is better than that without switchingmechanism

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank Malaysia Ministry of HigherEducation (LRGS Fund) for sponsoring the research anddeveloping this project Also special appreciation is due toProfessor DrMohamadKamal Abdul Rahim fromUniversitiTeknologiMalaysia Head of the LRGS Project and AssociateProfessor Dr Alyani Ismail Universiti PutraMalaysia for thesupport

References

[1] DIGI XBeeXBee PRO ZB RF Module Datasheet 2013[2] J B Howard ldquoCombination RFID and GPS functionality on

intelligent labelrdquo WO 2003050960 A2 2003[3] S Chand V R Bapat K H Hall et al ldquoPortable RFID reader

having a location determination systemrdquo EP 1752908 A220062006

[4] G O Garsia S I M Nunez L A Saavedra and M D I TapialdquoReal time location system (RTLS) with ZigBee technologyrdquo EP2196816 A12008 2008

[5] A Bandyopadhyay D Hakim B E Funk E A Kohn C AToelis and G Blankenship ldquoSystem and method for locatingtracking andor monitoring the status of personnel andorassets both indoorsrdquo US 20090043504 A1 2009

[6] RFID Journal ldquoHybrid tags includes active RFID GPS satelliteand sensorrdquo RFID Journal 2009 httpwwwrfidjournalcomarticlesview4635

[7] M Mohandes M Haleem M Deriche and K BalakrishnanldquoWireless sensor networks for pilgrims trackingrdquo IEEE Embed-ded Systems Letters vol 4 no 4 pp 106ndash109 2012

[8] N D Simoes J L Goncalves M L Caeiro M J Boavida andF D Cardoso ldquoZigBeeGPS tracking system for rowing racesrdquoin Proceedings of the International Conference on Computer as aTool (EUROCON rsquo11) Lisbon Portugal April 2011

[9] Y Xu R Jiang S Yan and D Xiong ldquoThe research of safetymonitoring system applied in school bus based on the internetof thingsrdquo Procedia Engineering vol 15 pp 2464ndash2468 2011

[10] L Zheng M Li C Wu et al ldquoDevelopment of a smartmobile farming service systemrdquo Mathematical and ComputerModelling vol 54 no 3-4 pp 1194ndash1203 2011

[11] F A Poad and W Ismail ldquoAutomated switching mechanismfor indoor and outdoor propagation with embedded RFID andGPS in wireless sensor network platformrdquo in Proceedings ofthe International Conference on Wireless Network pp 710ndash714London UK 2014

[12] J OrsquoToole J Tuttle M Tuttle et al ldquoRadio frequency identifi-cation sensor tag apparatusesrdquo US 20080030306 A1 2008

[13] V M Gupta and S Annambhotla ldquoRFID tracker and locatorrdquoUS 20090085745 A1 2009

[14] F Shebli I Dayoub and J M Rouvaen ldquoMinimizing energyconsumptionwithinwireless sensor networkrdquoUbiquitous Com-puting and Communication Journal vol 87 2007

[15] R Ding J Hou and B Xing ldquoResearch of wireless sensornetwork nodes based on ambient energy harvestingrdquo in Pro-ceedings of the 6th IEEE International Conference on IntelligentNetworks and Intelligent Systems (ICINIS rsquo13) pp 286ndash288Shenyang China November 2013

[16] A Cammarano C Petrioli andD Spenza ldquoPro-Energy a novelenergy prediction model for solar and wind energy-harvestingwireless sensor networksrdquo in Proceedings of the 9th IEEEInternational Conference on Mobile Ad-Hoc and Sensor Systems(MASS rsquo12) pp 75ndash83 Las Vegas Nev USA October 2012

[17] J Song and Y K Tan ldquoEnergy consumption analysis of ZigBee-based energy harvesting wireless sensor networksrdquo in Proceed-ings of the IEEE International Conference on CommunicationSystems (ICCS rsquo12) pp 468ndash472 November 2012

[18] San Jose Technology Inc Datasheet GPS Engine Board ModelFV M8 2014 httpswwwsparkfuncomdatasheetsGPSFV-M8 Specpdf

[19] W Ismail and R Abdulla ldquoPortable RFID Reader for RTLS(Real Time Location Systems)rdquo PCT International Filing NoPCTMY2011000167 Country Australia Pub No WO2012102600 2012

[20] W Ismail C Z Zulkifli and M G Rahman ldquoWireless pro-duction monitoring systemrdquo PCT International Filing PCTMY2013000037 2013

[21] ISOIEC 18000-4 Information Technology-Radio FrequencyIdentification for ItemManagement-Part7 Parameters for ActiveAir Interface Communications at 245 GHz ISO 18000-42008International Organization for Standardization GenevaSwitzerland 2008

[22] D-H Shih P-L Sun D C Yen and S-M Huang ldquoTaxonomyand survey of RFID anti-collision protocolsrdquo Computer Com-munications vol 29 no 11 pp 2150ndash2166 2006

[23] B Nilsson L Bengtsson P-A Wiberg and B SvenssonldquoProtocols for active RFIDmdashthe energy consumption aspectrdquo inProceedings of the IEEE 2nd International Symposium on Indus-trial Embedded Systems (SIES rsquo07) pp 41ndash48 July 2007

[24] H L van Eeden ldquoWireless Communication Systemrdquo US Patent0045923 2009

International Journal of Distributed Sensor Networks 15

[25] H Haar K Bischof and H Scharke ldquoRFID transponderrdquo WOPatent 109212 2006

[26] C A Balanis AntennaTheory Analysis and Design JohnWileyamp Sons New York NY USA 3rd edition 2005

[27] P Sorrells Optimizing Read Range in RFID System MicrochipTechnology Inc 2000 httpedncomdesignwireless-network-ing4341980Optimizing-read-range-in-RFID-systems

[28] A Ammu L Mapa and A H Jayatissa ldquoEffect of factors onRFID tag readability-statistical analysisrdquo in Proceedings of theIEEE International Conference on ElectroInformation Technol-ogy (EIT rsquo09) pp 355ndash358 June 2009

[29] R Zurawski Embedded System Handbook Second EditionNetwork Embedded Systems CRC Press 2009

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

Page 6: Research Article Propagation Analysis for Automated ...downloads.hindawi.com/journals/ijdsn/2015/392385.pdf · Hybrid tag includes active RFID, GPS, satellite, and sensors [ ] MHz

6 International Journal of Distributed Sensor Networks

Start

Initialization

Send collection command

Received ACK

Receive packet(1) Receive GPS data (2) Indoor location data (3) ACK packet

Packet received

Transmit ACK signal

Time out

Time out

Transmit mode

Transmitcommand to

RFID tag

Send to host computer via

RS-232

Receive mode

No

Yes

No

Yes

No

No

Received interrupt from

RS232

Receive interrupt from RTC alarm

Yes

Yes

Receive ACK

Disable interrupt

Yes

No

(a)

Collection command

Start

Initialization

Cyclic sleepmode

Receive GPSsignal

(UART)

Receive UARTinterrupt

GPS data valid

Transmit mode

Valid

Receive mode

Transmit ACK signal

Send ID data

Receive ACK

Time out

Time out

RetrieveRSSI

Retrievetag ID

Transmitmode

Send ID andlocation data

ReceiveACK

Receive RTCalarm interrupt

Networkcoverage

Disableinterruption

Send ID andlocation data

via GSM

Yes

No

No

Yes

Yes

No

No

Yes

No

Yes

NoNo

Yes

Yes

(b)

Figure 5 (a) The flow chart of algorithm implemented in the proposed RFID reader (b) The flow chart of algorithm implemented in theproposed RFID tag

Hostterminal

(PC) RS232 cable

RFIDreader

Antenna

RFID tag

RFID reader

Antenna

180∘

180∘

180∘

180∘

270∘

270∘

0∘360

0∘360

Distance d

H

E-field polarization

H-field polarization

Figure 6 Illustration of polarization measurements

level that provides the best variation of RSS for the proposedRFID system to operate in certain range of indoor locationThe fourth experiment is designed based on DOE methodto investigate the interactions and effects between powerlevel range and RSS for both existing and proposed RFIDsystem All experiments conducted in this work used the

same experimental setup as shown in Figure 6 There aretwo test beds involved for all experiments First test bed isfor nonembedded RFID system (stand-alone system withoutGPS) which consists of existing RFID reader and RFID tag[19] while the second test bed is for the embedded RFIDsystem which consists of a proposed active RFID reader and

International Journal of Distributed Sensor Networks 7

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(a)

NonembeddedEmbedded

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

(b)

Figure 7 H-plane pattern at (a) 2m and (b) 10m distance

RFID tag embedded with GPS The reader and tag are facingeach other at a specific distance which later communicates ina single hop ZigBee network

41 Radiation Pattern Measurement The radiation patternmeasurements are done for both existing RFID system [19]and proposed RFID system to evaluate and compare theperformance of the developed system based on automatedswitching algorithm presented at two different distanceswhich are 2m and 10m The far-field distance is obtained byusing (1) [26] based on characteristic of the antenna used inthis research work Consider

119877119891119891 gt21198712

120582 (1)

where 119877119891119891 is the distance between transmit and receiveantenna 119871 is an antenna aperture under test (825 cm) and120582 is the wavelength of transmission (1205824) indicating as120582 = 119862119891 where 119862 is the speed of light 3 times 108msminus2 andthe operating frequency 119891 = 245GHz Therefore 119877119891119891 isgreater than 042m when the antenna aperture value is notmore than 01m Hence 2m and 10mmeasurement distancesare in the far-field region The algorithm requires a readerto send information to tag and the tag will acknowledgethe packet received from reader and gives information thatis requested by the reader The measurements of RSS areperformed using whip antenna for both RFID reader and tagFor existing RFID system the RSS value is obtained from theX-CTU terminal by using ATDB command However for theproposed RFID system the RSS are retrieved from the algo-rithm implemented in the RFID system The power level of

the wireless communication module has been programmedto maximum value with the output power +10 dBm Theorientation of the antenna for H-field polarization is verticalwhile for E-field polarization it is horizontal as shown inFigure 6 The reader antenna is rotated in 119911-direction from0∘ to 360∘ degrees with 15∘ degree increment steps

Figures 7(a) and 7(b) show the comparison of radiationpatterns of existing RFID system and proposed RFID systemfor H-plane polarization at 2m and 10m distances respec-tively The result for H-field radiation pattern at 2m distanceis an omnidirectional It can be seen that the normalizedradiation pattern for proposed RFID system is slightly betterthan the existing RFID system in between 90∘ and 195∘degrees However the radiation pattern results for existingRFID system at 10m distance are better than proposedRFID system by about 09 dBm to 64 dBm differences Thedifferences in antenna pattern may be due to the immediatesurroundings since actual performance depends on manyfactors in the environment Obstructions in the propagationpath or other wireless networks or systems will also affectthe performance Figures 8(a) and 8(b) show the comparisonof radiation patterns for E-plane polarization at 2m and10mdistances respectivelyThenormalized radiation patternfor proposed RFID system is better than the existing RFIDsystem at 2m and 10m distances Based on these initialfindings it can be concluded that the radiation for shorterdistance is stronger than longer distance in both H-planeand E-plane and the embedded mechanism for RFID systemdoes affect the radiation pattern notmore than 15 differencefrom existing RFID system It shows that the proposed RFIDsystem contributes to stronger radiation compared to existingRFID system This may be due to the embedded power

8 International Journal of Distributed Sensor Networks

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(a)

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(b)

Figure 8 E-plane pattern at (a) 2m and (b) 10m distance

management that is implemented in the embedded activeRFID tag that provided power stability as mentioned by [27]thus increasing the performance of the system in the network

The second experiment is the continuity of the radiationpattern measurement which involved four factors with twolevels of treatment 24 The factors are referred to as type ofRFID (nonembedded embedded) power level (2 dBm10 dBm) range (5m 10m) polarization (E H) and angleof polarization (0∘ 90∘) The response of the experiment isin RSS and 2 replicates are used in this work to reduce themeasurement errors and noiseThe order in which the obser-vations are taken is selected randomly so that the design is acompletely randomized design To identify the interactionsand effects between type of RFID range polarization andangle of propagation with RSS a statistical method calledanalysis of variance (ANOVA) is used as presented in [28]and the effect estimate summaries for the proposed systemare shown in Table 2

In this work the main effect of A really dominates theprocess accounting for 63635 percent of the total variabilitywhereas themain effect of C accounts for about 7328 percentwhile the AB and BC interactions account for about 8808percent and 6616 percent respectively thus it is proved thatthe RSS value can be used as a prediction method to measuredistance and location of tags from reader in WSN platformsince the range contributes highly significantly compared toother factors The ANOVA in Table 3 is used to confirm themagnitude of these effects From Table 3 it can be seen thatthe main effects of A and C are highly significant since bothof the factors have very small 119875 values (lt0001) For 4-wayinteraction the ABCD interactions are also highly significant(low 119875 values) thus it can be concluded that there is a stronginteraction between four factors with RSS values

Figures 9 and 10 show the interactions and main effectsplot for radiation measurement that is used to interpret theresults from Tables 1 and 2

The graph shows that the RSS for existing RFID system isbetter than proposed RFID system at 10m distance howeverthe RSS for proposed RFID system is better than the existingRFID system at 2m distance In terms of angle the RSS for0∘ and 90∘ degree proposed RFID system are better thanexisting RFID system however the RSS value for 0∘ degreesis better than 90∘ degrees for both RFID system In terms ofpolarization the H-field polarization gives more significantvalues of RSS compared to E-field polarization for both RFIDsystems The findings show that the RFID system will havebetter RSS value if it is located nearer to the reader at 0∘degrees direction of propagation with H-field polarizationof antenna This statement is supported by Friis free spaceequation as shown in [26]

119875119903 (119889) =119875119905119866119905119866119903120582

2

(4120587)21198892119871 (2)

where119875119905 is the transmitted power119875119903(119889) is the received power119866119905 is the transmitter antenna gain 119866119903 is the receiver antennagain 119889 is the transmit and receive separation and 119871 is the sys-tem loss factor dependent upon line attenuation filter lossesand antenna losses and not related to propagation Accordingto Friis equation the electromagnetic (EM) signals whentraveling through wireless channels experience fading effectsdue to various effects but for ideal case the transmission thatis considered with a direct line of sight will give the bestvalue of received power The effects of impedance mismatchmisalignment of the antenna pointing and polarization and

International Journal of Distributed Sensor Networks 9

Table 2 Effect estimate summaries

Term constant Effect Sum of squares Percentage contributionRange (119860) minus1109 983460 63635Angle (119861) minus079 4960 0321Polarization (119862) 376 113250 7328Type of RFID (119863) 160 20480 1325Range lowast angle (119860119861) minus358 102250 6616Range lowast polarization (119860119862) minus073 4210 0272Range lowast type of RFID (119860119863) minus226 40950 2650Angle lowast polarization (119861119862) minus412 136120 8808Angle lowast type of RFID (119861119863) 034 0910 0058Polarization lowast type of RFID (119862119863) minus186 27750 1796Range lowast angle lowast polarization (119860119861119862) minus124 12250 0793Range lowast angle lowast type of RFID (119860119861119863) 140 15680 1015Range lowast polarization lowast type of RFID (119860119862119863) minus075 4500 0291Angle lowast polarization lowast type of RFID (119861119862119863) minus100 8000 0518Range lowast angle lowast polarization lowast type of RFID (119860119861119862119863) minus206 34030 2202

Table 3 ANOVA for the radiation pattern measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Range (119860) 1 983460 983461 429340 0000Angle (119861) 1 4960 4961 2170 0160Polarization (119862) 1 113250 113251 49440 0000Type of RFID (119863) 1 20480 20480 8940 0009Range lowast angle (119860119861) 1 102250 102245 44640 0000Range lowast polarization (119860119862) 1 4210 4205 1840 0194Range lowast type of RFID (119860119863) 1 40950 40951 17880 0001Angle lowast polarization (119861119862) 1 136120 136125 59430 0000Angle lowast type of RFID (119861119863) 1 0910 0911 0400 0537Polarization lowast type of RFID (119862119863) 1 27750 27751 12120 0003Range lowast angle lowast polarization (119860119861119862) 1 12250 12251 5350 0034Range lowast angle lowast type of RFID (119860119861119863) 1 15680 15680 6850 0019Range lowast polarization lowast type RFID (119860119862119863) 1 4500 4500 1960 0180Angle lowast polarization lowast type of RFID (119861119862119863) 1 8000 8000 3490 0080Range lowast angle lowast polarization lowast type of RFID (119860119861119862119863) 1 34030 34031 14860 0001Residual error 16 36650 2291Pure error 16 36650 2291Total 31 1545450

absorption can be included by adding additional factors asgiven in [26]

119875119903

119875119905

= 119866119905 (120579119905 120601119905) 119866119903 (120579119903 120601119903) (120582

4120587119877)

2

times (1 minus1003816100381610038161003816Γ1199051003816100381610038161003816

2) (1 minus1003816100381610038161003816Γ1199031003816100381610038161003816

2)1003816100381610038161003816119886119905 sdot 1198861199031003816100381610038161003816

2119890minus120572119877

(3)

where 119866119905(120579119905 120601119905) is the gain of the transmit antenna in thedirection 119866119903(120579119903 120601119903) is the gain of the receive antenna in thedirection (120579119903 120601119903) Γ119905 and Γ119903 are the reflection coefficients ofthe transmit and receive antennas respectively and 119886119905 and119886119903 are the polarization vectors of the transmit and receiveantennas respectively taken in the appropriate directions

120572 is the absorption coefficient of the intervening mediumHowever these additional factors will contribute to reductionof antenna received power Thus in the line of sight directionthe received power will become higher compared to non-lineof sight direction as experimented in this section

42 Power Level versus Range Measurement The power levelversus range measurement is done for distance between 5mand 100m indoorswith 5m incrementThedistance is limiteddue to the buildingrsquos built-up area which is limited to 100-meter lengthThe response is RSS values which are measuredfor existing and proposed RFID system at two levels of powerwhich are 10 dBm (max) and 2 dBm (min) At each point10 single values of receiving power are collected to calculate

10 International Journal of Distributed Sensor Networks

210

Range

090

Angle

Polarization

Type of RFID

Interactions plot

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

2 10

0 90

Range

Angle

Polarization

Type of RFID

E

E

H

H

N

P

N P

Figure 9 Interaction plot for radiation measurement

Range

Mea

nM

ean

Angle

Polarization Type of RFID

Main effects plot for RSSIData means

minus36

minus39

minus42

minus45

minus48

minus36

minus39

minus42

minus45

minus48

2 10 0 90

E H N P

Figure 10 Main effects plot for radiation measurement

the mean Figures 11 and 12 show the comparison betweenRSS values of existing and proposed RFID system based ondistance of 5 meters to 100 meters indoors at 10 dBm and2 dBm power levels The graph shows that the RSS valuesare decreased when the distance is increased at both powerlevels The existing RFID system gives better RSS valuesthan proposed RFID system at 10 dBm power level while the

proposed RFID system gives better RSS values than existingRFID system at 2 dBm power level To compare the perfor-mance of two types of RFID system based on power leveland range the third experiment is conducted based on DOEmethod with three factors having two levels of treatment 23The factors are referred to as type of RFID (nonembeddedembedded) power level (2 dBm 10 dBm) and range (5m

International Journal of Distributed Sensor Networks 11

00 20 40 60 80 100 120

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

RSSI

(dBm

)

Figure 11 Received signal strength values at 10 dBm in an indoorenvironment

00 20 40 60 80 100 120

RSSI

(dBm

)

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

minus80

Figure 12 Received signal strength values at 2 dBm in an indoorenvironment

10m) The response of the experiment is RSS and 2 replicatesare used in this work to reduce the measurement errorsand noise The order in which the observations are takenis selected randomly so that the design is a completelyrandomized design Later the statistical method ANOVA isperformed using Minitab 162 and the results are displayedin Tables 4 and 5 for effect estimate summary and ANOVAfor power level versus range measurement From the effectestimate summary presented in Table 4 the main effect of Creally dominates the process accounting for 91381 percentof the total variability whereas the main effect of A and theAC interaction account for about 6144 and 1727 percentrespectively thus the DOE for this experiment also provedthat the RSS value can be used as a prediction method tomeasure distance and location of tags from reader in wireless

sensor network platform since the range contributions arehighly significant compared to other factors The ANOVAin Table 4 is used to confirm the magnitude of these effectsFrom Table 5 the main effects of A B and C are highlysignificant since the factors have very small119875 values (lt0001)For 2-way interaction the AB and AC interaction are alsohighly significant (low 119875 values) thus it can be concludedthat there is a strong interaction between two factors withRSS values in this experimental study Figures 13 and 14 showthe interaction and effects plot for power level versus rangemeasurement that is used to interpret the results from Tables4 and 5 It can be seen that theRSS for embeddedRFID systemis nearly equal to nonembedded RFID system at 10 dBmpower level However at 2 dBm power level the RSS forembedded RFID system is better than nonembedded RFIDsystem In terms of range the RSS for embeddedRFID systemis better compared to nonembedded RFID system at both 5-meter and 100-meter range Based on the results obtainedit can be summarized that for indoor tracking purposesthe low power level gives better RSS values for embeddedRFID system compared to nonembedded RFID system Thisfinding can be supported by [29] which stated that the lowtransmit power is appropriate and sufficient for area withlimited range however if the high transmit power is use inthe area with limited range it will resulting in disturbancethat will cause degradation in radiation performanceThus todesign an efficient indoor localization system these findingsmust be taken into account as a guideline implementing thefull setup for real time location monitoring system in WSNplatform

5 Battery Life Testing

A single RFID tag requires a minimum of 27 V to 33 Vvoltage to operate as a transmitter and receiver in thenetwork Howver the embedded RFID tag proposed in thiswork requires minimum of 50V to power up the embeddeddevices Thus the embedded RFID tag can be powered upusing 9 VDC power supply if the tag is located at fixedlocation or Li-Po rechargeable batteries can be used if thetag is mobile The tag has been programmed to consumethe lowest amount of energy as much as possible sinceit introduced an automated switching mechanism betweenindoor and outdoor location tracking and operating in sleepmode at certain timeTheRFID tag only sends data if the validoutdoor location is detected by the GPS receiver however ifthe embedded RFID tag is located inside the building or theplace where the signal from satellite cannot be detected bythe GPS receiver it will turn to sleep mode and wait for theRFID reader interrogation every hourTheGSMmodemonlyworks when the embedded RFID tag is out of the wirelessnetwork coverage to consume the lowest amount of energy

The battery lifetime measurement is performed for threesystems which are stand-alone RFID tag (A) [20] embeddedRFID tag with switching mechanism (B) and embeddedRFID tag without switching mechanism (C) in an indoorenvironment (inside building) where the GPS signals are notavailable most of the time The results show that the battery

12 International Journal of Distributed Sensor Networks

Table 4 Effect estimate summary

Term constant Effect Sum of squares Percent contributionPower (119860) 724 209530 6144Type of RFID (119861) 179 12780 0375Range (119862) minus2791 3116430 91381Power lowast type of RFID (119860119861) minus141 7980 0234Power lowast range (119860119862) 384 58910 1727Type RFID lowast range (119861119862) 074 2180 0064Power lowast type of RFID lowast range (119860119861119862) minus056 1270 00372

Table 5 ANOVA for power level versus range measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Power (119860) 1 209530 209530 1294370 0000Type of RFID (119861) 1 12780 12780 78950 0000Range (119862) 1 3116430 3116430 19252080 0000Power lowast type of RFID (119860119861) 1 7980 7980 49300 0000Power lowast range (119860119862) 1 58910 58910 363900 0000Type RFID lowast range (119861119862) 1 2180 2180 13440 0006Power lowast type of RFID lowast range (119860119861119862) 1 1270 1270 7820 0023Residual error 8 1290 0160Pure error 8 1300 0160Total 15 3410360

for the tag drops from 824V to 711 V in the duration of 85hours for the proposed embedded RFID tag with switchingmechanism compared to other systems as shown in Figure 15

The results for system B are much better than systemC due to the switching introduced However the result forsystem A is better than system B due to the nonembeddedsystem According to these real implementation results thetag will be expected to stop operating approximately in24 hours when the battery reaches the minimum voltagerequired which is 50 V However this result does not includethe situation where the proposed embedded RFID tag islocated outside building which might increase the energyusage The estimation results also did not include otherexternal factors that might reduce the battery life Thus thetag is estimated to cover the operation for a day withouthaving to recharge the battery until the next day To reducethe energy consumption and maximize the time of usage anenergy harvesting technologywill be implemented later in theembedded RFID tag as our next target

6 Conclusion

This paper describes the implementation of an automatedswitching algorithm in the design of a 245GHz active inte-grated RFID system for indoor and outdoor location trackingutilizing WSN platform Four experimental measurementsbased on DOE method are performed in order to validatethe performance of the proposed RFID system as well as tostudy the relationship interaction and effects of each factor

to RSS values extracted from existing and proposed RFIDsystem The results obtained from first experiment show thatembedded RFID system does affect the radiation patternnot more than 15 differences from existing RFID system Itshows that the proposed RFID system contributes to strongerradiation compared to existing RFID system This may bedue to the embedded design and improvement of algorithmimplemented in the proposed RFID system which increasedthe power stability of the system which thus provided highersignal strength [28]

In the second experiment the results show that the pro-posed RFID tag has better RSS value when it is located nearerto the reader at 0∘ degrees direction of propagation with H-field polarization of antenna In the third experiment theexisting RFID system gives better RSS values than proposedRFID system at 10 dBm power level while the proposedRFID system gives better RSS values than existing RFIDsystem at 2 dBm power level In the fourth experiment it canbe summarized that for indoor tracking purposes the lowpower level gives better RSS values for the proposed RFIDtag compared to the existing RFID system The statisticalanalysis ANOVA proved that the RSS values are significantwith range thus it can be concluded that the RSS can beused as a prediction method to evaluate distance betweenRFID reader and RFID tag in WSN platform Based onthe experimental study and statistical analysis done for allsituations a conclusion has been made where the proposedRFID system with automated switching algorithm is betterthan the existing RFID system in terms of performance andvariability of the factors The battery life testing also gives

International Journal of Distributed Sensor Networks 13

Power

Range

Interactions plot

2 10 5 100

minus30

minus45

minus60

minus30

minus45

minus60

minus30

minus45

minus60

Range

Power

Type of RFID

210

5100

Type of RFID

E

N

EN

Figure 13 Interaction plot for power level versus range measurement

Power

Mea

nM

ean

Type RFID

Range

Main effects plot for RSSI

minus40

minus50

minus60

minus40

minus50

minus60

2 10

5 100

EN

Figure 14 Main effects plot for power level versus range measurement

significant findings where the proposed embedded RFIDtag with switching mechanism introduced is much betterthan embedded RFID tag without switching mechanismTo increase the performance of the embedded RFID tag interms of time of usage an energy harvesting technology willbe introduced later as a part of our next contribution Interms of hardware costing the proposed approach is moreacceptable compared to the system developed by [6] since thetechnology used in this work especially GSM communicationis much cheaper than satellite communication used in the

existing system [6] In addition this proposed system doesnot totally depend on GSM since the communication willbe done by RFID which is free as long as the embeddedRFID tag is working in the control area of WSN platformHowever the existing work [6] did not support long rangecommunication and thus when the tag is moving far fromthe reader the satellite communication needs to be used andthis will lead to high cost of communication As a conclusionthe proposed embedded RFID tag provided better signalpropagation compared to the existing RFID tag [19] and

14 International Journal of Distributed Sensor Networks

00 1 2 3 4 5 6 7 8 9

Batte

ry re

duct

ion

()

Time (hours)

Battery lifetime testing

Embedded RFID tag with switchingEmbedded RFID tag without switchingStandalone RFID tag

minus5

minus10

minus15

minus20

minus25

minus30

Figure 15 Battery life testing based on three types of RFIDdesigned

the battery lifetime for embedded RFID tag with switchingmechanism is better than that without switchingmechanism

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank Malaysia Ministry of HigherEducation (LRGS Fund) for sponsoring the research anddeveloping this project Also special appreciation is due toProfessor DrMohamadKamal Abdul Rahim fromUniversitiTeknologiMalaysia Head of the LRGS Project and AssociateProfessor Dr Alyani Ismail Universiti PutraMalaysia for thesupport

References

[1] DIGI XBeeXBee PRO ZB RF Module Datasheet 2013[2] J B Howard ldquoCombination RFID and GPS functionality on

intelligent labelrdquo WO 2003050960 A2 2003[3] S Chand V R Bapat K H Hall et al ldquoPortable RFID reader

having a location determination systemrdquo EP 1752908 A220062006

[4] G O Garsia S I M Nunez L A Saavedra and M D I TapialdquoReal time location system (RTLS) with ZigBee technologyrdquo EP2196816 A12008 2008

[5] A Bandyopadhyay D Hakim B E Funk E A Kohn C AToelis and G Blankenship ldquoSystem and method for locatingtracking andor monitoring the status of personnel andorassets both indoorsrdquo US 20090043504 A1 2009

[6] RFID Journal ldquoHybrid tags includes active RFID GPS satelliteand sensorrdquo RFID Journal 2009 httpwwwrfidjournalcomarticlesview4635

[7] M Mohandes M Haleem M Deriche and K BalakrishnanldquoWireless sensor networks for pilgrims trackingrdquo IEEE Embed-ded Systems Letters vol 4 no 4 pp 106ndash109 2012

[8] N D Simoes J L Goncalves M L Caeiro M J Boavida andF D Cardoso ldquoZigBeeGPS tracking system for rowing racesrdquoin Proceedings of the International Conference on Computer as aTool (EUROCON rsquo11) Lisbon Portugal April 2011

[9] Y Xu R Jiang S Yan and D Xiong ldquoThe research of safetymonitoring system applied in school bus based on the internetof thingsrdquo Procedia Engineering vol 15 pp 2464ndash2468 2011

[10] L Zheng M Li C Wu et al ldquoDevelopment of a smartmobile farming service systemrdquo Mathematical and ComputerModelling vol 54 no 3-4 pp 1194ndash1203 2011

[11] F A Poad and W Ismail ldquoAutomated switching mechanismfor indoor and outdoor propagation with embedded RFID andGPS in wireless sensor network platformrdquo in Proceedings ofthe International Conference on Wireless Network pp 710ndash714London UK 2014

[12] J OrsquoToole J Tuttle M Tuttle et al ldquoRadio frequency identifi-cation sensor tag apparatusesrdquo US 20080030306 A1 2008

[13] V M Gupta and S Annambhotla ldquoRFID tracker and locatorrdquoUS 20090085745 A1 2009

[14] F Shebli I Dayoub and J M Rouvaen ldquoMinimizing energyconsumptionwithinwireless sensor networkrdquoUbiquitous Com-puting and Communication Journal vol 87 2007

[15] R Ding J Hou and B Xing ldquoResearch of wireless sensornetwork nodes based on ambient energy harvestingrdquo in Pro-ceedings of the 6th IEEE International Conference on IntelligentNetworks and Intelligent Systems (ICINIS rsquo13) pp 286ndash288Shenyang China November 2013

[16] A Cammarano C Petrioli andD Spenza ldquoPro-Energy a novelenergy prediction model for solar and wind energy-harvestingwireless sensor networksrdquo in Proceedings of the 9th IEEEInternational Conference on Mobile Ad-Hoc and Sensor Systems(MASS rsquo12) pp 75ndash83 Las Vegas Nev USA October 2012

[17] J Song and Y K Tan ldquoEnergy consumption analysis of ZigBee-based energy harvesting wireless sensor networksrdquo in Proceed-ings of the IEEE International Conference on CommunicationSystems (ICCS rsquo12) pp 468ndash472 November 2012

[18] San Jose Technology Inc Datasheet GPS Engine Board ModelFV M8 2014 httpswwwsparkfuncomdatasheetsGPSFV-M8 Specpdf

[19] W Ismail and R Abdulla ldquoPortable RFID Reader for RTLS(Real Time Location Systems)rdquo PCT International Filing NoPCTMY2011000167 Country Australia Pub No WO2012102600 2012

[20] W Ismail C Z Zulkifli and M G Rahman ldquoWireless pro-duction monitoring systemrdquo PCT International Filing PCTMY2013000037 2013

[21] ISOIEC 18000-4 Information Technology-Radio FrequencyIdentification for ItemManagement-Part7 Parameters for ActiveAir Interface Communications at 245 GHz ISO 18000-42008International Organization for Standardization GenevaSwitzerland 2008

[22] D-H Shih P-L Sun D C Yen and S-M Huang ldquoTaxonomyand survey of RFID anti-collision protocolsrdquo Computer Com-munications vol 29 no 11 pp 2150ndash2166 2006

[23] B Nilsson L Bengtsson P-A Wiberg and B SvenssonldquoProtocols for active RFIDmdashthe energy consumption aspectrdquo inProceedings of the IEEE 2nd International Symposium on Indus-trial Embedded Systems (SIES rsquo07) pp 41ndash48 July 2007

[24] H L van Eeden ldquoWireless Communication Systemrdquo US Patent0045923 2009

International Journal of Distributed Sensor Networks 15

[25] H Haar K Bischof and H Scharke ldquoRFID transponderrdquo WOPatent 109212 2006

[26] C A Balanis AntennaTheory Analysis and Design JohnWileyamp Sons New York NY USA 3rd edition 2005

[27] P Sorrells Optimizing Read Range in RFID System MicrochipTechnology Inc 2000 httpedncomdesignwireless-network-ing4341980Optimizing-read-range-in-RFID-systems

[28] A Ammu L Mapa and A H Jayatissa ldquoEffect of factors onRFID tag readability-statistical analysisrdquo in Proceedings of theIEEE International Conference on ElectroInformation Technol-ogy (EIT rsquo09) pp 355ndash358 June 2009

[29] R Zurawski Embedded System Handbook Second EditionNetwork Embedded Systems CRC Press 2009

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

Page 7: Research Article Propagation Analysis for Automated ...downloads.hindawi.com/journals/ijdsn/2015/392385.pdf · Hybrid tag includes active RFID, GPS, satellite, and sensors [ ] MHz

International Journal of Distributed Sensor Networks 7

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(a)

NonembeddedEmbedded

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

(b)

Figure 7 H-plane pattern at (a) 2m and (b) 10m distance

RFID tag embedded with GPS The reader and tag are facingeach other at a specific distance which later communicates ina single hop ZigBee network

41 Radiation Pattern Measurement The radiation patternmeasurements are done for both existing RFID system [19]and proposed RFID system to evaluate and compare theperformance of the developed system based on automatedswitching algorithm presented at two different distanceswhich are 2m and 10m The far-field distance is obtained byusing (1) [26] based on characteristic of the antenna used inthis research work Consider

119877119891119891 gt21198712

120582 (1)

where 119877119891119891 is the distance between transmit and receiveantenna 119871 is an antenna aperture under test (825 cm) and120582 is the wavelength of transmission (1205824) indicating as120582 = 119862119891 where 119862 is the speed of light 3 times 108msminus2 andthe operating frequency 119891 = 245GHz Therefore 119877119891119891 isgreater than 042m when the antenna aperture value is notmore than 01m Hence 2m and 10mmeasurement distancesare in the far-field region The algorithm requires a readerto send information to tag and the tag will acknowledgethe packet received from reader and gives information thatis requested by the reader The measurements of RSS areperformed using whip antenna for both RFID reader and tagFor existing RFID system the RSS value is obtained from theX-CTU terminal by using ATDB command However for theproposed RFID system the RSS are retrieved from the algo-rithm implemented in the RFID system The power level of

the wireless communication module has been programmedto maximum value with the output power +10 dBm Theorientation of the antenna for H-field polarization is verticalwhile for E-field polarization it is horizontal as shown inFigure 6 The reader antenna is rotated in 119911-direction from0∘ to 360∘ degrees with 15∘ degree increment steps

Figures 7(a) and 7(b) show the comparison of radiationpatterns of existing RFID system and proposed RFID systemfor H-plane polarization at 2m and 10m distances respec-tively The result for H-field radiation pattern at 2m distanceis an omnidirectional It can be seen that the normalizedradiation pattern for proposed RFID system is slightly betterthan the existing RFID system in between 90∘ and 195∘degrees However the radiation pattern results for existingRFID system at 10m distance are better than proposedRFID system by about 09 dBm to 64 dBm differences Thedifferences in antenna pattern may be due to the immediatesurroundings since actual performance depends on manyfactors in the environment Obstructions in the propagationpath or other wireless networks or systems will also affectthe performance Figures 8(a) and 8(b) show the comparisonof radiation patterns for E-plane polarization at 2m and10mdistances respectivelyThenormalized radiation patternfor proposed RFID system is better than the existing RFIDsystem at 2m and 10m distances Based on these initialfindings it can be concluded that the radiation for shorterdistance is stronger than longer distance in both H-planeand E-plane and the embedded mechanism for RFID systemdoes affect the radiation pattern notmore than 15 differencefrom existing RFID system It shows that the proposed RFIDsystem contributes to stronger radiation compared to existingRFID system This may be due to the embedded power

8 International Journal of Distributed Sensor Networks

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(a)

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(b)

Figure 8 E-plane pattern at (a) 2m and (b) 10m distance

management that is implemented in the embedded activeRFID tag that provided power stability as mentioned by [27]thus increasing the performance of the system in the network

The second experiment is the continuity of the radiationpattern measurement which involved four factors with twolevels of treatment 24 The factors are referred to as type ofRFID (nonembedded embedded) power level (2 dBm10 dBm) range (5m 10m) polarization (E H) and angleof polarization (0∘ 90∘) The response of the experiment isin RSS and 2 replicates are used in this work to reduce themeasurement errors and noiseThe order in which the obser-vations are taken is selected randomly so that the design is acompletely randomized design To identify the interactionsand effects between type of RFID range polarization andangle of propagation with RSS a statistical method calledanalysis of variance (ANOVA) is used as presented in [28]and the effect estimate summaries for the proposed systemare shown in Table 2

In this work the main effect of A really dominates theprocess accounting for 63635 percent of the total variabilitywhereas themain effect of C accounts for about 7328 percentwhile the AB and BC interactions account for about 8808percent and 6616 percent respectively thus it is proved thatthe RSS value can be used as a prediction method to measuredistance and location of tags from reader in WSN platformsince the range contributes highly significantly compared toother factors The ANOVA in Table 3 is used to confirm themagnitude of these effects From Table 3 it can be seen thatthe main effects of A and C are highly significant since bothof the factors have very small 119875 values (lt0001) For 4-wayinteraction the ABCD interactions are also highly significant(low 119875 values) thus it can be concluded that there is a stronginteraction between four factors with RSS values

Figures 9 and 10 show the interactions and main effectsplot for radiation measurement that is used to interpret theresults from Tables 1 and 2

The graph shows that the RSS for existing RFID system isbetter than proposed RFID system at 10m distance howeverthe RSS for proposed RFID system is better than the existingRFID system at 2m distance In terms of angle the RSS for0∘ and 90∘ degree proposed RFID system are better thanexisting RFID system however the RSS value for 0∘ degreesis better than 90∘ degrees for both RFID system In terms ofpolarization the H-field polarization gives more significantvalues of RSS compared to E-field polarization for both RFIDsystems The findings show that the RFID system will havebetter RSS value if it is located nearer to the reader at 0∘degrees direction of propagation with H-field polarizationof antenna This statement is supported by Friis free spaceequation as shown in [26]

119875119903 (119889) =119875119905119866119905119866119903120582

2

(4120587)21198892119871 (2)

where119875119905 is the transmitted power119875119903(119889) is the received power119866119905 is the transmitter antenna gain 119866119903 is the receiver antennagain 119889 is the transmit and receive separation and 119871 is the sys-tem loss factor dependent upon line attenuation filter lossesand antenna losses and not related to propagation Accordingto Friis equation the electromagnetic (EM) signals whentraveling through wireless channels experience fading effectsdue to various effects but for ideal case the transmission thatis considered with a direct line of sight will give the bestvalue of received power The effects of impedance mismatchmisalignment of the antenna pointing and polarization and

International Journal of Distributed Sensor Networks 9

Table 2 Effect estimate summaries

Term constant Effect Sum of squares Percentage contributionRange (119860) minus1109 983460 63635Angle (119861) minus079 4960 0321Polarization (119862) 376 113250 7328Type of RFID (119863) 160 20480 1325Range lowast angle (119860119861) minus358 102250 6616Range lowast polarization (119860119862) minus073 4210 0272Range lowast type of RFID (119860119863) minus226 40950 2650Angle lowast polarization (119861119862) minus412 136120 8808Angle lowast type of RFID (119861119863) 034 0910 0058Polarization lowast type of RFID (119862119863) minus186 27750 1796Range lowast angle lowast polarization (119860119861119862) minus124 12250 0793Range lowast angle lowast type of RFID (119860119861119863) 140 15680 1015Range lowast polarization lowast type of RFID (119860119862119863) minus075 4500 0291Angle lowast polarization lowast type of RFID (119861119862119863) minus100 8000 0518Range lowast angle lowast polarization lowast type of RFID (119860119861119862119863) minus206 34030 2202

Table 3 ANOVA for the radiation pattern measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Range (119860) 1 983460 983461 429340 0000Angle (119861) 1 4960 4961 2170 0160Polarization (119862) 1 113250 113251 49440 0000Type of RFID (119863) 1 20480 20480 8940 0009Range lowast angle (119860119861) 1 102250 102245 44640 0000Range lowast polarization (119860119862) 1 4210 4205 1840 0194Range lowast type of RFID (119860119863) 1 40950 40951 17880 0001Angle lowast polarization (119861119862) 1 136120 136125 59430 0000Angle lowast type of RFID (119861119863) 1 0910 0911 0400 0537Polarization lowast type of RFID (119862119863) 1 27750 27751 12120 0003Range lowast angle lowast polarization (119860119861119862) 1 12250 12251 5350 0034Range lowast angle lowast type of RFID (119860119861119863) 1 15680 15680 6850 0019Range lowast polarization lowast type RFID (119860119862119863) 1 4500 4500 1960 0180Angle lowast polarization lowast type of RFID (119861119862119863) 1 8000 8000 3490 0080Range lowast angle lowast polarization lowast type of RFID (119860119861119862119863) 1 34030 34031 14860 0001Residual error 16 36650 2291Pure error 16 36650 2291Total 31 1545450

absorption can be included by adding additional factors asgiven in [26]

119875119903

119875119905

= 119866119905 (120579119905 120601119905) 119866119903 (120579119903 120601119903) (120582

4120587119877)

2

times (1 minus1003816100381610038161003816Γ1199051003816100381610038161003816

2) (1 minus1003816100381610038161003816Γ1199031003816100381610038161003816

2)1003816100381610038161003816119886119905 sdot 1198861199031003816100381610038161003816

2119890minus120572119877

(3)

where 119866119905(120579119905 120601119905) is the gain of the transmit antenna in thedirection 119866119903(120579119903 120601119903) is the gain of the receive antenna in thedirection (120579119903 120601119903) Γ119905 and Γ119903 are the reflection coefficients ofthe transmit and receive antennas respectively and 119886119905 and119886119903 are the polarization vectors of the transmit and receiveantennas respectively taken in the appropriate directions

120572 is the absorption coefficient of the intervening mediumHowever these additional factors will contribute to reductionof antenna received power Thus in the line of sight directionthe received power will become higher compared to non-lineof sight direction as experimented in this section

42 Power Level versus Range Measurement The power levelversus range measurement is done for distance between 5mand 100m indoorswith 5m incrementThedistance is limiteddue to the buildingrsquos built-up area which is limited to 100-meter lengthThe response is RSS values which are measuredfor existing and proposed RFID system at two levels of powerwhich are 10 dBm (max) and 2 dBm (min) At each point10 single values of receiving power are collected to calculate

10 International Journal of Distributed Sensor Networks

210

Range

090

Angle

Polarization

Type of RFID

Interactions plot

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

2 10

0 90

Range

Angle

Polarization

Type of RFID

E

E

H

H

N

P

N P

Figure 9 Interaction plot for radiation measurement

Range

Mea

nM

ean

Angle

Polarization Type of RFID

Main effects plot for RSSIData means

minus36

minus39

minus42

minus45

minus48

minus36

minus39

minus42

minus45

minus48

2 10 0 90

E H N P

Figure 10 Main effects plot for radiation measurement

the mean Figures 11 and 12 show the comparison betweenRSS values of existing and proposed RFID system based ondistance of 5 meters to 100 meters indoors at 10 dBm and2 dBm power levels The graph shows that the RSS valuesare decreased when the distance is increased at both powerlevels The existing RFID system gives better RSS valuesthan proposed RFID system at 10 dBm power level while the

proposed RFID system gives better RSS values than existingRFID system at 2 dBm power level To compare the perfor-mance of two types of RFID system based on power leveland range the third experiment is conducted based on DOEmethod with three factors having two levels of treatment 23The factors are referred to as type of RFID (nonembeddedembedded) power level (2 dBm 10 dBm) and range (5m

International Journal of Distributed Sensor Networks 11

00 20 40 60 80 100 120

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

RSSI

(dBm

)

Figure 11 Received signal strength values at 10 dBm in an indoorenvironment

00 20 40 60 80 100 120

RSSI

(dBm

)

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

minus80

Figure 12 Received signal strength values at 2 dBm in an indoorenvironment

10m) The response of the experiment is RSS and 2 replicatesare used in this work to reduce the measurement errorsand noise The order in which the observations are takenis selected randomly so that the design is a completelyrandomized design Later the statistical method ANOVA isperformed using Minitab 162 and the results are displayedin Tables 4 and 5 for effect estimate summary and ANOVAfor power level versus range measurement From the effectestimate summary presented in Table 4 the main effect of Creally dominates the process accounting for 91381 percentof the total variability whereas the main effect of A and theAC interaction account for about 6144 and 1727 percentrespectively thus the DOE for this experiment also provedthat the RSS value can be used as a prediction method tomeasure distance and location of tags from reader in wireless

sensor network platform since the range contributions arehighly significant compared to other factors The ANOVAin Table 4 is used to confirm the magnitude of these effectsFrom Table 5 the main effects of A B and C are highlysignificant since the factors have very small119875 values (lt0001)For 2-way interaction the AB and AC interaction are alsohighly significant (low 119875 values) thus it can be concludedthat there is a strong interaction between two factors withRSS values in this experimental study Figures 13 and 14 showthe interaction and effects plot for power level versus rangemeasurement that is used to interpret the results from Tables4 and 5 It can be seen that theRSS for embeddedRFID systemis nearly equal to nonembedded RFID system at 10 dBmpower level However at 2 dBm power level the RSS forembedded RFID system is better than nonembedded RFIDsystem In terms of range the RSS for embeddedRFID systemis better compared to nonembedded RFID system at both 5-meter and 100-meter range Based on the results obtainedit can be summarized that for indoor tracking purposesthe low power level gives better RSS values for embeddedRFID system compared to nonembedded RFID system Thisfinding can be supported by [29] which stated that the lowtransmit power is appropriate and sufficient for area withlimited range however if the high transmit power is use inthe area with limited range it will resulting in disturbancethat will cause degradation in radiation performanceThus todesign an efficient indoor localization system these findingsmust be taken into account as a guideline implementing thefull setup for real time location monitoring system in WSNplatform

5 Battery Life Testing

A single RFID tag requires a minimum of 27 V to 33 Vvoltage to operate as a transmitter and receiver in thenetwork Howver the embedded RFID tag proposed in thiswork requires minimum of 50V to power up the embeddeddevices Thus the embedded RFID tag can be powered upusing 9 VDC power supply if the tag is located at fixedlocation or Li-Po rechargeable batteries can be used if thetag is mobile The tag has been programmed to consumethe lowest amount of energy as much as possible sinceit introduced an automated switching mechanism betweenindoor and outdoor location tracking and operating in sleepmode at certain timeTheRFID tag only sends data if the validoutdoor location is detected by the GPS receiver however ifthe embedded RFID tag is located inside the building or theplace where the signal from satellite cannot be detected bythe GPS receiver it will turn to sleep mode and wait for theRFID reader interrogation every hourTheGSMmodemonlyworks when the embedded RFID tag is out of the wirelessnetwork coverage to consume the lowest amount of energy

The battery lifetime measurement is performed for threesystems which are stand-alone RFID tag (A) [20] embeddedRFID tag with switching mechanism (B) and embeddedRFID tag without switching mechanism (C) in an indoorenvironment (inside building) where the GPS signals are notavailable most of the time The results show that the battery

12 International Journal of Distributed Sensor Networks

Table 4 Effect estimate summary

Term constant Effect Sum of squares Percent contributionPower (119860) 724 209530 6144Type of RFID (119861) 179 12780 0375Range (119862) minus2791 3116430 91381Power lowast type of RFID (119860119861) minus141 7980 0234Power lowast range (119860119862) 384 58910 1727Type RFID lowast range (119861119862) 074 2180 0064Power lowast type of RFID lowast range (119860119861119862) minus056 1270 00372

Table 5 ANOVA for power level versus range measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Power (119860) 1 209530 209530 1294370 0000Type of RFID (119861) 1 12780 12780 78950 0000Range (119862) 1 3116430 3116430 19252080 0000Power lowast type of RFID (119860119861) 1 7980 7980 49300 0000Power lowast range (119860119862) 1 58910 58910 363900 0000Type RFID lowast range (119861119862) 1 2180 2180 13440 0006Power lowast type of RFID lowast range (119860119861119862) 1 1270 1270 7820 0023Residual error 8 1290 0160Pure error 8 1300 0160Total 15 3410360

for the tag drops from 824V to 711 V in the duration of 85hours for the proposed embedded RFID tag with switchingmechanism compared to other systems as shown in Figure 15

The results for system B are much better than systemC due to the switching introduced However the result forsystem A is better than system B due to the nonembeddedsystem According to these real implementation results thetag will be expected to stop operating approximately in24 hours when the battery reaches the minimum voltagerequired which is 50 V However this result does not includethe situation where the proposed embedded RFID tag islocated outside building which might increase the energyusage The estimation results also did not include otherexternal factors that might reduce the battery life Thus thetag is estimated to cover the operation for a day withouthaving to recharge the battery until the next day To reducethe energy consumption and maximize the time of usage anenergy harvesting technologywill be implemented later in theembedded RFID tag as our next target

6 Conclusion

This paper describes the implementation of an automatedswitching algorithm in the design of a 245GHz active inte-grated RFID system for indoor and outdoor location trackingutilizing WSN platform Four experimental measurementsbased on DOE method are performed in order to validatethe performance of the proposed RFID system as well as tostudy the relationship interaction and effects of each factor

to RSS values extracted from existing and proposed RFIDsystem The results obtained from first experiment show thatembedded RFID system does affect the radiation patternnot more than 15 differences from existing RFID system Itshows that the proposed RFID system contributes to strongerradiation compared to existing RFID system This may bedue to the embedded design and improvement of algorithmimplemented in the proposed RFID system which increasedthe power stability of the system which thus provided highersignal strength [28]

In the second experiment the results show that the pro-posed RFID tag has better RSS value when it is located nearerto the reader at 0∘ degrees direction of propagation with H-field polarization of antenna In the third experiment theexisting RFID system gives better RSS values than proposedRFID system at 10 dBm power level while the proposedRFID system gives better RSS values than existing RFIDsystem at 2 dBm power level In the fourth experiment it canbe summarized that for indoor tracking purposes the lowpower level gives better RSS values for the proposed RFIDtag compared to the existing RFID system The statisticalanalysis ANOVA proved that the RSS values are significantwith range thus it can be concluded that the RSS can beused as a prediction method to evaluate distance betweenRFID reader and RFID tag in WSN platform Based onthe experimental study and statistical analysis done for allsituations a conclusion has been made where the proposedRFID system with automated switching algorithm is betterthan the existing RFID system in terms of performance andvariability of the factors The battery life testing also gives

International Journal of Distributed Sensor Networks 13

Power

Range

Interactions plot

2 10 5 100

minus30

minus45

minus60

minus30

minus45

minus60

minus30

minus45

minus60

Range

Power

Type of RFID

210

5100

Type of RFID

E

N

EN

Figure 13 Interaction plot for power level versus range measurement

Power

Mea

nM

ean

Type RFID

Range

Main effects plot for RSSI

minus40

minus50

minus60

minus40

minus50

minus60

2 10

5 100

EN

Figure 14 Main effects plot for power level versus range measurement

significant findings where the proposed embedded RFIDtag with switching mechanism introduced is much betterthan embedded RFID tag without switching mechanismTo increase the performance of the embedded RFID tag interms of time of usage an energy harvesting technology willbe introduced later as a part of our next contribution Interms of hardware costing the proposed approach is moreacceptable compared to the system developed by [6] since thetechnology used in this work especially GSM communicationis much cheaper than satellite communication used in the

existing system [6] In addition this proposed system doesnot totally depend on GSM since the communication willbe done by RFID which is free as long as the embeddedRFID tag is working in the control area of WSN platformHowever the existing work [6] did not support long rangecommunication and thus when the tag is moving far fromthe reader the satellite communication needs to be used andthis will lead to high cost of communication As a conclusionthe proposed embedded RFID tag provided better signalpropagation compared to the existing RFID tag [19] and

14 International Journal of Distributed Sensor Networks

00 1 2 3 4 5 6 7 8 9

Batte

ry re

duct

ion

()

Time (hours)

Battery lifetime testing

Embedded RFID tag with switchingEmbedded RFID tag without switchingStandalone RFID tag

minus5

minus10

minus15

minus20

minus25

minus30

Figure 15 Battery life testing based on three types of RFIDdesigned

the battery lifetime for embedded RFID tag with switchingmechanism is better than that without switchingmechanism

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank Malaysia Ministry of HigherEducation (LRGS Fund) for sponsoring the research anddeveloping this project Also special appreciation is due toProfessor DrMohamadKamal Abdul Rahim fromUniversitiTeknologiMalaysia Head of the LRGS Project and AssociateProfessor Dr Alyani Ismail Universiti PutraMalaysia for thesupport

References

[1] DIGI XBeeXBee PRO ZB RF Module Datasheet 2013[2] J B Howard ldquoCombination RFID and GPS functionality on

intelligent labelrdquo WO 2003050960 A2 2003[3] S Chand V R Bapat K H Hall et al ldquoPortable RFID reader

having a location determination systemrdquo EP 1752908 A220062006

[4] G O Garsia S I M Nunez L A Saavedra and M D I TapialdquoReal time location system (RTLS) with ZigBee technologyrdquo EP2196816 A12008 2008

[5] A Bandyopadhyay D Hakim B E Funk E A Kohn C AToelis and G Blankenship ldquoSystem and method for locatingtracking andor monitoring the status of personnel andorassets both indoorsrdquo US 20090043504 A1 2009

[6] RFID Journal ldquoHybrid tags includes active RFID GPS satelliteand sensorrdquo RFID Journal 2009 httpwwwrfidjournalcomarticlesview4635

[7] M Mohandes M Haleem M Deriche and K BalakrishnanldquoWireless sensor networks for pilgrims trackingrdquo IEEE Embed-ded Systems Letters vol 4 no 4 pp 106ndash109 2012

[8] N D Simoes J L Goncalves M L Caeiro M J Boavida andF D Cardoso ldquoZigBeeGPS tracking system for rowing racesrdquoin Proceedings of the International Conference on Computer as aTool (EUROCON rsquo11) Lisbon Portugal April 2011

[9] Y Xu R Jiang S Yan and D Xiong ldquoThe research of safetymonitoring system applied in school bus based on the internetof thingsrdquo Procedia Engineering vol 15 pp 2464ndash2468 2011

[10] L Zheng M Li C Wu et al ldquoDevelopment of a smartmobile farming service systemrdquo Mathematical and ComputerModelling vol 54 no 3-4 pp 1194ndash1203 2011

[11] F A Poad and W Ismail ldquoAutomated switching mechanismfor indoor and outdoor propagation with embedded RFID andGPS in wireless sensor network platformrdquo in Proceedings ofthe International Conference on Wireless Network pp 710ndash714London UK 2014

[12] J OrsquoToole J Tuttle M Tuttle et al ldquoRadio frequency identifi-cation sensor tag apparatusesrdquo US 20080030306 A1 2008

[13] V M Gupta and S Annambhotla ldquoRFID tracker and locatorrdquoUS 20090085745 A1 2009

[14] F Shebli I Dayoub and J M Rouvaen ldquoMinimizing energyconsumptionwithinwireless sensor networkrdquoUbiquitous Com-puting and Communication Journal vol 87 2007

[15] R Ding J Hou and B Xing ldquoResearch of wireless sensornetwork nodes based on ambient energy harvestingrdquo in Pro-ceedings of the 6th IEEE International Conference on IntelligentNetworks and Intelligent Systems (ICINIS rsquo13) pp 286ndash288Shenyang China November 2013

[16] A Cammarano C Petrioli andD Spenza ldquoPro-Energy a novelenergy prediction model for solar and wind energy-harvestingwireless sensor networksrdquo in Proceedings of the 9th IEEEInternational Conference on Mobile Ad-Hoc and Sensor Systems(MASS rsquo12) pp 75ndash83 Las Vegas Nev USA October 2012

[17] J Song and Y K Tan ldquoEnergy consumption analysis of ZigBee-based energy harvesting wireless sensor networksrdquo in Proceed-ings of the IEEE International Conference on CommunicationSystems (ICCS rsquo12) pp 468ndash472 November 2012

[18] San Jose Technology Inc Datasheet GPS Engine Board ModelFV M8 2014 httpswwwsparkfuncomdatasheetsGPSFV-M8 Specpdf

[19] W Ismail and R Abdulla ldquoPortable RFID Reader for RTLS(Real Time Location Systems)rdquo PCT International Filing NoPCTMY2011000167 Country Australia Pub No WO2012102600 2012

[20] W Ismail C Z Zulkifli and M G Rahman ldquoWireless pro-duction monitoring systemrdquo PCT International Filing PCTMY2013000037 2013

[21] ISOIEC 18000-4 Information Technology-Radio FrequencyIdentification for ItemManagement-Part7 Parameters for ActiveAir Interface Communications at 245 GHz ISO 18000-42008International Organization for Standardization GenevaSwitzerland 2008

[22] D-H Shih P-L Sun D C Yen and S-M Huang ldquoTaxonomyand survey of RFID anti-collision protocolsrdquo Computer Com-munications vol 29 no 11 pp 2150ndash2166 2006

[23] B Nilsson L Bengtsson P-A Wiberg and B SvenssonldquoProtocols for active RFIDmdashthe energy consumption aspectrdquo inProceedings of the IEEE 2nd International Symposium on Indus-trial Embedded Systems (SIES rsquo07) pp 41ndash48 July 2007

[24] H L van Eeden ldquoWireless Communication Systemrdquo US Patent0045923 2009

International Journal of Distributed Sensor Networks 15

[25] H Haar K Bischof and H Scharke ldquoRFID transponderrdquo WOPatent 109212 2006

[26] C A Balanis AntennaTheory Analysis and Design JohnWileyamp Sons New York NY USA 3rd edition 2005

[27] P Sorrells Optimizing Read Range in RFID System MicrochipTechnology Inc 2000 httpedncomdesignwireless-network-ing4341980Optimizing-read-range-in-RFID-systems

[28] A Ammu L Mapa and A H Jayatissa ldquoEffect of factors onRFID tag readability-statistical analysisrdquo in Proceedings of theIEEE International Conference on ElectroInformation Technol-ogy (EIT rsquo09) pp 355ndash358 June 2009

[29] R Zurawski Embedded System Handbook Second EditionNetwork Embedded Systems CRC Press 2009

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

Page 8: Research Article Propagation Analysis for Automated ...downloads.hindawi.com/journals/ijdsn/2015/392385.pdf · Hybrid tag includes active RFID, GPS, satellite, and sensors [ ] MHz

8 International Journal of Distributed Sensor Networks

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(a)

0

02

04

06

08

10

1530

45

60

75

90

105

120

135

150165

180195

210

225

240

255

270

285

300

315

330345

NonembeddedEmbedded

(b)

Figure 8 E-plane pattern at (a) 2m and (b) 10m distance

management that is implemented in the embedded activeRFID tag that provided power stability as mentioned by [27]thus increasing the performance of the system in the network

The second experiment is the continuity of the radiationpattern measurement which involved four factors with twolevels of treatment 24 The factors are referred to as type ofRFID (nonembedded embedded) power level (2 dBm10 dBm) range (5m 10m) polarization (E H) and angleof polarization (0∘ 90∘) The response of the experiment isin RSS and 2 replicates are used in this work to reduce themeasurement errors and noiseThe order in which the obser-vations are taken is selected randomly so that the design is acompletely randomized design To identify the interactionsand effects between type of RFID range polarization andangle of propagation with RSS a statistical method calledanalysis of variance (ANOVA) is used as presented in [28]and the effect estimate summaries for the proposed systemare shown in Table 2

In this work the main effect of A really dominates theprocess accounting for 63635 percent of the total variabilitywhereas themain effect of C accounts for about 7328 percentwhile the AB and BC interactions account for about 8808percent and 6616 percent respectively thus it is proved thatthe RSS value can be used as a prediction method to measuredistance and location of tags from reader in WSN platformsince the range contributes highly significantly compared toother factors The ANOVA in Table 3 is used to confirm themagnitude of these effects From Table 3 it can be seen thatthe main effects of A and C are highly significant since bothof the factors have very small 119875 values (lt0001) For 4-wayinteraction the ABCD interactions are also highly significant(low 119875 values) thus it can be concluded that there is a stronginteraction between four factors with RSS values

Figures 9 and 10 show the interactions and main effectsplot for radiation measurement that is used to interpret theresults from Tables 1 and 2

The graph shows that the RSS for existing RFID system isbetter than proposed RFID system at 10m distance howeverthe RSS for proposed RFID system is better than the existingRFID system at 2m distance In terms of angle the RSS for0∘ and 90∘ degree proposed RFID system are better thanexisting RFID system however the RSS value for 0∘ degreesis better than 90∘ degrees for both RFID system In terms ofpolarization the H-field polarization gives more significantvalues of RSS compared to E-field polarization for both RFIDsystems The findings show that the RFID system will havebetter RSS value if it is located nearer to the reader at 0∘degrees direction of propagation with H-field polarizationof antenna This statement is supported by Friis free spaceequation as shown in [26]

119875119903 (119889) =119875119905119866119905119866119903120582

2

(4120587)21198892119871 (2)

where119875119905 is the transmitted power119875119903(119889) is the received power119866119905 is the transmitter antenna gain 119866119903 is the receiver antennagain 119889 is the transmit and receive separation and 119871 is the sys-tem loss factor dependent upon line attenuation filter lossesand antenna losses and not related to propagation Accordingto Friis equation the electromagnetic (EM) signals whentraveling through wireless channels experience fading effectsdue to various effects but for ideal case the transmission thatis considered with a direct line of sight will give the bestvalue of received power The effects of impedance mismatchmisalignment of the antenna pointing and polarization and

International Journal of Distributed Sensor Networks 9

Table 2 Effect estimate summaries

Term constant Effect Sum of squares Percentage contributionRange (119860) minus1109 983460 63635Angle (119861) minus079 4960 0321Polarization (119862) 376 113250 7328Type of RFID (119863) 160 20480 1325Range lowast angle (119860119861) minus358 102250 6616Range lowast polarization (119860119862) minus073 4210 0272Range lowast type of RFID (119860119863) minus226 40950 2650Angle lowast polarization (119861119862) minus412 136120 8808Angle lowast type of RFID (119861119863) 034 0910 0058Polarization lowast type of RFID (119862119863) minus186 27750 1796Range lowast angle lowast polarization (119860119861119862) minus124 12250 0793Range lowast angle lowast type of RFID (119860119861119863) 140 15680 1015Range lowast polarization lowast type of RFID (119860119862119863) minus075 4500 0291Angle lowast polarization lowast type of RFID (119861119862119863) minus100 8000 0518Range lowast angle lowast polarization lowast type of RFID (119860119861119862119863) minus206 34030 2202

Table 3 ANOVA for the radiation pattern measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Range (119860) 1 983460 983461 429340 0000Angle (119861) 1 4960 4961 2170 0160Polarization (119862) 1 113250 113251 49440 0000Type of RFID (119863) 1 20480 20480 8940 0009Range lowast angle (119860119861) 1 102250 102245 44640 0000Range lowast polarization (119860119862) 1 4210 4205 1840 0194Range lowast type of RFID (119860119863) 1 40950 40951 17880 0001Angle lowast polarization (119861119862) 1 136120 136125 59430 0000Angle lowast type of RFID (119861119863) 1 0910 0911 0400 0537Polarization lowast type of RFID (119862119863) 1 27750 27751 12120 0003Range lowast angle lowast polarization (119860119861119862) 1 12250 12251 5350 0034Range lowast angle lowast type of RFID (119860119861119863) 1 15680 15680 6850 0019Range lowast polarization lowast type RFID (119860119862119863) 1 4500 4500 1960 0180Angle lowast polarization lowast type of RFID (119861119862119863) 1 8000 8000 3490 0080Range lowast angle lowast polarization lowast type of RFID (119860119861119862119863) 1 34030 34031 14860 0001Residual error 16 36650 2291Pure error 16 36650 2291Total 31 1545450

absorption can be included by adding additional factors asgiven in [26]

119875119903

119875119905

= 119866119905 (120579119905 120601119905) 119866119903 (120579119903 120601119903) (120582

4120587119877)

2

times (1 minus1003816100381610038161003816Γ1199051003816100381610038161003816

2) (1 minus1003816100381610038161003816Γ1199031003816100381610038161003816

2)1003816100381610038161003816119886119905 sdot 1198861199031003816100381610038161003816

2119890minus120572119877

(3)

where 119866119905(120579119905 120601119905) is the gain of the transmit antenna in thedirection 119866119903(120579119903 120601119903) is the gain of the receive antenna in thedirection (120579119903 120601119903) Γ119905 and Γ119903 are the reflection coefficients ofthe transmit and receive antennas respectively and 119886119905 and119886119903 are the polarization vectors of the transmit and receiveantennas respectively taken in the appropriate directions

120572 is the absorption coefficient of the intervening mediumHowever these additional factors will contribute to reductionof antenna received power Thus in the line of sight directionthe received power will become higher compared to non-lineof sight direction as experimented in this section

42 Power Level versus Range Measurement The power levelversus range measurement is done for distance between 5mand 100m indoorswith 5m incrementThedistance is limiteddue to the buildingrsquos built-up area which is limited to 100-meter lengthThe response is RSS values which are measuredfor existing and proposed RFID system at two levels of powerwhich are 10 dBm (max) and 2 dBm (min) At each point10 single values of receiving power are collected to calculate

10 International Journal of Distributed Sensor Networks

210

Range

090

Angle

Polarization

Type of RFID

Interactions plot

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

2 10

0 90

Range

Angle

Polarization

Type of RFID

E

E

H

H

N

P

N P

Figure 9 Interaction plot for radiation measurement

Range

Mea

nM

ean

Angle

Polarization Type of RFID

Main effects plot for RSSIData means

minus36

minus39

minus42

minus45

minus48

minus36

minus39

minus42

minus45

minus48

2 10 0 90

E H N P

Figure 10 Main effects plot for radiation measurement

the mean Figures 11 and 12 show the comparison betweenRSS values of existing and proposed RFID system based ondistance of 5 meters to 100 meters indoors at 10 dBm and2 dBm power levels The graph shows that the RSS valuesare decreased when the distance is increased at both powerlevels The existing RFID system gives better RSS valuesthan proposed RFID system at 10 dBm power level while the

proposed RFID system gives better RSS values than existingRFID system at 2 dBm power level To compare the perfor-mance of two types of RFID system based on power leveland range the third experiment is conducted based on DOEmethod with three factors having two levels of treatment 23The factors are referred to as type of RFID (nonembeddedembedded) power level (2 dBm 10 dBm) and range (5m

International Journal of Distributed Sensor Networks 11

00 20 40 60 80 100 120

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

RSSI

(dBm

)

Figure 11 Received signal strength values at 10 dBm in an indoorenvironment

00 20 40 60 80 100 120

RSSI

(dBm

)

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

minus80

Figure 12 Received signal strength values at 2 dBm in an indoorenvironment

10m) The response of the experiment is RSS and 2 replicatesare used in this work to reduce the measurement errorsand noise The order in which the observations are takenis selected randomly so that the design is a completelyrandomized design Later the statistical method ANOVA isperformed using Minitab 162 and the results are displayedin Tables 4 and 5 for effect estimate summary and ANOVAfor power level versus range measurement From the effectestimate summary presented in Table 4 the main effect of Creally dominates the process accounting for 91381 percentof the total variability whereas the main effect of A and theAC interaction account for about 6144 and 1727 percentrespectively thus the DOE for this experiment also provedthat the RSS value can be used as a prediction method tomeasure distance and location of tags from reader in wireless

sensor network platform since the range contributions arehighly significant compared to other factors The ANOVAin Table 4 is used to confirm the magnitude of these effectsFrom Table 5 the main effects of A B and C are highlysignificant since the factors have very small119875 values (lt0001)For 2-way interaction the AB and AC interaction are alsohighly significant (low 119875 values) thus it can be concludedthat there is a strong interaction between two factors withRSS values in this experimental study Figures 13 and 14 showthe interaction and effects plot for power level versus rangemeasurement that is used to interpret the results from Tables4 and 5 It can be seen that theRSS for embeddedRFID systemis nearly equal to nonembedded RFID system at 10 dBmpower level However at 2 dBm power level the RSS forembedded RFID system is better than nonembedded RFIDsystem In terms of range the RSS for embeddedRFID systemis better compared to nonembedded RFID system at both 5-meter and 100-meter range Based on the results obtainedit can be summarized that for indoor tracking purposesthe low power level gives better RSS values for embeddedRFID system compared to nonembedded RFID system Thisfinding can be supported by [29] which stated that the lowtransmit power is appropriate and sufficient for area withlimited range however if the high transmit power is use inthe area with limited range it will resulting in disturbancethat will cause degradation in radiation performanceThus todesign an efficient indoor localization system these findingsmust be taken into account as a guideline implementing thefull setup for real time location monitoring system in WSNplatform

5 Battery Life Testing

A single RFID tag requires a minimum of 27 V to 33 Vvoltage to operate as a transmitter and receiver in thenetwork Howver the embedded RFID tag proposed in thiswork requires minimum of 50V to power up the embeddeddevices Thus the embedded RFID tag can be powered upusing 9 VDC power supply if the tag is located at fixedlocation or Li-Po rechargeable batteries can be used if thetag is mobile The tag has been programmed to consumethe lowest amount of energy as much as possible sinceit introduced an automated switching mechanism betweenindoor and outdoor location tracking and operating in sleepmode at certain timeTheRFID tag only sends data if the validoutdoor location is detected by the GPS receiver however ifthe embedded RFID tag is located inside the building or theplace where the signal from satellite cannot be detected bythe GPS receiver it will turn to sleep mode and wait for theRFID reader interrogation every hourTheGSMmodemonlyworks when the embedded RFID tag is out of the wirelessnetwork coverage to consume the lowest amount of energy

The battery lifetime measurement is performed for threesystems which are stand-alone RFID tag (A) [20] embeddedRFID tag with switching mechanism (B) and embeddedRFID tag without switching mechanism (C) in an indoorenvironment (inside building) where the GPS signals are notavailable most of the time The results show that the battery

12 International Journal of Distributed Sensor Networks

Table 4 Effect estimate summary

Term constant Effect Sum of squares Percent contributionPower (119860) 724 209530 6144Type of RFID (119861) 179 12780 0375Range (119862) minus2791 3116430 91381Power lowast type of RFID (119860119861) minus141 7980 0234Power lowast range (119860119862) 384 58910 1727Type RFID lowast range (119861119862) 074 2180 0064Power lowast type of RFID lowast range (119860119861119862) minus056 1270 00372

Table 5 ANOVA for power level versus range measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Power (119860) 1 209530 209530 1294370 0000Type of RFID (119861) 1 12780 12780 78950 0000Range (119862) 1 3116430 3116430 19252080 0000Power lowast type of RFID (119860119861) 1 7980 7980 49300 0000Power lowast range (119860119862) 1 58910 58910 363900 0000Type RFID lowast range (119861119862) 1 2180 2180 13440 0006Power lowast type of RFID lowast range (119860119861119862) 1 1270 1270 7820 0023Residual error 8 1290 0160Pure error 8 1300 0160Total 15 3410360

for the tag drops from 824V to 711 V in the duration of 85hours for the proposed embedded RFID tag with switchingmechanism compared to other systems as shown in Figure 15

The results for system B are much better than systemC due to the switching introduced However the result forsystem A is better than system B due to the nonembeddedsystem According to these real implementation results thetag will be expected to stop operating approximately in24 hours when the battery reaches the minimum voltagerequired which is 50 V However this result does not includethe situation where the proposed embedded RFID tag islocated outside building which might increase the energyusage The estimation results also did not include otherexternal factors that might reduce the battery life Thus thetag is estimated to cover the operation for a day withouthaving to recharge the battery until the next day To reducethe energy consumption and maximize the time of usage anenergy harvesting technologywill be implemented later in theembedded RFID tag as our next target

6 Conclusion

This paper describes the implementation of an automatedswitching algorithm in the design of a 245GHz active inte-grated RFID system for indoor and outdoor location trackingutilizing WSN platform Four experimental measurementsbased on DOE method are performed in order to validatethe performance of the proposed RFID system as well as tostudy the relationship interaction and effects of each factor

to RSS values extracted from existing and proposed RFIDsystem The results obtained from first experiment show thatembedded RFID system does affect the radiation patternnot more than 15 differences from existing RFID system Itshows that the proposed RFID system contributes to strongerradiation compared to existing RFID system This may bedue to the embedded design and improvement of algorithmimplemented in the proposed RFID system which increasedthe power stability of the system which thus provided highersignal strength [28]

In the second experiment the results show that the pro-posed RFID tag has better RSS value when it is located nearerto the reader at 0∘ degrees direction of propagation with H-field polarization of antenna In the third experiment theexisting RFID system gives better RSS values than proposedRFID system at 10 dBm power level while the proposedRFID system gives better RSS values than existing RFIDsystem at 2 dBm power level In the fourth experiment it canbe summarized that for indoor tracking purposes the lowpower level gives better RSS values for the proposed RFIDtag compared to the existing RFID system The statisticalanalysis ANOVA proved that the RSS values are significantwith range thus it can be concluded that the RSS can beused as a prediction method to evaluate distance betweenRFID reader and RFID tag in WSN platform Based onthe experimental study and statistical analysis done for allsituations a conclusion has been made where the proposedRFID system with automated switching algorithm is betterthan the existing RFID system in terms of performance andvariability of the factors The battery life testing also gives

International Journal of Distributed Sensor Networks 13

Power

Range

Interactions plot

2 10 5 100

minus30

minus45

minus60

minus30

minus45

minus60

minus30

minus45

minus60

Range

Power

Type of RFID

210

5100

Type of RFID

E

N

EN

Figure 13 Interaction plot for power level versus range measurement

Power

Mea

nM

ean

Type RFID

Range

Main effects plot for RSSI

minus40

minus50

minus60

minus40

minus50

minus60

2 10

5 100

EN

Figure 14 Main effects plot for power level versus range measurement

significant findings where the proposed embedded RFIDtag with switching mechanism introduced is much betterthan embedded RFID tag without switching mechanismTo increase the performance of the embedded RFID tag interms of time of usage an energy harvesting technology willbe introduced later as a part of our next contribution Interms of hardware costing the proposed approach is moreacceptable compared to the system developed by [6] since thetechnology used in this work especially GSM communicationis much cheaper than satellite communication used in the

existing system [6] In addition this proposed system doesnot totally depend on GSM since the communication willbe done by RFID which is free as long as the embeddedRFID tag is working in the control area of WSN platformHowever the existing work [6] did not support long rangecommunication and thus when the tag is moving far fromthe reader the satellite communication needs to be used andthis will lead to high cost of communication As a conclusionthe proposed embedded RFID tag provided better signalpropagation compared to the existing RFID tag [19] and

14 International Journal of Distributed Sensor Networks

00 1 2 3 4 5 6 7 8 9

Batte

ry re

duct

ion

()

Time (hours)

Battery lifetime testing

Embedded RFID tag with switchingEmbedded RFID tag without switchingStandalone RFID tag

minus5

minus10

minus15

minus20

minus25

minus30

Figure 15 Battery life testing based on three types of RFIDdesigned

the battery lifetime for embedded RFID tag with switchingmechanism is better than that without switchingmechanism

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank Malaysia Ministry of HigherEducation (LRGS Fund) for sponsoring the research anddeveloping this project Also special appreciation is due toProfessor DrMohamadKamal Abdul Rahim fromUniversitiTeknologiMalaysia Head of the LRGS Project and AssociateProfessor Dr Alyani Ismail Universiti PutraMalaysia for thesupport

References

[1] DIGI XBeeXBee PRO ZB RF Module Datasheet 2013[2] J B Howard ldquoCombination RFID and GPS functionality on

intelligent labelrdquo WO 2003050960 A2 2003[3] S Chand V R Bapat K H Hall et al ldquoPortable RFID reader

having a location determination systemrdquo EP 1752908 A220062006

[4] G O Garsia S I M Nunez L A Saavedra and M D I TapialdquoReal time location system (RTLS) with ZigBee technologyrdquo EP2196816 A12008 2008

[5] A Bandyopadhyay D Hakim B E Funk E A Kohn C AToelis and G Blankenship ldquoSystem and method for locatingtracking andor monitoring the status of personnel andorassets both indoorsrdquo US 20090043504 A1 2009

[6] RFID Journal ldquoHybrid tags includes active RFID GPS satelliteand sensorrdquo RFID Journal 2009 httpwwwrfidjournalcomarticlesview4635

[7] M Mohandes M Haleem M Deriche and K BalakrishnanldquoWireless sensor networks for pilgrims trackingrdquo IEEE Embed-ded Systems Letters vol 4 no 4 pp 106ndash109 2012

[8] N D Simoes J L Goncalves M L Caeiro M J Boavida andF D Cardoso ldquoZigBeeGPS tracking system for rowing racesrdquoin Proceedings of the International Conference on Computer as aTool (EUROCON rsquo11) Lisbon Portugal April 2011

[9] Y Xu R Jiang S Yan and D Xiong ldquoThe research of safetymonitoring system applied in school bus based on the internetof thingsrdquo Procedia Engineering vol 15 pp 2464ndash2468 2011

[10] L Zheng M Li C Wu et al ldquoDevelopment of a smartmobile farming service systemrdquo Mathematical and ComputerModelling vol 54 no 3-4 pp 1194ndash1203 2011

[11] F A Poad and W Ismail ldquoAutomated switching mechanismfor indoor and outdoor propagation with embedded RFID andGPS in wireless sensor network platformrdquo in Proceedings ofthe International Conference on Wireless Network pp 710ndash714London UK 2014

[12] J OrsquoToole J Tuttle M Tuttle et al ldquoRadio frequency identifi-cation sensor tag apparatusesrdquo US 20080030306 A1 2008

[13] V M Gupta and S Annambhotla ldquoRFID tracker and locatorrdquoUS 20090085745 A1 2009

[14] F Shebli I Dayoub and J M Rouvaen ldquoMinimizing energyconsumptionwithinwireless sensor networkrdquoUbiquitous Com-puting and Communication Journal vol 87 2007

[15] R Ding J Hou and B Xing ldquoResearch of wireless sensornetwork nodes based on ambient energy harvestingrdquo in Pro-ceedings of the 6th IEEE International Conference on IntelligentNetworks and Intelligent Systems (ICINIS rsquo13) pp 286ndash288Shenyang China November 2013

[16] A Cammarano C Petrioli andD Spenza ldquoPro-Energy a novelenergy prediction model for solar and wind energy-harvestingwireless sensor networksrdquo in Proceedings of the 9th IEEEInternational Conference on Mobile Ad-Hoc and Sensor Systems(MASS rsquo12) pp 75ndash83 Las Vegas Nev USA October 2012

[17] J Song and Y K Tan ldquoEnergy consumption analysis of ZigBee-based energy harvesting wireless sensor networksrdquo in Proceed-ings of the IEEE International Conference on CommunicationSystems (ICCS rsquo12) pp 468ndash472 November 2012

[18] San Jose Technology Inc Datasheet GPS Engine Board ModelFV M8 2014 httpswwwsparkfuncomdatasheetsGPSFV-M8 Specpdf

[19] W Ismail and R Abdulla ldquoPortable RFID Reader for RTLS(Real Time Location Systems)rdquo PCT International Filing NoPCTMY2011000167 Country Australia Pub No WO2012102600 2012

[20] W Ismail C Z Zulkifli and M G Rahman ldquoWireless pro-duction monitoring systemrdquo PCT International Filing PCTMY2013000037 2013

[21] ISOIEC 18000-4 Information Technology-Radio FrequencyIdentification for ItemManagement-Part7 Parameters for ActiveAir Interface Communications at 245 GHz ISO 18000-42008International Organization for Standardization GenevaSwitzerland 2008

[22] D-H Shih P-L Sun D C Yen and S-M Huang ldquoTaxonomyand survey of RFID anti-collision protocolsrdquo Computer Com-munications vol 29 no 11 pp 2150ndash2166 2006

[23] B Nilsson L Bengtsson P-A Wiberg and B SvenssonldquoProtocols for active RFIDmdashthe energy consumption aspectrdquo inProceedings of the IEEE 2nd International Symposium on Indus-trial Embedded Systems (SIES rsquo07) pp 41ndash48 July 2007

[24] H L van Eeden ldquoWireless Communication Systemrdquo US Patent0045923 2009

International Journal of Distributed Sensor Networks 15

[25] H Haar K Bischof and H Scharke ldquoRFID transponderrdquo WOPatent 109212 2006

[26] C A Balanis AntennaTheory Analysis and Design JohnWileyamp Sons New York NY USA 3rd edition 2005

[27] P Sorrells Optimizing Read Range in RFID System MicrochipTechnology Inc 2000 httpedncomdesignwireless-network-ing4341980Optimizing-read-range-in-RFID-systems

[28] A Ammu L Mapa and A H Jayatissa ldquoEffect of factors onRFID tag readability-statistical analysisrdquo in Proceedings of theIEEE International Conference on ElectroInformation Technol-ogy (EIT rsquo09) pp 355ndash358 June 2009

[29] R Zurawski Embedded System Handbook Second EditionNetwork Embedded Systems CRC Press 2009

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

Page 9: Research Article Propagation Analysis for Automated ...downloads.hindawi.com/journals/ijdsn/2015/392385.pdf · Hybrid tag includes active RFID, GPS, satellite, and sensors [ ] MHz

International Journal of Distributed Sensor Networks 9

Table 2 Effect estimate summaries

Term constant Effect Sum of squares Percentage contributionRange (119860) minus1109 983460 63635Angle (119861) minus079 4960 0321Polarization (119862) 376 113250 7328Type of RFID (119863) 160 20480 1325Range lowast angle (119860119861) minus358 102250 6616Range lowast polarization (119860119862) minus073 4210 0272Range lowast type of RFID (119860119863) minus226 40950 2650Angle lowast polarization (119861119862) minus412 136120 8808Angle lowast type of RFID (119861119863) 034 0910 0058Polarization lowast type of RFID (119862119863) minus186 27750 1796Range lowast angle lowast polarization (119860119861119862) minus124 12250 0793Range lowast angle lowast type of RFID (119860119861119863) 140 15680 1015Range lowast polarization lowast type of RFID (119860119862119863) minus075 4500 0291Angle lowast polarization lowast type of RFID (119861119862119863) minus100 8000 0518Range lowast angle lowast polarization lowast type of RFID (119860119861119862119863) minus206 34030 2202

Table 3 ANOVA for the radiation pattern measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Range (119860) 1 983460 983461 429340 0000Angle (119861) 1 4960 4961 2170 0160Polarization (119862) 1 113250 113251 49440 0000Type of RFID (119863) 1 20480 20480 8940 0009Range lowast angle (119860119861) 1 102250 102245 44640 0000Range lowast polarization (119860119862) 1 4210 4205 1840 0194Range lowast type of RFID (119860119863) 1 40950 40951 17880 0001Angle lowast polarization (119861119862) 1 136120 136125 59430 0000Angle lowast type of RFID (119861119863) 1 0910 0911 0400 0537Polarization lowast type of RFID (119862119863) 1 27750 27751 12120 0003Range lowast angle lowast polarization (119860119861119862) 1 12250 12251 5350 0034Range lowast angle lowast type of RFID (119860119861119863) 1 15680 15680 6850 0019Range lowast polarization lowast type RFID (119860119862119863) 1 4500 4500 1960 0180Angle lowast polarization lowast type of RFID (119861119862119863) 1 8000 8000 3490 0080Range lowast angle lowast polarization lowast type of RFID (119860119861119862119863) 1 34030 34031 14860 0001Residual error 16 36650 2291Pure error 16 36650 2291Total 31 1545450

absorption can be included by adding additional factors asgiven in [26]

119875119903

119875119905

= 119866119905 (120579119905 120601119905) 119866119903 (120579119903 120601119903) (120582

4120587119877)

2

times (1 minus1003816100381610038161003816Γ1199051003816100381610038161003816

2) (1 minus1003816100381610038161003816Γ1199031003816100381610038161003816

2)1003816100381610038161003816119886119905 sdot 1198861199031003816100381610038161003816

2119890minus120572119877

(3)

where 119866119905(120579119905 120601119905) is the gain of the transmit antenna in thedirection 119866119903(120579119903 120601119903) is the gain of the receive antenna in thedirection (120579119903 120601119903) Γ119905 and Γ119903 are the reflection coefficients ofthe transmit and receive antennas respectively and 119886119905 and119886119903 are the polarization vectors of the transmit and receiveantennas respectively taken in the appropriate directions

120572 is the absorption coefficient of the intervening mediumHowever these additional factors will contribute to reductionof antenna received power Thus in the line of sight directionthe received power will become higher compared to non-lineof sight direction as experimented in this section

42 Power Level versus Range Measurement The power levelversus range measurement is done for distance between 5mand 100m indoorswith 5m incrementThedistance is limiteddue to the buildingrsquos built-up area which is limited to 100-meter lengthThe response is RSS values which are measuredfor existing and proposed RFID system at two levels of powerwhich are 10 dBm (max) and 2 dBm (min) At each point10 single values of receiving power are collected to calculate

10 International Journal of Distributed Sensor Networks

210

Range

090

Angle

Polarization

Type of RFID

Interactions plot

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

2 10

0 90

Range

Angle

Polarization

Type of RFID

E

E

H

H

N

P

N P

Figure 9 Interaction plot for radiation measurement

Range

Mea

nM

ean

Angle

Polarization Type of RFID

Main effects plot for RSSIData means

minus36

minus39

minus42

minus45

minus48

minus36

minus39

minus42

minus45

minus48

2 10 0 90

E H N P

Figure 10 Main effects plot for radiation measurement

the mean Figures 11 and 12 show the comparison betweenRSS values of existing and proposed RFID system based ondistance of 5 meters to 100 meters indoors at 10 dBm and2 dBm power levels The graph shows that the RSS valuesare decreased when the distance is increased at both powerlevels The existing RFID system gives better RSS valuesthan proposed RFID system at 10 dBm power level while the

proposed RFID system gives better RSS values than existingRFID system at 2 dBm power level To compare the perfor-mance of two types of RFID system based on power leveland range the third experiment is conducted based on DOEmethod with three factors having two levels of treatment 23The factors are referred to as type of RFID (nonembeddedembedded) power level (2 dBm 10 dBm) and range (5m

International Journal of Distributed Sensor Networks 11

00 20 40 60 80 100 120

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

RSSI

(dBm

)

Figure 11 Received signal strength values at 10 dBm in an indoorenvironment

00 20 40 60 80 100 120

RSSI

(dBm

)

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

minus80

Figure 12 Received signal strength values at 2 dBm in an indoorenvironment

10m) The response of the experiment is RSS and 2 replicatesare used in this work to reduce the measurement errorsand noise The order in which the observations are takenis selected randomly so that the design is a completelyrandomized design Later the statistical method ANOVA isperformed using Minitab 162 and the results are displayedin Tables 4 and 5 for effect estimate summary and ANOVAfor power level versus range measurement From the effectestimate summary presented in Table 4 the main effect of Creally dominates the process accounting for 91381 percentof the total variability whereas the main effect of A and theAC interaction account for about 6144 and 1727 percentrespectively thus the DOE for this experiment also provedthat the RSS value can be used as a prediction method tomeasure distance and location of tags from reader in wireless

sensor network platform since the range contributions arehighly significant compared to other factors The ANOVAin Table 4 is used to confirm the magnitude of these effectsFrom Table 5 the main effects of A B and C are highlysignificant since the factors have very small119875 values (lt0001)For 2-way interaction the AB and AC interaction are alsohighly significant (low 119875 values) thus it can be concludedthat there is a strong interaction between two factors withRSS values in this experimental study Figures 13 and 14 showthe interaction and effects plot for power level versus rangemeasurement that is used to interpret the results from Tables4 and 5 It can be seen that theRSS for embeddedRFID systemis nearly equal to nonembedded RFID system at 10 dBmpower level However at 2 dBm power level the RSS forembedded RFID system is better than nonembedded RFIDsystem In terms of range the RSS for embeddedRFID systemis better compared to nonembedded RFID system at both 5-meter and 100-meter range Based on the results obtainedit can be summarized that for indoor tracking purposesthe low power level gives better RSS values for embeddedRFID system compared to nonembedded RFID system Thisfinding can be supported by [29] which stated that the lowtransmit power is appropriate and sufficient for area withlimited range however if the high transmit power is use inthe area with limited range it will resulting in disturbancethat will cause degradation in radiation performanceThus todesign an efficient indoor localization system these findingsmust be taken into account as a guideline implementing thefull setup for real time location monitoring system in WSNplatform

5 Battery Life Testing

A single RFID tag requires a minimum of 27 V to 33 Vvoltage to operate as a transmitter and receiver in thenetwork Howver the embedded RFID tag proposed in thiswork requires minimum of 50V to power up the embeddeddevices Thus the embedded RFID tag can be powered upusing 9 VDC power supply if the tag is located at fixedlocation or Li-Po rechargeable batteries can be used if thetag is mobile The tag has been programmed to consumethe lowest amount of energy as much as possible sinceit introduced an automated switching mechanism betweenindoor and outdoor location tracking and operating in sleepmode at certain timeTheRFID tag only sends data if the validoutdoor location is detected by the GPS receiver however ifthe embedded RFID tag is located inside the building or theplace where the signal from satellite cannot be detected bythe GPS receiver it will turn to sleep mode and wait for theRFID reader interrogation every hourTheGSMmodemonlyworks when the embedded RFID tag is out of the wirelessnetwork coverage to consume the lowest amount of energy

The battery lifetime measurement is performed for threesystems which are stand-alone RFID tag (A) [20] embeddedRFID tag with switching mechanism (B) and embeddedRFID tag without switching mechanism (C) in an indoorenvironment (inside building) where the GPS signals are notavailable most of the time The results show that the battery

12 International Journal of Distributed Sensor Networks

Table 4 Effect estimate summary

Term constant Effect Sum of squares Percent contributionPower (119860) 724 209530 6144Type of RFID (119861) 179 12780 0375Range (119862) minus2791 3116430 91381Power lowast type of RFID (119860119861) minus141 7980 0234Power lowast range (119860119862) 384 58910 1727Type RFID lowast range (119861119862) 074 2180 0064Power lowast type of RFID lowast range (119860119861119862) minus056 1270 00372

Table 5 ANOVA for power level versus range measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Power (119860) 1 209530 209530 1294370 0000Type of RFID (119861) 1 12780 12780 78950 0000Range (119862) 1 3116430 3116430 19252080 0000Power lowast type of RFID (119860119861) 1 7980 7980 49300 0000Power lowast range (119860119862) 1 58910 58910 363900 0000Type RFID lowast range (119861119862) 1 2180 2180 13440 0006Power lowast type of RFID lowast range (119860119861119862) 1 1270 1270 7820 0023Residual error 8 1290 0160Pure error 8 1300 0160Total 15 3410360

for the tag drops from 824V to 711 V in the duration of 85hours for the proposed embedded RFID tag with switchingmechanism compared to other systems as shown in Figure 15

The results for system B are much better than systemC due to the switching introduced However the result forsystem A is better than system B due to the nonembeddedsystem According to these real implementation results thetag will be expected to stop operating approximately in24 hours when the battery reaches the minimum voltagerequired which is 50 V However this result does not includethe situation where the proposed embedded RFID tag islocated outside building which might increase the energyusage The estimation results also did not include otherexternal factors that might reduce the battery life Thus thetag is estimated to cover the operation for a day withouthaving to recharge the battery until the next day To reducethe energy consumption and maximize the time of usage anenergy harvesting technologywill be implemented later in theembedded RFID tag as our next target

6 Conclusion

This paper describes the implementation of an automatedswitching algorithm in the design of a 245GHz active inte-grated RFID system for indoor and outdoor location trackingutilizing WSN platform Four experimental measurementsbased on DOE method are performed in order to validatethe performance of the proposed RFID system as well as tostudy the relationship interaction and effects of each factor

to RSS values extracted from existing and proposed RFIDsystem The results obtained from first experiment show thatembedded RFID system does affect the radiation patternnot more than 15 differences from existing RFID system Itshows that the proposed RFID system contributes to strongerradiation compared to existing RFID system This may bedue to the embedded design and improvement of algorithmimplemented in the proposed RFID system which increasedthe power stability of the system which thus provided highersignal strength [28]

In the second experiment the results show that the pro-posed RFID tag has better RSS value when it is located nearerto the reader at 0∘ degrees direction of propagation with H-field polarization of antenna In the third experiment theexisting RFID system gives better RSS values than proposedRFID system at 10 dBm power level while the proposedRFID system gives better RSS values than existing RFIDsystem at 2 dBm power level In the fourth experiment it canbe summarized that for indoor tracking purposes the lowpower level gives better RSS values for the proposed RFIDtag compared to the existing RFID system The statisticalanalysis ANOVA proved that the RSS values are significantwith range thus it can be concluded that the RSS can beused as a prediction method to evaluate distance betweenRFID reader and RFID tag in WSN platform Based onthe experimental study and statistical analysis done for allsituations a conclusion has been made where the proposedRFID system with automated switching algorithm is betterthan the existing RFID system in terms of performance andvariability of the factors The battery life testing also gives

International Journal of Distributed Sensor Networks 13

Power

Range

Interactions plot

2 10 5 100

minus30

minus45

minus60

minus30

minus45

minus60

minus30

minus45

minus60

Range

Power

Type of RFID

210

5100

Type of RFID

E

N

EN

Figure 13 Interaction plot for power level versus range measurement

Power

Mea

nM

ean

Type RFID

Range

Main effects plot for RSSI

minus40

minus50

minus60

minus40

minus50

minus60

2 10

5 100

EN

Figure 14 Main effects plot for power level versus range measurement

significant findings where the proposed embedded RFIDtag with switching mechanism introduced is much betterthan embedded RFID tag without switching mechanismTo increase the performance of the embedded RFID tag interms of time of usage an energy harvesting technology willbe introduced later as a part of our next contribution Interms of hardware costing the proposed approach is moreacceptable compared to the system developed by [6] since thetechnology used in this work especially GSM communicationis much cheaper than satellite communication used in the

existing system [6] In addition this proposed system doesnot totally depend on GSM since the communication willbe done by RFID which is free as long as the embeddedRFID tag is working in the control area of WSN platformHowever the existing work [6] did not support long rangecommunication and thus when the tag is moving far fromthe reader the satellite communication needs to be used andthis will lead to high cost of communication As a conclusionthe proposed embedded RFID tag provided better signalpropagation compared to the existing RFID tag [19] and

14 International Journal of Distributed Sensor Networks

00 1 2 3 4 5 6 7 8 9

Batte

ry re

duct

ion

()

Time (hours)

Battery lifetime testing

Embedded RFID tag with switchingEmbedded RFID tag without switchingStandalone RFID tag

minus5

minus10

minus15

minus20

minus25

minus30

Figure 15 Battery life testing based on three types of RFIDdesigned

the battery lifetime for embedded RFID tag with switchingmechanism is better than that without switchingmechanism

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank Malaysia Ministry of HigherEducation (LRGS Fund) for sponsoring the research anddeveloping this project Also special appreciation is due toProfessor DrMohamadKamal Abdul Rahim fromUniversitiTeknologiMalaysia Head of the LRGS Project and AssociateProfessor Dr Alyani Ismail Universiti PutraMalaysia for thesupport

References

[1] DIGI XBeeXBee PRO ZB RF Module Datasheet 2013[2] J B Howard ldquoCombination RFID and GPS functionality on

intelligent labelrdquo WO 2003050960 A2 2003[3] S Chand V R Bapat K H Hall et al ldquoPortable RFID reader

having a location determination systemrdquo EP 1752908 A220062006

[4] G O Garsia S I M Nunez L A Saavedra and M D I TapialdquoReal time location system (RTLS) with ZigBee technologyrdquo EP2196816 A12008 2008

[5] A Bandyopadhyay D Hakim B E Funk E A Kohn C AToelis and G Blankenship ldquoSystem and method for locatingtracking andor monitoring the status of personnel andorassets both indoorsrdquo US 20090043504 A1 2009

[6] RFID Journal ldquoHybrid tags includes active RFID GPS satelliteand sensorrdquo RFID Journal 2009 httpwwwrfidjournalcomarticlesview4635

[7] M Mohandes M Haleem M Deriche and K BalakrishnanldquoWireless sensor networks for pilgrims trackingrdquo IEEE Embed-ded Systems Letters vol 4 no 4 pp 106ndash109 2012

[8] N D Simoes J L Goncalves M L Caeiro M J Boavida andF D Cardoso ldquoZigBeeGPS tracking system for rowing racesrdquoin Proceedings of the International Conference on Computer as aTool (EUROCON rsquo11) Lisbon Portugal April 2011

[9] Y Xu R Jiang S Yan and D Xiong ldquoThe research of safetymonitoring system applied in school bus based on the internetof thingsrdquo Procedia Engineering vol 15 pp 2464ndash2468 2011

[10] L Zheng M Li C Wu et al ldquoDevelopment of a smartmobile farming service systemrdquo Mathematical and ComputerModelling vol 54 no 3-4 pp 1194ndash1203 2011

[11] F A Poad and W Ismail ldquoAutomated switching mechanismfor indoor and outdoor propagation with embedded RFID andGPS in wireless sensor network platformrdquo in Proceedings ofthe International Conference on Wireless Network pp 710ndash714London UK 2014

[12] J OrsquoToole J Tuttle M Tuttle et al ldquoRadio frequency identifi-cation sensor tag apparatusesrdquo US 20080030306 A1 2008

[13] V M Gupta and S Annambhotla ldquoRFID tracker and locatorrdquoUS 20090085745 A1 2009

[14] F Shebli I Dayoub and J M Rouvaen ldquoMinimizing energyconsumptionwithinwireless sensor networkrdquoUbiquitous Com-puting and Communication Journal vol 87 2007

[15] R Ding J Hou and B Xing ldquoResearch of wireless sensornetwork nodes based on ambient energy harvestingrdquo in Pro-ceedings of the 6th IEEE International Conference on IntelligentNetworks and Intelligent Systems (ICINIS rsquo13) pp 286ndash288Shenyang China November 2013

[16] A Cammarano C Petrioli andD Spenza ldquoPro-Energy a novelenergy prediction model for solar and wind energy-harvestingwireless sensor networksrdquo in Proceedings of the 9th IEEEInternational Conference on Mobile Ad-Hoc and Sensor Systems(MASS rsquo12) pp 75ndash83 Las Vegas Nev USA October 2012

[17] J Song and Y K Tan ldquoEnergy consumption analysis of ZigBee-based energy harvesting wireless sensor networksrdquo in Proceed-ings of the IEEE International Conference on CommunicationSystems (ICCS rsquo12) pp 468ndash472 November 2012

[18] San Jose Technology Inc Datasheet GPS Engine Board ModelFV M8 2014 httpswwwsparkfuncomdatasheetsGPSFV-M8 Specpdf

[19] W Ismail and R Abdulla ldquoPortable RFID Reader for RTLS(Real Time Location Systems)rdquo PCT International Filing NoPCTMY2011000167 Country Australia Pub No WO2012102600 2012

[20] W Ismail C Z Zulkifli and M G Rahman ldquoWireless pro-duction monitoring systemrdquo PCT International Filing PCTMY2013000037 2013

[21] ISOIEC 18000-4 Information Technology-Radio FrequencyIdentification for ItemManagement-Part7 Parameters for ActiveAir Interface Communications at 245 GHz ISO 18000-42008International Organization for Standardization GenevaSwitzerland 2008

[22] D-H Shih P-L Sun D C Yen and S-M Huang ldquoTaxonomyand survey of RFID anti-collision protocolsrdquo Computer Com-munications vol 29 no 11 pp 2150ndash2166 2006

[23] B Nilsson L Bengtsson P-A Wiberg and B SvenssonldquoProtocols for active RFIDmdashthe energy consumption aspectrdquo inProceedings of the IEEE 2nd International Symposium on Indus-trial Embedded Systems (SIES rsquo07) pp 41ndash48 July 2007

[24] H L van Eeden ldquoWireless Communication Systemrdquo US Patent0045923 2009

International Journal of Distributed Sensor Networks 15

[25] H Haar K Bischof and H Scharke ldquoRFID transponderrdquo WOPatent 109212 2006

[26] C A Balanis AntennaTheory Analysis and Design JohnWileyamp Sons New York NY USA 3rd edition 2005

[27] P Sorrells Optimizing Read Range in RFID System MicrochipTechnology Inc 2000 httpedncomdesignwireless-network-ing4341980Optimizing-read-range-in-RFID-systems

[28] A Ammu L Mapa and A H Jayatissa ldquoEffect of factors onRFID tag readability-statistical analysisrdquo in Proceedings of theIEEE International Conference on ElectroInformation Technol-ogy (EIT rsquo09) pp 355ndash358 June 2009

[29] R Zurawski Embedded System Handbook Second EditionNetwork Embedded Systems CRC Press 2009

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

Page 10: Research Article Propagation Analysis for Automated ...downloads.hindawi.com/journals/ijdsn/2015/392385.pdf · Hybrid tag includes active RFID, GPS, satellite, and sensors [ ] MHz

10 International Journal of Distributed Sensor Networks

210

Range

090

Angle

Polarization

Type of RFID

Interactions plot

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

minus32

minus40

minus48

2 10

0 90

Range

Angle

Polarization

Type of RFID

E

E

H

H

N

P

N P

Figure 9 Interaction plot for radiation measurement

Range

Mea

nM

ean

Angle

Polarization Type of RFID

Main effects plot for RSSIData means

minus36

minus39

minus42

minus45

minus48

minus36

minus39

minus42

minus45

minus48

2 10 0 90

E H N P

Figure 10 Main effects plot for radiation measurement

the mean Figures 11 and 12 show the comparison betweenRSS values of existing and proposed RFID system based ondistance of 5 meters to 100 meters indoors at 10 dBm and2 dBm power levels The graph shows that the RSS valuesare decreased when the distance is increased at both powerlevels The existing RFID system gives better RSS valuesthan proposed RFID system at 10 dBm power level while the

proposed RFID system gives better RSS values than existingRFID system at 2 dBm power level To compare the perfor-mance of two types of RFID system based on power leveland range the third experiment is conducted based on DOEmethod with three factors having two levels of treatment 23The factors are referred to as type of RFID (nonembeddedembedded) power level (2 dBm 10 dBm) and range (5m

International Journal of Distributed Sensor Networks 11

00 20 40 60 80 100 120

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

RSSI

(dBm

)

Figure 11 Received signal strength values at 10 dBm in an indoorenvironment

00 20 40 60 80 100 120

RSSI

(dBm

)

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

minus80

Figure 12 Received signal strength values at 2 dBm in an indoorenvironment

10m) The response of the experiment is RSS and 2 replicatesare used in this work to reduce the measurement errorsand noise The order in which the observations are takenis selected randomly so that the design is a completelyrandomized design Later the statistical method ANOVA isperformed using Minitab 162 and the results are displayedin Tables 4 and 5 for effect estimate summary and ANOVAfor power level versus range measurement From the effectestimate summary presented in Table 4 the main effect of Creally dominates the process accounting for 91381 percentof the total variability whereas the main effect of A and theAC interaction account for about 6144 and 1727 percentrespectively thus the DOE for this experiment also provedthat the RSS value can be used as a prediction method tomeasure distance and location of tags from reader in wireless

sensor network platform since the range contributions arehighly significant compared to other factors The ANOVAin Table 4 is used to confirm the magnitude of these effectsFrom Table 5 the main effects of A B and C are highlysignificant since the factors have very small119875 values (lt0001)For 2-way interaction the AB and AC interaction are alsohighly significant (low 119875 values) thus it can be concludedthat there is a strong interaction between two factors withRSS values in this experimental study Figures 13 and 14 showthe interaction and effects plot for power level versus rangemeasurement that is used to interpret the results from Tables4 and 5 It can be seen that theRSS for embeddedRFID systemis nearly equal to nonembedded RFID system at 10 dBmpower level However at 2 dBm power level the RSS forembedded RFID system is better than nonembedded RFIDsystem In terms of range the RSS for embeddedRFID systemis better compared to nonembedded RFID system at both 5-meter and 100-meter range Based on the results obtainedit can be summarized that for indoor tracking purposesthe low power level gives better RSS values for embeddedRFID system compared to nonembedded RFID system Thisfinding can be supported by [29] which stated that the lowtransmit power is appropriate and sufficient for area withlimited range however if the high transmit power is use inthe area with limited range it will resulting in disturbancethat will cause degradation in radiation performanceThus todesign an efficient indoor localization system these findingsmust be taken into account as a guideline implementing thefull setup for real time location monitoring system in WSNplatform

5 Battery Life Testing

A single RFID tag requires a minimum of 27 V to 33 Vvoltage to operate as a transmitter and receiver in thenetwork Howver the embedded RFID tag proposed in thiswork requires minimum of 50V to power up the embeddeddevices Thus the embedded RFID tag can be powered upusing 9 VDC power supply if the tag is located at fixedlocation or Li-Po rechargeable batteries can be used if thetag is mobile The tag has been programmed to consumethe lowest amount of energy as much as possible sinceit introduced an automated switching mechanism betweenindoor and outdoor location tracking and operating in sleepmode at certain timeTheRFID tag only sends data if the validoutdoor location is detected by the GPS receiver however ifthe embedded RFID tag is located inside the building or theplace where the signal from satellite cannot be detected bythe GPS receiver it will turn to sleep mode and wait for theRFID reader interrogation every hourTheGSMmodemonlyworks when the embedded RFID tag is out of the wirelessnetwork coverage to consume the lowest amount of energy

The battery lifetime measurement is performed for threesystems which are stand-alone RFID tag (A) [20] embeddedRFID tag with switching mechanism (B) and embeddedRFID tag without switching mechanism (C) in an indoorenvironment (inside building) where the GPS signals are notavailable most of the time The results show that the battery

12 International Journal of Distributed Sensor Networks

Table 4 Effect estimate summary

Term constant Effect Sum of squares Percent contributionPower (119860) 724 209530 6144Type of RFID (119861) 179 12780 0375Range (119862) minus2791 3116430 91381Power lowast type of RFID (119860119861) minus141 7980 0234Power lowast range (119860119862) 384 58910 1727Type RFID lowast range (119861119862) 074 2180 0064Power lowast type of RFID lowast range (119860119861119862) minus056 1270 00372

Table 5 ANOVA for power level versus range measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Power (119860) 1 209530 209530 1294370 0000Type of RFID (119861) 1 12780 12780 78950 0000Range (119862) 1 3116430 3116430 19252080 0000Power lowast type of RFID (119860119861) 1 7980 7980 49300 0000Power lowast range (119860119862) 1 58910 58910 363900 0000Type RFID lowast range (119861119862) 1 2180 2180 13440 0006Power lowast type of RFID lowast range (119860119861119862) 1 1270 1270 7820 0023Residual error 8 1290 0160Pure error 8 1300 0160Total 15 3410360

for the tag drops from 824V to 711 V in the duration of 85hours for the proposed embedded RFID tag with switchingmechanism compared to other systems as shown in Figure 15

The results for system B are much better than systemC due to the switching introduced However the result forsystem A is better than system B due to the nonembeddedsystem According to these real implementation results thetag will be expected to stop operating approximately in24 hours when the battery reaches the minimum voltagerequired which is 50 V However this result does not includethe situation where the proposed embedded RFID tag islocated outside building which might increase the energyusage The estimation results also did not include otherexternal factors that might reduce the battery life Thus thetag is estimated to cover the operation for a day withouthaving to recharge the battery until the next day To reducethe energy consumption and maximize the time of usage anenergy harvesting technologywill be implemented later in theembedded RFID tag as our next target

6 Conclusion

This paper describes the implementation of an automatedswitching algorithm in the design of a 245GHz active inte-grated RFID system for indoor and outdoor location trackingutilizing WSN platform Four experimental measurementsbased on DOE method are performed in order to validatethe performance of the proposed RFID system as well as tostudy the relationship interaction and effects of each factor

to RSS values extracted from existing and proposed RFIDsystem The results obtained from first experiment show thatembedded RFID system does affect the radiation patternnot more than 15 differences from existing RFID system Itshows that the proposed RFID system contributes to strongerradiation compared to existing RFID system This may bedue to the embedded design and improvement of algorithmimplemented in the proposed RFID system which increasedthe power stability of the system which thus provided highersignal strength [28]

In the second experiment the results show that the pro-posed RFID tag has better RSS value when it is located nearerto the reader at 0∘ degrees direction of propagation with H-field polarization of antenna In the third experiment theexisting RFID system gives better RSS values than proposedRFID system at 10 dBm power level while the proposedRFID system gives better RSS values than existing RFIDsystem at 2 dBm power level In the fourth experiment it canbe summarized that for indoor tracking purposes the lowpower level gives better RSS values for the proposed RFIDtag compared to the existing RFID system The statisticalanalysis ANOVA proved that the RSS values are significantwith range thus it can be concluded that the RSS can beused as a prediction method to evaluate distance betweenRFID reader and RFID tag in WSN platform Based onthe experimental study and statistical analysis done for allsituations a conclusion has been made where the proposedRFID system with automated switching algorithm is betterthan the existing RFID system in terms of performance andvariability of the factors The battery life testing also gives

International Journal of Distributed Sensor Networks 13

Power

Range

Interactions plot

2 10 5 100

minus30

minus45

minus60

minus30

minus45

minus60

minus30

minus45

minus60

Range

Power

Type of RFID

210

5100

Type of RFID

E

N

EN

Figure 13 Interaction plot for power level versus range measurement

Power

Mea

nM

ean

Type RFID

Range

Main effects plot for RSSI

minus40

minus50

minus60

minus40

minus50

minus60

2 10

5 100

EN

Figure 14 Main effects plot for power level versus range measurement

significant findings where the proposed embedded RFIDtag with switching mechanism introduced is much betterthan embedded RFID tag without switching mechanismTo increase the performance of the embedded RFID tag interms of time of usage an energy harvesting technology willbe introduced later as a part of our next contribution Interms of hardware costing the proposed approach is moreacceptable compared to the system developed by [6] since thetechnology used in this work especially GSM communicationis much cheaper than satellite communication used in the

existing system [6] In addition this proposed system doesnot totally depend on GSM since the communication willbe done by RFID which is free as long as the embeddedRFID tag is working in the control area of WSN platformHowever the existing work [6] did not support long rangecommunication and thus when the tag is moving far fromthe reader the satellite communication needs to be used andthis will lead to high cost of communication As a conclusionthe proposed embedded RFID tag provided better signalpropagation compared to the existing RFID tag [19] and

14 International Journal of Distributed Sensor Networks

00 1 2 3 4 5 6 7 8 9

Batte

ry re

duct

ion

()

Time (hours)

Battery lifetime testing

Embedded RFID tag with switchingEmbedded RFID tag without switchingStandalone RFID tag

minus5

minus10

minus15

minus20

minus25

minus30

Figure 15 Battery life testing based on three types of RFIDdesigned

the battery lifetime for embedded RFID tag with switchingmechanism is better than that without switchingmechanism

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank Malaysia Ministry of HigherEducation (LRGS Fund) for sponsoring the research anddeveloping this project Also special appreciation is due toProfessor DrMohamadKamal Abdul Rahim fromUniversitiTeknologiMalaysia Head of the LRGS Project and AssociateProfessor Dr Alyani Ismail Universiti PutraMalaysia for thesupport

References

[1] DIGI XBeeXBee PRO ZB RF Module Datasheet 2013[2] J B Howard ldquoCombination RFID and GPS functionality on

intelligent labelrdquo WO 2003050960 A2 2003[3] S Chand V R Bapat K H Hall et al ldquoPortable RFID reader

having a location determination systemrdquo EP 1752908 A220062006

[4] G O Garsia S I M Nunez L A Saavedra and M D I TapialdquoReal time location system (RTLS) with ZigBee technologyrdquo EP2196816 A12008 2008

[5] A Bandyopadhyay D Hakim B E Funk E A Kohn C AToelis and G Blankenship ldquoSystem and method for locatingtracking andor monitoring the status of personnel andorassets both indoorsrdquo US 20090043504 A1 2009

[6] RFID Journal ldquoHybrid tags includes active RFID GPS satelliteand sensorrdquo RFID Journal 2009 httpwwwrfidjournalcomarticlesview4635

[7] M Mohandes M Haleem M Deriche and K BalakrishnanldquoWireless sensor networks for pilgrims trackingrdquo IEEE Embed-ded Systems Letters vol 4 no 4 pp 106ndash109 2012

[8] N D Simoes J L Goncalves M L Caeiro M J Boavida andF D Cardoso ldquoZigBeeGPS tracking system for rowing racesrdquoin Proceedings of the International Conference on Computer as aTool (EUROCON rsquo11) Lisbon Portugal April 2011

[9] Y Xu R Jiang S Yan and D Xiong ldquoThe research of safetymonitoring system applied in school bus based on the internetof thingsrdquo Procedia Engineering vol 15 pp 2464ndash2468 2011

[10] L Zheng M Li C Wu et al ldquoDevelopment of a smartmobile farming service systemrdquo Mathematical and ComputerModelling vol 54 no 3-4 pp 1194ndash1203 2011

[11] F A Poad and W Ismail ldquoAutomated switching mechanismfor indoor and outdoor propagation with embedded RFID andGPS in wireless sensor network platformrdquo in Proceedings ofthe International Conference on Wireless Network pp 710ndash714London UK 2014

[12] J OrsquoToole J Tuttle M Tuttle et al ldquoRadio frequency identifi-cation sensor tag apparatusesrdquo US 20080030306 A1 2008

[13] V M Gupta and S Annambhotla ldquoRFID tracker and locatorrdquoUS 20090085745 A1 2009

[14] F Shebli I Dayoub and J M Rouvaen ldquoMinimizing energyconsumptionwithinwireless sensor networkrdquoUbiquitous Com-puting and Communication Journal vol 87 2007

[15] R Ding J Hou and B Xing ldquoResearch of wireless sensornetwork nodes based on ambient energy harvestingrdquo in Pro-ceedings of the 6th IEEE International Conference on IntelligentNetworks and Intelligent Systems (ICINIS rsquo13) pp 286ndash288Shenyang China November 2013

[16] A Cammarano C Petrioli andD Spenza ldquoPro-Energy a novelenergy prediction model for solar and wind energy-harvestingwireless sensor networksrdquo in Proceedings of the 9th IEEEInternational Conference on Mobile Ad-Hoc and Sensor Systems(MASS rsquo12) pp 75ndash83 Las Vegas Nev USA October 2012

[17] J Song and Y K Tan ldquoEnergy consumption analysis of ZigBee-based energy harvesting wireless sensor networksrdquo in Proceed-ings of the IEEE International Conference on CommunicationSystems (ICCS rsquo12) pp 468ndash472 November 2012

[18] San Jose Technology Inc Datasheet GPS Engine Board ModelFV M8 2014 httpswwwsparkfuncomdatasheetsGPSFV-M8 Specpdf

[19] W Ismail and R Abdulla ldquoPortable RFID Reader for RTLS(Real Time Location Systems)rdquo PCT International Filing NoPCTMY2011000167 Country Australia Pub No WO2012102600 2012

[20] W Ismail C Z Zulkifli and M G Rahman ldquoWireless pro-duction monitoring systemrdquo PCT International Filing PCTMY2013000037 2013

[21] ISOIEC 18000-4 Information Technology-Radio FrequencyIdentification for ItemManagement-Part7 Parameters for ActiveAir Interface Communications at 245 GHz ISO 18000-42008International Organization for Standardization GenevaSwitzerland 2008

[22] D-H Shih P-L Sun D C Yen and S-M Huang ldquoTaxonomyand survey of RFID anti-collision protocolsrdquo Computer Com-munications vol 29 no 11 pp 2150ndash2166 2006

[23] B Nilsson L Bengtsson P-A Wiberg and B SvenssonldquoProtocols for active RFIDmdashthe energy consumption aspectrdquo inProceedings of the IEEE 2nd International Symposium on Indus-trial Embedded Systems (SIES rsquo07) pp 41ndash48 July 2007

[24] H L van Eeden ldquoWireless Communication Systemrdquo US Patent0045923 2009

International Journal of Distributed Sensor Networks 15

[25] H Haar K Bischof and H Scharke ldquoRFID transponderrdquo WOPatent 109212 2006

[26] C A Balanis AntennaTheory Analysis and Design JohnWileyamp Sons New York NY USA 3rd edition 2005

[27] P Sorrells Optimizing Read Range in RFID System MicrochipTechnology Inc 2000 httpedncomdesignwireless-network-ing4341980Optimizing-read-range-in-RFID-systems

[28] A Ammu L Mapa and A H Jayatissa ldquoEffect of factors onRFID tag readability-statistical analysisrdquo in Proceedings of theIEEE International Conference on ElectroInformation Technol-ogy (EIT rsquo09) pp 355ndash358 June 2009

[29] R Zurawski Embedded System Handbook Second EditionNetwork Embedded Systems CRC Press 2009

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

Page 11: Research Article Propagation Analysis for Automated ...downloads.hindawi.com/journals/ijdsn/2015/392385.pdf · Hybrid tag includes active RFID, GPS, satellite, and sensors [ ] MHz

International Journal of Distributed Sensor Networks 11

00 20 40 60 80 100 120

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

RSSI

(dBm

)

Figure 11 Received signal strength values at 10 dBm in an indoorenvironment

00 20 40 60 80 100 120

RSSI

(dBm

)

Distance (m)

ExistingProposed

minus10

minus20

minus30

minus40

minus50

minus60

minus70

minus80

Figure 12 Received signal strength values at 2 dBm in an indoorenvironment

10m) The response of the experiment is RSS and 2 replicatesare used in this work to reduce the measurement errorsand noise The order in which the observations are takenis selected randomly so that the design is a completelyrandomized design Later the statistical method ANOVA isperformed using Minitab 162 and the results are displayedin Tables 4 and 5 for effect estimate summary and ANOVAfor power level versus range measurement From the effectestimate summary presented in Table 4 the main effect of Creally dominates the process accounting for 91381 percentof the total variability whereas the main effect of A and theAC interaction account for about 6144 and 1727 percentrespectively thus the DOE for this experiment also provedthat the RSS value can be used as a prediction method tomeasure distance and location of tags from reader in wireless

sensor network platform since the range contributions arehighly significant compared to other factors The ANOVAin Table 4 is used to confirm the magnitude of these effectsFrom Table 5 the main effects of A B and C are highlysignificant since the factors have very small119875 values (lt0001)For 2-way interaction the AB and AC interaction are alsohighly significant (low 119875 values) thus it can be concludedthat there is a strong interaction between two factors withRSS values in this experimental study Figures 13 and 14 showthe interaction and effects plot for power level versus rangemeasurement that is used to interpret the results from Tables4 and 5 It can be seen that theRSS for embeddedRFID systemis nearly equal to nonembedded RFID system at 10 dBmpower level However at 2 dBm power level the RSS forembedded RFID system is better than nonembedded RFIDsystem In terms of range the RSS for embeddedRFID systemis better compared to nonembedded RFID system at both 5-meter and 100-meter range Based on the results obtainedit can be summarized that for indoor tracking purposesthe low power level gives better RSS values for embeddedRFID system compared to nonembedded RFID system Thisfinding can be supported by [29] which stated that the lowtransmit power is appropriate and sufficient for area withlimited range however if the high transmit power is use inthe area with limited range it will resulting in disturbancethat will cause degradation in radiation performanceThus todesign an efficient indoor localization system these findingsmust be taken into account as a guideline implementing thefull setup for real time location monitoring system in WSNplatform

5 Battery Life Testing

A single RFID tag requires a minimum of 27 V to 33 Vvoltage to operate as a transmitter and receiver in thenetwork Howver the embedded RFID tag proposed in thiswork requires minimum of 50V to power up the embeddeddevices Thus the embedded RFID tag can be powered upusing 9 VDC power supply if the tag is located at fixedlocation or Li-Po rechargeable batteries can be used if thetag is mobile The tag has been programmed to consumethe lowest amount of energy as much as possible sinceit introduced an automated switching mechanism betweenindoor and outdoor location tracking and operating in sleepmode at certain timeTheRFID tag only sends data if the validoutdoor location is detected by the GPS receiver however ifthe embedded RFID tag is located inside the building or theplace where the signal from satellite cannot be detected bythe GPS receiver it will turn to sleep mode and wait for theRFID reader interrogation every hourTheGSMmodemonlyworks when the embedded RFID tag is out of the wirelessnetwork coverage to consume the lowest amount of energy

The battery lifetime measurement is performed for threesystems which are stand-alone RFID tag (A) [20] embeddedRFID tag with switching mechanism (B) and embeddedRFID tag without switching mechanism (C) in an indoorenvironment (inside building) where the GPS signals are notavailable most of the time The results show that the battery

12 International Journal of Distributed Sensor Networks

Table 4 Effect estimate summary

Term constant Effect Sum of squares Percent contributionPower (119860) 724 209530 6144Type of RFID (119861) 179 12780 0375Range (119862) minus2791 3116430 91381Power lowast type of RFID (119860119861) minus141 7980 0234Power lowast range (119860119862) 384 58910 1727Type RFID lowast range (119861119862) 074 2180 0064Power lowast type of RFID lowast range (119860119861119862) minus056 1270 00372

Table 5 ANOVA for power level versus range measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Power (119860) 1 209530 209530 1294370 0000Type of RFID (119861) 1 12780 12780 78950 0000Range (119862) 1 3116430 3116430 19252080 0000Power lowast type of RFID (119860119861) 1 7980 7980 49300 0000Power lowast range (119860119862) 1 58910 58910 363900 0000Type RFID lowast range (119861119862) 1 2180 2180 13440 0006Power lowast type of RFID lowast range (119860119861119862) 1 1270 1270 7820 0023Residual error 8 1290 0160Pure error 8 1300 0160Total 15 3410360

for the tag drops from 824V to 711 V in the duration of 85hours for the proposed embedded RFID tag with switchingmechanism compared to other systems as shown in Figure 15

The results for system B are much better than systemC due to the switching introduced However the result forsystem A is better than system B due to the nonembeddedsystem According to these real implementation results thetag will be expected to stop operating approximately in24 hours when the battery reaches the minimum voltagerequired which is 50 V However this result does not includethe situation where the proposed embedded RFID tag islocated outside building which might increase the energyusage The estimation results also did not include otherexternal factors that might reduce the battery life Thus thetag is estimated to cover the operation for a day withouthaving to recharge the battery until the next day To reducethe energy consumption and maximize the time of usage anenergy harvesting technologywill be implemented later in theembedded RFID tag as our next target

6 Conclusion

This paper describes the implementation of an automatedswitching algorithm in the design of a 245GHz active inte-grated RFID system for indoor and outdoor location trackingutilizing WSN platform Four experimental measurementsbased on DOE method are performed in order to validatethe performance of the proposed RFID system as well as tostudy the relationship interaction and effects of each factor

to RSS values extracted from existing and proposed RFIDsystem The results obtained from first experiment show thatembedded RFID system does affect the radiation patternnot more than 15 differences from existing RFID system Itshows that the proposed RFID system contributes to strongerradiation compared to existing RFID system This may bedue to the embedded design and improvement of algorithmimplemented in the proposed RFID system which increasedthe power stability of the system which thus provided highersignal strength [28]

In the second experiment the results show that the pro-posed RFID tag has better RSS value when it is located nearerto the reader at 0∘ degrees direction of propagation with H-field polarization of antenna In the third experiment theexisting RFID system gives better RSS values than proposedRFID system at 10 dBm power level while the proposedRFID system gives better RSS values than existing RFIDsystem at 2 dBm power level In the fourth experiment it canbe summarized that for indoor tracking purposes the lowpower level gives better RSS values for the proposed RFIDtag compared to the existing RFID system The statisticalanalysis ANOVA proved that the RSS values are significantwith range thus it can be concluded that the RSS can beused as a prediction method to evaluate distance betweenRFID reader and RFID tag in WSN platform Based onthe experimental study and statistical analysis done for allsituations a conclusion has been made where the proposedRFID system with automated switching algorithm is betterthan the existing RFID system in terms of performance andvariability of the factors The battery life testing also gives

International Journal of Distributed Sensor Networks 13

Power

Range

Interactions plot

2 10 5 100

minus30

minus45

minus60

minus30

minus45

minus60

minus30

minus45

minus60

Range

Power

Type of RFID

210

5100

Type of RFID

E

N

EN

Figure 13 Interaction plot for power level versus range measurement

Power

Mea

nM

ean

Type RFID

Range

Main effects plot for RSSI

minus40

minus50

minus60

minus40

minus50

minus60

2 10

5 100

EN

Figure 14 Main effects plot for power level versus range measurement

significant findings where the proposed embedded RFIDtag with switching mechanism introduced is much betterthan embedded RFID tag without switching mechanismTo increase the performance of the embedded RFID tag interms of time of usage an energy harvesting technology willbe introduced later as a part of our next contribution Interms of hardware costing the proposed approach is moreacceptable compared to the system developed by [6] since thetechnology used in this work especially GSM communicationis much cheaper than satellite communication used in the

existing system [6] In addition this proposed system doesnot totally depend on GSM since the communication willbe done by RFID which is free as long as the embeddedRFID tag is working in the control area of WSN platformHowever the existing work [6] did not support long rangecommunication and thus when the tag is moving far fromthe reader the satellite communication needs to be used andthis will lead to high cost of communication As a conclusionthe proposed embedded RFID tag provided better signalpropagation compared to the existing RFID tag [19] and

14 International Journal of Distributed Sensor Networks

00 1 2 3 4 5 6 7 8 9

Batte

ry re

duct

ion

()

Time (hours)

Battery lifetime testing

Embedded RFID tag with switchingEmbedded RFID tag without switchingStandalone RFID tag

minus5

minus10

minus15

minus20

minus25

minus30

Figure 15 Battery life testing based on three types of RFIDdesigned

the battery lifetime for embedded RFID tag with switchingmechanism is better than that without switchingmechanism

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank Malaysia Ministry of HigherEducation (LRGS Fund) for sponsoring the research anddeveloping this project Also special appreciation is due toProfessor DrMohamadKamal Abdul Rahim fromUniversitiTeknologiMalaysia Head of the LRGS Project and AssociateProfessor Dr Alyani Ismail Universiti PutraMalaysia for thesupport

References

[1] DIGI XBeeXBee PRO ZB RF Module Datasheet 2013[2] J B Howard ldquoCombination RFID and GPS functionality on

intelligent labelrdquo WO 2003050960 A2 2003[3] S Chand V R Bapat K H Hall et al ldquoPortable RFID reader

having a location determination systemrdquo EP 1752908 A220062006

[4] G O Garsia S I M Nunez L A Saavedra and M D I TapialdquoReal time location system (RTLS) with ZigBee technologyrdquo EP2196816 A12008 2008

[5] A Bandyopadhyay D Hakim B E Funk E A Kohn C AToelis and G Blankenship ldquoSystem and method for locatingtracking andor monitoring the status of personnel andorassets both indoorsrdquo US 20090043504 A1 2009

[6] RFID Journal ldquoHybrid tags includes active RFID GPS satelliteand sensorrdquo RFID Journal 2009 httpwwwrfidjournalcomarticlesview4635

[7] M Mohandes M Haleem M Deriche and K BalakrishnanldquoWireless sensor networks for pilgrims trackingrdquo IEEE Embed-ded Systems Letters vol 4 no 4 pp 106ndash109 2012

[8] N D Simoes J L Goncalves M L Caeiro M J Boavida andF D Cardoso ldquoZigBeeGPS tracking system for rowing racesrdquoin Proceedings of the International Conference on Computer as aTool (EUROCON rsquo11) Lisbon Portugal April 2011

[9] Y Xu R Jiang S Yan and D Xiong ldquoThe research of safetymonitoring system applied in school bus based on the internetof thingsrdquo Procedia Engineering vol 15 pp 2464ndash2468 2011

[10] L Zheng M Li C Wu et al ldquoDevelopment of a smartmobile farming service systemrdquo Mathematical and ComputerModelling vol 54 no 3-4 pp 1194ndash1203 2011

[11] F A Poad and W Ismail ldquoAutomated switching mechanismfor indoor and outdoor propagation with embedded RFID andGPS in wireless sensor network platformrdquo in Proceedings ofthe International Conference on Wireless Network pp 710ndash714London UK 2014

[12] J OrsquoToole J Tuttle M Tuttle et al ldquoRadio frequency identifi-cation sensor tag apparatusesrdquo US 20080030306 A1 2008

[13] V M Gupta and S Annambhotla ldquoRFID tracker and locatorrdquoUS 20090085745 A1 2009

[14] F Shebli I Dayoub and J M Rouvaen ldquoMinimizing energyconsumptionwithinwireless sensor networkrdquoUbiquitous Com-puting and Communication Journal vol 87 2007

[15] R Ding J Hou and B Xing ldquoResearch of wireless sensornetwork nodes based on ambient energy harvestingrdquo in Pro-ceedings of the 6th IEEE International Conference on IntelligentNetworks and Intelligent Systems (ICINIS rsquo13) pp 286ndash288Shenyang China November 2013

[16] A Cammarano C Petrioli andD Spenza ldquoPro-Energy a novelenergy prediction model for solar and wind energy-harvestingwireless sensor networksrdquo in Proceedings of the 9th IEEEInternational Conference on Mobile Ad-Hoc and Sensor Systems(MASS rsquo12) pp 75ndash83 Las Vegas Nev USA October 2012

[17] J Song and Y K Tan ldquoEnergy consumption analysis of ZigBee-based energy harvesting wireless sensor networksrdquo in Proceed-ings of the IEEE International Conference on CommunicationSystems (ICCS rsquo12) pp 468ndash472 November 2012

[18] San Jose Technology Inc Datasheet GPS Engine Board ModelFV M8 2014 httpswwwsparkfuncomdatasheetsGPSFV-M8 Specpdf

[19] W Ismail and R Abdulla ldquoPortable RFID Reader for RTLS(Real Time Location Systems)rdquo PCT International Filing NoPCTMY2011000167 Country Australia Pub No WO2012102600 2012

[20] W Ismail C Z Zulkifli and M G Rahman ldquoWireless pro-duction monitoring systemrdquo PCT International Filing PCTMY2013000037 2013

[21] ISOIEC 18000-4 Information Technology-Radio FrequencyIdentification for ItemManagement-Part7 Parameters for ActiveAir Interface Communications at 245 GHz ISO 18000-42008International Organization for Standardization GenevaSwitzerland 2008

[22] D-H Shih P-L Sun D C Yen and S-M Huang ldquoTaxonomyand survey of RFID anti-collision protocolsrdquo Computer Com-munications vol 29 no 11 pp 2150ndash2166 2006

[23] B Nilsson L Bengtsson P-A Wiberg and B SvenssonldquoProtocols for active RFIDmdashthe energy consumption aspectrdquo inProceedings of the IEEE 2nd International Symposium on Indus-trial Embedded Systems (SIES rsquo07) pp 41ndash48 July 2007

[24] H L van Eeden ldquoWireless Communication Systemrdquo US Patent0045923 2009

International Journal of Distributed Sensor Networks 15

[25] H Haar K Bischof and H Scharke ldquoRFID transponderrdquo WOPatent 109212 2006

[26] C A Balanis AntennaTheory Analysis and Design JohnWileyamp Sons New York NY USA 3rd edition 2005

[27] P Sorrells Optimizing Read Range in RFID System MicrochipTechnology Inc 2000 httpedncomdesignwireless-network-ing4341980Optimizing-read-range-in-RFID-systems

[28] A Ammu L Mapa and A H Jayatissa ldquoEffect of factors onRFID tag readability-statistical analysisrdquo in Proceedings of theIEEE International Conference on ElectroInformation Technol-ogy (EIT rsquo09) pp 355ndash358 June 2009

[29] R Zurawski Embedded System Handbook Second EditionNetwork Embedded Systems CRC Press 2009

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

Page 12: Research Article Propagation Analysis for Automated ...downloads.hindawi.com/journals/ijdsn/2015/392385.pdf · Hybrid tag includes active RFID, GPS, satellite, and sensors [ ] MHz

12 International Journal of Distributed Sensor Networks

Table 4 Effect estimate summary

Term constant Effect Sum of squares Percent contributionPower (119860) 724 209530 6144Type of RFID (119861) 179 12780 0375Range (119862) minus2791 3116430 91381Power lowast type of RFID (119860119861) minus141 7980 0234Power lowast range (119860119862) 384 58910 1727Type RFID lowast range (119861119862) 074 2180 0064Power lowast type of RFID lowast range (119860119861119862) minus056 1270 00372

Table 5 ANOVA for power level versus range measurement

Source Degree of freedom Sum of squares Mean of square 119865 119875

Power (119860) 1 209530 209530 1294370 0000Type of RFID (119861) 1 12780 12780 78950 0000Range (119862) 1 3116430 3116430 19252080 0000Power lowast type of RFID (119860119861) 1 7980 7980 49300 0000Power lowast range (119860119862) 1 58910 58910 363900 0000Type RFID lowast range (119861119862) 1 2180 2180 13440 0006Power lowast type of RFID lowast range (119860119861119862) 1 1270 1270 7820 0023Residual error 8 1290 0160Pure error 8 1300 0160Total 15 3410360

for the tag drops from 824V to 711 V in the duration of 85hours for the proposed embedded RFID tag with switchingmechanism compared to other systems as shown in Figure 15

The results for system B are much better than systemC due to the switching introduced However the result forsystem A is better than system B due to the nonembeddedsystem According to these real implementation results thetag will be expected to stop operating approximately in24 hours when the battery reaches the minimum voltagerequired which is 50 V However this result does not includethe situation where the proposed embedded RFID tag islocated outside building which might increase the energyusage The estimation results also did not include otherexternal factors that might reduce the battery life Thus thetag is estimated to cover the operation for a day withouthaving to recharge the battery until the next day To reducethe energy consumption and maximize the time of usage anenergy harvesting technologywill be implemented later in theembedded RFID tag as our next target

6 Conclusion

This paper describes the implementation of an automatedswitching algorithm in the design of a 245GHz active inte-grated RFID system for indoor and outdoor location trackingutilizing WSN platform Four experimental measurementsbased on DOE method are performed in order to validatethe performance of the proposed RFID system as well as tostudy the relationship interaction and effects of each factor

to RSS values extracted from existing and proposed RFIDsystem The results obtained from first experiment show thatembedded RFID system does affect the radiation patternnot more than 15 differences from existing RFID system Itshows that the proposed RFID system contributes to strongerradiation compared to existing RFID system This may bedue to the embedded design and improvement of algorithmimplemented in the proposed RFID system which increasedthe power stability of the system which thus provided highersignal strength [28]

In the second experiment the results show that the pro-posed RFID tag has better RSS value when it is located nearerto the reader at 0∘ degrees direction of propagation with H-field polarization of antenna In the third experiment theexisting RFID system gives better RSS values than proposedRFID system at 10 dBm power level while the proposedRFID system gives better RSS values than existing RFIDsystem at 2 dBm power level In the fourth experiment it canbe summarized that for indoor tracking purposes the lowpower level gives better RSS values for the proposed RFIDtag compared to the existing RFID system The statisticalanalysis ANOVA proved that the RSS values are significantwith range thus it can be concluded that the RSS can beused as a prediction method to evaluate distance betweenRFID reader and RFID tag in WSN platform Based onthe experimental study and statistical analysis done for allsituations a conclusion has been made where the proposedRFID system with automated switching algorithm is betterthan the existing RFID system in terms of performance andvariability of the factors The battery life testing also gives

International Journal of Distributed Sensor Networks 13

Power

Range

Interactions plot

2 10 5 100

minus30

minus45

minus60

minus30

minus45

minus60

minus30

minus45

minus60

Range

Power

Type of RFID

210

5100

Type of RFID

E

N

EN

Figure 13 Interaction plot for power level versus range measurement

Power

Mea

nM

ean

Type RFID

Range

Main effects plot for RSSI

minus40

minus50

minus60

minus40

minus50

minus60

2 10

5 100

EN

Figure 14 Main effects plot for power level versus range measurement

significant findings where the proposed embedded RFIDtag with switching mechanism introduced is much betterthan embedded RFID tag without switching mechanismTo increase the performance of the embedded RFID tag interms of time of usage an energy harvesting technology willbe introduced later as a part of our next contribution Interms of hardware costing the proposed approach is moreacceptable compared to the system developed by [6] since thetechnology used in this work especially GSM communicationis much cheaper than satellite communication used in the

existing system [6] In addition this proposed system doesnot totally depend on GSM since the communication willbe done by RFID which is free as long as the embeddedRFID tag is working in the control area of WSN platformHowever the existing work [6] did not support long rangecommunication and thus when the tag is moving far fromthe reader the satellite communication needs to be used andthis will lead to high cost of communication As a conclusionthe proposed embedded RFID tag provided better signalpropagation compared to the existing RFID tag [19] and

14 International Journal of Distributed Sensor Networks

00 1 2 3 4 5 6 7 8 9

Batte

ry re

duct

ion

()

Time (hours)

Battery lifetime testing

Embedded RFID tag with switchingEmbedded RFID tag without switchingStandalone RFID tag

minus5

minus10

minus15

minus20

minus25

minus30

Figure 15 Battery life testing based on three types of RFIDdesigned

the battery lifetime for embedded RFID tag with switchingmechanism is better than that without switchingmechanism

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank Malaysia Ministry of HigherEducation (LRGS Fund) for sponsoring the research anddeveloping this project Also special appreciation is due toProfessor DrMohamadKamal Abdul Rahim fromUniversitiTeknologiMalaysia Head of the LRGS Project and AssociateProfessor Dr Alyani Ismail Universiti PutraMalaysia for thesupport

References

[1] DIGI XBeeXBee PRO ZB RF Module Datasheet 2013[2] J B Howard ldquoCombination RFID and GPS functionality on

intelligent labelrdquo WO 2003050960 A2 2003[3] S Chand V R Bapat K H Hall et al ldquoPortable RFID reader

having a location determination systemrdquo EP 1752908 A220062006

[4] G O Garsia S I M Nunez L A Saavedra and M D I TapialdquoReal time location system (RTLS) with ZigBee technologyrdquo EP2196816 A12008 2008

[5] A Bandyopadhyay D Hakim B E Funk E A Kohn C AToelis and G Blankenship ldquoSystem and method for locatingtracking andor monitoring the status of personnel andorassets both indoorsrdquo US 20090043504 A1 2009

[6] RFID Journal ldquoHybrid tags includes active RFID GPS satelliteand sensorrdquo RFID Journal 2009 httpwwwrfidjournalcomarticlesview4635

[7] M Mohandes M Haleem M Deriche and K BalakrishnanldquoWireless sensor networks for pilgrims trackingrdquo IEEE Embed-ded Systems Letters vol 4 no 4 pp 106ndash109 2012

[8] N D Simoes J L Goncalves M L Caeiro M J Boavida andF D Cardoso ldquoZigBeeGPS tracking system for rowing racesrdquoin Proceedings of the International Conference on Computer as aTool (EUROCON rsquo11) Lisbon Portugal April 2011

[9] Y Xu R Jiang S Yan and D Xiong ldquoThe research of safetymonitoring system applied in school bus based on the internetof thingsrdquo Procedia Engineering vol 15 pp 2464ndash2468 2011

[10] L Zheng M Li C Wu et al ldquoDevelopment of a smartmobile farming service systemrdquo Mathematical and ComputerModelling vol 54 no 3-4 pp 1194ndash1203 2011

[11] F A Poad and W Ismail ldquoAutomated switching mechanismfor indoor and outdoor propagation with embedded RFID andGPS in wireless sensor network platformrdquo in Proceedings ofthe International Conference on Wireless Network pp 710ndash714London UK 2014

[12] J OrsquoToole J Tuttle M Tuttle et al ldquoRadio frequency identifi-cation sensor tag apparatusesrdquo US 20080030306 A1 2008

[13] V M Gupta and S Annambhotla ldquoRFID tracker and locatorrdquoUS 20090085745 A1 2009

[14] F Shebli I Dayoub and J M Rouvaen ldquoMinimizing energyconsumptionwithinwireless sensor networkrdquoUbiquitous Com-puting and Communication Journal vol 87 2007

[15] R Ding J Hou and B Xing ldquoResearch of wireless sensornetwork nodes based on ambient energy harvestingrdquo in Pro-ceedings of the 6th IEEE International Conference on IntelligentNetworks and Intelligent Systems (ICINIS rsquo13) pp 286ndash288Shenyang China November 2013

[16] A Cammarano C Petrioli andD Spenza ldquoPro-Energy a novelenergy prediction model for solar and wind energy-harvestingwireless sensor networksrdquo in Proceedings of the 9th IEEEInternational Conference on Mobile Ad-Hoc and Sensor Systems(MASS rsquo12) pp 75ndash83 Las Vegas Nev USA October 2012

[17] J Song and Y K Tan ldquoEnergy consumption analysis of ZigBee-based energy harvesting wireless sensor networksrdquo in Proceed-ings of the IEEE International Conference on CommunicationSystems (ICCS rsquo12) pp 468ndash472 November 2012

[18] San Jose Technology Inc Datasheet GPS Engine Board ModelFV M8 2014 httpswwwsparkfuncomdatasheetsGPSFV-M8 Specpdf

[19] W Ismail and R Abdulla ldquoPortable RFID Reader for RTLS(Real Time Location Systems)rdquo PCT International Filing NoPCTMY2011000167 Country Australia Pub No WO2012102600 2012

[20] W Ismail C Z Zulkifli and M G Rahman ldquoWireless pro-duction monitoring systemrdquo PCT International Filing PCTMY2013000037 2013

[21] ISOIEC 18000-4 Information Technology-Radio FrequencyIdentification for ItemManagement-Part7 Parameters for ActiveAir Interface Communications at 245 GHz ISO 18000-42008International Organization for Standardization GenevaSwitzerland 2008

[22] D-H Shih P-L Sun D C Yen and S-M Huang ldquoTaxonomyand survey of RFID anti-collision protocolsrdquo Computer Com-munications vol 29 no 11 pp 2150ndash2166 2006

[23] B Nilsson L Bengtsson P-A Wiberg and B SvenssonldquoProtocols for active RFIDmdashthe energy consumption aspectrdquo inProceedings of the IEEE 2nd International Symposium on Indus-trial Embedded Systems (SIES rsquo07) pp 41ndash48 July 2007

[24] H L van Eeden ldquoWireless Communication Systemrdquo US Patent0045923 2009

International Journal of Distributed Sensor Networks 15

[25] H Haar K Bischof and H Scharke ldquoRFID transponderrdquo WOPatent 109212 2006

[26] C A Balanis AntennaTheory Analysis and Design JohnWileyamp Sons New York NY USA 3rd edition 2005

[27] P Sorrells Optimizing Read Range in RFID System MicrochipTechnology Inc 2000 httpedncomdesignwireless-network-ing4341980Optimizing-read-range-in-RFID-systems

[28] A Ammu L Mapa and A H Jayatissa ldquoEffect of factors onRFID tag readability-statistical analysisrdquo in Proceedings of theIEEE International Conference on ElectroInformation Technol-ogy (EIT rsquo09) pp 355ndash358 June 2009

[29] R Zurawski Embedded System Handbook Second EditionNetwork Embedded Systems CRC Press 2009

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

Page 13: Research Article Propagation Analysis for Automated ...downloads.hindawi.com/journals/ijdsn/2015/392385.pdf · Hybrid tag includes active RFID, GPS, satellite, and sensors [ ] MHz

International Journal of Distributed Sensor Networks 13

Power

Range

Interactions plot

2 10 5 100

minus30

minus45

minus60

minus30

minus45

minus60

minus30

minus45

minus60

Range

Power

Type of RFID

210

5100

Type of RFID

E

N

EN

Figure 13 Interaction plot for power level versus range measurement

Power

Mea

nM

ean

Type RFID

Range

Main effects plot for RSSI

minus40

minus50

minus60

minus40

minus50

minus60

2 10

5 100

EN

Figure 14 Main effects plot for power level versus range measurement

significant findings where the proposed embedded RFIDtag with switching mechanism introduced is much betterthan embedded RFID tag without switching mechanismTo increase the performance of the embedded RFID tag interms of time of usage an energy harvesting technology willbe introduced later as a part of our next contribution Interms of hardware costing the proposed approach is moreacceptable compared to the system developed by [6] since thetechnology used in this work especially GSM communicationis much cheaper than satellite communication used in the

existing system [6] In addition this proposed system doesnot totally depend on GSM since the communication willbe done by RFID which is free as long as the embeddedRFID tag is working in the control area of WSN platformHowever the existing work [6] did not support long rangecommunication and thus when the tag is moving far fromthe reader the satellite communication needs to be used andthis will lead to high cost of communication As a conclusionthe proposed embedded RFID tag provided better signalpropagation compared to the existing RFID tag [19] and

14 International Journal of Distributed Sensor Networks

00 1 2 3 4 5 6 7 8 9

Batte

ry re

duct

ion

()

Time (hours)

Battery lifetime testing

Embedded RFID tag with switchingEmbedded RFID tag without switchingStandalone RFID tag

minus5

minus10

minus15

minus20

minus25

minus30

Figure 15 Battery life testing based on three types of RFIDdesigned

the battery lifetime for embedded RFID tag with switchingmechanism is better than that without switchingmechanism

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank Malaysia Ministry of HigherEducation (LRGS Fund) for sponsoring the research anddeveloping this project Also special appreciation is due toProfessor DrMohamadKamal Abdul Rahim fromUniversitiTeknologiMalaysia Head of the LRGS Project and AssociateProfessor Dr Alyani Ismail Universiti PutraMalaysia for thesupport

References

[1] DIGI XBeeXBee PRO ZB RF Module Datasheet 2013[2] J B Howard ldquoCombination RFID and GPS functionality on

intelligent labelrdquo WO 2003050960 A2 2003[3] S Chand V R Bapat K H Hall et al ldquoPortable RFID reader

having a location determination systemrdquo EP 1752908 A220062006

[4] G O Garsia S I M Nunez L A Saavedra and M D I TapialdquoReal time location system (RTLS) with ZigBee technologyrdquo EP2196816 A12008 2008

[5] A Bandyopadhyay D Hakim B E Funk E A Kohn C AToelis and G Blankenship ldquoSystem and method for locatingtracking andor monitoring the status of personnel andorassets both indoorsrdquo US 20090043504 A1 2009

[6] RFID Journal ldquoHybrid tags includes active RFID GPS satelliteand sensorrdquo RFID Journal 2009 httpwwwrfidjournalcomarticlesview4635

[7] M Mohandes M Haleem M Deriche and K BalakrishnanldquoWireless sensor networks for pilgrims trackingrdquo IEEE Embed-ded Systems Letters vol 4 no 4 pp 106ndash109 2012

[8] N D Simoes J L Goncalves M L Caeiro M J Boavida andF D Cardoso ldquoZigBeeGPS tracking system for rowing racesrdquoin Proceedings of the International Conference on Computer as aTool (EUROCON rsquo11) Lisbon Portugal April 2011

[9] Y Xu R Jiang S Yan and D Xiong ldquoThe research of safetymonitoring system applied in school bus based on the internetof thingsrdquo Procedia Engineering vol 15 pp 2464ndash2468 2011

[10] L Zheng M Li C Wu et al ldquoDevelopment of a smartmobile farming service systemrdquo Mathematical and ComputerModelling vol 54 no 3-4 pp 1194ndash1203 2011

[11] F A Poad and W Ismail ldquoAutomated switching mechanismfor indoor and outdoor propagation with embedded RFID andGPS in wireless sensor network platformrdquo in Proceedings ofthe International Conference on Wireless Network pp 710ndash714London UK 2014

[12] J OrsquoToole J Tuttle M Tuttle et al ldquoRadio frequency identifi-cation sensor tag apparatusesrdquo US 20080030306 A1 2008

[13] V M Gupta and S Annambhotla ldquoRFID tracker and locatorrdquoUS 20090085745 A1 2009

[14] F Shebli I Dayoub and J M Rouvaen ldquoMinimizing energyconsumptionwithinwireless sensor networkrdquoUbiquitous Com-puting and Communication Journal vol 87 2007

[15] R Ding J Hou and B Xing ldquoResearch of wireless sensornetwork nodes based on ambient energy harvestingrdquo in Pro-ceedings of the 6th IEEE International Conference on IntelligentNetworks and Intelligent Systems (ICINIS rsquo13) pp 286ndash288Shenyang China November 2013

[16] A Cammarano C Petrioli andD Spenza ldquoPro-Energy a novelenergy prediction model for solar and wind energy-harvestingwireless sensor networksrdquo in Proceedings of the 9th IEEEInternational Conference on Mobile Ad-Hoc and Sensor Systems(MASS rsquo12) pp 75ndash83 Las Vegas Nev USA October 2012

[17] J Song and Y K Tan ldquoEnergy consumption analysis of ZigBee-based energy harvesting wireless sensor networksrdquo in Proceed-ings of the IEEE International Conference on CommunicationSystems (ICCS rsquo12) pp 468ndash472 November 2012

[18] San Jose Technology Inc Datasheet GPS Engine Board ModelFV M8 2014 httpswwwsparkfuncomdatasheetsGPSFV-M8 Specpdf

[19] W Ismail and R Abdulla ldquoPortable RFID Reader for RTLS(Real Time Location Systems)rdquo PCT International Filing NoPCTMY2011000167 Country Australia Pub No WO2012102600 2012

[20] W Ismail C Z Zulkifli and M G Rahman ldquoWireless pro-duction monitoring systemrdquo PCT International Filing PCTMY2013000037 2013

[21] ISOIEC 18000-4 Information Technology-Radio FrequencyIdentification for ItemManagement-Part7 Parameters for ActiveAir Interface Communications at 245 GHz ISO 18000-42008International Organization for Standardization GenevaSwitzerland 2008

[22] D-H Shih P-L Sun D C Yen and S-M Huang ldquoTaxonomyand survey of RFID anti-collision protocolsrdquo Computer Com-munications vol 29 no 11 pp 2150ndash2166 2006

[23] B Nilsson L Bengtsson P-A Wiberg and B SvenssonldquoProtocols for active RFIDmdashthe energy consumption aspectrdquo inProceedings of the IEEE 2nd International Symposium on Indus-trial Embedded Systems (SIES rsquo07) pp 41ndash48 July 2007

[24] H L van Eeden ldquoWireless Communication Systemrdquo US Patent0045923 2009

International Journal of Distributed Sensor Networks 15

[25] H Haar K Bischof and H Scharke ldquoRFID transponderrdquo WOPatent 109212 2006

[26] C A Balanis AntennaTheory Analysis and Design JohnWileyamp Sons New York NY USA 3rd edition 2005

[27] P Sorrells Optimizing Read Range in RFID System MicrochipTechnology Inc 2000 httpedncomdesignwireless-network-ing4341980Optimizing-read-range-in-RFID-systems

[28] A Ammu L Mapa and A H Jayatissa ldquoEffect of factors onRFID tag readability-statistical analysisrdquo in Proceedings of theIEEE International Conference on ElectroInformation Technol-ogy (EIT rsquo09) pp 355ndash358 June 2009

[29] R Zurawski Embedded System Handbook Second EditionNetwork Embedded Systems CRC Press 2009

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

Page 14: Research Article Propagation Analysis for Automated ...downloads.hindawi.com/journals/ijdsn/2015/392385.pdf · Hybrid tag includes active RFID, GPS, satellite, and sensors [ ] MHz

14 International Journal of Distributed Sensor Networks

00 1 2 3 4 5 6 7 8 9

Batte

ry re

duct

ion

()

Time (hours)

Battery lifetime testing

Embedded RFID tag with switchingEmbedded RFID tag without switchingStandalone RFID tag

minus5

minus10

minus15

minus20

minus25

minus30

Figure 15 Battery life testing based on three types of RFIDdesigned

the battery lifetime for embedded RFID tag with switchingmechanism is better than that without switchingmechanism

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to thank Malaysia Ministry of HigherEducation (LRGS Fund) for sponsoring the research anddeveloping this project Also special appreciation is due toProfessor DrMohamadKamal Abdul Rahim fromUniversitiTeknologiMalaysia Head of the LRGS Project and AssociateProfessor Dr Alyani Ismail Universiti PutraMalaysia for thesupport

References

[1] DIGI XBeeXBee PRO ZB RF Module Datasheet 2013[2] J B Howard ldquoCombination RFID and GPS functionality on

intelligent labelrdquo WO 2003050960 A2 2003[3] S Chand V R Bapat K H Hall et al ldquoPortable RFID reader

having a location determination systemrdquo EP 1752908 A220062006

[4] G O Garsia S I M Nunez L A Saavedra and M D I TapialdquoReal time location system (RTLS) with ZigBee technologyrdquo EP2196816 A12008 2008

[5] A Bandyopadhyay D Hakim B E Funk E A Kohn C AToelis and G Blankenship ldquoSystem and method for locatingtracking andor monitoring the status of personnel andorassets both indoorsrdquo US 20090043504 A1 2009

[6] RFID Journal ldquoHybrid tags includes active RFID GPS satelliteand sensorrdquo RFID Journal 2009 httpwwwrfidjournalcomarticlesview4635

[7] M Mohandes M Haleem M Deriche and K BalakrishnanldquoWireless sensor networks for pilgrims trackingrdquo IEEE Embed-ded Systems Letters vol 4 no 4 pp 106ndash109 2012

[8] N D Simoes J L Goncalves M L Caeiro M J Boavida andF D Cardoso ldquoZigBeeGPS tracking system for rowing racesrdquoin Proceedings of the International Conference on Computer as aTool (EUROCON rsquo11) Lisbon Portugal April 2011

[9] Y Xu R Jiang S Yan and D Xiong ldquoThe research of safetymonitoring system applied in school bus based on the internetof thingsrdquo Procedia Engineering vol 15 pp 2464ndash2468 2011

[10] L Zheng M Li C Wu et al ldquoDevelopment of a smartmobile farming service systemrdquo Mathematical and ComputerModelling vol 54 no 3-4 pp 1194ndash1203 2011

[11] F A Poad and W Ismail ldquoAutomated switching mechanismfor indoor and outdoor propagation with embedded RFID andGPS in wireless sensor network platformrdquo in Proceedings ofthe International Conference on Wireless Network pp 710ndash714London UK 2014

[12] J OrsquoToole J Tuttle M Tuttle et al ldquoRadio frequency identifi-cation sensor tag apparatusesrdquo US 20080030306 A1 2008

[13] V M Gupta and S Annambhotla ldquoRFID tracker and locatorrdquoUS 20090085745 A1 2009

[14] F Shebli I Dayoub and J M Rouvaen ldquoMinimizing energyconsumptionwithinwireless sensor networkrdquoUbiquitous Com-puting and Communication Journal vol 87 2007

[15] R Ding J Hou and B Xing ldquoResearch of wireless sensornetwork nodes based on ambient energy harvestingrdquo in Pro-ceedings of the 6th IEEE International Conference on IntelligentNetworks and Intelligent Systems (ICINIS rsquo13) pp 286ndash288Shenyang China November 2013

[16] A Cammarano C Petrioli andD Spenza ldquoPro-Energy a novelenergy prediction model for solar and wind energy-harvestingwireless sensor networksrdquo in Proceedings of the 9th IEEEInternational Conference on Mobile Ad-Hoc and Sensor Systems(MASS rsquo12) pp 75ndash83 Las Vegas Nev USA October 2012

[17] J Song and Y K Tan ldquoEnergy consumption analysis of ZigBee-based energy harvesting wireless sensor networksrdquo in Proceed-ings of the IEEE International Conference on CommunicationSystems (ICCS rsquo12) pp 468ndash472 November 2012

[18] San Jose Technology Inc Datasheet GPS Engine Board ModelFV M8 2014 httpswwwsparkfuncomdatasheetsGPSFV-M8 Specpdf

[19] W Ismail and R Abdulla ldquoPortable RFID Reader for RTLS(Real Time Location Systems)rdquo PCT International Filing NoPCTMY2011000167 Country Australia Pub No WO2012102600 2012

[20] W Ismail C Z Zulkifli and M G Rahman ldquoWireless pro-duction monitoring systemrdquo PCT International Filing PCTMY2013000037 2013

[21] ISOIEC 18000-4 Information Technology-Radio FrequencyIdentification for ItemManagement-Part7 Parameters for ActiveAir Interface Communications at 245 GHz ISO 18000-42008International Organization for Standardization GenevaSwitzerland 2008

[22] D-H Shih P-L Sun D C Yen and S-M Huang ldquoTaxonomyand survey of RFID anti-collision protocolsrdquo Computer Com-munications vol 29 no 11 pp 2150ndash2166 2006

[23] B Nilsson L Bengtsson P-A Wiberg and B SvenssonldquoProtocols for active RFIDmdashthe energy consumption aspectrdquo inProceedings of the IEEE 2nd International Symposium on Indus-trial Embedded Systems (SIES rsquo07) pp 41ndash48 July 2007

[24] H L van Eeden ldquoWireless Communication Systemrdquo US Patent0045923 2009

International Journal of Distributed Sensor Networks 15

[25] H Haar K Bischof and H Scharke ldquoRFID transponderrdquo WOPatent 109212 2006

[26] C A Balanis AntennaTheory Analysis and Design JohnWileyamp Sons New York NY USA 3rd edition 2005

[27] P Sorrells Optimizing Read Range in RFID System MicrochipTechnology Inc 2000 httpedncomdesignwireless-network-ing4341980Optimizing-read-range-in-RFID-systems

[28] A Ammu L Mapa and A H Jayatissa ldquoEffect of factors onRFID tag readability-statistical analysisrdquo in Proceedings of theIEEE International Conference on ElectroInformation Technol-ogy (EIT rsquo09) pp 355ndash358 June 2009

[29] R Zurawski Embedded System Handbook Second EditionNetwork Embedded Systems CRC Press 2009

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

Page 15: Research Article Propagation Analysis for Automated ...downloads.hindawi.com/journals/ijdsn/2015/392385.pdf · Hybrid tag includes active RFID, GPS, satellite, and sensors [ ] MHz

International Journal of Distributed Sensor Networks 15

[25] H Haar K Bischof and H Scharke ldquoRFID transponderrdquo WOPatent 109212 2006

[26] C A Balanis AntennaTheory Analysis and Design JohnWileyamp Sons New York NY USA 3rd edition 2005

[27] P Sorrells Optimizing Read Range in RFID System MicrochipTechnology Inc 2000 httpedncomdesignwireless-network-ing4341980Optimizing-read-range-in-RFID-systems

[28] A Ammu L Mapa and A H Jayatissa ldquoEffect of factors onRFID tag readability-statistical analysisrdquo in Proceedings of theIEEE International Conference on ElectroInformation Technol-ogy (EIT rsquo09) pp 355ndash358 June 2009

[29] R Zurawski Embedded System Handbook Second EditionNetwork Embedded Systems CRC Press 2009

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

Page 16: Research Article Propagation Analysis for Automated ...downloads.hindawi.com/journals/ijdsn/2015/392385.pdf · Hybrid tag includes active RFID, GPS, satellite, and sensors [ ] MHz

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of