international journal of scientific research inijsrcseit.com/paper/v1i1.pdf · 2017-03-14 · i...

i

International Journal of Scientific Research in

Computer Science, Engineering and Information

Technology

ISSN : 2456-3307

Volume 1, Issue 1, July-August-2016

International Peer Reviewed, Open Access Journal

Bimonthly Publication

Published By

Technoscience Academy

(The International Open Access Publisher)

Email: [email protected]

Website: www.technoscienceacademy.com

mailto:[email protected]

http://www.technoscienceacademy.com/

ii

Advisory/Editorial Board

Prof. Shardul Agravat

Information Technology, C U Shah Unuiversity, Surendranagar, Gujarat, India

Prof. Vaishali Kalaria

Information Technology, RKU, Rajkot, Gujarat, India

Prof. H. B. Jethva

Computer Enginering, L. D. College of Engineering, Ahmedabad, Gujarat, India

Prof. Bakul Panchal

Information Technology, L. D. College of Engineering, Ahmedabad, Gujarat, India

Prof. Bhavesh Prajapati

Computer Science, Government MCA College Maninagar, Ahmedabad, Gujarat, India

Dr. Syed Umar

Computer Science and Engineering, KL University, Guntur, Andhra Pradesh, India

Prof. S. Jagadeesan

Computer Science, Nandha Engineering College Erode, Tamil Nadu, India

Prof. Joshi Rahul Prakashchandra

Information Technology, Parul Institute of Engineering & Technology, Vadodara, Gujarat,

India

Dr. Aftab Alam Tyagi

Department of Mathematics, SRM University NCR Campus, Uttar Pradesh, India

Prof (Dr.) Umesh Kumar

Department of Science & Technology, Govt. Women’s Polytechnic, Ranchi, Jharkhand, India

Dr. N. Pughazendi

Computer Science and Engineering, Panimalar Engineering College Chennai, Tamilnadu,

India

Sachin Narendra Pardeshi

Department of Computer Engineering, R.C.Patel Institute of Technology, Shirpur,

Maharashtra, India

Dr. Bangole Narendra Kumar

Department of Computer Science and Systems Engineering, Sree Vidyanikethan

Engineering College, Tirupati, Andhra Pradesh, India

Dr. Dhananjaya Reddy

M.Sc.(Maths),M.Sc.(Stat), M.Phil.(Maths), M.phil. (Statistics), B.Ed., PhD Department of

Mathematics,Government Degree College, Puttur, Chittoor, Andhra Pradesh, India

iii

Dr. Ajitesh Singh Baghel

Department of Computer Science, Awadhesh Pratap Singh University, Rewa, Madhya

Pradesh, India

Prof. Sarita Dhawale

Ashoka Center for Business & Computer Studies, Ashoka Marg, Ashoka Nagar, Nashik,

Maharashtra, India

International Advisory/Editorial Board

AbdulGaniyu Abdu Yusuf

Computer Science, National Biotechnology Development Agency (NABDA), Abuja, Nigeria

Dr. M. A. Ashabrawy

Computer Science and Engineering, Prince Sattm bin Abdulaziz University, Kingdom Saudi

Arabia(KSA), Saudi Arabia

Dr. V. Balaji

Bahir Dar University, Bahir Dar, Ethiopia

Lusekelo Kibona

Department of Computer Science, Ruaha Catholic University (RUCU) , Iringa, Tanzania

Md. Amir Hossain

IBAIS University/Uttara University, Dhaka, Bangladesh

Mohammed Noaman Murad

Department of Computer Science, Cihan University Erbil, Kurdistan Region, Iraq

Prof. Dr. H. M. Srivastava

Department of Mathematics and Statistics, University of Victoria, Victoria, British Columbia,

Canada

Prof. Sundeep Singh

Mississauga, Ontario, Canada

iv

CONTENTS

Sr.

No Article/Paper Page No

1 A Case Study on Mobile Adhoc Network Security for Hostile

Environment Dr. D. Devi Aruna, Dr. D.Vimal Kumar

01-06

2 A Qualitative Comparison of Various Routing Protocols in WSN Kabeer Khan, Abdul Waris, Hamayun Safi

07-13

3 Detecting BOT Victim in Client Networks Abinaya. E, Balamurugan. K

14-18

4 A New Approach for Transistor-Clamped H-Bridge Multilevel Inverter

with voltage Boosting Capacity Suparna Buchke, Prof. Kaushal Pratap Sengar

19-23

5 An Improved Performance of Greedy Perimeter Stateless Routing

protocol of Vehicular Adhoc Network in Urban Realistic Scenarios Ritesh Gupta, Parimal Patel

24-29

6 A Brief Survey of Acoustic Wireless Sensor Network Mansoor Ullah, Abbas Khan, Muhammad Adil

30-34

7 A Survey on Secure Cloud Storage with Techniques Like Data

Deduplication and Convergent Key management P. Balasubhramanyam Reddy, G. Nagappan

35-39

8 Emergency Information Access using QR Code Technology in Medical

Field P. Deepika, Sushanth. B , Tarun Kumar. S. P, Vignesh. M

40-43

9 Improving Classifier Performance Using Feature Selection with

Ensemble Learning Bhavesh Patankar, Dr. Vijay Chavda

44-48

10 The Use of Wireless Sensor Networks for Forest Fire Monitoring - A

Survey Mehwish Zaheer, Rabia Riaz, Shakeeb Ahmad

49-53

10 Address Allocation Algorithm with Cooperative Communication in

MANET Parameswaran T, Dr.Palanisamy C, Logeshwari N

54-59

12 A Survey on WSN-based Forest Fire Detection Techniques Waqas Ali, Abdullah, Ishfaq-ur-rashid

60-65

13 Congestion Detection and Mitigation Protocols for Wireless Sensor

Networks Muhammad Zeeshan, Fazlullah Khan, Syed Roohullah Jan

66-71

CSEIT16111 | Received: 13 July 2016 | Accepted: 24 July 2016 | July-August 2016 [(1)1: 01-06]

International Journal of Scientific Research in Computer Science, Engineering and Information Technology

© 2016 IJSRCSEIT | Volume 1 | Issue 1 | ISSN : 2456-3307

1

A Case Study on Mobile Adhoc Network Security for Hostile Environment

Dr. D. Devi Aruna, Dr. D.Vimal Kumar

Department of Computer Science, Nehru Arts and Science College, Coimbatore, Tamil Nadu, India

ABSTRACT

A mobile adhoc network (MANET) is a peer to peer wireless network where nodes can communicate with each

other without infrastructure. Due to this nature of MANET; it is possible that there could be some malicious and

selfish nodes that try to compromise the routing protocol functionality and makes MANET vulnerable to Denial of

Service attack in military communication environments. Hence security is an important challenge while deploying

MANET. This research effort examines the case study for a Layerwise Security (LaySec) framework that provides

security for an ad-hoc network operating in a military environment. LaySec incorporates three security features

(Secure neighbor authentication and Layerwise Security techniques and multipath routing) into its framework while

maintaining network performance sufficient to operate in hostile environment. layerwise security protocol has been

implemented and simulated on Qualnet 5.0. Based on the simulation result, it is observed that the proposed approach

has shown better results in terms of Quality of Service parameters like Average packet delivery ratio, Average

throughput, Average end to end delay and Routing Overhead.

Keywords : Mobile Adhoc Network, Layer Wise Security Protocols, Denial of Service Attack.

I. INTRODUCTION

Recent years Mobile ad hoc Networks start gaining

attention from the industrial and academic research

community due to their wide deployment and inherent

nature of solving practical real world

applications[1][4].The ease of deployment without the

existing infrastructure makes ad hoc networks an

attractive choice for dynamic situations such as

military operations, disaster recovery, and so forth.

Especially, military communication environments

have been considered as one of the original

motivations for MANET, due to the need for

battlefield survivability and rapid deployment of self-

organizing mobile infrastructure. This research work

evaluates the case study for mobile adhoc network

with concentration to defend against Denial of Service

attack in MANET layers. A military case study

scenarios is introduced: the scenario modifies its

channel and physical layer settings for army military

devices in an unknown and unstable MANET military

environment system with concentration to defend

against Denial of Service attack[2].

The paper is organized in such a way that Chapter 2

discusses Review of Literature, Chapter 3 discusses

problem statement,, Chapter4 discusses proposed

method, Chapter5 discusses Experimental evaluation

and Chapter6gives the conclusion.

II. METHODS AND MATERIAL

2. Review of Literature

2.1 Denial of Service attack

This chapter briefly describes the Denial of Service

Attacks for MANET.

An attacker attempts to avoid authorized and

legitimate users from the services offered by the

network. The typical way is to flood packets to any

centralized resource present in the network so that the

resource is no longer available to nodes in the network,

as a result of which the network no longer operate in

the manner in which it is designed to operate. This

Volume 1 | Issue 1 | 2016 | www.ijsrcseit.com 2

may lead to a failure in the delivery of guaranteed

services to the end users. DoS attacks can be launched

against any layer in the network protocol stack. On the

physical and MAC layers, an adversary could employ

jamming signals which disrupt the ongoing

transmissions on the wireless channel. On the network

layer, an adversary could take part in the routing

process and exploit the routing protocol to disrupt the

normal functioning of the network. For example, an

adversary node could participate in a session but

simply drop a certain number of packets, which may

lead to degradation in the QoS being offered by the

network. On the higher layers, an adversary could

bringdown critical services such as the key

management service. For example, consider the

following: In figure1 assume a shortest path that exists

from S to X and C and X cannot hear each other, that

nodes B and C cannot hear each other, and that M is a

malicious node attempting a denial of service attack.

Suppose S wishes to communicate with X and that S

has an unexpired route to X in its route cache. S

transmits a data packet towards Xwith the source route

S --> A --> B--> M --> C --> D--> X contained in the

packet’s header. When M receives the packet, it can

alter the source route in the packet’s header, such as

deleting D from the source route. Consequently, when

C receives the altered packet, it attempts to forward

the packet toX. Since X cannot hear C,the

transmission is unsuccessful [2][3].

S ↔A↔ B↔ M ↔C↔ D↔ X

Equation 1 : Denial of Service attack

3. Problem Statement

This research investigates how to integrate security

policies of a MANET with secure neighbor

authentication that will allow the MANET to function

securely in a military environment without degrading

network performance. The specific problem to be

addressed is how to use secure neighbor authentication

of nodes in a multipath routing algorithm in MANET

protected from Denial of service attack in military

environment. Most of such performance analysis are

normally done on commercial settings. For instance,

wireless LAN technologies in the 2.4 GHz ISM

frequency band are generally assumed, offering data

rates up to 2 Mbps within the range of 250 m. This

paper is motivated by the observation that such

propagation and network models assumed by the

current ad hoc networking simulations are quite

different from real world military environments. In

fact, a few hundred MHz frequency band (i.e., VHF or

even HF) is used with very low data transmission rates

(e.g., 384 Kbps) for the military scenarios

4. Proposed Methodology

This approach aims in improving the performance in

terms of QoS characteristics as metrics. The

methodology is proposed in order to assure Layerwise

security for Mobile Ad hoc Networks. The specific

contributions are structured in six phases.

Phase I. Integration of SNAuth with SPMAODV

Phase II. SNAuth-SPMAODV with SIP for

Application and Network layer Security

Phase III. SNAuth-SPMAODV with WTLS for

Transport and Network Layer Security

Phase IV. SNAuth-SPMAODV with IPSec for

Network Layer Security

Phase V. SNAuth-SPMAODV with CCMP-AES for

Link and Network Layer Security

Phase VI. SNAuth-SPMAODV with DSSS for

Physical and Network Layer Security

Integration of SNAuth with SPMAODV SPMAODV

provides multiple paths between sender and receiver

nodes that can be used to offset the dynamic and

unpredictable configuration of ad-hoc networks. They

can also provide load balancing by spreading traffic

along multiple routes, fault-tolerance by providing

route resilience, and higher aggregate bandwidth. The

proper selection of routes using a strict-priority

multipath protocol can increase further the network

throughput. The main idea of this phase to integrate

strict priority multipath AODV with secure neighbor

authentication that facilitate neighboring nodes

exchange messages to discover and authenticate each

other. Thus this phase provides security mechanism

like message integrity, mutual authentication, and non-

repudiation; defend against Denial of Service attacks

and increase network throughput.

SNAuth-SPMAODV with SIP for Application and

Network layer Security Secure Neighbor

Authentication Strict Priority Multipath Ad hoc On-

demand Distance Vector Routing) with Session

Initiation Protocol (SIP) provides application layer and


network layer security and it is robust against Denial

of Service attack. It reduces dependency on single

nodes and routes; it discovers multiple paths between

sender and receiver nodes and it has the advantages of

a multipath protocol without introducing extra packets

into the network offering robustness in a secured

MANET. It can be used to offset the dynamic and

unpredictable configuration of adhoc networks. They

can also provide load balancing by spreading traffic

along multiple routes, fault-tolerance by providing

route resilience, and higher aggregate bandwidth in

hostile environment [15].

SNAuth-SPMAODV with WTLS for Transport and

Network Layer Security The primary focus of this

phase is to provide transport layer security for

authentication, securing end-to-end communications

through data encryption and to provide security

services for both routing information and data message

at network layer. It also handles delay and packet loss.

The proposed model combines SNAuth-SPMAODV

Routing with Wireless Transport Layer Security

(WTLS) to defend against Denial of Service (DoS)

attack and it also provides authentication, privacy and

integrity of packets in routing, end-to-end

communications through data encryption, packet loss

and transport and network layers of MANET [14].

SNAuth-SPMAODV with WTLS is found to be a

good security solution even with its known security

problems[9].

SNAuth-SPMAODV with IPSec for Network Layer

Security Secure Neighbor Authentication Strict

Priority Multipath Ad hoc On-demand Distance

Vector Routing) with IPSec is robust against Denial of

Service attack and it also provides security services for

both routing information and data message at network

layer in MANET.The proposed method uses a hybrid

version of the IPSec protocol, which includes both AH

and ESP modes. IPSec is a protocol suit for securing

IP based communication focusing on authentication,

integrity, confidentiality and support perfect security

forward. The significant importance of the

aforementioned protocol is that it offers flexibility,

which cannot be achieved at higher or lower layer

abstractions in addition to the symmetric

cryptographic schemes [11]. These are 1000 times

faster than asymmetric cryptographic schemes, a fact

that makes IPSec appropriate to be used in handheld

resources constrained devices such as PDAs. SNAuth-

SPMAODV with CCMP-AES for Link and Network

Layer Security.

SNAuth-SPMAODV combines with CCMP-AES

model to defend against Denial of Service attack and it

provide confidentiality and authentication of packets

in both network and data link layers of MANETs[2].

The primary focus of this phase is to provide security

mechanisms applied in transmitting data frames in a

node-to node manner through the security protocol

CCMP-AES working in data link layer. It keeps data

frame from eavesdropping, interception, alteration, or

dropping from unauthorized party along the route from

the source to the destination.

SNAuth-SPMAODV with DSSS for Physical and

Network Layer Security SNAuth-SPMAODV

combines with DSSS to defend against Denial of

Service attack. The physical layer protocol in

MANETs is reliable for bit-level transmission between

network nodes and network layer is responsible to

provide security services for both routing information

and data message [10]. The proposed model combines

SNAuth-SPMAODV routing protocol and spread

spectrum technology Direct Sequence Spread

Spectrum (DSSS) to defend against signal jamming

denial-of-service attacks in physical layer and network

layer for MANET.

III. RESULTS AND DISCUSSION

A. Experimentation and Evaluation

Using the QualNet network simulator [7],

comprehensive simulations are made to evaluate the

protocol. Qualnet provides a scalable simulation

environment for multi-hop wireless ad hoc networks,

with various medium access control protocols such as

CSMA and IEEE 802.11. channel and physical layer

settings are modified to apply more realistic military

scenarios. Note that PRC 999K device is used as a

reference model. 802.11 DCF and UDP protocols are

used for MAC and a transport protocols, respectively.

Also, CBR traffic is utilized in the study. As the TCP

based application protocols such as telnet or FTP show

unstable performance in mobile wireless

communication, it cannot evaluate precise

performance of routing protocol itself. CBR

application model sends one packet per second, which

represents relatively low traffic patterns in military


environments. Each packet size is 512 Bytes. In

military environments, operational network size is

very large as compare to conventional case. Nodes in

the simulation are assumed to move according to the

“random way point” mobility model. Pause time is

fixed to 20 seconds. The attackers are positioned

around the center of the routing mesh in all

experiments. To evaluate the performance of proposed

method by 4 measurements: Packet delivery radio,

average end-to-end delay, routing overhead and

Throughput.This simulated environment is defined by

the following parameters as shown in Table 1 and

Table 2.

Table 1: Simulation Metrics of Laysec Framework for

Military Scenario

Table 2: Physical Layer Model for Hostile

Environments

Parameters Military devices

Frequency 30-300 MHz

Propagation limits -120 dBm

Radio propagation

model

Two-Ray

Data rates 200 Kbps

Transmit power 45 dBm

Receive sensitivity -150 dBm

Reference model PRC-999K device

B. Performance Evaluation

The performance analysis of Layerwise security

framework with SNAuth-SPMAODV has been

conducted using the simulation setup for Hostile

Environment as outlined in Table 2 and 3. The

simulation scenarios consist of different network

density or size is assessed by deploying a different

number of mobile nodes over a space of 1500m x

1500m.

Average Packet Delivery Ratio (PDR)

In Figure 1, the Average Packet Delivery Ratio of

AODV, SNAuth-SPMAODV and Layerwise Security

Framework with SNAuth-SPMAODV for different

network sizes of 100 to 600 nodes are placed in a

topology area of 1500m x 1500m. Packet delivery

ratio shows how successfully a protocol performs

delivering packets from source to destination.

Figure 1 : Average Packet Delivery Ratio

Average Throughput

Figure 2 shows the network throughput is the average

rate of successful message delivery over a

communication channel.

Figure 2 : Average Throughput Ratio

0

50

100

150

200

250

300

100 200 300 400 500 600

Avg

.Pac

ket

De

liver

y R

atio

(%)

Number of Nodes

Layerwise Security Framework

SNAuth-SPMAODV

AODV with attack

02000400060008000

10000120001400016000

100 200 300 400 500 600

Av

g.T

hro

ug

hp

ut(

bit

/s)

Number of Nodes


SNAuth-SPMAODV

AODVwith attack


Average End-to-End Delay

Figure 3 shows an average end-to-end delay of AODV,

SNAuth-SPMAODV and Layerwise Security

Framework with SNAuth-SPMAODV according to

the increase of network density. Layerwise Security

Framework with SNAuth-SPMAODV exhibits the

lowest end-to-end delay most of the time. AODV has

much higher end-to-end delay than proposed method.

Layerwise Security Framework with SNAuth-

SPMAODV keeps up good performance in delay as

the network density becomes high. Layerwise Security

Framework with SNAuth-SPMAODV performs

poorly in sparse networks. (eg 200 to 300 nodes)

Figure 3 : Average End to End Delay

Routing Overhead

Figure 5 illustrates the routing overhead generated by

the proposed framework when the number of nodes is

varied. The figure shows that the generated routing

overhead in AODV, SNAuth-SPMAODV and


SPMAODV increases with increased number of nodes.


SPMAODV performs well compared to AODV and

SNAuth-SPMAODV

Figure 5 : Routing Overhead

IV.CONCLUSION

Mobile ad hoc networks (MANETs) can be applied to

many situations without the use of any existing

network infrastructure or centralized administration. In

hostile environment, there is a need for the network to

route packets through dynamically mobile nodes.

MANETs can be considered as the solution for this

highly mobile and dynamic military network.

However it is not appropriate to directly apply

conventional mobile ad hoc networks scheme to

military network, since military communication

system is different from conventional counter parts

both in device’s physical layer specification and

networking environment. Therefore consider these

particularities of military communication system

through simulation, and evaluate the performance of

Layerwise security framework on the assumed

military environment. In simulation results, the

proposed methods provide good performance with

every measurement metric in high network density

environment

V. REFERENCES

[1] Arunkumar B. R., Reddy L.C., and Hiremath

P.S., 2008, "A Survey of Mobile Ad Hoc

Network Routing Protocols" Journal of

Intelligent System Research, 8(6), 49-64.

[2] Bajaj. L., Takai.M., Ruja.R., Tang.K.,

Bagrodia.R., and Gerla.M.,1999,"GlomoSim: A

Scalable Networks Simulation Environments",

UCLA Computer Science Departments

Technical Report 900027.

0

0.2

0.4

0.6

0.8

1

1.2

100 200 300 400 500 600

Av

g.E

nd

to

En

d D

ela

y(s

)

Number of Nodes


SNAuth-SPMAODV

AODV with attack

0

20000

40000

60000

80000

100 200 300 400 500 600

Rou

tin

g O

ver

hea

d(p

ack

ets)

Number of Nodes Layerwise Security Framework

SNAuth-SPMAODV

AODV with attack


[3] Biswas K., Ali L., 2001, "Security Threats in

Mobile Ad Hoc Network" Department of

Interaction and System Design School of

Engineering, 1-39.

[4] Boomaranimalany.A., Dhulipala.S., and

Chandrasekaran R.M, 2009, "Throughput and

Delay Comparison of MANET Routing

Protocols"International Journal Open Problems

Computational Mathematics, ICSRS

Publications, 2(3), 461-468.

[5] Chenna. P and Dr. ChandraSekhar.P., 2007,

"Performance Analysis of Adhoc Network

Routing Protocols", International Symposium on

Ad Hoc and Ubiquitous Computing,

ISAUHC'06, 17, 186 – 187.

[6] Dwivedi.A.K., kushwaha.S., and Vyas O.P.,

2009,"Performance of Routing Protocols for

Mobile Ad hoc and wireless sensor networks: A

Comparative study", International Journal of

Recent Trends in Engineering, 2(4) ,101-105.

[7] Garg.N. and Mahapatra.R.P, 2009, "MANET

Security Issues". International Journal of

Computer Science and Network Security, 9(8),

241-246.

[8] Islam.S, 2006, "Implementation & Comparison

of IPSec Protocols for Secure Datab

Communication in Ad-Hoc Networks", Royal

Institute of Technology.

[9] Jang H.C., Lien Y.N., and Tsai T.C., 2009

,"Rescue Information System for Earth-quake

Disasters Based on MANET Emergency

Communication Platform" Proceedings of the

2009 International Conference on Wireless

Communications and Mobile Computing:

Connecting the World wirelessly, 623–627.

[10] Junaid.M., Dr Muid Mufti and Ilyas M.U., 2006,

"Vulnerabilities of IEEE 802.11i Wireless LAN

CCMP Protocol", In the Proceedings Of World

Academy Of Science, Engineering And

Technology, 11, 228-233.

[11] Pravin P.G., and Katkar G.G., 2010,"Mobile Ad

Hoc Networking: Imperatives and Challenges",

IJCA Special Issue on MANETs, 153–158.

[12] Reidt S., and Wolthusen S.D, 2008, "Exploiting

UAVs Capabilities in Tactical MANETS".

Proceedings of the 2nd Annual Conference of

ITA ,322–323.

[13] Salsano,, Veltri S., and Papalilo D., 2002,"SIP

Security Issues: The SIP authentication

procedure and its processing load" IEEE

Network,38-44.

[14] Taneja K., and Patel R.B., 2007, "Mobile Ad

hoc Networks: Challenges and Future"

Proceedings of National Conference on

Challenges & Opportunities in Information

Technology pp. 133-135.

[15] Vaidya.B. and Lim H., 2009 "Secure

Framework for Multipath Multimedia Streaming

Over Wireless Ad Hoc Network". Proceedings

of the 2009 IEEE Conference on Wireless

Communications & Networking

Conference,2678–2683.

[16] D.Devi Aruna and Dr.P.Subashini.,2014,"

Layerwise Security Framework with Snauth-

SPMAODV to Defend Denial of Service Attack

in Mobile Adhoc Networks for Hostile

Environment" International Journal of

Innovative Research in Science & Engineering.

[17] Qualnet Documentation, "Qualnet 5.0 Model

Library, Network Security", Available: Http://

Www.Scalablenetworks.Com/Products/Qualnet/

Downlaod




7

A Qualitative Comparison of Various Routing Protocols in WSN

Kabeer Khan, Abdul Waris, Hamayun Safi

Department of Computer Science, Abdul Wali Khan University Mardan, Pakistan

ABSTRACT

Wireless Sensor Network (WSN) consists of a large number of small nodes with the capabilities of sensing various

types of physical and environmental conditions, data processing, and wireless communication. In Wireless Sensor

Network(WSN) the sensor nodes collects the data from its surrounding and transmit the gathered data to a particular

user, the transmission of gathered data by sensor nodes depends on the application that is used. The nodes have

limited processing power, limited transmission range and storage capabilities as well as limited energy capabilities.

In this paper we discuss the routing protocols of wireless sensor network and also discuss the classification and

comparison of routing protocols. The architecture of routing protocols categories in three main category

Hierarchical, Location-Based and data centric protocols according to some important factors and will summarize in

the way these protocols operates. Finally, we will provide a comparative study on these various protocols.

Keywords: Wireless Sensor Network, Routing Protocols, Location-based Routing, Data Centric.

I. INTRODUCTION

Wireless Sensor Network(WSN) consist of hundred to

thousand low power sensor nodes, they are deployed

in field, have capability to gather data and send to base

station for taking decision about specific region for

specific purpose. Basic component of wireless sensor

node is sensor, processer, radio transceiver, power unit.

Sensors are responsible to sense the deployed region

for capturing the data. Processer received data form

storage unit process it and transmit to nearest neighbor

which may be node are base station. Radio transceiver

has the ability to transmit and received data form

neighbors nodes

Figure 1: Components of a Sensor Node

Power unit is responsible for managing energy

consumption of the node. These various components

of a node are shown in Figure 1 above.


A. Wireless Sensor Network Protocols

Wireless Sensor Network (WSN) has a wide range of

applications to industry, science, transportation, civil

infrastructure and security and many other fields, such

as some forests are very dangerous where human

approach fails, so from that areas we collect our

desired information using wireless sensor application.

In this paper we stud the routing protocol of wireless

sensor network and we categorize the routing

protocols in three basic categories on the basis of

network structures. This paper organized as follows. In

the first phase we discussed about the hierarchical

protocols, second phase is about location based and

third phase describes data centric protocols.

Routing Protocols in WSN

Wireless sensor network routing protocols are

different from traditional routing protocol [1-8]. On

Volume 1 | Issue 1 | 2016 | www.ijsrcseit.com

8

the basis of network structure routing protocols are

classified in many different categories, like data

centric, hierarchical and location-based protocols and

also compared the routing protocols.

Figure 2: classification of WSN Protocols

A. Hierarchical protocols In this paper different

hierarchical based routing protocols described.

Which is (LEACH, TEEN, APTEEN, PEGASIS)

energy efficient and maintain the energy of sensor

node. A hierarchical approach network is divides

in to cluster and cluster head. Cluster node

captures the data and sends the data to cluster

head. Cluster head received the data from cluster

aggregate the data and send to base station. In

hierarchical based routing protocols data send

form one node to another node and cover large

distance. This approach moves the data faster to

base station. Representative protocols of

hierarchical routing are is following:

1) LEACH

2) 2)TEEN

3) APTEEN

4) PEGASIS

1) LEACH: Low Energy Adaptive Clustering

Hierarchy (LEACH) is most popular hierarchical

routing protocols for sensor network. Leach

performs data fusion to compress the data when

data send to from cluster to base station. In this

why leach is most popular protocol to reduce the

energy consumption and enhance the lifetime of

the node. Leach protocol divide the total operation

in two phase. One is setup phase and other is

steady state phase. In set-up phase cluster head is

selected for each cluster. Cluster head is selected

from sensor nodes at the time of certain

probability. The cluster head is selected random

number between 0 and 1.the node become cluster

head for the current round if the number is less

than the threshold,

Where p is desired percentage of cluster head, r

current round, G are those node which is not selected

cluster head in 1/p round.

In steady state phase cluster head send the data to

leader node. Leader node compress, aggregate the data

and send to base station.

2) TEEN: Threshold energy efficient network (teen)

protocol a hierarchical clustering protocol. The

sensor network based on hierarchical grouping

cluster nodes from clusters and this process goes

on second level until base station is reached. In

this type of protocol cluster head send two types

of data to neighbor’s nodes. One is hard threshold

and other is soft threshold. The node transmits the

data if one of the following conditions satisfies:

a) Sense value > hard threshold

b) Sense value ~ hard threshold >= soft threshold

If hard threshold satisfy that condition if sensed value

is greater than hard threshold. This means the node

send those data which are interested and reduce the

number of transmission. In soft threshold any small

change in the sense value transmits the data to forward.

Figure 3: Architecture of TEEN protocol


9

3) APTEEN: Adaptive threshold energy efficient

network protocol (APTEEN) is improve version of

teen protocol which capturing both periodic data

collections and reacting to time critical event. The

architecture of APTEEN is same like

TEEN.APTEEN support different type query.

Historical analysis of past data values.

Snapshot of the current network view.

Monitoring of an event for a period of time.

4) PEGASIS: Power Efficient Gathering in Sensor

Information System (PEGASIS), which is chain

based power efficient algorithm. In PEGASIS

only one node chose to transmit data to base

station other nodes capture the data and send the

data to neighbor node. PEGASIS chain of sensor

node every sensor node transmit data to neighbor

node and received data for neighbor node. For

example:

Figure 4 : Architecture of PEGASIS

In the above example C0 send data to C1. Node C1

combine the data of C0 and own data then transmit to

leader node.C2 is leader node which send token to C4,

node C4 transmit data to C3. C3 combine the data of

C4 with its own data and transmit to leader node.

Node C2 wait for neighbor node data if received the

data form neighbor node then combine the data and

send only one message to base station [9-28].

B. Location Based Protocol:

In sensor networks the information about the location

of nodes are very necessary. By means of their

location the sensor nodes are located in Location

Based Protocol.

The energy estimate can consumed by all the routing

protocol to calculate the distance between two

particular nodes. Location protocol can increase the

lifetime of network. In Location Based routing

protocol the position of sensor nodes are estimated to

route data in the network. Location Based routing

protocol needs some location for the sensor node these

location can be obtained from GPS (Global

Positioning System) signals, Received Radio signals

etc.

GEAR is an example of Location Based routing

protocol.

i. Geographic and Energy-Aware Routing

(GEAR): GEAR is an energy efficient routing

protocol. This protocol is used to find the location

of sensor node in the network. Localization

hardware just like GPS, GIS etc. are fitted in

nodes through this the nodes will know about their

current position, their energy as well as they will

know about their neighbors. It uses energy aware

methods geographical information for sending the

packets towards its destination. At that point

GEAR use recursive geographic forwarding to

spread the packets inside the target region.

Figure 5: Operation of GEAR Protocol

ii. Geographic Adaptive Fidelity (GAF):

It is an energy aware routing protocol. Initially GAF

was proposed for MANETS and mobile ad hoc

networks. But later it can also applied to sensor

networks. GAF is a location based routing protocol. In

GAF nodes use location information through any

system just like GPS, GIS and received radio signal

etc. to locate itself with its nearest neighbors.

Nodes consume energy while transmitting data i.e. at

sending time as well as at the receiving time. In idle

state some amount of energy is used but it is less in

comparison to the active state.

From Discovery to Active state transition:

For finding the equivalent nodes each node exchange

discovery messages. Nodes belongs to the same grid


10

are equivalent. Discovery messages contains

information about the nodes i.e. node id, grid id, node

state and energy level [29-40]. After predefined time

Td the nodes enters into the active state if it doesn’t

receive any other discovery message.

Figure 6: State Operational Model

From Discovery to Sleeping state transition:

In this state if node receives any other discovery

message from another node which have higher energy

level than a node that is enter to the sleep state. At one

time only one node will be in active state the

remaining will be in sleep state. In order to keep the

routing fidelity, the sleeping neighbors will adjust

their sleeping time (Ts). If the active node expires then

another sleeping node become active.

From Active to Sleeping state transition:

Active time show that at what time a node will be in

active state. After active time (Ta), another node

which have higher energy from the rest of nodes in the

grid become active and the current one will go to sleep

state.

From Sleeping to Discovery state transition:

Before wake up a node has to complete its sleep time

and then enter to discovery state. If the node have

higher energy then it will enter to active state

otherwise re-enter into sleep state.

From Active to Discovery state transition:

When a node enters to the discovery phase after a

predefined time (Td) and rebroadcast the discovery

message for time td. If it receives a message from

another node having higher residual energy then it

enter into sleep state else re-enter into active state.

1) Trajectory Based Forwarding:

It is a method to forward packets in a dense ad hoc

network that makes it possible to route a packet over a

predefined path. The source specify the trajectory in a

packet but doesn’t explicitly the path on the hop-by-

hop basis. Based on the location information of its

neighbor a forwarding sensor makes a greedy decision

to determine the next hop that is the closest to the

trajectory fixed by the source sensor.

2) Minimum Energy Communication Network

(MECN) and Small-MECN:

Minimum Energy Communication Network (MECN)

is a Location-Based protocol. This protocol is used for

achieving minimum energy for randomly deployed ad

hoc networks. Which attempt to set up and maintain a

minimum energy network with mobile sensors. This

protocol has two phases.

In the first phase the protocol takes the position of

a two dimensional plane and constructs a sparse

graph which is also called an enclosure graph. It

consist of all the enclosures of each transmit node

in the graph. Enclose graph contains globally

optimal links in terms of energy consumptions.

The second phase finds optimal links on the

enclosure graph. It uses distributed shortest path

algorithm with power consumption as a cost

metric.

Small Minimum Energy Communication

Network (SMECN):This protocol is used to

improve MECN. In this protocol minimal graph

is regarded as with its minimum energy

property. In this protocol every sensor discovers

its immediate neighbors by broadcasting a

discovery message using some initial power that

is updated incrementally.

C. Data centric protocol

In sensor network the data centric protocol is different

from traditional in carrying data .data centric protocol

is query-based i-e the sink send queries to certain

region and wait for the required sensing data in the

sensor located region. Data is being requested through


11

queries. Naming data is essential to specify properties

of data based on attributes .the most important thing it

reduce redundancy in data transmission. The base

station sends queries to the specific sensor region and

waits for the information about that filed transmitting

by the nodes. Thus is very efficient in term of energy

consumption .in data centric protocol the sensor

themselves are less important than their own data

centric protocol is divided into many categories.

1. SPIN

(Sensor Protocol for Information via Negotiation):-it

was design to improve classic flooding protocol. The

plan behind spin is to name the data using high level

descriptor or Meta data. Meta data are changed

between sensors before transmission via a technique of

data advertisement. In spin all information is

broadcasted to each node in the network user can

easily query to any node and can get the information

very soon [41-44].

2. Directed Diffusion

It develop after spin .it is a data dissemination and

aggregation protocol an application aware protocol in

which sensor node generate data and name it by

attribute value paired. The operation is shown below

Figure 7: Directed Diffusion Operation


Comparison of Various Routing Protocols

Comparison of Various Routing Protocols


12

IV.CONCLUSION

In recent year, wireless sensor network is the most

interesting field for the researcher to contribute the

main aim of the routing protocol to improve lifetime

of sensor node for the purpose to improve network

lifetime. This paper is about the classification of

routing protocols into three main categories. Such as,

data centric, hierarchical and location based. In

hierarchical node are divide in cluster and cluster head.

In data centric protocol, all data come from the nodes

to gateway. Then gateway send the data to base station,

in this way gateway is more overloading. In location-

based protocols, need the information of node to

calculate the distance between two nodes to estimate

energy consumption.

V. REFERENCES

[1] https://www.academia.edu/7579636/A_TUTORIA

L_OF_ROUTING_PROTOCOLS_IN_WIRELES

S_SENSOR_NETWORKS_

[2] M. A. Jan, P. Nanda, X. He, and R. P. Liu, “A

Lightweight Mutual Authentication Scheme for

IoT Objects,”, “Submitted”, 2016.

[3] Khan, F., & Nakagawa, K. (2013). Comparative

study of spectrum sensing techniques in cognitive

radio networks. In Computer and Information

Technology (WCCIT), 2013 World Congress on (pp. 1-8). IEEE.

[4] Khan, F., Bashir, F., & Nakagawa, K. (2012). Dual

head clustering scheme in wireless sensor

networks. In Emerging Technologies (ICET), 2012

International Conference on (pp. 1-5). IEEE.

[5] Khan, F., Kamal, S. A., & Arif, F. (2013). Fairness

improvement in long chain multihop wireless ad

hoc networks. In 2013 International Conference

on Connected Vehicles and Expo (ICCVE) (pp.

556-561). IEEE.

[6] Khan, F. (2014). Secure communication and

routing architecture in wireless sensor networks.

In 2014 IEEE 3rd Global Conference on Consumer Electronics (GCCE) (pp. 647-650).

IEEE.

[7] M. A. Jan, P. Nanda, X. He and R. P. Liu,

“PASCCC: Priority-based application-specific

congestion control clustering protocol” Computer

Networks, Vol. 74, PP-92-102, 2014.

[8] Khan, S., & Khan, F. (2015). Delay and

Throughput Improvement in Wireless Sensor and

Actor Networks. In 5th National Symposium on

Information Technology: Towards New Smart World (NSITNSW) (pp. 1-8).

[9] Khan, F., Jan, S. R., Tahir, M., Khan, S., & Ullah,

F. (2016). Survey: Dealing Non-Functional

Requirements at Architecture Level. VFAST Transactions on Software Engineering, 9(2), 7-13.

[10] Khan, F., & Nakagawa, K. (2012). Performance

Improvement in Cognitive Radio Sensor

Networks. the IEICE Japan.

[11] Khan, F., Khan, S., & Khan, S. A. (2015, October).

Performance improvement in wireless sensor and

actor networks based on actor repositioning.

In 2015 International Conference on Connected Vehicles and Expo (ICCVE) (pp. 134-139). IEEE.

[12] M. A. Jan, P. Nanda, X. He and R. P. Liu, “A

Sybil Attack Detection Scheme for a Centralized

Clustering-based Hierarchical Network” in

Trustcom/BigDataSE/ISPA, Vol.1, PP-318-325,

2015, IEEE.

[13] Jabeen, Q., Khan, F., Khan, S., & Jan, M. A.

(2016). Performance Improvement in Multihop

Wireless Mobile Adhoc Networks. the Journal

Applied, Environmental, and Biological Sciences (JAEBS), 6(4S), 82-92.

[14] Khan, F. (2014, May). Fairness and throughput

improvement in multihop wireless ad hoc

networks. In Electrical and Computer Engineering

(CCECE), 2014 IEEE 27th Canadian Conference on (pp. 1-6). IEEE.

[15] Khan, S., Khan, F., Arif, F., Q., Jan, M. A., &

Khan, S. A. (2016). Performance Improvement in

Wireless Sensor and Actor Networks. Journal of

Applied Environmental and Biological

Sciences, 6(4S), 191-200.

[16] Khan, F., & Nakagawa, K. (2012). B-8-10

Cooperative Spectrum Sensing Techniques in

Cognitive Radio Networks, 2012(2), 152.

[17] Khan, F., Jan, S. R., Tahir, M., & Khan, S. (2015,

October). Applications, limitations, and

improvements in visible light communication

systems. In2015 International Conference on

Connected Vehicles and Expo (ICCVE)(pp. 259-

262). IEEE.

[18] Jabeen, Q., Khan, F., Hayat, M. N., Khan, H., Jan,

S. R., & Ullah, F. (2016). A Survey: Embedded

Systems Supporting By Different Operating

Systems. International Journal of Scientific Research in Science, Engineering and Technology

(IJSRSET), Print ISSN, 2395-1990.

[19] Jan, S. R., Ullah, F., Ali, H., & Khan, F. (2016).

Enhanced and Effective Learning through Mobile

Learning an Insight into Students Perception of

Mobile Learning at University Level. International

Journal of Scientific Research in Science, Engineering and Technology (IJSRSET), Print

ISSN, 2395-1990.

[20] Jan, S. R., Khan, F., & Zaman, A. The perception

of students about mobile learning at University

level.


13


Sybil Attack Detection Scheme for a Forest

Wildfire Monitoring Application,” Elsevier Future Generation Computer Systems (FGCS),

“Accepted”, 2016.

[22] Jan, S. R., Shah, S. T. U., Johar, Z. U., Shah, Y., &

Khan, F. (2016). An Innovative Approach to

Investigate Various Software Testing Techniques

and Strategies. International Journal of Scientific

Research in Science, Engineering and Technology


[23] Khan, I. A., Safdar, M., Ullah, F., Jan, S. R., Khan,

F., & Shah, S. (2016). Request-Response

Interaction Model in Constrained Networks. In

International Journal of Advance Research and

Innovative Ideas in Education, Online ISSN-2395-4396

[24] Azeem, N., Ahmad, I., Jan, S. R., Tahir, M., Ullah,

F., & Khan, F. (2016). A New Robust Video

Watermarking Technique Using H. 264/AAC

Codec Luma Components Based On DCT. In International Journal of Advance Research and

Innovative Ideas in Education, Online ISSN-2395-

4396

[25] Jan, S. R., Khan, F., Ullah, F., Azim, N., & Tahir,

M. (2016). Using CoAP Protocol for Resource

Observation in IoT. International Journal of

Emerging Technology in Computer Science &

Electronics, ISSN: 0976-1353

[26] Azim, N., Majid, A., Khan, F., Jan, S. R., Tahir,

M., & Jabeen, Q. (2016). People Factors in Agile

Software Development and Project Management.

In International Journal of Emerging Technology

in Computer Science & Electronics (IJETCSE) ISSN: 0976-1353

[27] Azim, N., Majid, A., Khan, F., Tahir, M., Safdar,

M., & Jabeen, Q. (2016). Routing of Mobile Hosts

in Adhoc Networks. In International Journal of


Electronics (IJETCSE) ISSN: 0976-1353.

[28] Azim, N., Khan, A., Khan, F., Majid, A., Jan, S. R.,

& Tahir, M. (2016) Offsite 2-Way Data Replication toward Improving Data Refresh

Performance. In International Journal of

Engineering Trends and Applications, ISSN: 2393

– 9516

[29] Tahir, M., Khan, F., Jan, S. R., Azim, N., Khan, I.

A., & Ullah, F. (2016) EEC: Evaluation of Energy

Consumption in Wireless Sensor Networks. . In

International Journal of Engineering Trends and Applications, ISSN: 2393 – 9516

[30] M. A. Jan, P. Nanda, M. Usman, and X. He,

“PAWN: A Payload-based mutual Authentication

scheme for Wireless Sensor Networks,”

Concurrency and Computation: Practice and Experience, “accepted”, 2016.

[31] Azim, N., Qureshi, Y., Khan, F., Tahir, M., Jan, S.

R., & Majid, A. (2016) Offsite One Way Data

Replication towards Improving Data Refresh

Performance. In International Journal of Computer Science Trends and Technology, ISSN:

2347-8578

[32] Safdar, M., Khan, I. A., Ullah, F., Khan, F., & Jan,

S. R. (2016) Comparative Study of Routing

Protocols in Mobile Adhoc Networks. In

International Journal of Computer Science Trends

and Technology, ISSN: 2347-8578

[33] Tahir, M., Khan, F., Babar, M., Arif, F., Khan, F.,

(2016) Framework for Better Reusability in

Component Based Software Engineering. In the

Journal of Applied Environmental and Biological

Sciences (JAEBS), 6(4S), 77-81.

[34] Khan, S., Babar, M., Khan, F., Arif, F., Tahir, M.

(2016). Collaboration Methodology for Integrating

Non-Functional Requirements in Architecture. In the Journal of Applied Environmental and

Biological Sciences (JAEBS), 6(4S), 63-67

[35] Jan, S.R., Ullah, F., Khan, F., Azim, N., Tahir, M.,

Khan, S., Safdar, M. (2016). Applications and

Challenges Faced by Internet of Things- A Survey.

In the International Journal of Engineering Trends and Applications, ISSN: 2393 – 9516

[36] Tahir, M., Khan, F., Jan, S.R., Khan, I.A., Azim, N.

(2016). Inter-Relationship between Energy

Efficient Routing and Secure Communication in

WSN. In International Journal of Emerging

Technology in Computer Science & Electronics (IJETCSE) ISSN: 0976-1353.




14

Detecting BOT Victim in Client Networks

Abinaya. E, Balamurugan. K

Department of Information Technology, St Peter Engineering College, Avadi, Tamil Nadu, India

ABSTRACT

In this paper we discuss my research in detecting bot victim in client networks. Botnets are collections of Internet

hosts (―bots‖) that, through malware infection, have fallen under the control of a single entity (―botmaster‖). Botnets

perform network scanning for different reasons: propagation, enumeration, penetration. One common type of

scanning, called ―horizontal scanning,‖ systematically probes the same protocol port across a given range of IP

addresses, sometimes selecting random IP addresses as targets. To infect new hosts in order to recruit them as bots,

some botnets, e.g., Conficker perform a horizontal scan continuously using self-propagating worm code that exploits

a known system vulnerability. In this project, we focus on a different type of botnet scan—one performed under the

explicit command and control of the botmaster, occurring over a well-delimited interval.

Keywords: Horizontal Scanning, Botmaster, Bots, P2P, IRC, BotGraph, DPI, Clustering

I. INTRODUCTION

Existing system contains a fundamental disadvantage

of centralized C&C servers are that they represent a

single point of failure. In order to overcome this

problem, botmasters have recently started to build

botnets with a more resilient C&C architecture, using

a peer-to-peer (P2P) structure or hybrid

P2P/centralized C&C structures. Detecting botnets is

of great importance. However, designing an effective

P2P-botnet detection system is faced with several

challenges. I am confident that this software package

can be readily used by non-programming personal

avoiding human handled chance of error.

Peer-to-peer (P2P) botnets have a random organization

and operate without a C&C server. Bot software

maintains a list of trusted computers (including other

infected machines), information drop locations and

locations where the machines can update their

malware. More advanced botnets use encryption in

order to hide communications between bots.

The purpose of decentralization is to help evade

detection and make it harder for security researchers to

access communications than is the case with a

conventional botnet topology. The lack of a command-

and-control server makes it less likely that detection of

a single bot can lead to investigators taking down the

entire network.


2. Related Works

Botnets have been an active area of research for

almost a decade, starting with early generation botnets

that used IRC channels to implement centralized

Command & Control (C&C) infrastructures.

Botnets commonly scan large segments of Internet

address space, seeking hosts to either infect or

compromise, or for the purpose of network mapping

and service discovery. Analyzing and detecting these

events can improve our understanding of evolving

botnet characteristics and spreading techniques, our

ability to distinguish them from benign traffic sources,

and our ability to mitigate attacks.

Since Sality is one of the largest known botnets but

relatively undocumented in research literature, another

contribution of our study is to shed light on the

scanning behavior of this new-generation botnet.


15

2.1 P2P as botnet command and control: a deeper

insight

The research community is now focusing on the

integration of peer-to-peer (P2P) concepts as

incremental improvements to distributed malicious

software networks (now generically referred to as

botnets). While much research exists in the field of

P2P in terms of protocols, scalability, and availability

of content in P2P file sharing networks, less exists

(until this last year) in terms of the shift in C&C from

central C&C using clear-text protocols, such as IRC

and HTTP, to distributed mechanisms for C&C where

the botnet becomes the C&C, and is resilient to

attempts to mitigate it. In this paper we review some

of the recent work in understanding the newest botnets

that employ P2P technology to increase their

survivability, and to conceal the identities of their

operators. We extend work done to date in explaining

some of the features of the Nugache P2P botnet, and

compare how current proposals for dealing with P2P

botnets would or would not affect a pure-P2P botnet

like Nugache. Our findings are based on a

comprehensive 2-year study of this botnet.

.

Figure 1. Structure of the Botnet

2.2 Experiences in Malware Binary DE obfuscation

Malware authors employ a myriad of evasion

techniques to impede automated reverse engineering

and static analysis sorts. The most popular

technologies include `code obfuscators' that serve to

rewrite the original binary code to an equivalent form

that provides identical functionality while defeating

signature-based detection systems. These systems

significantly complicate static analysis, making it

challenging to uncover the malware intent and the full

spectrum of embedded capabilities. While code

obfuscation techniques are commonly integrated into

contemporary commodity packers, from the

perspective of a reverse engineer, DE obfuscation is

often a necessary step that must be conducted

independently after unpacking the malware binary.

2.3 Internet Traffic Classification Using Bayesian

Analysis Techniques

Accurate traffic classification is of fundamental

importance to numerous other network activities, from

security monitoring to accounting, and from Quality of

Service to providing operators with useful forecasts

for long-term provisioning. We apply a Na¨ve Bayes

estimator to categorize traffic by application.

Uniquely, our work capitalizes on hand-classified

network data, using it as input to a supervised Na¨ve

Bayes estimator. In this paper we illustrate the high

level of accuracy achievable with the Na¨ve Bayes

estimator. We further illustrate the improved accuracy

of renewed variants of this estimator.

2.3.1 BotGraph : Large Scale Spamming Botnet

Detection

Network security applications often require analyzing

huge volumes of data to identify abnormal patterns or

activities. The emergence of cloud-computing models

opens up new opportunities to address this challenge

by leveraging the power of parallel computing. In this

paper, we design and implement a novel system called

BotGraph to detect a new type of botnet spamming

attacks targeting major Web email providers. Bot-

Graph uncovers the correlations among botnet

activities by constructing large user-user graphs and

looking for tightly connected subgraph components.

This enables us to identify stealthy botnet users that

are hard to detect when viewed in isolation.

2.3.2 BotGraph : Large Scale Spamming Botnet

Detection

Network security applications often require analyzing

huge volumes of data to identify abnormal patterns or

activities. The emergence of cloud-computing models

opens up new opportunities to address this challenge

by leveraging the power of parallel computing. In this


16

paper, we design and implement a novel system called

BotGraph to detect a new type of botnet spamming

attacks targeting major Web email providers. Bot-

Graph uncovers the correlations among botnet

activities by constructing large user-user graphs and

looking for tightly connected subgraph components.

This enables us to identify stealthy botnet users that

are hard to detect when viewed in isolation.

2.4 Understanding Churn in Peer-to-Peer

Networks

The dynamics of peer participation, or churn, are an

inherent property of Peer-to-Peer (P2P) systems and

critical for design and evaluation. Accurately

characterizing churn requires precise and unbiased

information about the arrival and departure of peers,

which is challenging to acquire? Prior studies show

that peer participation is highly dynamic but with

conflicting characteristics. Therefore, churn remains

poorly understood, despite its significance.

.

2.5 Boosting the Scalability of Botnet Detection

Using Adaptive Traffic Sampling

Botnets pose a serious threat to the health of the

Internet. Most current network-based botnet detection

systems require deep packet inspection (DPI) to detect

bots. Because DPI is a computational costly process,

such detection systems cannot handle large volumes of

traffic typical of large enterprise and ISP networks. In

this paper we propose a system that aims to efficiently

and effectively identify a small number of suspicious

hosts that are likely bots. Their traffic can then be

forwarded to DPI-based botnet detection systems for

fine-grained inspection and accurate botnet detection.

Bot Attack

Figure 2. Bot Attack in P2P system over 2014 end and

2015 start

2.5.1 P2P Botnet Detection using Behavior

Clustering & Statistical Tests

Most recent research on botnet detection focuses on

centralized botnets and primarily relies on two

assumptions: prior knowledge of potential C&C

channels and capability of monitoring them. However,

when botnets switch to a P2P (peer-to-peer) structure

and utilize multiple protocols for C&C, the above

assumptions no longer hold. Consequently, the

detection of P2P botnets is more difficult. In this

paper, we relax the above two assumptions and focus

on C&C channel detection for P2P botnets that use

multiple protocols (randomly chosen) for C&C.


Proposed Work

Sality is one of the largest botnets ever identified by

researchers. Its behavior represents ominous advances

in the evolution of modern malware: the use of more

sophisticated stealth scanning strategies by millions of

coordinated bots, targeting critical voice

communications infrastructure. This project offers a

detailed dissection of the botnet's scanning behavior,

including general methods to correlate, visualize, and

extrapolate botnet behavior across the global Internet.

Since bots are malicious programs used to perform

profitable malicious activities, they represent valuable

assets for the botmaster, who will intuitively try to

maximize utilization of bots. This is particularly true

for P2P bots because in order to have a functional

overlay network (the botnet), a sufficient number of

peers needs to be always online.

We need flow clustering-based analysis approach to

identify hosts that are mostly likely running P2P

applications. Approach does not rely on any transport

layer used by which can be easily violated by P2P

applications.

This project offers a detailed dissection of the botnet's

scanning behavior, including general methods to

correlate, visualize, and extrapolate botnet behavior

across the global Internet

The implementation can be done using Java, and the

following codes i.e., Coarse Grained Peer-To-Peer

Detection, File Uploading and Sending Bot Detection,


17

Clustering and Eliminating, Detection of Attacker IP

Address

Figure 3. Project Model

3.1 Coarse Grained Peer-To-Peer Detection

This component is responsible for detecting P2P

clients by analyzing the remaining network flows after

the Traffic Filter component. For each host h within

the monitored network we identify two flow sets,

denoted as Stcp(h) and Sudp(h), which contain the

flows related to successful outgoing TCP and UDP

connection, respectively. We consider as successful

those TCP connections with a completed SYN,

SYN/ACK, ACK handshake, and those UDP (virtual)

connections for which there was at least one ―request‖

packet and a consequent response packet.

3.2 File Uploading and Sending

This module is used to upload required file from

storage device to user account and send the file into

destination account. There are many different types of

files: data files, text files, program files, directory

files, and so on. Different types of files store different

types of information.

3.3 Bot Detection

Since bots are malicious programs used to perform

profitable malicious activities, they represent valuable

assets for the botmaster, who will intuitively try to

maximize utilization of bots. This is particularly true

for P2P bots because in order to have a functional

overlay network (the botnet), a sufficient number of

peers needs to be always online. In other words, the

active time of a bot should be comparable with the

active time of the underlying compromised system.

3.4 Clustering and Eliminating

The distance between two flows is subsequently

defined as the Euclidean distance of their two

corresponding vectors. We then apply a clustering

algorithm to partition the set of flows into a number of

clusters. Each of the obtained clusters of flows, Cj (h),

represents a group of flows with similar size.

3.4 Clustering and Eliminating Bot using Coarse

grained Botnet detection technique

For each Cj (h), we consider the set of destination IP

addresses related to the flows in the clusters, and for

each of these IPs we consider its BGP prefix (using

BGP prefix announcements).

3.5 Detection of Attacker IP Address

In this module used to determine the geographical

location of website visitors based on the IP addresses

for applications such as fraud detection. We can find

the IP address of the attacker.

IV.CONCLUSION

We also identify the performance bottleneck of our

system and optimize its scalability. We presented a

novel botnet detection system that is able to identify

stealthy P2P botnets, whose malicious activities may

not be observable.

V. FUTURE ENHANCEMENT

To summarize, although our system greatly enhances

and complements the capabilities of existing P2P

botnet detection systems, it is not perfect. We should

definitely strive to develop more robust defense

techniques, where the aforementioned discussion


18

outlines the potential improvements of our system.

Botnet developers are constantly improving their

development in order to produce more and more

stealthy malware for all kinds of attacks to make

profit. While various approaches have been studied or

used for botnet attacks, the risk of exploiting widely

used browser extensions and their automatic browser

extension update mechanisms for command and

control channel has not been practically investigated.

In this study, we show that it is not difficult to

construct stealthy botnet via browser extensions.

VI.REFERENCES

[1] S. Stover, D. Dittrich, J. Hernandez, and S.

Dietrich, "Analysis of the storm and nugache

trojans: P2P is here," in Proc. USENIX, vol. 32.

2007, pp. 18–27.

[2] P. Porras, H. Saidi, and V. Yegneswaran, "A

multi-perspective analysis of the storm

(peacomm) worm," Comput. Sci. Lab., SRI Int.,

Menlo Park, CA, USA, Tech. Rep., 2007.

[3] P. Porras, H. Saidi, and V. Yegneswaran.

(2009). Conficker C Analysis Online].

Available:

http://mtc.sri.com/Conficker/addendumC/index.

html

[4] G. Sinclair, C. Nunnery, and B. B. Kang, "The

waledac protocol: The how and why," in Proc.

4th Int. Conf. Malicious Unwanted Softw., Oct.

2009, pp. 69–77.

[5] R. Lemos. (2006). Bot Software Looks to

Improve Peerage Online]. Available:

http://www.securityfocus.com/news/11390

[6] Y. Zhao, Y. Xie, F. Yu, Q. Ke, and Y. Yu,

"Botgraph: Large scale spamming botnet

detection," in Proc. 6th USENIX NSDI, 2009,

pp. 1–14.

[7] G. Gu, R. Perdisci, J. Zhang, and W. Lee,

"Botminer: Clustering analysis of network

traffic for protocol- and structure-independent

botnet detection," in Proc. USENIX Security,

2008, pp. 139–154.

[8] T.-F. Yen and M. K. Reiter, "Are your hosts

trading or plotting? Telling P2P file-sharing and

bots apart," in Proc. ICDCS, Jun. 2010, pp. 241–

252.

[9] S. Nagaraja, P. Mittal, C.-Y. Hong, M. Caesar,

and N. Borisov, "BotGrep: Finding P2P bots

with structured graph analysis," in Proc.

USENIX Security, 2010, pp. 1–16.

[10] J. Zhang, X. Luo, R. Perdisci, G. Gu, W. Lee,

and N. Feamster, "Boosting the scalability of

botnet detection using adaptive traffic

sampling," in Proc. 6th ACM Symp. Inf.,

Comput.Commun. Security,




19

A New Approach for Transistor-Clamped H-Bridge Multilevel Inverter with voltage Boosting Capacity

Suparna Buchke, Prof. Kaushal Pratap Sengar

TIT, Madhya Pradesh, India

ABSTRACT

Multilevel converters offer high power capability, resulting with lower output harmonics and lower commutation

losses. Their main disadvantage is their complexity, requiring a great number of power devices and passive

components, and a rather complex control circuitry. This paper presents a new topology of the multilevel inverter

with feature like output voltage boosting capability along with capacitor voltage balancing .The proposed multilevel

inverter uses conventional transistor clamped H-bridge (TCHB) with an bidirectional switch and four auxillary

switches producing a boost output voltage . The single unit of new topology produces five-level output with output

voltage double the input DC voltage where as a single unit of conventional H-bridge produces three-level ouput

voltage similar to input DC voltage. A novel universal control scheme is used which results in balanced distribution

of power among H-bridge cells. This control scheme can also be used for the charge balance control with multiple

input DC sources in any given topology. The analysis of the output voltage harmonics is carried out and compared

with previous topology and the conventional cascaded H-bridge inverter topology. The proposed multilevel inverter

topology is modelled using matlab/simulink. From the results the proposed inverter provides more output voltage.

Keywords : Multilevel Inverter, Cascaded H-Bridge, Multicarrier Phase Width Modulation, Transistor Clamped

Inverter, Cascaded Neutral –Point Clamped Inverter.

I. INTRODUCTION

There are various application varying from medium

voltage to high voltage high power application which

requires DC to AC conversion using multilevel

inverters. The research on multilevel inverter is

ongoing further to reduce the number of switching

devices count to reduce the manufacturing cost,

capacitor voltage balancing .The inverters with number

of voltage levels equal to three or above than that are

known as the multilevel inverters. Multilevel inverters

are capable of producing high power high voltage as

the unique structure of the multilevel voltage source

inverter allows to reach high voltages with low

harmonics without the use of transformers or series

connected synchronized switching devices. As the

number of voltage levels increases , the harmonic

content of the output voltage waveform decreases. The

synthesized multilevel outputs are superior in quality

which results in reduced filter requirements [1].

There are three major multilevel voltage source

inverter topologies neutral-point lamped inverter (i.e

diode clamped) , flying capacitor (capacitor-clamped)

and cascaded H-bridge multilevel inverter . There are

also various other topologies which have been

proposed and have successfully adopted in various

industrial applications. The novel universal multi-

carrier PWM control scheme is used .This paper

mainly focuses mainly on the cascaded H-bridge

inverter topology. the cascaded multilevel inverter has

the potential to be the most reliable out of three

topologies . It has the best fault tolerance owing to its

modularity a feature that enables the inverter to

continue operate at lower power levels after cells

failure[2]. Due to the modularity of the cascaded

multilevel inverter it can be stacked easily for high

power and high voltage applications. The cascaded


20

multilevel inverter mainly consists of several identical

H-bridge cells, which are cascaded, in series from the

output side. The cascaded H-bridge (CHB) may further

be classified as symmetrical if the DC bus voltage is

equal in all the series power cells and as asymmetrical

if the DC bus voltage is not same for each power cell.

The symmetrical CHB is more advantageous over the

asymmetrical CHB in terms of modularity,

maintenance and cost. In case of the asymmetrical

CHB DC bus voltage is varied in each power as per the

requirement to increase the voltage levels [2]. In case

of the symmetrical CHB the voltage, level can be

increased without varying the DC voltage with same

number of power cells. The transistor clamped

topology is popular now a days a provides provision to

increase the output levels by taking different voltage

levels from the series stacked capacitors [1]. In this

paper the new configuration of the symmetrical H-

bridge is proposed which produces a five-level output

voltage similar to conventional transistor clamped

topology Instead of three-level as in case of

conventional H-bridge. However, this new proposed

topology produces the boost output voltage in

comparison to conventional transistor clamped

topology, which also produces the five-level output but

the output voltage equal to the DC voltage.


Proposed Inverter Configuration

The conventional cascaded H-bridge inverter consists

of DC voltage for each H-bridge and only four

switching devices. The value of the DC voltage in each

bridge depends whether the configuration is symmetric

or unsymmetric. Fig.1 shows the conventional H-

bridge. The general block diagram for the proposed

inverter is shown in fig.2 and the general configuration

of the proposed inverter topology is shown in fig.3

which represents a single cell which produces the five-

level output with boost output voltage. It consist of

total of 8 switches in a single cell along with an

additional bidirectional switch consisting of S11 and

S11’ which is connected between the first leg of the H-

bridge and the capacitor midpoint, enabling five output

voltage levels (+2Vdc , +Vdc , 0 , -Vdc , -2Vdc) based

on the switching combination . The switches

S21,S31,S41,S51 forms the H-bridge and the switches

Sa1,Sa2,Sa3,Sa4 are connected in the same leg which

plays a role in boosting the voltage and the input DC

voltage is connected with positive terminal between

the switches Sa1 and Sa2 and the negative terminal

between the switches Sa3 and Sa4. The capacitor

voltage divider is formed by C1 and C2.

Figure 1. Conventional cascaded H-bridge

Figure 2. General block diagram of new topology

Figure 3. Topology of five-level transistor clamped H-

bridge with boost output voltage for each cell


21

Figure 4. Configuration of the proposed 1-phase

transistor clamped cascaded H-bridge inverter using

two cells

Operation of proposed inverter topology

The working of the single cell of the proposed inverter

topology is explained telling how the required five

level output is produced:

1. Maximum positive output that can be produced is

the double of the input DC voltage i.e 2Vdc which

is produced when S21 is on connecting the load

positive terminal to the load and S51 is on

connecting the load negative terminal to the Vdc

thus the total output voltage is 2Vdc. The output

voltage level Vdc is obtained when Sa1, S11, S51

and Sa2 gets turned on other switches remaining

off.

2. Maximum negative output is -2Vdc, which is

produced when switches S41 and S31 gets turned

on connecting the negative and positive terminal of

the load respectively to the input source. The

negative level –Vdc is obtained when switches

Sa1, Sa3, S11, S41 are turned on other switches

remaining off.

The look up table for the proposed inverter is given in

the figure given below.

Voltage

level

+2Vdc +Vdc 0 -Vdc -

2Vdc

Sa1 0 0 0 1 0

Sa2 0 1 0 0 0

Sa3 0 0 0 1 0

Sa4 0 1 0 0 0

S11 0 1 0 1 0

S21 1 0 1 0 0

S31 0 0 0 0 1

S41 0 0 1 1 1

S51 1 1 0 0 0

Table 1. Look up table for the proposed TCHB

Volta

ge

level

+4

V

+3

V

+2

V

+1

V

0

V

-1V -

2V

-3V -4V

Sa1 0 0 0 0 0 1 1 0 0

Sa2 0 0 1 1 0 0 0 0 0

Sa3 0 0 0 0 0 1 1 0 0

Sa4 0 0 1 1 0 0 0 0 0

S11 0 0 1 1 0 1 1 0 0

S21 1 1 0 0 1 0 0 0 0

S31 0 0 0 0 0 0 0 1 1

S41 0 0 0 0 1 1 1 1 1

S51 1 1 1 1 0 0 0 0 0

Sb1 0 0 0 0 0 0 1 1 0

Sb2 0 1 1 0 0 0 0 0 0

Sb3 0 0 0 0 0 0 1 1 0

Sb4 0 1 1 0 0 0 0 0 0

S12 0 1 0 0 0 0 1 1 0

S22 1 0 1 1 1 1 0 0 0

S32 0 0 0 0 0 0 0 0 1

S42 0 0 0 1 1 1 1 1 1

S52 1 1 1 0 0 0 0 0 0

Table 2. Lookup table for single phase proposed

transistor clamped H-bridge inverter

PWM Control Scheme

Multilevel inverter has to synthesize a staircase

waveform by using the modulation technique to have

the controlled output voltage. There is variety of

modulation techniques available. The control

technique can be classified as the pulse width

modulation, which is considered as the most efficient


22

method. This PWM is further divided into various

PWM techniques such as single pulse PWM, space

vector PWM, multiple pulse PWM, phase

displacement control []. For this proposed topology,

we are using the multicarrier based control technique,

which can be applied, to all the topologies of the

multilevel inverter. For any given number of levels in

the output voltage the number of carrier to be used is

given as N-1.

Where N is the number of levels in the output voltage.

Simply a reference signal is taken which is a sinusoidal

signal of 50Hz frequency and this reference is

compared with the carrier signal which are the

triangular wave .The modulation index we are using in

this modulation technique is 0.95.The advantage of

this scheme is that it offers the charge balance control

in the input DC sources and voltage across the

capacitor are also balanced[4].

Table 3. Multicarrier based control scheme for the

proposed topology

Figure 4. 5-level output voltage waveform of single

phase cascaded H-bridge inverter with tw

Figure 5. 9-level output voltage waveform of single

phase proposed transistor clamped H-bridge inverter

with two bridges.


Comparison of Proposed Topology with Cascaded

H-Bridge Topology

The purpose of research for the multilevel inverter

includes to get a quality power output with the reduced

number of switching devices, balancing of the

capacitors, reduced number of clamping diodes in

order to reduce the overall cost of the multilevel

inverter. In the proposed multilevel inverter topology,

the number of switches is more in comparison to the

conventional CHB but we get the five-level in the

output voltage, which results in reduced THD. In

addition, the input DC voltage source required is half

of the voltage source required in the conventional

CHB. much superior than the cascaded H-bridge

topology in terms of the number of level in the output

voltage, magnitude of the output voltage, total

harmonic distortion. To produce the same output

voltage the cascaded H-bridge has to use the two cells

whereas only one cell is required with the proposed

topology. Fig.3 is showing the single-phase inverter

consisting of two cells of the proposed topology each

cell is having input 100V DC voltage and the output ac

voltage is 400V each of which is producing 200V. The

total harmonic distortion produced by the proposed

inverter is 11.49% only, which is very low as

compared to the conventional cascaded H-bridge

inverter having THD of 37.64%, which is 26.15%

more than the proposed topology. In order to produce

the nine levels in the output voltage the cascaded H-

bridge requires three cells whereas the proposed

topology requires only two cells.


23

Figure 6. THD in % for single phase cascaded H-

bridge multilevel inverter

Figure 7. THD in % for double phase cascaded H-

bridge multilevel inverter

IV.CONCLUSION

The proposed multilevel inverter topology is much

superior to the conventional cascaded H-bridge

topology in terms of the number of level in the output

voltage, magnitude of the output voltage, total

harmonic distortion (THD). To produce the same

output voltage the cascaded H-bridge has to use the

two cells whereas only one cell is required with the

proposed topology or in other words input DC voltage

sorce required in proposed topologi is half of that

required in conventional CHB.

V. REFERENCES

[1] Mahajan Sagar Bhaskar Ranjana,"MULTILEVEL

INVERTER WITH LEVEL SHIFTING SPWM

TECHNIQUE USING FEWER NUMBER OF

SWITCHES FOR SOLAR

APPLICATIONS",еISSN: 2319-1163,IJRET:

Intеrnational Journal of Resеarch in Engineеring

and Tеchnology,Volumе: 04 Issuе: 10,Oct-2015.

[2] Mr. D. Santhosh Kumar Yadav,"Analysis of

Cascadеd Multilevеl Invertеrs with Seriеs

Connеction of H-Bridgе in PV Grid",Intеrnational

Journal of Enhancеd Resеarch in Sciencе

Tеchnology & Engineеring, ISSN: 2319-7463,Vol.

4 Issuе 4, April-2015, pp: (101-106).

[3] Krishna Kumar Gupta,"Multilevеl Invertеr

Topologiеs With Reducеd Devicе Count: A

Reviеw",IEEE TRANSACTIONS ON POWER

ELECTRONICS, VOL. 31, NO. 1, JANUARY

2016.

[4] Ebrahim Babaеi,"A Singlе-Phasе Cascadеd

Multilevеl Invertеr Basеd on a New Basic Unit

With Reducеd Numbеr of Powеr Switchеs",IEEE

TRANSACTIONS ON INDUSTRIAL

ELECTRONICS, VOL. 62, NO. 2, FEBRUARY

2015.

[5] Vahid Dargahi, “A New Family of Modular

Multilevеl Convertеr Basеd on Modifiеd Flying-

Capacitor Multicеll Convertеrs",IEEE

TRANSACTIONS ON POWER ELECTRONICS,

VOL. 30, NO. 1, JANUARY 2015.

[6] Sourabh Rathorе, Mukеsh Kumar Kirar and S. K

Bhardwaj "SIMULATION OF CASCADED H-

BRIDGE MULTILEVEL"




24

An Improved Performance of Greedy Perimeter Stateless Routing protocol of Vehicular Adhoc Network in Urban Realistic Scenarios

Ritesh Gupta, Parimal Patel

Department of Computer Engineering, S. P. B Patel Engineering College, Mehsana, Gujarat, India

ABSTRACT

Vehicular Ad Hoc Networks (VANETs) or Inter-Vehicle Communication (IVC) is an extension to a popular Mobile

Ad Hoc Networks (MANETs) technology. VANET is developed to provide comfort communication between the

vehicle while driving. In VANET there is a continuous wireless data transmission occurs either between Road Side

Units (RSUs) or On Board Units (OBUs) in the vehicles. To keep the transmission smooth it required a good routing

protocol. Right from the inception of VANET technology in 2000s the work done only on basic routing protocol.

Mobility model is one of the key parameter while designing the vehicular network. In this paper the Simulation of

Urban Mobility (SUMO) and Mobility Model Generator for VANET (MOVE) are used for creating scenarios and

traffic. The real time maps are edited in JAVA open street map editor (JOSM) and the simulation is done in NS-2.

The performance is evaluated by using the two routing protocol on the basic of packet delivery ratio and end to end

delay for Urban scenarios.

Keywords : VANET (Vehicular Adhoc Network.) SUMO (Simulation of Urban Mobility), MOVE

I. INTRODUCTION

Vehicular Ad Hoc Network (VANET) is a fast

growing technology in today’s world. The

fundamental idea behind implementing VANET is to

offer information sharing, supportive driving,

providing navigation and safety to human life in fast

moving vehicles. The communication takes place

either between vehicle-to-vehicle (V2V) or between

vehicles-to infrastructure (V2I). On Board Unit

(OBU) that is fixed on vehicle is responsible for

collecting data from various sensors, which gives

condition of that vehicle. OBU send this data either to

other vehicle or to Road Side Unit (RSU). On the

other hand, RSU is a fixed infrastructure situated

along the sides of road whose work is to broadcast the

information to other vehicles. However, due to high

mobility and dynamic topology of VANET

discovering and maintaining routes is very challenging

task in VANET. To achieve an effective vehicular

communication, vehicular network must be available

all time in real time. A small delay in sending or

receiving of message may lead to devastating results.

Due to rapid changing topology, there are numerous

technical hitches in designing a Routing Protocol of

VANET.[1] Routing is the process of moving packets

from a source to a destination and Routing Protocols

are the one who decide how those packets are going to

move. Routing occurs at Layer3 (network layer) of the

OSI reference model via some logical addressing.

Routing protocols plays a key role in path discovery

so; it becomes important for routing protocol to give

effective result in real time.

In this paper, as shown in figure 1 of process flow, we

have taken the urban realistic scenario for

simulation .The real maps are taken from the open

street map for urban realistic scenario. The maps are

edited in Java Open Street Map editor (JOSM) to

remove the unwanted areas like buildings, rivers etc.

after the editing of real maps the output file is given to

the SUMO (Simulation of Urban Mobility) for

simulating the real traffic scenario of vehicular

network. The output of this SUMO is used in network

Simulator (NS-2) for the analysis of various QoS

parameters.


25

Figure 1: Process flow for capturing real time

mobility model


Introduction to Routing Protocols

VANET Routing protocol has significant role in

performance because of sending &receiving packets

between sources to destinations. There are number of

routing protocols has developed for wireless Adhoc

network. VANET routing protocol [1][2]basically

classified into two types: Proactive and reactive

routing protocols.

In proactive routing protocol, it maintains the route

information at all nodes and update the table

accordingly. In reactive routing protocol, it

maintaining the route information for nodes on

demand.

In this paper, the simulation and comparison is

performed on the basis of two different routing

protocols. [13] GPSR (Greedy Perimeter Stateless

Routing) & MGPSR (Modified Greedy Perimeter

Stateless Routing) protocols.

a) GPSR:

Greedy perimeter stateless routing (GPSR) is the best

known position based routing protocol for VANETs.

GPSR makes greedy forwarding decisions using only

information about a router’s immediate neighbors in

the network topology.

GPSR consists of two methods for forwarding

packets:

1. Greedy Forwarding

2. Perimeter Forwarding

Greedy Forwarding is used to send data to the

closest nodes to destination. Perimeter Forwarding

is used where Greedy Forwarding fails

1. Greedy Forwarding

Find neighbors who are the closer to the

destination

Forward the packet to the neighbor closest to the

destination

Figure 2 : Greedy Forwarding Method

Figure 3: Greedy Forwarding does not always work

2. Perimeter Forwarding

Apply the right-hand rule to traverse the edges of

a void

Pick the next anticlockwise edge

Figure 3 : Perimeter Forwarding with Void: Right-

Hand Rule


26

Figure 4 : Perimeter Forwarding Pick the next

anticlockwise edge

b) MGPSR:

MGPSR is the extension to the GPSR protocol for

computing effective communication among the nodes

which substantially increases network lifetime of

nodes.

Figure 5 : Modified Greedy Perimeter Stateless Routing

(MGPSR)


Simulation In Josm (Java Open Street Map Editor)

[17] JOSM is a desktop editing application, written in

java. It supports loading standalone GPX tracks and

GPX track from OSM database as well as loading and

editing existing nodes, ways, metadata tags and

relations from the OSM.

Figure 6 : Road Map for Vehicles of Urban Based

Scenario

The map in Figure 6 is taken from

http://openstreetmap.org, which is available free for

downloading via their export map feature.

Figure 7: Map of Urban Area – City Based Scenario Figure 8: Map of Urban Area – City Based Scenario for road in JOSM


27

As shown in figure 7 & 8, the downloaded maps are

saved in “.osm” file format that can be edited in

JOSM and from “map.osm”. In figure 5 the urban area

of city based scenario contains buildings, Trees, traffic

and other unwanted streets are removed. In Figure 8

all these unwanted parts are edited in JOSM, and only

the roads are remain for the traffic simulation. So, that

the file size become small and to lessen the

unnecessary computation. We can import that file in

[16] SUMO and create traffic environment.

Simulation in Sumo (Simulation of Urban

Mobility)

To generate vehicle traffic in [16] SUMO the tools

like "net convert”, “poly converts" and

"randomTrips.py" are used.

Net convert can imports road networks from

different sources (openstretmap.org) and generates

road networks that can be used in SUMO. It will

identify the Nodes, Junctions, and Signals etc and

build the network file that is compatible with

SUMO.

Poly convert imports geometrical shapes

(polygons - buildings) from different sources &

converts them to a representation that visualized in

SUMO-GUI.

RandomTrips.Py is used to generate random

routes.

From the above steps, we get the SUMO configuration

(medical.sumo.cfg) file in which we have to give path

of both the network file and route file. The

configuration file is used to like merge the network

file and route file

Figure 9 : Road network of urban area showing the

simulation of vehicles in Sumo (Traffic Simulator for

50ms delay)

Figure 10 : Imported map from JOSM in SUMO

Figure 11 : simulation of view of Traffic in SUMO

(Traffic Simulator for 100ms delay)

Simulation & Results

The mobility model of SUMO is given to the network

simulator 2 (NS-2) for simulation. We have done the

simulation using two different routing protocols for

urban realistic scenario. The simulation is done using

different number of nodes.

Figure 12 (a): Simulation in NS2 for urban realistic

scenario for 100 nodes


28

Figure 12 (b) : Simulation in NS2 for urban realistic

scenario for 120 nodes

In Figure 13, The Packet Delivery Ratio increases

with different number of nodes

Figure 13 : Comparison graph of PDR vs Number of Nodes

In Figure 14, The Packet Delivery Ratio increases with different node speed

Figure 14 : Comparison graph of PDR vs Node speed

In Figure 15, the average end to end delay decreases with different number of nodes

Figure 15 : Comparison graph of E2ED vs Number of Nodes

In Figure 16, the average end to end delay decreases with

different node speed

Figure 16 : Comparison graph of E2ED vs Node speed

IV.CONCLUSION

The simulation of urban mobility (SUMO) gives the

better mobility model after compiling in java Open

Street Map editor (JOSM) for urban realistic

scenarios. In this paper we simulate the GPSR &

MGPSR routing protocol for analyzing the packet

delivery ration and end to end delay parameters. As a

result Performance of Modified MGPSR gives better

results than GPSR for urban realistic scenario.

Modified GPSR proposed to get better performance in

all conditions and achieve better performance in high

packet delivery ratio and less end to end delay which

substantially increases network lifetime of vehicular

nodes and also increases effective communication

among the vehicles.For the future work, the

performance can be evaluated on the basis of different

Qos parameters like routing overhead throughput,

efficiency etc.


29

V. REFERENCES

[1] VANET Routing Protocols: Issues and

Challenges. Surmukh Singh, Sunil Agrawal.

UIET, Punjab University Chandigarh, India.

Proceedings of 2014 RAECS UIET Punjab

University Chandigarh, 06 – 08 IEEE March,

2014.

[2] Mr. Qi-wu Wu*, Mr. Wen Wen, Mr. Qingzi Liu,

"Comparative Study of VANET Routing

Protocols", Institution of Engineering and

Technology (IET), November 2014

[3] M. Behrisch, L. Bieker, J. Erdmann, and D.

Krajzewicz, “SUMO - Simulation of Urban

MObility: An Overview,” in SIMUL 2011, The

Third International Conference on Advances in

System Simulation, 2011.

[4] Performance Evaluation of GPSR Routing

Protocol for VANETs using Bi-directional

Coupling. Dharani N.V., Shylaja B.S., Sree

Lakshmi Ele. International Journal of Computer

Networks (IJCN), Volume (7): Issue (1): 2015.

[5] Design and evaluation of GBSR-B, an

improvement of GPSR for VANETs. C. T.

Barba , L. U. Aguiar and M. A. Igartua. IEEE

LATIN AMERICA TRANSACTIONS, VOL.

11, NO. 4, JUNE 2013.

[6] Scenario Based Performance Analysis of AODV

and GPSR Routing Protocols in a VANET. Raj

Bala, C. Rama Krishna. Deptt Of Computer

Science and Engg. NITTTR,

Chandigarh ,International Conference on

Computational Intelligence & Communication

Technology ©2015 IEEE

[7] An Enhanced GPSR Routing protocol based on

the buffer length of nodes for the congestion

problem in VANETs. Computer Science and

Education. Cambridge University, UK.Tianli Hu,

Minghui Liwang, Yuliang Tang. Department of

Communication Engineering Xiamen University.

The 10th International Conference on Computer

Science & Education (ICCSE 2015) July 22-

24,2015. Fizwilliam College,Cambridge

University,UK ©2015 IEEE

[8] Improved GPSR Routing Algorithm and its

Performance Analysis. Cheng Fenhua, Jin Min.

College of Software, Hunan University. Hunan

Science Vocational College Changsha China.

©2010 IEEE

[9] Performance Evaluation of Greedy Perimeter

Stateless Routing Protocol in Ad Hoc Networks.

Mohamed Adnene Zayene, Nabil Tabbane,

Refaat Elidoudi Multimedia Mobile Radio

Networks Research Unit Higher School of

communication of Tunis City of Communication

Technologies, Tunisia. 2009 Fourth International

Conference on Computer Sciences and

Convergence Technology ©2009 IEEE

[10] Research on One Kind of Improved GPSR

Algorithm. Liangli Lai, Qianping Wang, Qun

Wang School of Computer Science and

Technology China University of Mining and

Technology, CUMT Xuzhou, China. 2012

International Conference on Computer Sciences

and Electronics ©2012 IEEE

[11] An Improved GPSR Routing Strategy in VANET

Lili Hu, Zhizhong Ding, Huijing Shi Department

of Communication Engineering Hefei University

of Technology Hefei, China ©2012 IEEE.

[12] Simulated Analysis of Location and Distance

Based Routing in VANET with IEEE802.11p.

Akhtar Husaina and S.C. Sharma. Electronics and

Computer Discipline, DPT, Indian Institute of

Technology, Roorkee, India. 2015 Published by

Elsevier.

[13] Brad Karp, H.T. Kung, GPSR: Greedy Perimeter

Stateless Routing for Wireless Networks,

Retrieved March 4, 2008 from

http://www.eecs.harvard.edu/~htk/publication/20

00-mobi-karp-kung.pdf

[14] Study of various routing protocols in VANET

Nagaraj, U., Kharat, M.U., Dhamal, P.

International Journal of Computer Science and

Technology(IJCST), Volume (2): Issue (4): pp.

45–52 ©2011

[15] Vehicular communication: a survey Sourav

Kumar Bhoi, Pabitra Mohan Khilar The

Institution of Engineering and Technology(IET)

Volume (3): Issue (3): pp. 204-217 ©2014

[16] http://www.sumo.dlr.de

[17] http://openstreetmap.org

CSEIT16116 | Received: 28 July 2016 | Accepted: 4 August 2016 | July-August 2016 [(1)1: 30-34]



30

A Brief Survey of Acoustic Wireless Sensor Network

Mansoor Ullah, Abbas Khan, Muhammad Adil


ABSTRACT

In our earth 75% covered by water that could be rivers and ocean also. The underwater sensor network are enabling

technology and become more and more popular for monitoring Large scale of Area in oceans. Underwater sensor

Networks consist of a variable number of sensors that are deployed to perform Such as monitoring tasks over a

given area in Which The UWSNs applications like pollution monitoring, disaster prevention, Submarine detection

etc. In this paper, we discuss the internal architecture of underwater sensor. Here we Discuss architectures for two-

dimensional and three-dimensional underwater sensor network, we also discussed the application and main problem

or issue in underwater sensor network.

Keywords : Wireless Sensor Network, Acoustic, Underwater.

I. INTRODUCTION

As we Know earth Mostly covered by water. Which Is

The Most concern area and recently humans are

showing interest towards exploring it. Water is Hard to

investigate and find How’s the environment. in water

we used the Acoustic Sensors Which use the

mechanical waves. The UWSN consist of a variable

number of sensors that are deployed to perform the

monitoring tasks over a given Region. As in recent

times Many disasters Happened in the past Due To

which humans are need to greatly monitor the oceanic

environments for Different needs i.e scientific,

environmental, military, tsunamis, etc., in order to

perform these monitoring task We need to deploy

sensor nodes under water.

The UWSN Mostly operates on RF communication.

Yet, RF communication is not an Best possible

communication channel for underwater applications

because of great degree of restricted RF wave’s

propagation underwater. The Water has great

resistance propagation Power So It need HIGH

Performance antennas, bandwidth for RF. Thus, links

in underwater networks are based on acoustic wireless

communications [1] Acoustic communications are the

common physical layer technology in underwater

networks. The acoustic communication is more

reliable and Fault tolerant and bandwidth is limited.

underwater acoustic rates are between 5kb/s and

20kb/s, which is extremely slow compared to over air

RF rate(in Gb/s)[1].


A. Internal architecture of underwater sensor

The internal architecture of underwater sensor is

shown in figure 1. In internal architecture the CPU-on

board controller, sensor interface HW, acoustic

modem, memory, power supply and sensor are the

principle parts in an underwater or acoustic wireless

sensor network. These parts are mostly found in each

such application of an acoustic wireless sensor

network and constitute the main body.

Figure 1. Internal architecture of underwater sensor


31

It consists of the main controller that is interfaced with

sensor through a sensor interface circuitry. The CPU

or controller get the information from the sensor and

put it in the memory, process it and send to another

sensor through the acoustic modem. Sometimes all the

sensor component are protected by the Bottom-

mounted instrument frames that are design to permit

azimuthally omnidirectional communications, and

protect the sensor and modem from potential impact of

trawling gear[1].

In this paper, we discussed the literature survey of

underwater sensor network.

B. Literature Survey

The terrestrial sensor network and underwater sensor

network are different in many Perspectives. The

difference between terrestrial sensor network and

underwater sensor network is as follows:

Signal: In the terrestrial sensor network there are

radio signal used. But in underwater sensor

network there are acoustic signal will be used.

Power: In underwater sensor network High Energy

power required is more compare to terrestrial

sensor network because the signal will travelling

in water medium which have lots of resistance to

cover in that complex environment and high

distance among sensors.

Memory: In terrestrial sensor memory is limited

but underwater sensor may need to do some data

caching so, which require more memory.

Cost: Underwater sensors are more costly On the

other hand terrestrial sensors are not more costly.

The most important thing is use of some special

routing protocol, which can work efficiently. In this

research point of view UWSN use some special

routing protocol, which is very efficient.

Flooding based routing protocols: In the flooding

based routing protocols the node start sending too

many packets to all other node within transmission

range. There are many protocol in flooding based

family like HH-VBF(hop-by-hop vector based

forwarding protocol),DBR(depth based routing

protocol),FBR(focus beam routing protocol),HH-

DAB(hop-by hop dynamic address based routing

Protocol).

Multipath based routing protocols: In multipath based

there are more than one path are More paths available

to transmit their packets. In multipath based routing

include MPT etc.

Cluster based routing protocol: In this types of scheme

there are group of nodes .In which one is cluster head

node and cluster member node. These Protocols are

also use Data Fusion For Removing redundancy.

Cluster based include MCCP (minimum cost

clustering protocol), DUCS (distributed underwater

clustering scheme) PASCCC [2-24], etc.

C. Applications of Underwater Sensor Network

A. Fastest way for finding underwater information:

Underwater sensor is Now the most recent and

speediest method for discovering data in light of

its need and significance in a few circumstances

i.e catastrophes, marines and so on which is useful

for both the people additionally for scientists [1].

B. Disaster Prevention : The Most important is

disaster prevention characteristics of UWSN

system able to perform seismic activity which

produce tsunami warnings [1].

C. Ocean Sampling Networks: it brings refined new

automated vehicles i.e robots with advanced ocean

sea models to enhance our capacity to watch and

predict the ocean future conditions. We can

organized the sensor in various depth in ocean.so

we can sense the sea region at various depths [1]

D. Environmental Monitoring : Environment

Monitoring is a standout amongst the most

essential use of UWSN. They sense the

characteristics and properties of any object which

include pollution monitoring, Water quality and

habitant monitoring also.

E. Mine Reconnaissance : The simultaneous

operation of multiple AUVs (Autonomous

underwater vehicle) a robot with acoustic sensor

can be used to perform rapid environmental and

detect mine like object [1].

D. Underwater WSN Architecture

UWSN have diverse characterization i.e. One order

separates between static, semi-Mobile, and Mobile.

Another Type of UWSN strategy is to Divide UWSNs

into One-dimensional two-dimensional (spread sea

depths) and three-dimensional (includes depth as a


32

measurement). UWSN can likewise be single-jump,

multi-hop, or Both i.e Mixed (single-jump individual

sensors, multi-hop clusters). Structures can be

assembled into short-term, time-critical applications,

and long haul, non-time-critical applications. RF,

optical, and acoustic wave based models are another

approach to look at the accessible UWSNs [1].

Fig. 2 demonstrates the most widely recognized

UWSN design. Each and every Device is moored at

the sea depths. They are small in size, battery worked,

and acoustic modems for transmission. . So having

acoustic modems, The Cluster heads utilize two

acoustic handsets, i.e vertical and an even Trancievers.

The group head or uw-sink to speak with the sensor

nodes [1] utilizes the flat (horizontol) tranciever:

i. Send Queries and setup information to the sensors.

This correspondence will happening between

underwater sink and Cluster head to sensors.

ii. Collect checked information, which monitored.

This correspondence will happening between

sensors to group head or sink. The information

exchange from Node to group head can be single-

Jump (every Node imparted to the Cluster head

specifically) or multi-jump. The vertical

Transceiver is utilized by the uw-sinks to transfer

information to a surface station. Vertical antenna

must be long range antenna for profound water

applications as the sea can be as deep as 10 km.

The surface station is furnished with an acoustic

Transceiver that can deal with numerous parallel

interchanges with the conveyed uw-sinks. At long

last base or surface station will send the detected

information to on-shore base station through RF

signal [1].

Figure 2. 2D architecture of underwater sensor network

Not at all like TWSNs, direct communication with sea

surface separate the cluster head from others. There it

increases the overall network lifetime and which

became energy efficient as well. Also, the group head

is possibly the most security-helpless part in UWSNs

military applications, since it is an only point of failure

node.

3D UWSN Architecture. Three dimensional

underwater systems are utilized to use to detect and

observe phenomena that cannot be easily observed by

means of sea base sensor Device, i.e, to perform

agreeable inspecting of 3D sea environment.

In 3D Architecture ,sensors In this type of Network,

the sensors are sent in the form of clusters, and are

anchored at various depths because of which

correspondence is far superior than 2D. The depth of

sensor can be managed by changing the length of wire

that interface the sensor to the anchor, by means of an

electronically controlled engine that live on sensor. [1]

3D Architecture all nodes can be straightforwardly

convey to the surface group heads sent packets to the

base. In the previous case, all nodes are of the same

type, yet correspondence may be more vitality Energy

serious than that of the cluster head approach. The

grouped methodology is to single point of failure.

Military applications are to a great degree of sensitive

due to single point failure.

E. Problem in Underwater Sensor Network

More costly devices: Underwater sensor devices

are more costly.

Hardware Protection requirement: The uAs the

devices are expensive so its require to protect

against water damage

Need High Energy for communication: In

underwater sensor communication require more

power because the data transfer will done in water

medium. It is hard to propagate the signals easily

which needs lots of energy and bandwidth [25-36].

Propagation delay: The propagation delay is

major problem in UWSN Because of water

resistance

Limited battery power: UWSNs suffer from a

sensor’s fouling and corrosion. Electronics

component the battery, tend to degrade faster

under extremely low temperatures such as the one

found in deep underwater.


33

Bandwidth size limitation: In the underwater

sensor network bandwidth is another big problem.

Because bandwidth size is limited.

III. CONCLUSION

In this paper, we presented the underwater sensor

system. We Show the primary applications, the

importance of underwater sensor networks in recent

times, the design of UWSN sensor system, routing

protocols and principle challenges issues UWSN

system. We plan to continue our UWSN study. Also

we expect the time on to make efficient routing in

underwater sensor network.

IV.REFERENCES

[1] Survey paper on Underwater Wireless Sensor

Network Jaydip M. Kavar, Dr.K.H Wandra-

Student, CE dept. C.U.Shah College of Engineering

andTechnology,Wadhwancity,Gujarat,India1

underwater Sensor Network Applications: A

Comprehensive Survey


study of spectrum sensing techniques in cognitive

radio networks. In Computer and Information

Technology (WCCIT), 2013 World Congress on

(pp. 1-8). IEEE.

[3] Khan, F., Bashir, F., & Nakagawa, K. (2012). Dual

head clustering scheme in wireless sensor networks.

In Emerging Technologies (ICET), 2012

International Conference on (pp. 1-5). IEEE.

[4] Khan, F., Kamal, S. A., & Arif, F. (2013). Fairness

improvement in long chain multihop wireless ad

hoc networks. In 2013 International Conference on

Connected Vehicles and Expo (ICCVE) (pp. 556-

561). IEEE.

[5] Khan, F. (2014). Secure communication and routing

architecture in wireless sensor networks. In 2014

IEEE 3rd Global Conference on Consumer

Electronics (GCCE) (pp. 647-650). IEEE.



congestion control clustering protocol” Computer

Networks, Vol. 74, PP-92-102, 2014.


Throughput Improvement in Wireless Sensor and

Actor Networks. In 5th National Symposium on

Information Technology: Towards New Smart

World (NSITNSW) (pp. 1-8).

[8] Khan, F., Jan, S. R., Tahir, M., Khan, S., & Ullah,

F. (2016). Survey: Dealing Non-Functional

Requirements at Architecture Level. VFAST

Transactions on Software Engineering, 9(2), 7-13.


Improvement in Cognitive Radio Sensor Networks.

the IEICE Japan.

[10] Khan, F., Khan, S., & Khan, S. A. (2015, October).

Performance improvement in wireless sensor and

actor networks based on actor repositioning. In

2015 International Conference on Connected

Vehicles and Expo (ICCVE) (pp. 134-139). IEEE.

[11] M. A. Jan, P. Nanda, X. He and R. P. Liu, “A Sybil

Attack Detection Scheme for a Centralized



2015, IEEE.




Applied, Environmental, and Biological Sciences

(JAEBS), 6(4S), 82-92.


improvement in multihop wireless ad hoc networks.

In Electrical and Computer Engineering (CCECE),

2014 IEEE 27th Canadian Conference on (pp. 1-6).

IEEE.


Khan, S. A. (2016). Performance Improvement in

Wireless Sensor and Actor Networks. Journal of

Applied Environmental and Biological Sciences,

6(4S), 191-200.



Cognitive Radio Networks. 電子情報通信学会ソ

サイエティ大会講演論文集, 2012(2), 152.

[16] Khan, F., Jan, S. R., Tahir, M., & Khan, S. (2015,

October). Applications, limitations, and




262). IEEE.

[17] Jabeen, Q., Khan, F., Hayat, M. N., Khan, H., Jan,

S. R., & Ullah, F. (2016). A Survey: Embedded

Systems Supporting By Different Operating

Systems. International Journal of Scientific




Enhanced and Effective Learning through Mobile

Learning an Insight into Students Perception of

Mobile Learning at University Level. International


34

Journal of Scientific Research in Science,

Engineering and Technology (IJSRSET), Print

ISSN, 2395-1990.

[19] Jan, S. R., Khan, F., & Zaman, A. The perception of

students about mobile learning at University level.

[20] M. A. Jan, P. Nanda, X. He, and R. P. Liu, “A Sybil

Attack Detection Scheme for a Forest Wildfire

Monitoring Application,” Elsevier Future

Generation Computer Systems (FGCS),


[21] Jan, S. R., Shah, S. T. U., Johar, Z. U., Shah, Y., &

Khan, F. (2016). An Innovative Approach to

Investigate Various Software Testing Techniques

and Strategies. International Journal of Scientific



[22] Khan, I. A., Safdar, M., Ullah, F., Jan, S. R., Khan,

F., & Shah, S. (2016). Request-Response




4396

[23] Azeem, N., Ahmad, I., Jan, S. R., Tahir, M., Ullah,

F., & Khan, F. (2016). A New Robust Video

Watermarking Technique Using H. 264/AAC

Codec Luma Components Based On DCT. In



4396

[24] Jan, S. R., Khan, F., Ullah, F., Azim, N., & Tahir,

M. (2016). Using CoAP Protocol for Resource

Observation in IoT. International Journal of


Electronics, ISSN: 0976-1353

[25] Azim, N., Majid, A., Khan, F., Jan, S. R., Tahir, M.,

& Jabeen, Q. (2016). People Factors in Agile

Software Development and Project Management. In

International Journal of Emerging Technology in

Computer Science & Electronics (IJETCSE) ISSN:

0976-1353


M., & Jabeen, Q. (2016). Routing of Mobile Hosts

in Adhoc Networks. In International Journal of


Electronics (IJETCSE) ISSN: 0976-1353.

[27] Azim, N., Khan, A., Khan, F., Majid, A., Jan, S. R.,

& Tahir, M. (2016) Offsite 2-Way Data Replication

toward Improving Data Refresh Performance. In

International Journal of Engineering Trends and

Applications, ISSN: 2393 – 9516

[28] Tahir, M., Khan, F., Jan, S. R., Azim, N., Khan, I.

A., & Ullah, F. (2016) EEC: Evaluation of Energy

Consumption in Wireless Sensor Networks. . In

International Journal of Engineering Trends and

Applications, ISSN: 2393 – 9516


“PAWN: A Payload-based mutual Authentication

scheme for Wireless Sensor Networks,”

Concurrency and Computation: Practice and

Experience, “accepted”, 2016.

[30] Azim, N., Qureshi, Y., Khan, F., Tahir, M., Jan, S.

R., & Majid, A. (2016) Offsite One Way Data


Performance. In International Journal of Computer

Science Trends and Technology, ISSN: 2347-8578

[31] Safdar, M., Khan, I. A., Ullah, F., Khan, F., & Jan,

S. R. (2016) Comparative Study of Routing


International Journal of Computer Science Trends

and Technology, ISSN: 2347-8578

[32] Tahir, M., Khan, F., Babar, M., Arif, F., Khan, F.,

(2016) Framework for Better Reusability in


Journal of Applied Environmental and Biological



(2016). Collaboration Methodology for Integrating

Non-Functional Requirements in Architecture. In

the Journal of Applied Environmental and

Biological Sciences (JAEBS), 6(4S), 63-67

[34] Jan, S.R., Ullah, F., Khan, F., Azim, N., Tahir, M.,

Khan, S., Safdar, M. (2016). Applications and

Challenges Faced by Internet of Things- A Survey.

In the International Journal of Engineering Trends

and Applications, ISSN: 2393 – 951

[35] Tahir, M., Khan, F., Jan, S.R., Khan, I.A., Azim, N.

(2016). Inter-Relationship between Energy Efficient

Routing and Secure Communication in WSN. In

International Journal of Emerging Technology in

Computer Science & Electronics (IJETCSE) ISSN:

0976-1353.


Lightweight Mutual Authentication Scheme for IoT

Objects,”, “Submitted”, 2016.

CSEIT16117 | Received: 01 August 2016 | Accepted: 11 August 2016 | July-August 2016 [(1)1: 35-39]



35

A Survey on Secure Cloud Storage with Techniques Like Data Deduplication and Convergent Key management

P. Balasubhramanyam Reddy, G. Nagappan

Department of Computer Science and Engineering, Saveetha Engineering College, Thandalam, Chennai, Tamil Nadu, India

ABSTRACT

Data deduplication is a method for removing duplicate copies of data, It has been largely used in cloud storage to

reduce storage memory and upload bandwidth. It gives a challenge to do secure deduplication in cloud storage. In

encryption methods the keys can be produced but cannot manage huge number of keys. In the first attempt to

formally address the problem of achieving efficient and reliable key management in secure deduplication. The

general approach in which each user holds an independent master key for encrypting the convergent keys and

employing them to the cloud. Such a baseline key management scheme generates an enormous number of keys with

the increasing number of users and requires users to allegiance to protect the master keys. The De-key is the

process ,which creates new construction in which users do not need to manage any keys on their own but instead of

it secure distribute of the convergent key shares across multiple servers. Security analysis demonstrates that De-key

is secure in the proposed security model. Proof is that in realistic environment the De-key used in ramp secret

sharing .which can Demonstrate.

Keywords: De-Duplication, Convergent Encryption, Key Management, Auditing.

I. INTRODUCTION

The advantage of cloud storage motivates enterprises

and organizations to outsource data storage to third-

party cloud providers. One critical challenge of

today’s cloud storage services is the management of

the increasing volume of data. According to the report

of IDC, the volume of data in the will expected to

reach 50-60 trillion giga bytes in 2020. To make data

management scalable, de-duplication has been a well-

known technique to reduce storage space and upload

bandwidth in cloud storage. Instead of keeping

multiple data copies with the same content duplication

redundant data by keeping only one physical copy and

referring other redundant data to that copy. Each such

copy can be defined based on different granularities: it

may refer to either a whole file, or amore fine-grained

fixed-size or variable-size. The commercial cloud

storage services, such as Drop box, Mazy and Memo

pal, have been applying deduplication to user data to

save maintenance cost ,from the user side , data from

outside may have doubt in security and privacy

concerns. In this trust third-party cloud providers to

properly enforce confidentiality, integrity checking,

and access control mechanisms against any insider and

outsider attacks. The de-duplication is improving

storage and bandwidth efficiency, is incompatible with

traditional encryption. Especially different users to

encrypt their data with their own keys. Thus, identical

data copies of different users will lead to different

cipher texts, making de-duplication impossible

Convergent encryption provides a viable option to

enforce data confidentiality while realizing de-

duplication. It encrypts/decrypts data copy with a

convergent key, which is derived by computing the

cryptographic hash value of the content of the data

copy itself. After key generation and data encryption,

users retain the keys and send the cipher text to the

cloud.

Due to encryption is deterministic; the same data,

which already exists copies, will generate the same

convergent key and the same cipher text. This allows

the cloud to perform de-duplication on the ciphertexts.

The ciphertexts can only be decrypted by the

corresponding data owners with their convergent keys.


36

In baseline is approach suffers two critical deployment

issues. First, it is inefficient, as it will generate an

enormous number of keys with the increasing number

of users. each user must associate an encrypted

convergent key with each block of its outsource

decrypted data copies, so as to later restore the data

copies. Although different users may share the same

data copies, they must have their own set of

convergent keys so that no other users can access their

files. As a result, the number of convergent keys being

introduced linearly scales with the number of blocks

being stored and the number of users. This key

management overhead becomes more prominent if we

exploit fine-grained block-level de-duplication.

Second, the baseline approach is unreliable, as it

requires each user to dedicatedly protect his own

master key. If the master key is accidentally lost, then

the user data cannot be recovered; if it is compromised

by attackers, then the user data will be leaked. us to

explore how to efficiently and reliably manage

enormous convergent keys, while still achieving

secure de-duplication. To this end, we propose a new

construction called De-key, which provides efficiency

and reliability guarantees for convergent key

management on both user and cloud storage sides.


RELATED WORK

A. Traditional Encryption

To protect the confidentiality of outsourced data,

various cryptographic solutions have been proposed in

the literature. The idea is to builds untraditional

encryption, in which each user encrypts data with an

independent secret key. Some studies which is used to

propose the use of threshold secret sharing to maintain

the robustness of key management.

These do not consider deduplication. Using traditional

encryption, different users will simply encrypt

identical data copies with their own keys, but this will

lead to different cipher texts and hence make de-

duplication impossible.

B. Convergent Encryption

Convergent encryption ensures data privacy in de-

duplication Bellaire Formalize this primitive as

message-locked encryption, and explores its

application in space-efficient secure outsourced

storage. There are also several implementations of

convergent implementations of different convergent

encryption variants for secure de-duplication. It is

known that some commercial cloud storage providers,

such as Betas, also deploy convergent encryption.

However, as stated before, convergent encryption

leads to a significant number of convergent keys.

C. Proof of Ownership

Halevietal. propose ‘‘proofs of ownership’’ (POW)

ford duplication systems, such that a client can

efficiently prove to the cloud storage server that he/she

owns a file without uploading the file itself. Several

POW constructions based on the Merle Hash Tree are

proposed to enable client-side de-duplication, which

include the bounded leakage setting. Pietro and

Sorniotti propose another efficient POW scheme by

choosing the projection of a file onto some randomly

selected bit-positions as the file proof. Note that all the

above schemes do not consider data.

Figure 1. Impact of number of KM-CSPs n on

encoding/decoding times, where r = 2 and n - k =2.

Figure 2. Impact of confidentiality level r on the

encoding/decoding times where n=6


37

Architecture

Figure 3. low block diagrams of core modules in two

different approaches. (a) Baseline approach (keeping

the hash key with an encryption scheme).(b) De-key

(keeping the hash key with (n; k, r -RSSS).

Fig. 3 presents the flow block diagrams of core

modules in the baseline approach and De-key that we

implement. In this figure, we omit the ordinary file

transfer and de-duplication modules for simplification.

To make full use of the multi-core feature of

contemporary processors, we assume that these

modules running in parallel on different cores in a

pipeline style. In the baseline approach, we simply

encrypt each hash key H0 with the user’s master key,

while in De-key, we generate n shares of H0.We

choose 4 KB as the default data block size. A larger

data block size (e.g., 8 KB instead of 4 KB) results in

better encoding/decoding performance due to fewer

chunks being managed, but has less storage reduction

offered by de-duplication. Which each data block,

abash key of size 32 bytes is generated using the hash

function SHA-256, which belongs to the family of

SHA-2that is now recommended by the US National

Institute of Standards and Technology (NIST). In

addition, we adopt the symmetric-key encryption

algorithm AES-256in Cipher-Block Chaining (CBC)

mode as the default encryption algorithm. Both SHA-

256 and AES-256 are implemented using the EVP

library of OpenSSL Version1.0.1e.

We implement the RSSS based on Jerasure .Regarding

to the encoding and decoding modules in Fig. 1b, the

choice of code symbol size w (in bits) deserves our

discussion here. For an erasure code, a code symbol of

size w bits refers to a basic unit of encoding and

decoding operations, both of which are performed in a

finite field. In the RSSS, we choose the erasure code

whose generator matrix is a Cauchy matrix, and thus,

w should meet the condition. However, when each

hash key is divided into pieces with a size of multiple

w, its size (i.e., 32 bytes) is often not a multiple of w.

We thus often need to pad additional zeros to fill in the

Pieces, resulting in different storage blow up ratios.

Figure 4. (a), (b)

Fig. 4a shows the storage blowups ratios versus

different values of w for (6, 4, 2)-RSSS. We see that

for some w, the storage blowups ratio can be much

higher than the theoretical value calculated by n.

However, we find that if the minimum w is chosen,

the practical storage blowup can often be closely

matched to the theoretical value. In addition, we

evaluate the corresponding encoding and decoding

times on an Intel Xeon E5530 (2.40 GHz)server with

Linux 3.2.0-23-generic OS, and the results are shown

in Fig. 2b. We find that the encoding and decoding

times increase with w. Therefore, our De-key

implementation always chooses the minimum w that

meets w.


38


In discuss of implementation details of De-key. De-

key builds on the Ramp secret sharing scheme(RSSS)

to distribute the shares of convergent keys across

multiple key servers.

A. RSSS with Pseudo Randomness

In De-key, the RSSS secret is the hash key H0 of a

data block B, where H0=hash(B) .Recall the Share

function of the (n; k; r)-RSSS embeds r random pieces

to achieve a confidentiality level of r. One challenges

that randomization conflicts with de-duplication, since

the random pieces cannot be de-duplicated with each

other. Instead of directly adopting RSSS, we here

replace these random pieces with pseudorandom

pieces in our De-key implementation.

It generates the r pseudorandom pieces as follows. Let

M=[r/(k-r)]. The first generating m additional hash

valuesasH1 = hash(B+1); H2 = hash(B+2); . . .;

Hm=hash(B+ m). We then fill in the r pieces with the

generated m additional hash values H1;H2; . . .;Hm.

These r pieces are pseudorandom because

1. H1;H2; . . .;Hm cannot be guessed by attackers

along as the corresponding data block B is

unknown; and

2. H1;H2; . . .;Hm together with H0 cannot be

deduced from each other as long as the

corresponding data block B is unknown.

The parameters n, k, and r determine the following

four factors,

Confidentiality level: It is decided by the

parameter r.

Reliability level : It depends on the parameters n

and k, and can be defined by n _ k.

Storage blow-up : It determines the key

management overhead and depends on the

parameters n, k, and r.

It can be theoretically calculated by n /k-r.

Performance: It refers to the encoding

performance and decoding performance when

using the k-of-n erasure code in the Share and

Recover functions, respectively.

Fig. 1 presents the flow block diagrams of core

modules in the baseline approach and De-key that we

implement. In this figure, we omit the ordinary file

transfer and de-duplication modules for simplification.

To make full use of the multi-core feature of

contemporary processors, we assume that these

modules running in parallel on different cores in a

pipeline style. In the baseline approach, we simply

encrypt each hash key H0 with the user’s master-key,

while in De-key, we generate n shares of H0.

The 4 KB is chosen as the default data block size. A

larger data block size results in better

encoding/decoding performance due to fewer chunks

being managed, but has less storage reduction offered

by de-duplication. For each data block, abash key of

size 32 bytes is generated using the hash.

Function SHA-256, which belongs to the family of

SHA-2that is now recommended by the US National

Institute of Standards and Technology (NIST) . In

addition, we adopt the symmetric-key encryption

algorithm AES-256in Cipher-Block Chaining (CBC)

mode as the default encryption algorithm. Both SHA-

256 and AES-256 are implemented using the EVP

library of Opens’ Version1.0.10.

The implementation of RSSS based on Jerasure

Version 1.2. Regarding to the encoding and decoding

modules in Fig. 1b, the choice of code symbol size w

(in bits) deserves our discussion here. For an erasure

code, a code symbol of size w bits refers to a basic

unit of encoding and decoding operations, both of

which are performed in a finite field GF(2w). In the (n,

k, r)-RSSS, we choose the erasure code .Theshould

meet the condition 2w > n+k . However, when each

hash key is divided into (k- r) pieces with a size of

multiple w, its size (i.e., 32 bytes) is often not a

multiple of w multiplied with (k-r) we thus often need

to pad additional zeros to fill in the (k-r) pieces,

resulting in different storage blow up ratios.

IV.CONCLUSION

The De-key is an efficient and reliable convergent key

management scheme for secure de-duplication. De-

key applies de-duplication among convergent keys and

distributes convergent key shares across multiple key

servers, while preserving semantic security of

convergent keys and confidentiality of outsourced data.


39

We implement De-key using the Ramp secret sharing

scheme and demonstrate that it incurs small

encoding/decoding overhead compared to the network

transmission overhead in the regular upload/download

operations.

The audit of the file sharing and time can be recorded

and space can be utilise in various methods and make

it less expensive de-duplication can also be tried in

data warehousing although backup ,replication there

yet to we can implement this technology we can help

to make more free space and make It a low cost.

V. REFERENCES

[1] A. Shamir, "How to Share a Secret,". ACM, vol.

22,no. 11, pp. 612-613, 1979.

[2] M.W. Storer, K. Greenan, D.D.E. Long, and E.L.

Miller, "Secure Data De-duplication," in Proc.

Storages, 2008, pp. 1-10.

[3] Y. Tang, P.P. Lee, J.C. Lui, and R. Perlman, "Secure

Overlay Cloud Storage with Access Control and

Assured Deletion,"IEEE Trans. Dependable Secure

Computer., vol. 9, no. 6, pp. 903-916,Nov./Dec.

2012.

[4] G. Wallace, F. Douglis, H. Qian, P. Shilane, S.

Smaldone,M.hamness, and W. Hsu, "Characteristics

of Backup Workloads in Production Systems," in

Proc. 10th USENIX Conf. FAST,2012, pp. 1-16.

[5] Q. Wang, C. Wang, K. Ren, W. Lou, and J. Li,

"Enabling PublicAuditability and Data Dynamics for

Storage Security in Cloud Computing," IEEE Trans.

Parallel Distrib. Syst., vol. 22, no. 5,pp. 847-859,

May 2011.

[6] W. Wang, Z. Li, R. Owens, and B. Bhargava,

"Secure and Efficient Access to Outsourced Data," in

Proc. ACM CCSW,Nov. 2009, pp. 55-66.

[7] Z. Wilcox-O’Hearn and B. Warner, "Tahoe: The

Least-AuthorityFilesystem," in Proc. ACM

StorageSS, 2008, pp. 21-26

[8] A.Yun, C. Shi, and Y. Kim, "On Protecting Integrity

and Confidentiality of Cryptographic File System for

Outsourced Storage," in Proc. ACM CCSW, Nov.

2009, pp. 67-76.

[9] G.R. Blakley and C. Meadows, "Security of Ramp

Schemes," inProc. Adv. CRYPTO, vol. 196, Lecture

Notes in Computer ScienceG.R. Blakley and D.

Chaum, Eds., 1985, pp. 242-268.

[10] A.T. Clements, I. Ahmad, M. Vilayannur, and J. Li,

"DecentralizedDeduplication in San Cluster File

Systems," in Proc.USENIX ATC, 2009, p. 8.

[11] J.R. Douceur, A. Adya, W.J. Bolosky, D. Simon, and

M. Theimer,"Reclaiming Space from Duplicate Files

in a ServerlessDistributed.File System," in Proc.

ICDCS, 2002, pp. 617-624.

[12] J. Gantz and D. Reinsel, The Digital Universe in

2020: Big Data,Bigger Digital Shadows, Biggest

Growth in the Far East, Dec. 2012.

[13] R. Geambasu, T. Kohno, A. Levy, and H.M. Levy,

"Vanish:Increasing Data Privacy with Self-

Destructing Data," in Proc.`USENIX Security Symp.,

Aug. 2009, pp. 316-299.

[14] S. Halevi, D. Harnik, B. Pinkas, and A. Shulman-

Peleg,"Proofs of Ownership in Remote Storage

Systems," in Proc.ACM Conf. Comput. Commun.

Security, Y. Chen, G. Danezis,and V. Shmatikov,

Eds., 2011, pp. 491-500.

[15] D. Harnik, B. Pinkas, and A. Shulman-Peleg, "Side

Channels in Cloud Services: De-duplication in Cloud

Storage," IEEE SecurityPrivacy, vol. 8, no. 6, pp. 40-

47, Nov./Dec. 2010.

[16] S. Kamara and K. Lauter, "Cryptographic Cloud

Storage," inProc. Financial Cryptography: Workshop

Real-Life Cryptograph.Protocols Standardization,

2010, pp. 136-149.

[17] M. Li, "On the Confidentiality of Information

Dispersal Algorithmsand their Erasure Codes," in

Proc. CoRR, 2012, pp. 1-4abs/1206.4123.

[18] D. Meister and A. Brinkmann, "Multi-Level

Comparison of DataDeduplication in a Backup

Scenario," in Proc. SYSTOR, 2009,pp. 1-12




40

Emergency Information Access using QR Code Technology in Medical Field

P. Deepika , Sushanth. B , Tarun Kumar. S. P, Vignesh. M

Software Engineering, Information Technology, Easwari Engineering College, Chennai, Tamil Nadu, India

ABSTRACT

Health monitoring has become the most important factor in today’s medical era. During the time of emergency, it

would be difficult for the physicians to know the past health history of the victim to proceed with further

treatments .This project presents a health monitoring system where a person himself/herself can enter their own

health and emergency information into our servers and it can be accessed by anyone using the QR code technology

at the time of an emergency. The system is implemented in the android operating system which is the most widely

used operating system all over the world. This system helps to keep track on the individual’s health information,

henceforth giving a way for the physicians to access the information during the time of emergency. This not only

saves the life of the victim but also helps the physicians to work at ease.

Keywords : Health Monitoring, QR Code Technology, Medical Records

I. INTRODUCTION

Road safety is one issue that needs special attention as

there's one death reported every 4 minutes on the

streets of India, also, India holds the highest number of

deaths due to road accidents. Nearly 5 lakh road

accidents were reported in 2013 in which more than 1

lakh people lost their lives. A large chunk of the

victims were aged between 30 and 44 years. The

major deaths are due to the delay in the start of

treatments of patients admitted in the hospitals. This is

mainly due to the lack of previous medical

information of the patient. As they do not know the

medical information of a patient the hospitals cannot

proceed with any major treatment but just the first aid.

Our project focuses on providing the medical

information of a person at the case of emergencies.

The objective of this project is to develop a system

where a person can enter his/her medical information.

The system mainly focuses on the ability to quickly

access information in case of any emergency. The

users will be able to see the details of the person who

needs any kind of medical attention. The system

provides the information of the person, which includes

his recent medical records and also personal details

Table 1.1 Analysis of accidents that had occurred in

India.


2. Existing System

The Existing Systemis used for basic hospital

management services and health care. The medical

and lab reports are shared within various departments

of the hospital and with the patients in the form of QR

codes. The existing system [3] is specific to only

Years Total

Accidents

Accidents

involving death

and personal

injury

Number of

persons

killed

2008 825561 106994 5007

2009 950 120 104212 4236

2010 1 053 346 111121 4323

2011 1 106 201 116804 4045

2012 1 228 928 131845 3835

2013 1 296 634 153552 3750


41

certain hospitals, that is, the patient can only retrieve

medical records provided by a specific hospital.

This system focuses mainly on the sharing of reports

within a hospital organisation. Hence at any medical

emergencies, when a person is admitted to another

hospital, the retrieval of his previous medical records

becomes difficult.

3. Proposed System

We are creating an Android application, which uses a

login form to authenticate the user into his personal

account where he provides all the personal details and

information of his medical records. The details are

then saved in the database and a QR code is generated

which contains the required details of the user. In the

case of emergencies, the QR code can be scanned and

the details stored in the database are retrieved. This

saves the time to start the treatment of a patient

admitted at an emergency. This saves time taken to

complete all medical procedures in order to start

operating the patient. It is also a safe and secure data

storage and retrieval. By applying this method, it not

only saves the life of the victim but also helps the

physicians to work at ease.

4. QR CODE TECHNOLOGY

QR code[1], abbreviated from Quick Response Code,

is the trademark for a type of matrix barcode or two-

dimensional barcode. A QR code uses four

standardized encoding modes (numeric, alphanumeric,

byte/binary, and kanji) to efficiently store data;

extensions may also be used.A QR code consists of

black modules (square dots) arranged in a square grid

on a white background, which can be read by an

imaging device (such as a camera, scanner, etc.) and

processed using Reed–Solomon error correction until

the image can be appropriately interpreted. The

required data are then extracted from patterns that are

present in both horizontal and vertical components of

the image.

Figure 1. QR code

4.1 QR code representation

Nowadays, when smart phones equipped with

cameras are very common, conveying message via

QR code has become popular. As the aim was to

transfer data from a document to a mobile phone in

a feasible way it was a rational choice to apply

this standard to our purposes. This standard of

graphical data representation, established in 1994,

can hold even 2953 Bytes on a 177 by 177 modules

pattern. It possesses an attribute in encoding data

resistant for slight code distortions. There were set

up four error correction levels and the higher the

level, the less is storage capacity. [5] The levels L,

M, Q and H allow retrieving the whole message

when up to 7, 15, 25 and 30% respectively of the QR

image is destroyed. The priority was in getting as

much space for data as possible, not particularly in

damage resistance. That is why the level L was

acclaimed as sufficient.

5. ANDROID

Android is a software stack for mobile devices that

includes an operating system, middleware and key

applications. Android is a software platform and

operating system for mobile devices based on the

Linux operating system and developed by Google and

the Open Handset Alliance. It allows developers to

write managed code in a Java-like language that

utilizes Google-developed Java libraries, but does not

support programs developed in native code.

Android's source code is released by Google

under open source licenses, although most Android

devices ultimately ship with a combination of open

source and proprietary software, including proprietary

software developed and licensed by Google. Initially

http://en.wikipedia.org/wiki/Source_code

http://en.wikipedia.org/wiki/Open_source


42

developed by Android, Inc., which Google backed

financially and later bought in 2005, Android was

unveiled in 2007[5] , along with the founding of

the Open Handset Alliance a consortium of hardware,

software, and telecommunication companies devoted

to advancing open standards for mobile devices.

Android is popular with technology companies which

require a ready-made, low-cost and customizable

operating system for high devices. Android's open

nature has encouraged a large community of

developers and enthusiasts to use the open-source code

as a foundation for community-driven projects, which

add new features for advanced users or bring android

to devices which were officially, released running

other operating systems[7]. The operating system's

success has made it a target for patent litigation as part

of the so-called "Smartphone" between technology

companies.

6. System Archecture

The architecture of the system is simplified and

represented in the figure b. This schematic

representation of the architecture shows the processes,

services and related activities that happen in the entire

system. This is a consolidated representation of what

happens at what point of time in which device in the

system.

6.1. Client side process:

We used android for our client side development.

Android smart phone runs with the help of android

framework, which provides environment to run the

application in mobile devices. [6] Android framework

consists of Application framework, Libraries, Android

runtime, Applications and Linux Kernel. Android

runtime provides the environment to run the

application performing all those inbuilt activities to

run the application.

6.2 .Working of rest api:

REST API provides the interface the interface for

android to connect with the server side. REST is a set

of principles describing how standards can be used to

develop web applications. Its main purpose is to

anticipate on common implementation issues and

organize the relationship between logical clients and

servers. When implementing REST over HTTP, the

logical REST client is typically a web browser and the

logical REST server is a web server.

6.3. JSON:

JSON is a lightweight data format used in place of

XML. JSON is used to store and send data through

HTTP protocol. As we use REST API to connect the

client and server, we send the JSON data through

HTTP request and response methods. It uses human

readable form to transmit data over network. We can

store JSON data in array format. In our proposed

project, we are storing the user details securely using

the JSON encrypt and decrypt method. JSON allow us

to overcome the cross domain issue.

6.4. PHP:

PHP is the web application programming language we

used for our server side development. We can simply

the mix the PHP with HTML language. Using ZEND

Framework, we developed PHP programming

language.

6.5. Database:

As we are using PHP as our server side, we need

database to store data. MySQL database is the best

database, which supports PHP programming language

well.

Figure 2. System Architecture

6.6. Overall process working:

Android is our client side system which has android

application framework, it provides environment to run

the application. Client side needs to connect with

server side, so we used REST API to provide an

http://en.wikipedia.org/wiki/Open_Handset_Alliance

http://en.wikipedia.org/wiki/Computer_hardware

http://en.wikipedia.org/wiki/Open_standard


43

interface for android to connect with PHP using SLIM

framework. REST API offers HTTP request and

response method. JSON is used instead of XML to

transmit data from client to server side.

In server side process, PHP acts as the web application

provider and MySQL is the database we use to store

the user details. We send JSON data through HTTP

response and request method.

III. FUTUREENHANCEMENT

Our Idea can be further more enhanced by bringing in

hospitals their selves adding the information of a

patient into our servers. Similarly, the information

provided by the user can be verified by the nearby

hospitals. The medinfo profile IDs can be added to the

ID cards of major institutes and organisations.

IV.CONCLUSION

In this paper, we have presented the concept of sharing

emergency information through QR codes. The

customer has to enter all his personal and medical

information by him/herself. Consumer will be more

loyal towards the service provider. The QR code can

be scanned through any QR code scanner app across

any platforms. Hereby, we ensure that the number of

deaths due to accidents will be reduced.

V. REFERENCES

[1] Czuszynski, K., Ruminski, J,2014, "Interaction

with medical data using QR- codes", Seventh

International Conference on Human System

Interactions (HSI), pp. 101-105.

[2] Dimitris Tychalas, Athanasios Kakarountas, 2010,

"Planning and development of an electronic health

record client based on the android platform", 14th

Panhellenic Conference on Informatics, pp. 3 - 6.

[3] Hung-Ming Chen, Yong-ZanLiou, Shih-Ying

Chen, Jhuo-Syun Li, 2013, "Design of mobile

healthcare service with health records format

evaluation", IEEE 17th International Symposium

on Consumer Electronics, pp. 257 – 258.

[4] Liu Y, Yang J, and Liu M,2008, "Recognition of

QR- code with mobile phones," in Control and

Decision Conference. CCDC 2008. Chinese.

IEEE, 2008, pp. 203–206.

[5] Mohamed Amine Ben Yahmed, Mohamed Amine

Bounenni, ZeinebChelly, Amir Jlassi, 2013, "A

New Mobile Health Application for an ubiquitous

information system", 6th Joint IFIP Wireless and

Mobile Networking Conference, pp. 1 - 4.

[6] Mungyu Bae, Suk Kyu Lee, SeunghoYoo and

Hwangnam Kim, 2013,"FASE: Fast authentication

system for E-health", Fifth International

Conference on Ubiquitous and Future Networks,

pp. 648 – 649.

[7] SudhaG, GanesanR,2013,"Secure transmission

medical data for pervasive healthcare system using

android", International Conference on

Communications and Signal Processing, pp. 433 –

436.

CSEIT16119 | Received: 13 July 2016 | Accepted : 29 July 2016 | July-August 2016 [(1)1: 44-48]



44

Improving Classifier Performance Using Feature Selection with Ensemble Learning

Bhavesh Patankar*1, Dr. Vijay Chavda2

*1 Research Scholar, Department of Computer Science, Hemchandracharya North Gujarat University, Patan, Gujarat, India.

2NPCCSM, Kadi SarvaVishwaVidyalaya, Gandhinagar, Gujarat, India.

ABSTRACT

One of the critical task in data mining is classification. It is very much important in classification to achieve

maximum accuracy. In the field of data mining, numerous classifiers are present for the classification task. Each

classification techniques have their pros and cons. Some of the techniques work well with certain data sets while

other techniques work well with other data sets. There have been many techniques evolved for improving

classification accuracy. One of such technique is pre-processing which helps in improving quality of the data.

Another method is to combine the classifiers, which will in turn improve the classification accuracy. In this paper,

empirical study is been done on various techniques for improving classification accuracy. One of the technique is

feature selection, which will select best features from the available features in the data set. Other approach is

ensemble learning which combines many classifiers to improve the classification accuracy.

Keywords: Classification; Pre-processing; Feature Selection; Ensemble Learning;

I. INTRODUCTION

In data mining, it is evident that classification

accuracy is the critical factor for classification

techniques. Many classification techniques are been

evolved in data mining, but not every technique is

suitable for all data sets. They are various techniques

available in order to improve the classification

accuracy. Sometimes, data, which used to do

classification, is not as of required quality. Therefore,

it is good to improve the quality of the data, which

will result in improving the classification accuracy. In

data mining, pre-processing is one of the task, which

deals with the data set. It has been seen that a wide

variety of techniques are available for data pre-

processing like noise reduction, data cleaning which

includes filling missing values, feature selection,

dimensionality reduction, etc [1]. Ensemble techniques

have appeared as an influential technique for

improving the strength as well as the accuracy of both

solutions (i.e. supervised and unsupervised). In

addition, as massive amounts of data constantly

produced from different sights, it is vital to combine

different concepts for smart decision-making. In the

past few years, there have been various studies on the

problem of combining models into a single model, and

the success of ensemble techniques seen in multiple

disciplines, including anomaly detection, intrusion


45

detection, recommendation systems and web

applications [2].

Many papers are been reviewed to figure out various

parameters to be taken into consideration in order to

improve the classification accuracy. It is good to have

pre-processing step before the classification done in

order to achieve the increasing accuracy of the

classification. The available source data set is been

converted into more qualitative data set. In some cases,

it may occur that data set can contains high

dimensions; many of the dimensions may be irrelevant

for our classification approach. Hence, it becomes

necessary to perform Feature selection to utilize the

best features for achieving the greater accuracy in

classification. Many techniques recommended

reducing noise and outliers for the improvement of

classification accuracy.

II. FEATURE SELECTION

Achieving greater accuracy is very much important in

any data mining process. An aim of feature selection is

selecting a subset of relevant features for generating

strong learning models. Camelia Vidrighin et al.[3]

have considered the wrapper approach, as a

combination of three steps: model generation, model

evaluation and model validation. They have focused

on uniting feature selection with filling the missing

values in order to improve the performance of the

learning schemes. Analysis on various approaches for

feature selection have been done and based on the

result best models have been identified which have

consistently improved the accuracy of classification.

Feature selection can be termed as combination of

search technique to find out the best features out of the

available features in the given data set. The simplest

algorithm, which minimizes the error rate, is been

considered. As seen earlier wrapper methods use

predictive model to get the relevant feature subsets.

Wrapper methods are considered computationally very

much intensive, but generally provide best feature sets

from the given data set for the given classification

model. Filter methods use proxy measure to select the

optimum feature set. Filter techniques are generally

computationally less intensive than wrapper

techniques. Hence they produced feature set which are

not tuned to specific models and so classification

accuracy from filters are generally lesser than what we

can achieve from wrapper methods.

III. ENSEMBLE LEARNING

Ensemble learning techniques are learning algorithms

that generate a set of classifiers and then classify new

data points by considering a (weighted) vote of their

estimates. The novel ensemble technique is Bayesian

averaging, however more recent techniques include

error-correcting output coding, boosting, and bagging.

Dietterich et. al. [4] have reviewed these methods and

explained why ensembles often perform better than

any single classifier. They have reviewed some

previous studies comparing ensemble methods and

some new experiments is been shown to expose the

causes that Adaboost does not overfit rapidly.

It is known that a neural network ensemble unites a

finite number of neural networks or other types of

interpreters, which are trained concurrently for a

common classification assignment. After the

experimentation, on comparing with a single neural

network, the ensemble is able to efficiently improve

the classification accuracy of the classifier. Zhao et. al.

[5] have surveyed many ensemble techniques on

different data sets to see the effect of it. And in the


46

survey they have found that ensemble of neural

network always perform better than the single neuron.

Lira et. al. [6] have developed an ANN-based

automatic classifier for power system disturbance

waveforms. In the training process, actual voltage

waveforms applied and then Signals processed in two

steps that is decomposition and Principal Component

analysis (PCA) which results in reducing the input

space of the classifier to a much lower dimension.

Classification task was carried out using a

combination of six Multilayer perceptrons with

different. The result of experiment with real data

indicate that the random committee is clearly an

effective way in order to improve disturbance

classification accuracy when it compared with the

average and the separate models. Natesan et. al. [7]

have worked on secure communication between two

parties. They have proposed an Adaboost algorithm

for network intrusion detection system with single

weak classifier. The classifiers as Naive Bayes, Bayes

Net and Decision tree are been used as weak

classifiers. Experiments carried out with the help of

benchmark data set to reveal that boosting algorithm

can significantly improve weak classifiers

classification accuracy. Finally, the results were very

much effective. Base classifiers Naive Bayes and

Decision Tree have shown comparatively better

performance as a weak classifier with Adaboost.

IV. EXPERIMENTAL RESULTS AND

DISCUSSION

Experiment are been carried out using Weka. Weka

(Waikato Environment for Knowledge Analysis) is a

widespread machine-learning tool developed in JAVA

language. It is evident that it is one of the free open

source softwares available under the GNU General

Public License. Considering the experiment, it

executed on base classifier and then accuracy is

measured. Consequently, the experiment carried out

on the classifier with feature selection followed by

boosting and then the accuracy is measured. Data sets

used in the experiment is been collected from UCI

machine repository. At the end, results are been

compared and conclusion is drawn.

Following datasets from the UCI Machine Learning

Repository are been collected to initiate the

experiment.

Sr.No

Dataset Information

Dataset Instanc

es

Attribu

tes

1 Iris

150 5

2 Diabetes 768 9

3 Ionosphere 351 35

Table 1. Data set information

The experiment is been performed using Multilayer

perceptron, J48 and Naïve Bayes classifier. While

carrying out the experiment the data sets are been

chosen and not a single filter is applied on them.

Firstly experiment is performed using single base

classifier on the data set without feature selection

applied then experiment is carried out using single

base classifier with adaboost and data set with feature

selection applied on it. The experiment is been carried

out using weka 3.8.0.

Accuracy of the base single classifier and base

classifier with adaboost and feature selection is

measured which is displayed in given below table.


47

Classifier

Datasets

Iris Diabetes Ionosp

here

Multilayer

Perceptron 97.3

75.39 91.16

Multilayer

Perceptron with

AdaBoost and

feature selection

95.33 75.52 94.30

J48 96.00 73.82 91.45

J48 with

AdaBoost and

feature selection

94.67 73.58 94.30

Naïve Bayes 96.00 76.30 82.62

Naïve Bayes

with AdaBoost

and feature

selection

96.00 77.47 92.30

Table 2. Accuracy measures of Multilayer perceptron,

J48 and Naïve Bayes on Iris, Diabetes and Ionosphere

data set with feature selection and adaboost and

without feature selection and without adaboost.

Figure 1. Comparison of Multilayer perceptron, J48 and

Naïve Bayes with feature selection and adaboost and

without feature selection and adaboost on Iris data set.



without feature selection and adaboost on Diabetes data set.



without feature selection and adaboost on Ionosphere data

set.

V. CONCLUSION

In this paper, it is evident that classification accuracy

improved with the help of feature selection and

ensemble technique like Adaboost, which is been used

in this experiment. Here, Best First method with CFS

Subset Evaluation is been used to select the optimum

feature in order to improve the classification accuracy.

After that ensemble technique is used which combines

the multiple classifier in order to improve the

classification accuracy. Here Adaboost ensemble

technique is been used for the improvement of the

classification accuracy. From the results of the

experiment, it is clear that in most of the cases feature

93

93.5

94

94.5

95

95.5

96

96.5

97

97.5

MultilayerPerceptron

J48 Naïve Bayes

Iris Iris with Adaboost and feature selection

71

72

73

74

75

76

77

78


J48 Naïve Bayes

Diabetes Diabetes with Adaboost and feature selection

75

80

85

90

95

100


J48 Naïve Bayes

Ionosphere

Ionosphere with Adaboost and feature selection


48

selection with ensemble technique definitely improves

the classification accuracy of the classifier. Future

work include using different feature selection

approach than what is been used in this paper. In

addition, instead of AdaBoost any other ensemble

technique can be utilize to see the result.

VI. REFERENCES

[1] Moeinzadeh, H, Nasersharif, B, Rezaee, A.,

Pazhoumand-dar, H., “Improving Classification

Accuracy Using Evolutionary Fuzzy

Transformation”, 11th Annual Conference on

Genetic and Evolutionary Computation

Conference (GECCO 2009), Montreal, Canada,

2009 (1)

[2] Han, Jiawei, Micheline Kamber, and Jian Pei.

Data mining, southeast asia edition: Concepts

and techniques. Morgan kaufmann, 2006.

[3] Bratu, Camelia Vidrighin, Tudor Muresan, and

Rodica Potolea. "Improving classification

accuracy through feature selection." Intelligent

Computer Communication and Processing, 2008.

ICCP 2008. 4th International Conference on.

IEEE, 2008.

[4] Dietterich, Thomas G. "Ensemble methods in

machine learning." International workshop on

multiple classifier systems. Springer Berlin

Heidelberg, 2000.

[5] Zhao, Ying, Jun Gao, and Xuezhi Yang. "A

survey of neural network ensembles." 2005

International Conference on Neural Networks

and Brain. Vol. 1. IEEE, 2005.

[6] Lira, Milde MS, et al. "Combining multiple

artificial neural networks using random

committee to decide upon electrical disturbance

classification." 2007 International Joint

Conference on Neural Networks. IEEE, 2007.

Nilsson,R., Statistical Feature Selection, with

Applications in Life Science, PhD Thesis,

Linkoping University, 2007.

[7] Natesan, P., P. Balasubramanie, and G.

Gowrison. "Improving the attack detection rate

in network intrusion detection using adaboost

algorithm." Journal of Computer Science 8.7

(2012): 1041.




49

The Use of Wireless Sensor Networks for Forest Fire Monitoring – A Survey

Mehwish Zaheer*, Rabia Riaz, Shakeeb Ahmad

Department of Computer science, Abdul Wali Khan University Mardan, Pakistan

ABSTRACT

Wireless sensor network consists of small sensor nodes, deployed to capture various events of interest. For example,

temperature, oxygen, humidity sensor nodes are deployed in remote, hostile and geographical areas where the

presence of human being is infeasible. These nodes are powered by small battery, to communicate with each other

for monitoring various environments. These networks have found their applications in various domains such as

forest fire monitoring, industrial monitoring, military surveillance, inventory tracking, agriculture monitoring and

health care monitoring. Forest fire is the disaster having many negative effects in social, economic and ecological

matters. Forest fire cost million dollars in damage and claim many human lives every year.

Keywords: Wireless Sensor Networks, Health Care Monitoring, AVHRR, MODIS, WSN

I. INTRODUCTION

Wireless sensor networks are spatially distributed

systems that primarily work to collect data from

physical environments. The most important

fundamental element of these networks are sensor

nodes [1-4]. These sensors nodes are small

autonomous hardware devices that are capable of

carrying out some processing, collecting sensory

information, communicating with other connected

nodes in the network and produce a measurable

response to change in a physical condition [5-7] such

as sound [8-11], temperature [12], humidity or

pressure [13-15].

Wireless sensor network introduces a wide range of

possible application such as agriculture monitoring,

forest fire monitoring and medical monitoring. Forest

fire are the unrestricted fires happening in the wide

areas [16-22] in causing significant damage to natural

and human resources.

The goal of literature and recent studies is to detect

and predict forest fire immediately and actually, in

order to minimize the loss of forests, people and wild

in the forest fire. The network of a sensor nodes are

deployed densely in a forest sensor nodes sense the

forest periodically and collect measure data

(temperature, relative humidity) and sends to the

respective cluster nodes [23]. It has been shown in the

literature that about 20% of CO2 emission in the

atmosphere is due to forest fires. There are many

causes of forest fire including lightning, human

carelessness and exposure of fuel to extreme heat and

aridity.

In this paper, we study the existing literature about

wireless sensor network that are used for forest fire

monitoring. We explain the advantages and

disadvantages of recent studies and we present our

own conclusion.


A. Background of Detection System

Some of the early methods for the forest fire detection

were based on manned observation towers such as

Camera Surveillance System

Satellite Images

Satellite images have proved more efficient then

camera surveillance by two satellite the advanced very

high resolution radio meter (AVHRR) [23-26],


50

launched in a 1998 and the moderate resolution

imaging spectroradiometer (MODIS) launched in 1999

have been used. Unfortunately, these satellites can

provide images of the regions of the earth every two

days and that is a long time for fire scanning; besides

the quality of satellite images can be affected by

weather conditions.

The resolution of the wireless sensor network

technology in current years has made it possible to use

this technology for early forest fire detection. These

sensors need to be self- organized and follow an

efficient algorithm, interfaced with other technologies

or networks [30-34]. A number of present literature

considered using wireless sensor network in forest fire

systems [27-29]. These techniques have their own

limitations and disadvantages that are discussed in the

next section.

B. Disadvantages of Satellite and Camera

Surveillance Systems

The accuracy and reliability of satellite-based system

are largely impacted by the weather conditions.

In satellite system the detection accuracy is not that

much accurate. In such systems, a fire can be detected

only after it has spread largely [35] So it shows that

early systems cannot provide timely detection. Camera

surveillance systems cannot be applied to large forest

areas easily and is cost effective [36, 37]. The most

critical issue in a forest fire detection system is to

provide highly rapid response in order to minimize the

scale of disaster. Camera surveillance system and

satellite images do no provide timely detection due to

long period of scan. Therefore, we need real-time fire

detection with high accuracy and reliability.

We studied that wireless sensor networks can

potentially provide such solutions.

C. Advantages of WSN over Satellite Systems:

A pair of AA batteries is used in Wireless sensor

networks that can operate for a long period to provide

a constant monitoring during the fire sensor.

Wireless sensor networks can be easily deployed and

are low cost. These networks can detect events quickly

and accurately. Sensor nodes can be deployed

anywhere even when there is no human access

possible [38-40]. There is no need to build towers or

set up complicated communication links such as

microwave and satellite.

Based on the recent studies the key issues of this

network for forest fire monitoring are:

Localization: all the previous work used a GPS or

fixed the nodes in a known place.

Coverage: the nodes deployed randomly a full

coverage almost impossible.

Network life span: For sensor nodes working on

batteries, it is impossible to go back to each node

in the forest and recharge it again.

Fire detection method: this is the heart of the

application; it should be precise and reliable.

D. Architecture of Proposed Scheme

Forest fire detection system based on wireless sensor

network consists of small sensor nodes, base station,

communication system, internet access and structure

of monitoring hardware and software system. A large

number of nodes are randomly deployed in a forest

area and construct a self-organized network to monitor

the forest fire [41]. The nodes collect the data send it

to the sink.

E. Related Work

Wireless sensor network used for forest fire detection

consist of small sensor nodes that are used to monitor

the forest environment. Sensor nodes periodically

sense the forest when some emergency situation take

place it detects that critical data (temperature,

humidity, CO2) from the region and covert it to digital

form and forward it to base station. Base station or

sink is a device having high power energy.

III. CONCLUSION

This is a survey paper in which some various studies

about wireless sensor networks in the field of forest

fire monitoring is discussed. First, this study provides

that WSN technology is a very promising green

technology for the future in detecting efficiently the

forest fires. In this paper we present the deployment

and implementation of a wireless sensor network

system for detecting forest fires. Motes in the system

periodically sense the environment and capture the

sensed data, send it to the base station. To capture

temperature and humidity in the forest in a more


51

timely and precise way, we pointed out some

advantages of wireless sensor network.

The forest fire prevention is just one example of its

applications, this technology can also be used in major

areas such as intelligent transportation, environmental

detection, alarm floods, monitoring animal habitat,

monitoring health status of bridges, monitoring the

security situation under hole. Its development and

application have a profound impact on various fields

of living and produced.

IV. REFERENCES

[1] Khan, F., Bashir, F., & Nakagawa, K. (2012).

Dual Head Clustering Scheme in Wireless

Sensor Networks. in the IEEE International

Conference on Emerging Technologies (pp. 1-8).

Islamabad: IEEE Islamabad.

[2] M. A. Jan, P. Nanda, X. He, Z. Tan and R. P.

Liu, “A robust authentication scheme for

observing resources in the internet of things

environment” in 13th International Conference

on Trust, Security and Privacy in Computing

and Communications (TrustCom), pp. 205-211,

2014, IEEE.



Networks. in the Institute of Electronics,

Information and Communication Engineers

(IEICE) , 8.

[4] M. A. Jan, P. Nanda and X. He, “Energy

Evaluation Model for an Improved Centralized

Clustering Hierarchical Algorithm in WSN,” in

Wired/Wireless Internet Communication,

Lecture Notes in Computer Science, pp. 154–

167, Springer, Berlin, Germany, 2013.

[5] Khan, F., Kamal, S. A., & Arif, F. (2013).

Fairness Improvement in long-chain Multi-hop

Wireless Adhoc Networks. International

Conference on Connected Vehicles & Expo (pp.

1-8). Las Vegas: IEEE Las Vegas, USA.


“Enhancing lifetime and quality of data in

cluster-based hierarchical routing protocol for

wireless sensor network”, 2013 IEEE

International Conference on High Performance

Computing and Communications & 2013 IEEE

International Conference on Embedded and

Ubiquitous Computing (HPCC & EUC), pp.

1400-1407, 2013.

[7] Q. Jabeen, F. Khan, S. Khan and M.A Jan.


Wireless Mobile Adhoc Networks. in the

Journal Applied, Environmental, and Biological

Sciences (JAEBS), vol. 6(4S), pp. 82-92. Print

ISSN: 2090-4274 Online ISSN: 2090-4215,

TextRoad.


Study of Spectrum Sensing Techniques in

Cognitive Radio Networks. in IEEE World

Congress on Communication and Information

Technologies (p. 8). Tunisia: IEEE Tunisia.

[9] Khan, F. (2014). Secure Communication and

Routing Architecture in Wireless Sensor

Networks. the 3rd

Global Conference on

Consumer Electronics (GCCE) (p. 4). Tokyo,

Japan: IEEE Tokyo.



congestion control clustering protocol”

Computer Networks, Vol. 74, PP-92-102, 2014.



networks. In Electrical and Computer

Engineering (CCECE), 2014 IEEE 27th

Canadian Conference on (pp. 1-6). IEEE.

[12] Mian Ahmad Jan and Muhammad Khan, “A

Survey of Cluster-based Hierarchical Routing

Protocols”, in IRACST–International Journal of

Computer Networks and Wireless

Communications (IJCNWC), Vol.3, April. 2013,

pp.138-143.

[13] Khan, S., Khan, F., & Khan, S.A.(2015). Delay

and Throughput Improvement in Wireless

Sensor and Actor Networks. 5th National

Symposium on Information Technology:

Towards New Smart World (NSITNSW) (pp. 1-

8). Riyadh: IEEE Riyad Chapter.

[14] Khan, F., Khan, S., & Khan, S. A. (2015,

October). Performance improvement in wireless

sensor and actor networks based on actor

repositioning. In 2015 International Conference


134-139). IEEE.

[15] Khan, S., Khan, F., Jabeen. Q., Arif. F., & Jan.

M. A. (2016). Performance Improvement in


52

Wireless Sensor and Actor Networks. in the

Journal Applied, Environmental, and Biological

Sciences Print ISSN: 2090-4274 Online ISSN:

2090-4215

[16] Mian Ahmad Jan and Muhammad Khan,

“Denial of Service Attacks and Their

Countermeasures in WSN”, in IRACST–

International Journal of Computer Networks and

Wireless Communications (IJCNWC), Vol.3,

April. 2013.





2015, IEEE.

[18] Jabeen, Q., Khan, F., Hayat, M.N., Khan, H.,

Jan., S.R., Ullah, F., (2016) A Survey :

Embedded Systems Supporting By Different

Operating Systems in the International Journal

of Scientific Research in Science, Engineering

and Technology(IJSRSET), Print ISSN : 2395-

1990, Online ISSN : 2394-4099, Volume 2 Issue

2, pp.664-673.

[19] Syed Roohullah Jan, Syed Tauhid Ullah Shah,

Zia Ullah Johar, Yasin Shah, Khan, F., " An

Innovative Approach to Investigate Various

Software Testing Techniques and Strategies",

International Journal of Scientific Research in

Science, Engineering and

Technology(IJSRSET), Print ISSN : 2395-1990,

Online ISSN : 2394-4099, Volume 2 Issue 2,

pp.682-689, March-April 2016.

URL : http://ijsrset.com/IJSRSET1622210.php

[20] Khan, F., Jan, SR, Tahir, M., & Khan, S., (2015)

Applications, Limitations, and Improvements in

Visible Light Communication Systems” In 2015

International Conference on Connected Vehicles

and Expo (ICCVE) (pp. 259-262). IEEE.

[21] Syed Roohullah Jan, Khan, F., Muhammad

Tahir, Shahzad Khan,, (2016) “Survey: Dealing

Non-Functional Requirements At Architecture

Level”, VFAST Transactions on Software

Engineering, (Accepted 2016)

[22] M. A. Jan, “Energy-efficient routing and secure

communication in wireless sensor networks,”

Ph.D. dissertation, 2016.



IoT Objects,” IEEE Transactions on

Dependable and Secure Computing (TDSC),

“Submitted”, 2016.



Wildfire Monitoring Application,” Elsevier

Future Generation Computer Systems (FGCS),


[25] Puthal, D., Nepal, S., Ranjan, R., & Chen, J.

(2015, August). DPBSV--An Efficient and

Secure Scheme for Big Sensing Data Stream.

InTrustcom/BigDataSE/ISPA, 2015 IEEE (Vol.

1, pp. 246-253). IEEE.


(2015). A Dynamic Key Length Based

Approach for Real-Time Security Verification

of Big Sensing Data Stream. In Web

Information Systems Engineering–WISE

2015 (pp. 93-108). Springer International

Publishing.


(2016). A dynamic prime number based efficient

security mechanism for big sensing data

streams.Journal of Computer and System

Sciences.

[28] Puthal, D., & Sahoo, B. (2012). Secure Data

Collection & Critical Data Transmission in

Mobile Sink WSN: Secure and Energy efficient

data collection technique.

[29] Puthal, D., Sahoo, B., & Sahoo, B. P. S. (2012).

Effective Machine to Machine Communications

in Smart Grid Networks. ARPN J. Syst. Softw.©

2009-2011 AJSS Journal, 2(1), 18-22.


“PAWN: A Payload-based mutual

Authentication scheme for Wireless Sensor

Networks,” “accepted”, 2016.

[31] M. Usman, M. A. Jan, and X. He,

“Cryptography-based Secure Data Storage and

Sharing Using HEVC and Public Clouds,”

Elsevier Information sciences, “accepted”, 2016.

[32] Jan, S. R., Khan, F., & Zaman, A. THE

PERCEPTION OF STUDENTS ABOUT

MOBILE LEARNING AT UNIVERSITY

LEVEL. NO. CONTENTS PAGE NO., 97.



Cognitive Radio Networks. 電子情報通信学会

https://scholar.google.com/scholar?oi=bibs&cluster=5779524721848057200&btnI=1&hl=en

https://scholar.google.com/scholar?oi=bibs&cluster=5779524721848057200&btnI=1&hl=en


53

ソサイエティ大会講演論文集, 2012(2), 152.

[34] Safdar, M., Khan, I. A., Ullah, F., Khan, F., &

Jan, S. R. Comparative Study of Routing

Protocols in Mobile Adhoc Networks.

[35] Shahzad Khan, Fazlullah Khan, Fahim Arif,

Qamar Jabeen, M.A Jan and S. A Khan (2016).

“Performance Improvement in Wireless Sensor

and Actor Networks”, Journal of Applied

Environmental and Biological Sciences, Vol.

6(4S), pp. 191-200, Print ISSN: 2090-4274

Online ISSN: 2090-4215, TextRoad.

[36] M. Usman, M. A. Jan, X. He and P. Nanda,

“Data Sharing in Secure Multimedia Wireless

Sensor Networks,” in 15th IEEE International

Conference on Trust, Security and Privacy in

Computing and Communications (IEEE

TrustCom-16), “accepted”, 2016.

[37] Junguo ZHANG, Wenbin LI, Ning HAN,

Jiangming KAN Forest fire detection system

based on a ZigBee wireless sensor network.

[38] Forest Fire Detection with Wireless Sensor

Networks Çağdaş Döner*, Gökhan Şimşek,

Kasım Sinan Yıldırım and Aylin Kantarcı

Computer Engineering Department, Ege

University.

[39] The 3rd International Conference on Sustainable

Energy Information Technology (SEIT 2013)

Using Wireless Sensor Networks for Reliable

Forest Fires Detection Kechar Bouabdellaha,

Houache Noureddine, Sekhri Larbi Laboratory

of Industrial Computing and Networking,

Faculty of Sciences, Oran University, PO Box

1524 El M'naouar, Algeria.

[40] Forest Fire Modeling and Early Detection using

Wireless Sensor Networks MOHAMED

HEFEEDA Simon Fraser University, Canada.

[41] IRACST – Engineering Science and Technology:

An International Journal (ESTIJ), ISSN: 2250-

3498, Vol.2, No. 2, April 2012 Wireless Sensor

Network for Forest Fire Sensing and Detection

in Tamilnadu.




54

Address Allocation Algorithm with Cooperative Communication in MANET

Parameswaran T1, Dr.Palanisamy C

2, Logeshwari N

3

1Teaching Fellow, Department of CSE, Anna University Regional Campus, Coimbatore. India

2Professor and HOD, Department of IT, Bannari Amman Institute of Technology, Sathyamangalam. India

3PG Scholar, Department of CSE, Anna University Regional Campus, Coimbatore. India

ABSTRACT

Wireless sensor network (WSN) has devices with radio transceivers that cooperate to form and continue a fully

connected network of sensor nodes. In existing systems, topology control algorithms allow each node in a wireless

multi-hop network to adjust the power. It generates the transmission and decides neighbors where it communicates

while preserving goals like connectivity or coverage. When a node drains its inadequate energy, it does not reach

neighboring nodes resulting in a disjointed network and stopping important communications. With the limited

energy, the node does not able to continue the environmental monitoring performance that is important to the

efficient operation of the system. In proposed system, MANETs controlled the provision of pre-registered or

approved nodes and it has the opportunity for pre-deployed exchange of security parameters like public keys,

session keys. Each node in a MANET moves in any direction and changes its links frequently. A low-overhead

identity based distributed dynamic address configuration scheme for secure allocation of IP addresses is used to

allow the nodes of a managed mobile ad hoc network. MANET reduces the power usage for each packet

transmission and mobile node movement. MANET also improves the security of transmission in mobile networks.

Finally, this process conduct the performance metrics are: network overhead, delay time and security level.

Keywords: MANET, low-overhead identity, Wireless Sensor Network (WSN), Topology Control algorithm, IP

Address.

I. INTRODUCTION

Wireless sensor network (WSN) has devices with

radio transceivers that cooperate to form and continue

a fully connected network of sensor nodes.A wireless

sensor network (WSN) comprises spatially distributed

autonomous sensors to examine physical or

environmental conditions to pass data through the

network to main location.

Neighbor discovery protocols (NDPs) survey is made

in [2]. Generally, protocol is divided using four

principles. They are: randomness, over half

Occupation, rotation resistant intersection, and co

prime cycles. The birthday protocols functions as

Agents of NDPs by change where the node listen

NDP used to find the future information.

The growth of wireless sensor networks was

encouraged by military applications. Wireless Sensor

Networks (WSNs) is a class of wireless ad hoc

networks where sensor nodes gather, process and

communicate data attained from the physical

environment to Base-Station (BS).


2. LITERATURE SURVEY

In [1], a distributed algorithm is presented for creating

minimum weight directed spanning trees with root

node in connected directed graph. A processor

presents at each node. With weights and origins of

edges incoming to nodes, the processors follow the

algorithm and exchange messages with their neighbors

until all arborescence are built.

Cone Based Distributed topology control (CBTC)


55

algorithm is designed in [3]. The algorithm fails to

consider the nodes with GPS information and it

depends on directional information. Roughly speaking

is important design of the algorithm where the node

sends with minimum power to guarantee in all cone of

quantity around.

R3E increases the packet delivery ratio when

preserving the high energy efficiency and low delivery

latency. In [4], two localized topology control

algorithms are designed for heterogeneous networks.

They are: Directed Relative Neighborhood Graph

(DRNG) and Directed Local Spanning Sub graph.

Each node builds its neighbor set by changing the

transmission power and describes the network

topology with local information.

A protocol optimized in [5] for less energy usage in

mobile wireless networks which support peer-to-peer

communications. An easy local optimization scheme is

used at all nodes to guarantee link of network and

attains the global less energy solution for stationary

networks. In [6], distributed channel access protocol

joins the channel reservation and the iterative/global

transmission power control techniques in ad hoc

networks. The designed protocol solves the

convergence problem of global power control in ad

hoc networks. The designed access criteria use the

local admission control depending on the adequate

criteria for admissibility and global power control for

balancing the SIR of the links. In [7], a minimum

spanning tree (MST)-based algorithm termed as local

minimum spanning tree (LMST) is designed for

topology control in wireless multi hop networks. In

algorithm, each node creates LMST separately and

preserves on-tree nodes that are one-hop away in final

topology.

2.1 Related Works

In [8] to develop the benefits of user cooperation in

cooperative WANETs, distributed energy-efficient

selective diversity (EESD) topology control is planned

to enhance the energy efficiency. It equally considers

the network capacity and energy consumption using

bits per Joule. EESD creates the transmission

coalitions via cooperative manner selection by

considering the cost of channel information exchange.

Game theory is briefed in [9] to solve the power

control issue in a CDMA-based distributed sensor

network. A non-cooperative game with incomplete

information is designed by Nash equilibrium. With

this equilibrium, a distributed algorithm is planned for

optimal power control and verified the system is

power stable when the nodes observe with the transmit

power thresholds.

The energy efficiency problem is addressed and

designed a comprehensive study of topology control

techniques in [10] for increasing the lifetime of battery

powered WSNs. Initially, a topology control

algorithms are designed to present insights into how

energy efficiency is attained using the design. In

addition, algorithms are derived from the energy

preservation approach that implemented and computed

using the trade-offs they provide to help the designers

in choosing a method that suits the applications.

3. ADDRESS ALLOCATION ALGORITHM IN

MANET

Nodes are within the each other’s radio range that

communicate where the nodes are not in each other’s

radio range communicate via intermediate nodes

where the packets are transmitted from source to

destination. Number of nodes is increased in the

network and the time taken to attain an IP address,

number of packet replaces in less address allocation.

The existing node in the network creates distinctive IP

addresses from its own IP address for new authorized

nodes. Mobile ad hoc networks are divided into

stateless allocation and state full allocation methods. It

is evident in many existing dynamic address allocation

methods for MANET based on DAD.

Figure 1. Architecture Diagram


56

From figure 1, in initialization phase creates the node

and address the neighbor node. From the initialization

phase, it is send to the power adjustment phase. In

power adjustment phase, it checks for the network

coverage. The energy is shared through the node using

the game theory. Then, it sends for generating IP

address. After generating the address, communication

takes place through mobile. In mobile communication,

authentication verification takes place. After

authentication, nodes update the position and increase

the security of the node transmission. After that, the

performance is analysed.

3.1 MODULES

1. Nodes deployed and create source and destination

nodes

2. Node searching neighbor node for cooperative

communication

3. Game theory for minimize the energy

consumption

4. Generate IP address for each mobile node in

network

5. Updated position of mobile nodes provide secured

communication

6. Performance Analysis

3.1.1 Nodes Deployed and Create Source and

Destination Nodes

The nodes are deployed in the network with the help

of NS2. It also creates the source and destination node

with higher efficient.

3.1.2 Node Searching Neighbor Node for

Cooperative Communication

Based on the location of a node with respect to others,

some nodes end up with a larger communication

radius. By taking that all nodes initiates with the same

energy supply and make transmissions at the same

rate. Node A has the largest energy cost and the

shortest lifetime. For the cooperative communication,

Cooperative topology control algorithm is designed.

Initially, it is separated into two types. They are:

topology construction and topology maintenance.

Topology construction is the charge of initial

reduction and the topology maintenance is the charge

of maintenance of the reduced topology where the

features such as connectivity and coverage are

protected. The initial topology is employed when the

location of nodes is chance where the administrator

without the control over the design of the network.

Simultaneously, the topology is reduced and the

network starts allocation in the selected nodes by

spending the energy. Topology control is executed in

following steps to protect the desired properties like

connectivity, coverage, density.

Step 1: Change the transmission range of the nodes

Step 2: Turn off nodes from the network

Step 3: Create a communication backbone

Step 4: Clustering

Step 5: Add new nodes to the network to preserve

connectivity (Federated Wireless sensor networks)

3.1.3 Game Theory for Minimize the Energy

Consumption

Each node updates its transmission power periodically,

the algorithm functions in rounds. At staring of each

round, each node broadcasts its remaining energy. If it

not broadcasted before, the Energy Info Shared Flag

denotes it. Game theory is an ordinal potential game

looking for the optimal global potential function yield

Nash equilibrium.

3.1.4 Generate IP Address for Each Mobile Node in

Network

The network initiates from a single node and develops

as more nodes by adding one by one. These nodes are

free to move around and it joins or leaves the network

at any point of time. A node has to inform its parent

before departing the network. In case of graceless

departure, a node moves away from the network

inadvertently or even deliberately. As IPv6 provides a

large address space, it is also not that necessary for an

address to be reused.

3.1.5 Updated Position of Mobile Nodes Provide

Secured Communication

When authentication is successful, the parent node

modernizes and transmits OK message to the children.

On getting the OK message, the child node confirms

the authentication of parent. If authentication is

successful, it sends CONFIRM message and then

switches-off. On receiving CONFIRM message, the

parent node verifies the authentication of the said

children.


57


PERFORMANCE ANALYSIS

The performance quality is analysed for cooperative

communication using the cooperative topology control

algorithm. The metrics of parameters is given below:

Network overhead.

Security

Delay time.

4.1 Number of nodes vs. Network overhead

Network overhead is the metadata and network routing

information sent by an application that uses a portion

of the available bandwidth of a communications

protocol. The additional data creating the protocol

headers and application-specific information is

denoted as overhead. Network overhead is the ratio of

non-application bytes divided to the total number of

bytes in the message. Network overhead is measured

in terms of percentage (%).

Figure 4.1. Number of nodes vs. Network overhead

Figure 4.1 demonstrates the network overhead of

cooperative communication and dynamic address

allocation algorithm. X axis represents the no. of

nodes whereas Y axis denotes network overhead of the

cooperative topology control and dynamic address

allocation algorithm. When the no. of nodes is

increased, the performance of network overhead gets

automatically increases accordingly.

4.2 Number of nodes vs. Security

Network security involves the approval of access to

data in a network that are controlled by the network

administrator. Network security includes the large

number of computer networks both public and private

information in businesses, government agencies and

individuals. It secures the network and also protects

the operations carried out. Network security comprises

the policies implemented to avoid and

examine authorized access, misuse, alteration or denial

of a computer network and network-accessible

resources. The simple way of protecting a network

resource is by assigning a unique name and a

password. It is also measured in terms of percentage

(%).

Figure 4.2 Number of nodes vs. Security

Figure 4.2 illustrates the security of cooperative

communication and dynamic address allocation

algorithm. X axis represents the no. of nodes whereas

Y axis denotes security of the cooperative topology

control and dynamic address allocation algorithm.

When the no. of nodes is increased, the performance

of network security gets automatically increases

accordingly. In the proposed algorithm, the network

security is high.

4.3 Number of nodes vs. Delay time

Delay time is an essential design and performance

feature of the processed computer

network or telecommunications network. The delay of

a network denotes the time required for a bit of data to

travel across the network from one node or end point

to another. It is measured in terms of milliseconds

(ms). Delay changes based on the location of the exact

pair of communicating nodes. Processing delay is

defined as time taken by the routers to process the

packet header. Queuing delay is defined as time taken

by the packet for routing queues. Transmission

0

10

20

30

40

50

60

70

5 10 15 20 25

Net

work

over

hea

d

(%)

No. of Nodes

Proposed-

AAA

Existing-

CTCA

0

20

40

60

80

100

5 10 15 20 25

Sec

uri

ty (

%)

No. of Nodes

Proposed-

AAA

Existing-

CTCA


58

delay is defined as time taken to push the packet's bits

onto the link. Propagation delay is defined as time

taken for a signal to reach its destination.

Figure 4.3. Number of nodes vs. Delay Time

Figure 4.3 demonstrates the delay time of cooperative

communication and dynamic address allocation

algorithm. X axis represents the no. of nodes whereas

Y axis denotes delay time of the cooperative topology

control and dynamic address allocation algorithm.

When the no. of nodes is increased, the performance

of delay time gets automatically increases accordingly.

For an effective cooperative communication, the delay

time should be as low as possible.

IV.CONCLUSION

Cooperative communication provides the

communication directly or indirectly using neighbor

nodes, the Neighbor searching technique is used for

neighbor selection. Quality of topology by CTCA

algorithm to optimal solution attained with centralized

algorithm. A distributed algorithm called Cooperative

Topology Control with Adaptation (CTCA) executes

more information or options presented at each node.

The Cooperative Topology Control with Adaptation

(CTCA) algorithm executes better than other

distributed algorithms. Game theory is implemented to

minimize the energy consumption.

V. FUTURE ENHANCEMENT

Planned to improve the security and reduce the

Network overhead, delay time in MANET, it updates

the position and provide efficient communication

between nodes in the network depending on network

coverage. Use secure address allocation algorithm for

packet transmission to reduce the energy consumption

of nodes.

VI.REFERENCES

[1] Pierre A. Humblet, "A Distributed Algorithm for

Minimum Weight Directed Spanning Trees",

IEEE Transactions on Communications,

Volume 31, Issue 6, June 2003.

[2] Wei Sun, Zheng Yang, Xinglin Zhang, and

Yunhao Liu, "ENERGY-Efficient Neighbor

Discovery in Mobile Ad Hoc and Wireless

Sensor Networks: A Survey", IEEE

Communications Surveys and Tutorials,

Volume 16, Issue 3, Third Quarter 2014.

[3] Li (Erran) Li, Joseph Y. Halpern, Paramvir

Bahl, Yi-Min Wang, and Roger Wattenhofer, "A

Cone-Based Distributed Topology-Control

Algorithm for Wireless Multi-Hop Networks",

IEEE/ACM Transactions on Networking,

Volume 13, Issue 1, February 2005.

[4] Ning Li, and Jennifer C. Hou, "Localized

Topology Control Algorithms for

Heterogeneous Wireless Networks", IEEE/ACM

Transactions on Networking, Volume 13, Issue

6, December 2005.

[5] VolkanRodoplu, and Teresa H. Meng,

"Minimum Energy Mobile Wireless Networks"

IEEE Journal on Selected Areas in

Communications, Volume 17, Issue 8, August

1999.

[6] AzrinaAbd Aziz, Y. Ahmet S. Ekercioglu, Paul

Fitzpatrick, and MiloshIvanovich, "A

Distributed Channel Access Protocol for Ad

Hoc Networks with Feedback Power Control"

IEEE Communications Surveys and Tutorials,

Volume 15, Issue 1, First Quarter 2013.

[7] Ning Li, Jennifer C. Hou, and LuiSha,"Design

and Analysis of an MST-Based Topology

Control Algorithm" IEEE Transactions on

Wireless Communications, Volume 4, Issue 3,

MAY 2005.

[8] BingyiGuo, Quansheng Guan, F. Richard Yu.,

Shengming Jiang and Victor C. M. Leung,

"Energy-Efficient Topology Control with

Selective Diversity in Cooperative Wireless Ad

Hoc Networks: A Game-Theoretic Approach"

IEEE Transactions on Wireless

Communications, Volume 13, Issue 11,

November 2014.

[9] ShamikSengupta, MainakChatterjee, and Kevin

A. Kwiat, "A Game Theoretic Framework for

Power Control in Wireless Sensor Networks"

0

10

20

30

40

50

60

70

5 10 15 20 25

Net

wo

rk o

ver

hea

d

(%)

No. of Nodes

Proposed

-AAA

Existing-

CTCA


59

IEEE Transactions on Computers, Volume 59,

Issue 2, February 2010.

[10] AzrinaAbd i hmet e ercioglu Paul

Fitzpatrick, and MiloshIvanovich., "A Survey on

Distributed Topology Control Techniques for

Extending the Lifetime of Battery Powered

Wireless Sensor Networks", IEEE

Communications Surveys and Tutorials,

Volume 15, Issue 1, First Quarter 2013.

[11] C.EPerkins, E.M.Royer, and S.R.Das:"IP

Address Auto Configuration for Ad Hoc

Networks", Technical Report draft-ietf-manet-

autoconfig-00.txt, Internet Engineering Task

Force, MANET Working Group, July 2000.

[12] S.Thomson, and T.Narten:"Ipv6 Stateless Auto

Configuration", RFC 2462, December 1998.

[13] A.Misra, S.Das, A.McAulley, and

S.K.Das:"Auto configuration, Registration, and

Mobility Management for Pervasive

Computing,"IEEE Personal Communications,

Volume8, Issue 4, Auguest 2001.Pages 24-31.

[14] R.Droms,"Dynamic Host Configuration

Protocol," Network Working Group, RFC 2131,

Mar 1997.

[15] N.Vaidhya,"Weak Duplicate Address Detection

in Mobile Ad Hoc Networks," ACM

International Symposium on Mobile Ad Hoc

Networking and Computing (MobiHoc02), June

2002, pp.201-216.

[16] Jeff.Bleng,"Efficient Network Layer Addressing

for MANET s," in Proc.of International

Conference on Wireless Networks

(ICWN'020Las Vegas, USA.

[17] C.Perkins et al.,"IP Address Auto configuration

for Ad Hoc Networks,"IETF draft, 2001.

[18] J.Broch, D.Maltz, D.Johnson, Y.Hu, and

J.Jetcheva."A Performance Comparison of

Multi-Hop Wireless Ad Hoc Routing Protocols,"

Proceedings of the Fourth Annual ACM/IEEE

Inter-national Conference on Mobile Computing

and Networking, pp.85-97, October 1998.

[19] M.Mohsin and R.Prakash,"IP Address

Assignment in a Mobile Ad Hoc Network,"Proc

MILCOM, Vol.2, Oct 2002, pp.856-61.

[20] P.Patchipulusu,"Dynamic Address Allocation

Protocols for Mobile Ah Hoc Networks," M.Sc

thesis, Comp.Sci.Texas A&M Univ., 2001.

Parameswaran.T has received his B.E

degree in Electronics and

Communication Engineering from

Velalar College of Engineering and

Technology, Erode, and M.E degree in

Software Engineering from College of

Engineering Guindy, Anna University

Chennai in 2005 and 2008 respectively.

He is currently pursuing his Ph.D Anna University Chennai.

He is currently working as Teaching Fellow in the

Department of Computer Science and Engineering, Anna

University Regional Campus, Coimbatore, Tamilnadu, India.

Palanisamy.C has received his B.E

degree in Electronics and

Communication Engineering from

University of Madras, Chennai and

M.E degree (Gold Medalist) in

Communication Systems from

Thiagarajar College of Engineering,

Madurai, and Madurai Kamaraj

University in 1998 and 2000 respectively. He has received

his Ph.D from the faculty of Information and

Communication Engineering, Anna University, Chennai in

2009. He has more than 15 years of academic and research

experience and currently he holds the post of Professor and

Head of the Department of Information Technology,

Bannari Amman Institute of Technology, Sathyamangalam,

and Tamilnadu, India. He has published more than 40

research papers in various journals and conferences. He has

organized more than 15 workshops and holds 2 funded

projects. He is a lifetime member of ISTE. He Won Best

M.E Thesis Award at Thiagarajar College of Engineering,

Madurai and best paper award titled, "A Neural Network

Based Classification Model Using Fourier and Wavelet

Features ” Proceedings of the 2nd Int. Conf. on Cognition

and Recognition 2008, (ICCR 2008), Organised by P. E. S.

College of Engineering, Mandaya, Karnataka, India, pp.

664-670, 2008.His research interests include Data mining,

image processing, and mobile networks.

Logeshwari .N has received her

B.TECH degree in Information

Technology from Madras Institute of

Technology, Chrompet, and Anna

University Chennai in 2010 and

2014.She is currently pursuing her

M.E Degree in Anna University

Regional Campus Coimbatore. Her

area of Interest is networks.




60

A Survey on WSN-based Forest Fire Detection Techniques

Waqas Ali, Abdullah, Ishfaq-ur-rashid


ABSTRACT

In this paper, we will present a survey on existing studies of forest fire detection system. Every year, thousands of

forest fires across the world cause disasters including thousands of hectares of forests and hundreds of houses.

Various methods are implemented in this area. We will explain in detail the advantages and disadvantages of each

method. In addition, at the end we will show the comparison between the methods that are used for forest fire

detection system.

Keywords: Forest Fire Detection System, GPS, WSN, CCD, MEMS

I. INTRODUCTION

Forest fires generally occur due to human uncontrolled

behavior in social activities and change in weather

conditions. Forest fires may result in human and

animal deaths. They are fatal threat in the world: it is

reported [1] that a total of 77,534 wildfires burned

6,790,692 acres in USA for 2004. Unfortunately,

forest fires are usually only observed when it has

already spread over a large area, making its control

difficult and even impossible at some times. Forest

fires have also a huge impact on atmosphere (30% of

carbon dioxide in the atmosphere comes from forest

fires).

Every year thousands of hectares of forests are

destroyed by fire .carbon monoxide produced from the

areas that are destroyed by fires are more than the

overall automobile traffic. There are many methods

for the detection of forest fires like satellite based

monitoring, wireless sensor networks based detection

etc. The objective is to detect the conditions that

results in forest fires. In this paper we will show the

different techniques that are implemented for the

detection of forest fires and we will briefly discuss its

advantages and disadvantages.


A. Optical Sensors and Camera Surveillance

These systems are also used to detect fire in the forest.

But every technology has it pros and cons. In a

camera-based system, CCD cameras and IR detectors

are installed on top of towers. In case of fire or smoke

activity, the cameras and detectors sense this abnormal

event and report it to a control center. However, the

accuracy of such a system is highly affected by terrain,

time of day, and weather conditions such as clouds,

light reflections, and smoke from innocent industrial

or social activities. Optical sensors or camera systems

in general need to be improved in order to reduce the

number of false alarms due to various dynamic

phenomena, such as wind-tossed trees, cloud shadows,

reflections, and human activity. This kind of

technology only provides a line of sight vision; where

high trees or the hills and mountains can block the

vision; and it might be impossible to provide images

for ignition place. The performance of the camera can

be affected by weather conditions and in darkness. To

cover large area these system was developed with

minimum number of towers; each tower has to detect

smoke in range of 15–80Km, where it requires a long

delay after the ignition to produce a watchable smoke

cloud that can be detected by the camera. These

systems were tried for short distances but for large


61

areas they are inefficient because the installation of

camera have to be manual and have to be in

appropriate position. However, these systems are very

expensive and the cost of one camera tower is in

thousand dollars.

B. Satellite Based Systems

Another alternative technology for detecting forest

fires is the use of satellites and satellite images.

Usually, satellites provide a complete image of the

earth every 1–2 days. This long scan period, however,

is not acceptable for detecting forest fires quickly.

Additionally, the smallest fire size that can be detected

by such a system is around 0.1 hectare, which also

prevents fire detection just at the time when the fire

starts, and fire localization error is about 1 km, which

is not very accurate. Two main satellites launched for

forest fire detection purposes, the advanced very high

resolution radiometer (AVHRR) [2], launched in 1998,

and the moderate resolution imaging Spector

radiometer (MODIS), launched in 1999 these satellites

are used to detect the forest fire. Unfortunately, these

satellites can provide images of the regions of the

earth every two days and that is a long time for fire

scanning; besides the quality of satellite images can be

affected by weather conditions [3]. Any existing

satellite-based observations for forest fires suffer from

severe limitations resulting in a failure in speedy and

effective control for forest areas. Some of the

limitations in an approach based on direct observation

of forest fires from geostationary (GEO) or Low Earth

Orbit (LEO) satellite are as follows: it might be

impossible to provide a full satellite coverage or even

intermittent coverage.

Geo and Leo satellites are located on orbits over

22,800 miles above the earth‟s surface. The satellite

might not be equipped with transponders, antennas,

amplification reception, regeneration, frequency

translation, and downlink transmission suited for

detection of forest fires. In fact, there may not yet be

formal allocation of the appropriate frequency and

bandwidth for forest fire detection.

C. Wireless Sensor Networks

In recent years, wireless sensor networks (WSNs)

have gained worldwide attention, particularly with the

proliferation of Micro-Electro-Mechanical system

(MEMS) technology that has facilitated the

development of smart sensors. These sensor nodes are

inexpensive and small with limited processing and

computing resources. These sensor nodes can sense

and gather information from the environment and

transmits the sensed data to the user. These sensor

nodes have limited battery power and limited memory

and are normally deployed in difficult to access

locations where humans cannot go easily. In wireless

sensor network, a radio is implemented on every

sensor node, which is used for wireless

communication between nodes and base station. In

WSNs the deployment of sensor nodes are of two

types: Random deployment and replanned deployment.

In random deployment, nodes are deployed normally

from the helicopter or plane and are used in large

wireless sensor network in which the number of nodes

is in thousands. In pre-planned deployment, nodes are

deployed in pre-planned manner and are used in small

wireless sensor network in which the number of sensor

nodes is less than the number of sensor nodes in

random deployment. Maintenance of random

deployment is difficult as compared to the

maintenance of pre-planned deployment. Wireless

sensor network have many application like

Environmental monitoring, Acoustic detection,

Seismic detection, Military surveillance, Process

monitoring etc.

Forest fire detection is the main problem faced by

number of countries all over the world. In early

detection of forest fire, camera surveillance and

satellite based monitoring were used but that was

inefficient in a number of manners. In recent years,

Wireless sensor network was used to detect the forest

fire. A number of research have considered using

wireless sensor network for wood fire system Son et

al.[4] presents in their paper a forest fire detection

system in the south Korean mountains using wireless

sensor networks. WSNs can be connected to the

internet so that the information can be used for future

risks. The developed system consists of WSNs,

middleware and web application. The protocol they

used for routing was MCF (minimum cost path

forwarding) which required a routing table for each

sensor node to find a minimum path to the base station.

Sensor nodes sense the temperature, humidity and

smoke to forward it to the base station node and then

to the gateway. The gateway is connected to the

middle ware and web application, which analyse the


62

collected data and information on daily basis, and

looks for the likelihood of an event. Son‟s network is

more concerned with detection of the fire and decision

making than the network communication reliability.

They did not discuss the network coverage and

distribution of the sensors.

Hurting et al. [5] in their paper presents FireWxNet, a

multi-tiered portable wireless system for monitoring

weather conditions in rugged wild land fire

environments .In early stages the main aim of their

studies was to investigate the behavior of the forest

fire rather than the detection of the fire. In their

network, they used wireless sensor networks and web

cameras. Wireless sensor network for weather status

and web cameras for images of the fire. Their system

uses a tiered structure beginning with directional

radios to stretch deployment capabilities into the

wilderness far beyond current infrastructures. They

stated that for vision, they used web cameras and for

location information, they used sensor nodes with

small GPS.

Doolin and Sitar [6] proposed wireless sensor network

for wildfire monitoring in which they used

environmental sensors for sensing the temperature,

humidity, and barometric pressure. In addition, with

every sensor node a GPS device is used which is one

of the problems in this network because using a GPS

device with every sensor node will make the network

more expensive. GPS device will consume power so

they will reduce the network lifetime. The nodes will

send the sensed data to the base station. The base

station was connected to the MySQL database and

clients for alarm monitoring. The main problem in this

network is the deployment of sensor nodes are pre-

planned so deploying every sensor manually will be

impossible for large forests. Another problem is the

distance between the sensor nodes are too far so in

case of node failure the connection between sensors

and base station might be lost.

Yu et al. [7] in their paper proposed real-time forest

fire detection with wireless sensor network. To

prolong the network life time they have used neural

wireless sensor network and for routing the data they

relied on clustering algorithm in which the nodes

sends the sensed data to the cluster head and then to

the base station. They have used in-network

processing approach to reduce the communication

between the sensor and saves energy consumption.

Aslan [3] presented a framework for the use WSNs in

forest fire detection and monitoring. Their framework

incorporates the design of four main components of a

wireless sensor network: the deployment scheme, the

logical topology and architecture of the network, the

intra-cluster communication scheme, and the inter-

cluster communication scheme. They used cluster

scheme as network topology. Sensor deployment

scheme was represented as the distance between

sensors, minimum collision, and minimum number of

sensors deployed with full coverage. The

communication between nodes and clusters divided

into initialization phase, risk free phase, fire threat

phase, and progressed fire phase. Nodes enter or

change their phase according to danger rate calculation,

which depends on NFDRS (National Fire Danger

Rating System), temperature, and humidity ranges.

The aim of the intra-cluster communication scheme is

the power balancing for cluster heads.

Lloret et al. [8] in their paper presents a mesh network

of wireless sensors with internet protocol (IP) cameras

in order to detect and verify fire in rural and forest

areas in Spain. In the proposed network they suggested

that the sensor will first detect the fire and then it will

sends information to the base station. The base station

will then sends the response and will switch „on‟ the

camera closest to the event to catch real images and

avoiding false alarms. Their paper is based on testing

the performance of four IP cameras and energy

consumption. The problem in their network is that IP

cameras are not efficient in dark, foggy, and severe

weather conditions and also the transfer of captured

images that will be very huge and that will consume a

lot of energy and will occupy a lot of space. In

addition, we know in sensor network we have limited

memory and limited energy. The installation of IP

cameras should be manual and will be in appropriate

position.

Conrad et al. [9] produced a business case for the

Enhanced Forest Fire Detection System with a GPS

project in Pennsylvania. They say that every year in

Pennsylvania 2554 acres are damaged because of

forest fires, which causes economic loss and potential

loss of human life and environment. They proposed

using fire sensors and GPS devices for the detection of


63

fire. They want to use the existing technology and

replacing the existing technology with more good ones.

When smoke was detected, the sensors will send a

signal to GPS satellite, and then the GPS satellite

would duplicate the signal to the handheld GPS device

and the central monitoring database to display the fire

location on the installed map for that area. The

problem here is that this project will require a lot of

money and also using GPS devices will reduce the

network lifetime because they will consume much

more energy. Garcia et al. [10] present a simulation

environment that can create a model for a fire by

analysing the data reported by sensor nodes and by

using some geographical information about the area.

The use of topography of the environment

distinguishes the study from some other solutions. The

estimation of the spread of a fire is sent to hand-held

devices of fire fighters to help them in fighting against

the fire in field.

III. COMPARISON AND CONCLUSION

Hefeeda et al. [11] developed a wireless sensor

network for forest fire detection based on Fire

Weather Index (FWI) system, which is one of the most

comprehensive forest fire danger rating systems in

USA. The system determines the spread risk of a fire

according to several index parameters. It collects

weather data via the sensor nodes, and the data

collected is analysed at a centre according to FWI. A

distributed algorithm is used to minimize the error

estimation for spread direction of a forest fire [12-45].

Table 1: Comparison among Various Schemes

IV. REFERENCES

[1] http://www.nifc.gov/fireinfo/2004/index.html,

“Wildland Fire Season 2004 Statistics and

Summaries,” National Interagency Coordination

Center.

[2] NOAA satellite and information service,

“Advanced Very High Resolution Radiometer

AVHRR,” 2012,

http://noaasis.noaa.gov/NOAASIS/ml/avhrr.htm

l.

[3] Y. Aslan, A framework for the use of wireless

sensor networks in the forest fire detection and

monitoring [M.S. thesis], Department of

Computer Engineering, The Institute of

Engineering and Science Bilkent University,

2010.

[4] Son, Y. Her, and K. Kim, “A Design and

Implementation of Forest-Fires Surveillance

System based on Wireless Sensor Networks for

South Korea Mountains,” International Journal

of Computer Science and Network Security, vol.

6, no. 9, pp. 124– 130, 2006.

[5] C.Hartung, R. Han, C. Seielstad, and S.

Holbrook, “FireWxNet: Amulti-tiered portable

wireless systemfor monitoring weather

conditions in wildland fire environments,” in

Proceedings of the 4th International Conference

on Mobile Systems, Applications and Services

(MobiSys ‟06), pp. 28–41, ACM, Uppsala,

Sweden, June 2006

[6] D.Doolin and N. Sitar, Wireless Sensors for

Wild Fire Monitoring, Smart Structure and

Material, San Diego, Calif, USA, 2005.

[7] L. Yu, N. Wang, and X. Meng, “Real-time

forest fire detection with wireless sensor

networks,” in Proceedings of the International

Conference on Wireless Communications,

Networking and Mobile Computing

(WCNM‟05), pp. 1214–1217, September 2005.

[8] J. Lloret, M. Garcia, D. Bri, and S. Sendra, “A

wireless sensor network deployment for rural

and forest fire detection and verification,”

Sensors, vol. 9, no. 11, pp. 8722–8747, 2009

[9] A. Conrad, Q. Liu, J. Russell, and J. Lalla,

“Enhanced Forest Fire Detection System with

GPS Pennsylvania,” 2009.

[10] E. M. Garc´ıa, M. ´ A. Serna, A. Berm´udez, and


64

R. Casado, “Simulating a WSN-based wildfire

fighting support system,” in Proceedings of the

International Symposium on Parallel and

Distributed Processing with Applications

(ISPA‟08), pp.896–902, December 2008.

[11] M.Hefeeda and M. Bagheri, “Wireless sensor

networks for early detection of forest fires,” in

Proceedings of the IEEE International

Conference on Mobile Adhoc and Sensor

Systems (MASS ‟07), October 2007.


study of spectrum sensing techniques in

cognitive radio networks. In Computer and

Information Technology (WCCIT), 2013 World

Congress on (pp. 1-8). IEEE.


Dual head clustering scheme in wireless sensor

networks. In Emerging Technologies (ICET),

2012 International Conference on (pp. 1-5).

IEEE.

[14] Khan, F., Kamal, S. A., & Arif, F. (2013).

Fairness improvement in long chain multihop

wireless ad hoc networks. In 2013 International

Conference on Connected Vehicles and Expo

(ICCVE) (pp. 556-561). IEEE.



In 2014 IEEE 3rd Global Conference on

Consumer Electronics (GCCE) (pp. 647-650).

IEEE.






Throughput Improvement in Wireless Sensor

and Actor Networks. In 5th National Symposium

on Information Technology: Towards New

Smart World (NSITNSW) (pp. 1-8).

[18] Khan, F., Jan, S. R., Tahir, M., Khan, S., &

Ullah, F. (2016). Survey: Dealing Non-

Functional Requirements at Architecture

Level. VFAST Transactions on Software

Engineering, 9(2), 7-13.









134-139). IEEE.





2015, IEEE.




Applied, Environmental, and Biological




networks. In Electrical and Computer

Engineering (CCECE), 2014 IEEE 27th

Canadian Conference on (pp. 1-6). IEEE.


Khan, S. A. (2016). Performance Improvement

in Wireless Sensor and Actor Networks. Journal

of Applied Environmental and Biological

Sciences, 6(4S), 191-200.



Cognitive Radio Networks. 電子情報通信学会

ソサイエティ大会講演論文集, 2012(2), 152.

[26] Khan, F., Jan, S. R., Tahir, M., & Khan, S.

(2015, October). Applications, limitations, and




262). IEEE.

[27] Jabeen, Q., Khan, F., Hayat, M. N., Khan, H.,

Jan, S. R., & Ullah, F. (2016). A Survey:

Embedded Systems Supporting By Different

Operating Systems. International Journal of

Scientific Research in Science, Engineering and

Technology (IJSRSET), Print ISSN, 2395-1990.


Enhanced and Effective Learning through

Mobile Learning an Insight into Students

Perception of Mobile Learning at University

Level. International Journal of Scientific

Research in Science, Engineering and

Technology (IJSRSET), Print ISSN, 2395-1990.


65

[29] Jan, S. R., Khan, F., & Zaman, A. The

perception of students about mobile learning at

University level.






[31] Jan, S. R., Shah, S. T. U., Johar, Z. U., Shah, Y.,

& Khan, F. (2016). An Innovative Approach to

Investigate Various Software Testing

Techniques and Strategies. International

Journal of Scientific Research in Science,

Engineering and Technology (IJSRSET), Print

ISSN, 2395-1990.

[32] Khan, I. A., Safdar, M., Ullah, F., Jan, S. R.,

Khan, F., & Shah, S. (2016). Request-Response



Innovative Ideas in Education, Online ISSN-

2395-4396

[33] Azeem, N., Ahmad, I., Jan, S. R., Tahir, M.,

Ullah, F., & Khan, F. (2016). A New Robust

Video Watermarking Technique Using H.

264/AAC Codec Luma Components Based On

DCT. In International Journal of Advance

Research and Innovative Ideas in Education,

Online ISSN-2395-4396

[34] Jan, S. R., Khan, F., Ullah, F., Azim, N., &

Tahir, M. (2016). Using CoAP Protocol for

Resource Observation in IoT. International

Journal of Emerging Technology in Computer

Science & Electronics, ISSN: 0976-1353

[35] Azim, N., Majid, A., Khan, F., Jan, S. R., Tahir,

M., & Jabeen, Q. (2016). People Factors in

Agile Software Development and Project

Management. In International Journal of


Electronics (IJETCSE) ISSN: 0976-1353


M., & Jabeen, Q. (2016). Routing of Mobile

Hosts in Adhoc Networks. In International

Journal of Emerging Technology in Computer

Science & Electronics (IJETCSE) ISSN: 0976-

1353.

[37] Azim, N., Khan, A., Khan, F., Majid, A., Jan, S.

R., & Tahir, M. (2016) Offsite 2-Way Data

Replication toward Improving Data Refresh


Engineering Trends and Applications, ISSN:

2393 – 9516

[38] Tahir, M., Khan, F., Jan, S. R., Azim, N., Khan,

I. A., & Ullah, F. (2016) EEC: Evaluation of

Energy Consumption in Wireless Sensor

Networks. . In International Journal of


2393 – 9516




Networks,” Concurrency and Computation:

Practice and Experience, “accepted”, 2016.

[40] Azim, N., Qureshi, Y., Khan, F., Tahir, M., Jan,

S. R., & Majid, A. (2016) Offsite One Way Data



Computer Science Trends and Technology,

ISSN: 2347-8578

[41] Safdar, M., Khan, I. A., Ullah, F., Khan, F., &

Jan, S. R. (2016) Comparative Study of Routing


International Journal of Computer Science

Trends and Technology, ISSN: 2347-8578

[42] Tahir, M., Khan, F., Babar, M., Arif, F., Khan,

F., (2016) Framework for Better Reusability in


Journal of Applied Environmental and

Biological Sciences (JAEBS), 6(4S), 77-81.


(2016). Collaboration Methodology for

Integrating Non-Functional Requirements in

Architecture. In the Journal of Applied

Environmental and Biological Sciences

(JAEBS), 6(4S), 63-67

[44] Jan, S.R., Ullah, F., Khan, F., Azim, N., Tahir,

M., Khan, S., Safdar, M. (2016). Applications

and Challenges Faced by Internet of Things- A

Survey. In the International Journal of


2393 – 9516



IoT Objects,”, “Submitted”, 2016.

CSEIT161113 | Received: 02 August 2016 | Accepted : 07 August 2016 | July-August 2016 [(1)1: 66-71]



66

Congestion Detection and Mitigation Protocols for Wireless Sensor Networks

Muhammad Zeeshan, Fazlullah Khan, Syed Roohullah Jan

Deptartment of Computer Science, Abdul Wali Khan University Mardan, Pakistan

ABSTRACT

Congestion control is tremendously important area within wireless sensor networks (WSN). With the appearance of

new network applications, the non-stop increasing traffic is starting to experience unexpected situation of network

congestion. Congestion in wireless sensor network affects the nonstop flow of data, loss of information, delay and

reduces the energy of nodes due to overhead of retransmission. Therefore, congestion needs to be control in wireless

sensor network in order to extend system lifetime, improve fairness and quality of service and to attain high energy

efficiency. This paper revives different routing protocols used in wireless sensor networks to mitigate and control

congestion and to provide consistency for different applications and prolong the life of the wireless sensor network.

Keywords : WSN (Wireless sensors networks), Congestion, Congestion Control.

I. INTRODUCTION

Wireless sensor networks (WSN)[1] consists of

various wireless devices mount with various types of

sensors, shown in Fig.1 to assemble information such

as temperature, pressure, humidity, sound, vibration

and wind speed from the surroundings. Wireless

sensor network widely applied to environment such as

environmental monitoring, target tracking, habitat

monitoring, healthcare, telecommunication monitoring,

military surveillance and factory monitoring. When

the event is occurring for which the system is installed

the sensors activate and begin to send the data to the

base station.

Figure 1: Wireless Sensor network

If the event is occurring frequently or with high value

or many sensors capture the same event occurring they

will send more packets to the base station as a result

congestion starts from that point and spreads along its

links as shown in Fig 2. Data crossing that sub-

network (area of congestion) would suffers from

prolonged delays (buffer waiting) eventually leading

to timeouts (loss rate).

Figure 2 : Congestion Scenario in WSN

Initially the researchers are interested in the design of

routing scheme to enable data transfer in WSN. But

later on they realized that there must be such a

mechanism to address the situation when there is

chance of congestion or congestion has occurred. In

this paper we give an overview of the congestion

control and detection protocols/Techniques in

Wireless Sensor Networks. We have to first avoid the

congestion so that the congestion did not occur and if


67

the congestion occurs then we will try to mitigate the

congestion using different techniques.

In Section “A” we have discussed FUSION which is

congestion detection and avoidance mechanism uses

hop by hop congestion detection like CODA [3] but

the difference is that it uses implicit notification or

congestion notification bit in the header of packet. It

also include prioritized MAC scheme to ensure

fairness i.e. unlike CSMA[2] which gives equal

chance to every node to transmit their data but in this

technique the high priority nodes which have more

data then we assign extra time to drain their buffer. In

Section “B” we have discuss the CODA (Congestion

detection and avoidance) technique which uses the

buffer overflow and channel sensing for congestion

detection. Once the congestion is detected we use the

back pressure and close loop method to inform the

other nodes that the congestion is occurred and also to

inform them to limit their sending data rate. In section

“C” we have discuss anther protocol for congestion

detection and avoidance named CCF(Congestion

control and Fairness) protocol which states that every

node is able to control the rate of its downstream

nodes This allows the root node to reduce the

generation rates of all downstream nodes. By reducing

the transmission rates of all downstream nodes when

this node's queue is full or about to become full, it

allows the queue to empty. CCF guarantees simple

fairness so that each node receives the same

throughput. In Section "D" we have discussed another

congestion control protocol which is priority based

congestion control protocol for wireless sensor

network. The node priority index is brought in to show

the importance of every one node. In Section “E” we

have discussed another congestion control protocol

named ECODA (Enhanced Congestion Detection and

Avoidance) protocol which is the superior version of

CODA which uses dual buffer thresholds to detect

congestion and a queue scheduler that arrange the

packets for transmission and to drop the packets with

low priority when the buffer is about to full or accedes

the upper limit of the buffer. On node level it uses the

back pressure method and bottle neck data sending

rate control for congestion mitigation. ECODA cares

of the high priority packets in the manner if the

congestion has occurring the flexible queue scheduler

select the low precedence packets for dropping to

defeat the congestion.


A. Related Work

FUSION MECHANISIM

FUSION IS DESIGN TO PROVIDE UPSTREAM

CONGESTION CONTROL MECHANISM IT CONSISTS OF

THREE CONGESTION MITIGATION TECHNIQUES

APPLY IN DIFFERENT LAYERS, THAT IS

i. Hop-by-hop flow control

Using hop by hop flow control a sensor node present

congestion detection and congestion mitigation

congestion is discover through both queue occupancy

and channel sampling techniques. The hop by hop

flow control scheme in FUSION is similar to

backpressure scheme in CODA[9]. The only

difference in fusion is that each sensor node sets a

congestion bit in the header of every outgoing packet

instead of using backpressure messages.

ii. Rate limiting of source traffic in the traffic in

the transit sensor nodes to provide fairness.

When a sensor node overhear a packet from its parent

node (the node closer to the sink) with the congestion

bit is set, it stops forwarding data toward the sink,

Rate limiting is a defensive scheme to avoid

congestion.

iii. Prioritized MAC(Medium Access Control)

FUSION also includes a prioritized MAC scheme to

guarantee that congested nodes receive prioritized

access to the channel. In traditional CSMA mechanism

each node has the same opportunity to send the data

over the channel but in WSN the parent’s nodes that

are closed to the sink may gather more traffic and can

overflows if it does not have more chances to send out

its packets. To avoid this problem FUSION uses a

random back-off time for each node is introduced

which it related to its local congestion state so that the

congested nodes may drain its buffer faster.

CODA (Congestion Detection and Avoidance)

CODA is an energy conserving and efficient control

technique that is designed to solve congestion in the

upstream direction i.e., the sensor to sink direction.


68

It involves three mechanism which is following:

B. Congestion Detection

Accurate and efficient congestion detection plays an

important role in the congestion control of wireless

networks. The detection method in CODA is the

receiver based congestion detection. CODA uses a

combination of the present and past channel loading

conditions, and the current buffer occupancy, to infer

accurate detection of congestion at each receiver with

low cost. Sensor networks must know the state of the

channel since the transmission medium is shared and

may be congested with traffic between other nodes in

the neighborhood. Listening to the channel to measure

local loading incurs high energy costs if performed

continually. Therefore, CODA uses a sampling

scheme that activates local channel monitoring at the

appropriate time to minimize cost while forming a

accurate estimate.

Figure 3: Congestion detection and notification in

CODA

i. Open-loop, hop-by-hop backpressure

Once congestion is detected node broadcast

backpressure message and this message circulate

towards the source as shown in Fig. 3. In CODA, a

node broadcasts backpressure mechanism as long as it

detects congestion. The node detecting congestion will

report its upstream neighbors to reduce rate of data

flow. A node that receives a backpressure message

will adjust its sending rate by Adaptive Increase

Multiplicative Decrease, the AIMD rate adjustment

technique or by dropping packets based on the local

congestion plan. When an upstream node (towards the

source) receives the backpressure message, it decides

whether it has to further propagate the backpressure

upstream based on the local network situation.

ii. Closed-loop, multisource regulation

The cost of closed loop flow control is high compared

to open loop flow control because it required feedback

signaling. CODA runs Closed-loop congestion control

mechanism on the sink to regulate multiple sources in

the case of continual congestion. Essentially when the

transmission rate of a source surpass maximum

theoretical throughput (Smax) the source informs the

sink by setting a bit in every packet that it transmits to

the sink as long as the transmission rate remains

higher than Smax. in response sink starts sending

ACKs to the source until the sink detects congestion.

When the sink detects congestion, it stops sending

ACKs until the congestion is alleviated, to implicitly

notify to drop its rate.

Disadvantages of CODA

1. Unidirectional control from sensors to sink.

2. Decreased reliability.

3. The delay and response time increases under

heavy closed loop congestion

CCF(Congestion Control and Fairness)

In this section we propose a distributed and scalable

algorithm that purges congestion within a sensor

network, and that ensures the fair delivery of packets

to a central node, or base station. We say that fairness

is achieved when equal number of packets is received

from each node. Congestion control and fairness [4]

for many-to-one routing method is a distributed and

scalable algorithm which eradicates the congestion in

a wireless sensor network. CCF provides hop-by-hop

upstream congestion control that not only eliminates

congestion but also ensure fair delivery of packets to

the base station [6] in CCF fairness is achieved

when an equal number of packets are received from

each node to the base station or sink by maintaining a

separate queue for each their preceding hop node. CCF

detects congestion based on packet service time and

control congestion based on hope-by-hope manner. In

CCF the congestion of any node can be evaluated by

the number of available child nodes (downstream

nodes) the average rate at which the packets can be

sent by it. When the congestion is experienced it

informs the downstream nodes to reduce their data rate

and vice versa by implicit notification like closed loop

hop-by-hop backpressure in CODA [5] .


69

Limitations:

i. CCF uses only packet service time to detect

congestion and therefore it cannot detect either

under utilization nodes or links.

ii. CCF does not provide any reliability mechanism

i.e. preserve equal resource for each sensor node.

PCCP ( Priority Based Congestion Control Protocol):

In Priority based Congestion Control [7] node priority

index is introduced to reflect the importance of each

node. PCCP is a more rapid and more energy well-

organized congestion control algorithm than CCF.

PCCP maintains a priority index which is used as an

indicator of the importance of each node. The packet

inter arrival time and the packet service time are used

together to find out the degree of congestion.

Intelligent congestion detection (ICD)

Implicit congestion notification (ICN)

Priority based rates adjustment (PRA)

C. Intelligent congestion detection (ICD):

Not like in TCP in which congestion is detected at the

endpoints in PCCP congestion is locally detected in

the intermediary nodes, based on the mean packet

inter-arrival time (ta) which is the mean of the time

between two successively arriving packets and the

packet service time (ts) [8] which is the time between

the MAC layer and the successful broadcast of the last

bit ICD can be define as

d (i) = ts / ta

The congestion degree helps in estimating the current

congestion level at each intermediate node.

D. Implicit Congestion Notification (ICN)

PCCP uses the technique of ICN to attached

congestion information in the header of the data

packets. The congestion information is stored in the

header of the packets that are forwarded when trigged

by either of the following two events.

When the threshold is exceeded by the number of

packets forwarded by a node.

When a congestion notification is heard by a node

from its parent node.

At every node "i" PCCP attached ta, ts and overall

priority value.

E. Priority based rates adjustment (PRA)

The additive increase and multiplicative decrease

(AIMD) used in conventional transport control

protocols such as TCP is not of much help in adjusting

the transmission rate as the congestion notification bit

holding limited information. Therefore it is important

for the nodes to be notified as to precisely how much

to increase or decrease the rate the congestion degree.

The priority index and the global priority values help

in providing more information for exact rate

adjustment.

ECODA (Enhanced Congestion Detection and

Avoidance)

ECODA is an energy efficient congestion control

scheme for sensor networks. In this method, the

given method is followed.

i. Dual Buffer threshold Congestion detection:

The dual buffer thresholds and weighted buffer

difference are used to detect the congestion. The Fig

shows the details of buffer state such as “accept state

”, “filter state” and “reject state)”. The different buffer

states are pretend different channel loading which is

used to accept or reject packets in different states.

Figure 4: Dual Buffer threshold Congestion

The packet at each node has to send for buffer

monitoring and attached its weighted buffer changing

rate (WR) and weighted queue length (WQ) with

outgoing packets. The corresponding congestion level

bit in the outgoing packet header is set if a node’s

buffer occupancy exceeds a certain threshold.


70

The ECODA [10] care of high priority packets, if the

node data is most essential among its neighbors then if

congestion take place, other nodes should lower down

their data sending rate to mitigate node’s congestion

so that the high priority packets may reach to the

station in time.

Two thresholds Qmin and Qmax are used to show

different buffer states. Different buffer states imitate

different channel loading, corresponding strategy is

adopted to accept or reject packets in different states.

ii. Flexible queue scheduler and weighted

fairness:

The Flexible Queue Scheduler is used to drop a low

priority packet rather than the high priority packet

when a high priority packet arrives if the queue in a

sensor node is nearly full and dominated with low

priority packets. At the same time, the high priority

packet may be dropped due to queue overflow with

tail-dropping.

For managing packets with different strategy, two

thresholds are used such as Qmin and Qmax .

The scheduler works based on the following steps:

a. If ( 0 ≤ N ≤ Qmin) :

All incoming packets are buffered, because queue

utilization is low.

b. If ( Qmin ≤ N ≤ Qmax ) :

Some packets with low priority are dropped or

overwritten by succeeding packets with high

priority.

c. If ( Qmax ≤ N ≤ Q ):

Some packets with high dynamic priority is

dropped or overwritten, then the expected average

buffer length increases at a rate of two variables

that can be tuned to achieve best possible system

performance.

iii. A bottleneck-node-based source sending rate

control scheme

Both transient (temporary) and persistent (continiual)

congestion handle by ECODA. It makes use of hop-

by-hop backpressure mechanism to handle transient

congestion. It uses bottleneck node based source

sending rate control and multipath load balancing to

handle persistent congestion. This mechanism does not

required explicit ACK from sink; every node

determines routing path status from sink and sender

find better path to forward data. Bottleneck node

identified and source data sending rate adjust more

accurately using this mechanism.

III. CONCLUSION

In this paper we have reviewed different wireless

sensor routing protocols / algorithm for congestion

control, reliability and packet loss. There are several

protocols for congestion control in wireless sensor

network. This paper present an overview of several

congestion control techniques like FUSION, CODA,

CCF, PCCP and ECODA. Also the comparative

analysis of congestion control techniques is presented

using several parameters like throughput, delay and

energy consumption which shows among the these

techniques Enhanced Congestion Detection and

Avoidance (ECODA) is best for congestion control as

it attain high throughput, less delay & less

consumption of energy compare to other techniques

[12-20].

Comparision Table of congestion control protocols

IV.REFERENCES

[1] Hull, B., Jamieson, K., Balakrishnan, H.:

Mitigating congestion in wireless sensor

networks. In: 2nd International Conference on

Embedded Networked Sensor Systems.

Maryland (2004)

[2] Akyildiz, W. Su, Y. Sankarasubramaniam,

and E. Cayirci, "A survey on sensor


71

networks," IEEE Communications Magazine,

Vol: 40 Issue: 8, pp. 102-114, August 2002

[3] P802.11, IEEE Draft Standard forWirelessLAN

Medium Access Control (MAC) and Physical

Layer (PHY) Specifications, D2.0, July 1995.

[4] Wan C -Y, Eisenman S B, Campbell A T,

CODA: Congestion detection and avoidance in

sensor networks. In the Proceedings of the First

International Conference on Embedded

Networked Sensor Systems (SenSys’03), Los

Angeles, CA, USA, 2003,pp. 266–279

[5] V. Vijayaraja, Dr. R. Rani Hemalini,

“Congestion in Wireless Sensor Networks and

various techniques for Mitigation Congestion- A

review”. IEEE International Conference on

Computational Intelligence and computing

Research.

[6] Chien- Yin Wan, Shane B. Eiseaman, Andrew

T. Champbell, “ CODA: Congestion Detection

and Avoidance”.

[7] V. Vijayaraja, Dr. R. Rani Hemalini,

“Congestion in Wireless Sensor Networks and

various techniques for Mitigation Congestion- A

review”. IEEE International Conference on

Computational Intelligence and computing

Research.

[8] Yaghmaee, Mohammad Hossein, and Donald

Adjeroh. "A new priority based congestion

control protocol for wireless multimedia sensor

networks." In World of Wireless, Mobile and

Multimedia Networks, 2008. WoWMoM 2008.

2008 International Symposium on a, pp. 1-8.

IEEE, 2008.

[9] Yaghmaee, Mohammad Hossein, and Donald

Adjeroh. "A new priority based congestion

control protocol for wireless multimedia sensor

networks." In World of Wireless, Mobile and

Multimedia Networks, 2008. WoWMoM 2008.

2008 International Symposium on a, pp. 1-8.

IEEE, 2009.

[10] Wan ,C-Y, Eisenman, SB, Campbell, AT..

CODA: Congestion detection and avoidance in

sensor networks. In: Proceedings of

ACMSensys’03, November5–7, 2003,

LosAngeles, California, USA; 2003.

[11] Tao LQ, Yu FQ. ECODA: enhanced congestion

detection and avoidance for multiple class of

traffic in sensor networks. Transactions on

Consumer Electronics 2010.






Dual head clustering scheme in wireless sensor

networks. In Emerging Technologies (ICET),

2012 International Conference on (pp. 1-5).

IEEE.



In 2014 IEEE 3rd Global Conference on

Consumer Electronics (GCCE) (pp. 647-650).

IEEE.

[15] Khan, F., Jan, S. R., Tahir, M., Khan, S., &

Ullah, F. (2016). Survey: Dealing Non-

Functional Requirements at Architecture Level.

VFAST Transactions on Software Engineering,

9(2), 7-13





2015, IEEE.









134-139). IEEE.









Networks,” Concurrency and Computation:

Practice and Experience, “accepted”, 2016.

international journal of scientific research inijsrcseit.com/paper/v1i1.pdf · 2017-03-14 · i...

Documents