research article energy-aware sink node...

Research ArticleEnergy-Aware Sink Node Localization Algorithm forWireless Sensor Networks

Mohamed Mostafa Fouad123 Vaclav Snasel24 and Aboul Ella Hassanien35

1Arab Academy for Science Technology and Maritime Transport Cairo Egypt2IT4Innovations VSB-Technical University of Ostrava Ostrava Czech Republic3Scientific Research Group in Egypt (SRGE) Cairo Egypt4Faculty of Electrical Engineering and Computer Science VSB-Technical University of Ostrava Ostrava Czech Republic5Faculty of Computers and Information Cairo University Cairo Egypt

Correspondence should be addressed to Mohamed Mostafa Fouad mohamed mostafaaastedu

Received 27 February 2015 Accepted 29 June 2015

Academic Editor Federico Barrero

Copyright copy 2015 Mohamed Mostafa Fouad et al This is an open access article distributed under the Creative CommonsAttribution License which permits unrestricted use distribution and reproduction in any medium provided the original work isproperly cited

Wireless sensor networks (WSNs) are a family of wireless networks that usually operate with irreplaceable batteries The energysources limitation raises the need for designing specific protocols to prolong the operational lifetime of such networks Theseprotocols are responsible for messages exchanging through the wireless communications medium from the sensors to the basestation (sink node) Therefore the determination of the optimal location of the sink node becomes crucial to assure both theprolongation of the networkrsquos operation and the quality of the provided services This paper proposes a novel algorithm based ona Particle Swarm Optimization (PSO) approach for designing an energy-aware topology control protocol The deliverable of thealgorithm is the optimal sink node location within a deployment area The proposed objective function is based on a number oftopology control protocolrsquos characteristics such as numbers of neighbors per node the nodesrsquo residual energy and how they are farfrom the center of the deployment area The simulation results show that the proposed algorithm reveals significant effectivenessto both topology construction and maintenance phases of a topology control protocol in terms of the number of active nodes thetopology construction time the number of topology reconstructions and the operational networkrsquos lifetime

1 Introduction

The increasing needs for ubiquitous devices to interact withthe physical world have developed the importance of wirelesssensor networks (WSNs) in a number of applications Theseapplications may include military [1] remote environmentalmonitoring [2] smart road monitoring [3] and remotepatient monitoring [4] applications The major challengesattached with such applications are related to the wirelesssensor networksrsquo limitations where a sensor node has alimited energy source a small memory footprint and lowcomputational capability processor Furthermore the deploy-ment strategies of the sensors may add extra limitationsfor example the random deployment of large numbers ofsensors may develop other issues related to the wireless

networkrsquos scalability data reliability security privacy andefficient coverage

Determining the sensor node location is an importantmatter in WSNs the more accurate the node placementreinforcement the more the sensing accuracy The nodesrsquolocations information becomes more important when thesensors are deployed randomlywhichmeans that they shouldautomatically reconfigure themselves without any humanintervention or control After the network establishmenteach node captures some measurements from its environ-ment and broadcasts these data through its neighbors tothe sink node However assigning a sink node addressin self-configuration topology raises another problem thatinfluences the performance of the WSN in terms of energydelay and operational lifetime Therefore the sink node

Hindawi Publishing CorporationInternational Journal of Distributed Sensor NetworksVolume 2015 Article ID 810356 7 pageshttpdxdoiorg1011552015810356

2 International Journal of Distributed Sensor Networks

location should be accurately selected so other nodes willnot use much power to deliver their data to that locationAs a consequence the networkrsquos lifetime will be maximizedFurthermore any proposed optimization should not trade offthe area coverage or networkrsquos connectivity characteristics

A great number of researches assumed that within asingle network the sink node should always be deployedwithin the center point of the area of interest [5] for examplethe work done in [6 7] which applied the P-Median Problem(PMP) model which is a well-known NP-hard problemdefined by Hakimi [8] It defined the sink node locationwhere the total weighted distance of reaching all nodesis minimized On the other hand [9] gives a comparisonbetween a number of sink placement strategies Among theother discussed strategies is the Geographic Sink Placement(GSP) strategy in which the sink is placed at the centerof gravity of a sector of a circle Although it gives fastsuitable solutions it cannot guarantee the optimal sink nodelocation Also the authors introduced another heuristic basedapproach entitled Genetic Algorithm-Based Sink Placement(GASP) The GASP is suitable to produce nearly optimalsolution for large-scale networks

Chen and Li [5] investigated two proposed sink nodeplacement strategies the energy-oriented and lifetime-oriented strategies in both the single-hop and multiple-hopWSNs respectively The two strategies adopted a routing-cost based Ant routing algorithm According to their con-clusion the lifetime-oriented strategy had outperformed theenergy-oriented strategy in terms of networkrsquos lifetime Otherresearches used the integer linear programming to find eitherthe optimal position for the sink node or relay nodes forexample the work of Hou et al [12] that developed an effi-cient polynomial-time heuristic algorithm (SPINDS) whichattempted to increase the network lifetime by iterativelymoving a relay node to another enhanced location Guneyet al [13] developed two mixed-integer linear programmingformulations as well those had easily computed good feasiblesolutions for the sink location and efficient routingThey useda Tabu Search heuristic to identify the best sensor locationsthat could guarantee the total networkrsquos coverage

Selecting best locations for cluster headswithin clusteringbased routing algorithms is more similar to selecting alocation for a sink node within a wireless sensor networkUsually both are sharing the same aim which is trying toreduce the required transmission energy used by sensornodes within a network

Yadav et al [14] suggest a PSO-based solution to theoptimal clustering problem through the use of residualenergy and transmission distance of sensor nodes Theyconsider a new operator to be applied inside their algorithmthat checks the validity of reached location of the head ofthe cluster within the current iteration of the PSO algorithmIf the location is not valid then the algorithm returns tothe nearest valid node location that has the highest residualenergy to become the current head of the cluster Althoughthe results are promising the main limitation within thealgorithm is the need for a centralized authority node

Moreover there are other nature inspired algorithmsthat were used to optimally formulate and optimize routing

through clusters Pan et al [15] propose an algorithm basedon Uneven Clustering Multihop Algorithm (UCMA) andan Improved Ant Colony Optimization (IACO) algorithmThe first former algorithm is used for grouping sensor nodesinto unequal clusters along with selecting their cluster headThe decision of the UCMA is based on the nodesrsquo locationsand their residual energies On the other hand the IACO isapplied for the routing discovery between the head of thecluster and the sink node The performance evaluations oftheir proposed algorithms show that the application of theIACO algorithm had accelerated the iterative conversion rateand efficiently reduces the energy consumption rate

This paper proposes an energy-aware sink node localiza-tion algorithm for a topology control protocol using a ParticleSwarm Optimization technique The simulation results showthat the proposed energy-aware algorithm minimizes thenumbers of active nodes and subsequently increases thewhole networkrsquos lifetime

The remainder of this paper is organized as followsSection 2 gives a brief overview of topology control protocolsIn Section 3 is the proposed optimization technique Thesimulation results and a number of performance evaluationsare given in Section 4 Finally Section 5 points out theconcluding remarks

2 Topology Control Protocols

A large number of protocols and algorithms have beenproposed to overcome the constrained resources of thewireless sensor networks Many of these protocols andalgorithms were developed to extend the lifetime of thewireless network through wisely managing the use of theirconstrained resources Topology control (TC) protocols are ofthe algorithms designed to construct minimized topologiesThese topologies proved their efficiency for both energyconsumption and radio interferences reductions

A topology control protocol is an iterative process thatdynamically reduces the initial topology of a wireless sensornetwork through controlling nodesrsquo transmission rangesThemain advantage of the TC protocols is the assets saving of thewireless sensor networks such as the networkrsquos connectivityand coverage [16 17] Figure 1 illustrates with brief descrip-tions the topology control protocolrsquos phases

Although the TC protocols reduce energy consumptionratio they still suffer from a number of drawbacks for exam-ple securing topology control protocols could negativelyaffect the performance of such protocols [18] therefore thecurrent existing security schemes should be adopted to suitTC protocols [19] On the other hand most of the designedTC protocols (such as A3 [17] and A3Cov [20]) supposethat even if the deployment is randomized the sink nodelocation should be fixed within the center of the deploymentarea which is not the optimal case in real-life scenariosTherefore this paper provides a contribution that is of worthin enhancing the lifetime and the performance of the WSNsIt proposes a PSO-based algorithm to find the adequateposition for the sink node within a topology control protocol(the A3 protocol)

International Journal of Distributed Sensor Networks 3

(i) Each node uses its maximum transmission power therefore the resulting topology is a fully connected graph

Initialization topology

establishment phase

(i) Aiming to reduce the initial topology through controlling nodesrsquoactivities

Topology construction phase (i) Monitoring the network

status and taking a decision to switch to a new topology construction phase

Topology maintenance phase

Figure 1 Topology control protocols phases

3 Proposed Sink Node LocalizationAlgorithm Using PSO

Particle Swarm Optimization or PSO is a computationalmethod which could be defined within the heuristic methodscategories As a member of Swarm Intelligence methods (AntColony Optimization Genetic Algorithm etc) this methodtries to find best solutions from a set of candidate solutions(particles) based on predefined criteria [21] The differentialfeature of the PSO is that each particle memorizes both itsposition and velocity within the search area Then it takes adecision according to a predefined objective function howwell is its current position Through a set of iterations theparticle position is updated using two ldquobestrdquo values The firstone is the best solution the particle achieved so far and theother tracked value is the best value obtained so far by anyparticle within the solution space

The first aim of this paper is to define the fitness functionof the proposed PSO-based algorithm It is composed of anumber of features that affect the stability of any topologycontrol protocols such as the number of adjacent neighborsthe neighborsrsquo residual energy and the Euclidean distance tothe center of the deployment area [5] Since the nodes in ourexperiments are stationary it was assumed that there are 119873particles which are distributed randomly within the deploy-ment area [22] Each particle inherits specific propertiesfrom its nearest node such as position residual energy andneighborrsquos list Figure 2 shows an illustrated example whereparticle (1199011) inherits the properties of its nearest sensor node(1199044) Then and according to the fitness values the particleswill shift their position towards the currently detected 119892BestThe following steps give a complete view of the proposedPSO-based algorithm

Input A set of sensor nodes 119878 = 1199041 1199042 119904119872 where119872 isthe number of nodes Each sensor node 119904

119894has a number of

characteristics 119904119894= (pos

119894 119890119894 119899119894) where pos

119894represents the

position of the node 119904119894within the deployment area 119890

119894defines

its residual energy and 119899119894is the number of neighbors existing

within its communication radiusAnother important input is a set of particles 119875 =1199011 1199012 119901119873 where 119873 is the number of particles and

gBest

Stationary sensor node

Dynamic particle

s1

s2 s

3

s4

s5

s6

sM

p1

p2

p3

pN

Figure 2 Illustration of particles that simulate sensorsrsquo shifts to thefittest sink node location

|119875| le |119878| Each 119901119894= (V119894 pos119894 119901Best

119894 119892Best) where V

119894is a

vector that represents the particle 119901119894velocity pos

119894is another

vector that saves particlersquos position within the deploymentarea and finally 119901Best

119894and 119892Best refer to the current best

solution the particle 119901119894has achieved and the best solution

within the search space respectively

Output The fittest node 119904 isin 119878 that will act as a sink nodewhere its location guarantees the networkrsquos performance interms of connectivity coverage and operational lifetime

Step 1 Initialize V119894for all particles to zero

Step 2 Adjust the initial fitness values of 119901Best119901119894and 119892Best

to zero

Step 3 Each particle inherits the nearest node characteristics

Step 4 Use the following equation to compute the fitnessvalue 119891(119901

119894) for each particle 119901

119894

119891 (119901119894) = 12057211003816100381610038161003816119873 (119901119894)1003816100381610038161003816 + 1205722 sum119901isin119873(119901119894)

119901119890 + 1205723119889119901 (1)


where 1205721 1205722 and 1205723 are random numbers ranged in [0 1]While119873(119901

119894) refers to the sensors neighbors for the particle119901

119894

119901119890 refers to the residual energy within a neighbor node 119901 isin119873(119901119894) and 119889

119901is the Euclidean distance between the position

of the particle 119901 and the center of the deployment area

Step 5 Update 119901Best119894using

119901Best119894=

119901119894119891 (119901119894) gt 119891 (119901Best

119894)

119901Best119894

otherwise(2)

Step 6 Select the optimized 119901Best119901119894value among all particles

to update the value of 119892Best using

119892Best = max 119901Best119901| 119901 isin 119875 (3)

Step 7 Calculate the new velocity per each particle within thecurrent iteration using

V119894 (119905 + 1) = 120596V119894 (119905) + 11988811199031 (119901Best119894 minus V119894 (119905))

+ 11988821199032 (119892Bestminus V119894 (119905)) (4)

While 119905 denotes the iteration counter and V119894represents

the particle velocity 120596 parameter is a constant inertia-weightthat controls velocity of the exploration within the searchspace Also 1199031 and 1199032 are random numbers in the range [0 1]Whereas 1198881 represents the cognitive coefficient 1198882 representsthe social coefficient towards the best solution [11]

Step 8 Each particle updates its position based on the newvelocity by means of the following equation

pos119894(119905 + 1) = pos

119894(119905) + V119894 (119905 + 1) (5)

Step 9 While either a stopping criterion or a predefinednumber of iterations are still not satisfied repeat from Step 3otherwise go to Step 10 The intended stopping criterionwithin the PSO part of the proposed algorithm is when 119892Bestvalue fixed into a certain threshold

Step 10 Select the nearest node to the final obtained 119892Bestparticle as the fittest position suiting enough to act as a sinknode for the current scenario

Although the PSO proved that it is one of the bestoptimization techniques to solve many problems it stillsuffers from the trapping within local optima specially in lowdimensional search spaceTherefore the proposed algorithmuses the Gaussian jump to escape from the local minima [23]within Step 9 For limited iterations when 119892Best value fixedinto a certain threshold each particle updates its positionby a Gaussian jump (shift) using (6) and then the algorithmreturns to start from Step 3 Consider

pos1015840119894= pos

119894+ gaussian() (6)

where pos1015840119894is the new position shift of particle 119901

119894and

gaussian() is a random number based on the Gaussiandistribution [23]

Table 1 The adjusted PSO parameters

Parameter ValueFitness function probability 1205721 04Fitness function probability 1205722 01Fitness function probability 1205723 05The inertia-weight 120596 08 [10]The acceleration constants (1198881 1198882) 2 [11]Random numbers (1199031 1199032) 05

Table 2 Atarraya simulation parameters

Parameter ValueDeployment area 600m lowast 600mNumber of nodes 100 200 300 400 500 600 700Sensor node model Mica MoteNode communication range 100mNode sensing range 20mNode location distribution UniformNode energy distribution UniformMax energy 2000 milliamperes-hour (mA-h )

4 Experiments and Performance Evaluation

Theproposed PSO-based algorithmwas coded and evaluatedusing a Java based simulation tool called Atarraya [24]WhileTable 1 lists the adjusted PSO parameters for the experimentsTable 2 shows a summary of the most important simulationparameters that were adjusted for the experimental scenariosThe nodes within the simulation are assumed to mimic thecharacteristics of Crossbowrsquos Mica Mote sensors with theenergy model defined in [25] The experiments that testthe whole networkrsquos lifetime use the dynamic global time-based topology recreation (DGTTRec) topology maintenanceprotocol which was proved as the best maintenance policyfor the A3 construction protocol [20]The adjusted triggeringcriterion for the construction of a new reduced topology iswhen the time threshold is exceeded which has been set to1000 seconds

The performance metric used within the experimentsis the number of active nodes provided by the topologyconstruction that guarantee coverage and the total networkrsquoslifetime The evaluation section is divided into two partsthe first part tests the impact of the proposed optimizationtechnique on the A3 topology as a construction protocolwhile the second part evaluates the performance of addinga maintenance policy to the already optimized topology Thepaper will refer to the original A3 topology control protocol(without any optimization feature) as the basic topologyprotocol (BTP) throughout the context

41 The Influences over the Topology Construction PhaseThis part tests the impacts of setting the sink node locationusing the proposed PSO-based algorithm on the topologyconstruction process Figure 3 shows the consumed time pereach topology construction scenario (the experiments didnot consider the sink node PSO-based selection time within


Table 3 The mean values of the number of active nodes scenarios

Number of nodes per each scenario 100 200 300 400 500 600 700Topology with PSO algorithm 348 4075 4675 438 495 505 5225The basic topology 3325 415 4775 494 5225 62 53

Table 4 The mean values of the topology construction time scenarios


30

45

60

100 200 300 400 500 600 700

Tim

e uni

ts

Number of nodes

Topology with PSO algorithmBasic topology protocol

Figure 3 Topology construction time

30

45

60

100 200 300 400 500 600 700

Num

ber o

f act

ive n

odes

Number of nodes


Figure 4 Number of active nodes

the scope of this paper) Although the results cannot be gen-eralized to different deployment scenarios the experimentsof uniform distribution of nodes within the deployment areashow that the sink node PSO-based localization algorithmneeds shorter time to construct a topology (the time preser-vation ranged from 10 to 15 of time units) than the BTP

Moreover the proposed algorithm utilizes a lesser num-ber of active nodes with an average of 11 of networkrsquosnodes for the topology construction compared to the BTPthat utilizes nearly 12 of active nodes as it is illustratedin Figure 4 This significant small reduction of active nodeswill assure the prolongation of the operational lifetime of

0500

10001500200025003000350040004500

100 200 300 400 500 600 700

Num

ber o

f top

olog

yre

cons

truc

tions

Number of nodes


Figure 5 Number of topology maintenance reconstructions

the sensor network since it will save more nodes for futuremaintenance phases This reduction cuts down the messagesrsquocomplexity between the nodes as well

Furthermore Tables 3 and 4 show some statistical anal-ysis for testing the performance of the proposed algorithmthrough a number of experiments conducted per each nodesrsquodeployment scenario (four experiments per each scenario)While Table 3 shows a reduction of number of active nodesachieved by the proposed algorithm that reached 613Table 4 defines a 563 decrease in the total mean timerequired to construct a topology by the proposed algorithm

42 The Influences over the Whole Networkrsquos PerformanceAs it is previously proved that the proposed optimizationalgorithm is an efficient technique that minimizes the num-ber of active nodes per topology construction it is alsoproved that the sink node localization within TC proto-cols is influencing the number of topology maintenancereconstruction executions which give more extensions tothe network operational lifetime Figure 5 shows that theproposed algorithm preserves the network health through anumber of topologymaintenance procedures with an averageof 6 compared to 48 for the topology control without anyoptimization feature

As a consequence of the optimization granted from boththe number of active nodes and the number of topologyreconstruction executions the sensor network will operate


100600

11001600210026003100360041004600

100 200 300 400 500 600 700

Tim

e uni

ts (th

ousa

nds)

Number of nodes


Figure 6 Operational networkrsquos lifetime

for a lifetime long enough to fulfill the application require-ments Figure 6 demonstrates a chart that shows the networkrsquoslifetime changes over different networkrsquos capacitiesThe chartshows that the proposed PSO-based algorithm had clearadvantagewithin a range from 300 to 600 of networkrsquos capaci-tiesThe study of the convergence between our algorithm andthe basic topology protocol in high networkrsquos capacities (over700 nodes) is currently under investigation

5 Conclusions

The wireless sensor networks energy model is affected bynodesrsquo distribution and the sink nodersquos location In thispaper a Particle Swarm Optimization approach has beenused to select the optimal location for the sink node withina topology control protocol The proposed fitness functionfor that topology control included the number of neighborstheir residual energy and the distance to the center of thedeployment area The simulation results confirm that theproposed optimization approach improves the performanceof both phases of the topology control protocol the topologyconstruction phase and the topology maintenance phaseSince the topology construction time is shortened by a rangeof 10 to 15 along with the number of active nodes thosepledge the coverage and networkrsquos connectivity thus theapproach provides a layout for a prolongation of the networkrsquoslifetime

Conflict of Interests

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

Acknowledgments

This paper has been elaborated in the framework of theproject New Creative Teams in Priorities of ScientificResearch Reg no CZ1072300300055 supported byOperational Programme Education for Competitiveness andcofinanced by the European Social Fund and the state budget

of the Czech Republic and supported by the IT4InnovationsCentre of Excellence project (CZ1051100020070) fundedby the European Regional Development Fund and thenational budget of the Czech Republic via the Research andDevelopment for Innovations Operational Programme

References

[1] M P Durisic Z Tafa G Dimic and V Milutinovic ldquoAsurvey of military applications of wireless sensor networksrdquo inProceedings of the 1st Mediterranean Conference on EmbeddedComputing (MECO rsquo12) pp 196ndash199 Montenegro Bar June2012

[2] N El-Bendary M M M Fouad R A Ramadan S Banerjeeand A E Hassanien ldquoSmart environmental monitoring usingwireless sensor networksrdquo in Wireless Sensor Networks FromTheory to Applications CRC Press 2013

[3] AMohamedMMM Fouad E Elhariri et al ldquoRoadMonitoran intelligent road surface condition monitoring systemrdquo inIntelligent Systems pp 377ndash387 Springer Berlin Germany 2015

[4] M M M Fouad N El-Bendary R A Ramadan and A EHassanien ldquoWireless sensor Networks a medical perspectiverdquoinWireless Sensor Networks FromTheory to Applications CRCPress 2013

[5] F Chen and R Li ldquoSink node placement strategies for wirelesssensor networksrdquo Wireless Personal Communications vol 68no 2 pp 303ndash319 2013

[6] A Efrat S Har-Peled and J S Mitchell ldquoApproximation algo-rithms for two optimal location problems in sensor networksrdquoin Proceedings of the 2nd International Conference on BroadbandNetworks (BroadNets rsquo05) IEEE October 2005

[7] J Luo and J-P Hubaux ldquoJoint mobility and routing for lifetimeelongation in wireless sensor networksrdquo in Proceedings ofthe 24th Annual Joint Conference of the IEEE Computer andCommunications Societies (INFOCOM rsquo05) vol 3 pp 1735ndash1746 IEEE March 2005

[8] S L Hakimi ldquop-median theorems for competitive locationsrdquoAnnals of Operations Research vol 6 no 4 pp 75ndash98 1986

[9] W Y Poe and J B Schmitt ldquoMinimizing the maximum delay inwireless sensor networks by intelligent sink placementrdquo TechRep 36207 Distributed Computer Systems Lab University ofKaiserslautern Kaiserslautern Germany 2007

[10] A E Charalampakis and C K Dimou ldquoIdentification of Bouc-Wen hysteretic systems using particle swarm optimizationrdquoComputers and Structures vol 88 no 21-22 pp 1197ndash1205 2010

[11] Y Shi and R C Eberhart ldquoEmpirical study of particle swarmoptimizationrdquo in Proceedings of the 1999 Congress on Evolution-aryComputation (CEC rsquo99) vol 3 IEEEWashingtonDCUSA1999

[12] Y T Hou Y Shi H D Sherali and S F Midkiff ldquoOn energyprovisioning and relay node placement for wireless sensornetworksrdquo IEEE Transactions on Wireless Communications vol4 no 5 pp 2579ndash2590 2005

[13] E Guney N Aras I K Altnel and C Ersoy ldquoEfficientsolution techniques for the integrated coverage sink locationand routing problem in wireless sensor networksrdquo Computersand Operations Research vol 39 no 7 pp 1530ndash1539 2012

[14] R K Yadav V Kumar and R Kumar ldquoA discrete particleswarm optimization based clustering algorithm for wirelesssensor networksrdquo in Emerging ICT for Bridging the FuturemdashProceedings of the 49th Annual Convention of the Computer


Society of India CSI Volume 2 vol 338 of Advances in IntelligentSystems and Computing pp 137ndash144 Springer 2015

[15] Y Z Pan F Liu and N Zhang ldquoA WSNs routing protocolbased on clustering and improved ACO for the SDPMrdquo inRemote Sensing and Smart City vol 64 pp 137ndash147 WIT PressSouthampton UK 2015

[16] P Santi ldquoTopology control in wireless ad hoc and sensornetworksrdquo ACM Computing Surveys vol 37 no 2 pp 164ndash1942005

[17] P M Wightman andM A Labrador ldquoA3 a topology construc-tion algorithm for wireless sensor networksrdquo in Proceedings ofthe IEEE Global Telecommunications Conference (GLOBECOMrsquo08) pp 1ndash6 New Orleans La USA December 2008

[18] M M M Fouad A R Dawood and M-S M Mostafa ldquoStudyof the effects of pairwise key pre-distribution scheme on theperformance of a topology control protocolrdquo in Proceedings ofthe 7th IEEE International Conference onDistributed Computingin Sensor Systems (DCOSS rsquo11) pp 1ndash5 IEEE Barcelona SpainJune 2011

[19] M M M Fouad M-S M Mostafa and A R Dawood ldquoSOPKsecond opportunity pairwise key scheme for topology controlprotocolsrdquo in Proceedings of the 3rd International Conference onIntelligent SystemsModelling and Simulation (ISMS rsquo12) pp 632ndash638 IEEE Kota Kinabalu Malaysia February 2012

[20] P M Wightman and M A Labrador ldquoA3Cov a new topologyconstruction protocol for connected area coverage in WSNrdquoin Proceedings of the Wireless Communications and NetworkingConference (IEEE WCNC rsquo11) pp 522ndash527 Quintana-RooMexico 2011

[21] A Biswas and B Biswas ldquoSwarm intelligence techniques andtheir adaptive nature with applicationsrdquo in Complex SystemModelling and Control Through Intelligent Soft Computationsvol 319 of Studies in Fuzziness and Soft Computing pp 253ndash273Springer 2015

[22] B Singh and D K Lobiyal ldquoA novel energy-aware cluster headselection based on particle swarm optimization for wirelesssensor networksrdquo Human-Centric Computing and InformationSciences vol 2 no 1 article 13 18 pages 2012

[23] R A Krohling ldquoGaussian particle Swarm with jumpsrdquo inProceedings of the IEEE Congress on Evolutionary Computation(CEC rsquo05) vol 2 pp 1226ndash1231 IEEE September 2005

[24] M A Labrador and P M Wightman Topology Control inWireless Sensor Networksmdashwith a Companion Simulation Toolfor Teaching and Research Springer Berlin Germany 2009

[25] Y Cai M Li W Shu and M Y Wu ldquoAcos an area-basedcollaborative sleeping protocol for wireless sensor networksrdquoAdHocamp SensorWireless Networks vol 3 no 1 pp 77ndash97 2007

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks



location should be accurately selected so other nodes willnot use much power to deliver their data to that locationAs a consequence the networkrsquos lifetime will be maximizedFurthermore any proposed optimization should not trade offthe area coverage or networkrsquos connectivity characteristics

A great number of researches assumed that within asingle network the sink node should always be deployedwithin the center point of the area of interest [5] for examplethe work done in [6 7] which applied the P-Median Problem(PMP) model which is a well-known NP-hard problemdefined by Hakimi [8] It defined the sink node locationwhere the total weighted distance of reaching all nodesis minimized On the other hand [9] gives a comparisonbetween a number of sink placement strategies Among theother discussed strategies is the Geographic Sink Placement(GSP) strategy in which the sink is placed at the centerof gravity of a sector of a circle Although it gives fastsuitable solutions it cannot guarantee the optimal sink nodelocation Also the authors introduced another heuristic basedapproach entitled Genetic Algorithm-Based Sink Placement(GASP) The GASP is suitable to produce nearly optimalsolution for large-scale networks

Chen and Li [5] investigated two proposed sink nodeplacement strategies the energy-oriented and lifetime-oriented strategies in both the single-hop and multiple-hopWSNs respectively The two strategies adopted a routing-cost based Ant routing algorithm According to their con-clusion the lifetime-oriented strategy had outperformed theenergy-oriented strategy in terms of networkrsquos lifetime Otherresearches used the integer linear programming to find eitherthe optimal position for the sink node or relay nodes forexample the work of Hou et al [12] that developed an effi-cient polynomial-time heuristic algorithm (SPINDS) whichattempted to increase the network lifetime by iterativelymoving a relay node to another enhanced location Guneyet al [13] developed two mixed-integer linear programmingformulations as well those had easily computed good feasiblesolutions for the sink location and efficient routingThey useda Tabu Search heuristic to identify the best sensor locationsthat could guarantee the total networkrsquos coverage

Selecting best locations for cluster headswithin clusteringbased routing algorithms is more similar to selecting alocation for a sink node within a wireless sensor networkUsually both are sharing the same aim which is trying toreduce the required transmission energy used by sensornodes within a network

Yadav et al [14] suggest a PSO-based solution to theoptimal clustering problem through the use of residualenergy and transmission distance of sensor nodes Theyconsider a new operator to be applied inside their algorithmthat checks the validity of reached location of the head ofthe cluster within the current iteration of the PSO algorithmIf the location is not valid then the algorithm returns tothe nearest valid node location that has the highest residualenergy to become the current head of the cluster Althoughthe results are promising the main limitation within thealgorithm is the need for a centralized authority node

Moreover there are other nature inspired algorithmsthat were used to optimally formulate and optimize routing

through clusters Pan et al [15] propose an algorithm basedon Uneven Clustering Multihop Algorithm (UCMA) andan Improved Ant Colony Optimization (IACO) algorithmThe first former algorithm is used for grouping sensor nodesinto unequal clusters along with selecting their cluster headThe decision of the UCMA is based on the nodesrsquo locationsand their residual energies On the other hand the IACO isapplied for the routing discovery between the head of thecluster and the sink node The performance evaluations oftheir proposed algorithms show that the application of theIACO algorithm had accelerated the iterative conversion rateand efficiently reduces the energy consumption rate

This paper proposes an energy-aware sink node localiza-tion algorithm for a topology control protocol using a ParticleSwarm Optimization technique The simulation results showthat the proposed energy-aware algorithm minimizes thenumbers of active nodes and subsequently increases thewhole networkrsquos lifetime

The remainder of this paper is organized as followsSection 2 gives a brief overview of topology control protocolsIn Section 3 is the proposed optimization technique Thesimulation results and a number of performance evaluationsare given in Section 4 Finally Section 5 points out theconcluding remarks

2 Topology Control Protocols

A large number of protocols and algorithms have beenproposed to overcome the constrained resources of thewireless sensor networks Many of these protocols andalgorithms were developed to extend the lifetime of thewireless network through wisely managing the use of theirconstrained resources Topology control (TC) protocols are ofthe algorithms designed to construct minimized topologiesThese topologies proved their efficiency for both energyconsumption and radio interferences reductions

A topology control protocol is an iterative process thatdynamically reduces the initial topology of a wireless sensornetwork through controlling nodesrsquo transmission rangesThemain advantage of the TC protocols is the assets saving of thewireless sensor networks such as the networkrsquos connectivityand coverage [16 17] Figure 1 illustrates with brief descrip-tions the topology control protocolrsquos phases

Although the TC protocols reduce energy consumptionratio they still suffer from a number of drawbacks for exam-ple securing topology control protocols could negativelyaffect the performance of such protocols [18] therefore thecurrent existing security schemes should be adopted to suitTC protocols [19] On the other hand most of the designedTC protocols (such as A3 [17] and A3Cov [20]) supposethat even if the deployment is randomized the sink nodelocation should be fixed within the center of the deploymentarea which is not the optimal case in real-life scenariosTherefore this paper provides a contribution that is of worthin enhancing the lifetime and the performance of the WSNsIt proposes a PSO-based algorithm to find the adequateposition for the sink node within a topology control protocol(the A3 protocol)




establishment phase












119894 119890119894 119899119894) where pos



119894defines



gBest


Dynamic particle

s1

s2 s

3

s4

s5

s6

sM

p1

p2

p3

pN


|119875| le |119878| Each 119901119894= (V119894 pos119894 119901Best

119894 119892Best) where V

119894is a


119894is another








to zero




119894

119891 (119901119894) = 12057211003816100381610038161003816119873 (119901119894)1003816100381610038161003816 + 1205722 sum119901isin119873(119901119894)

119901119890 + 1205723119889119901 (1)




119894





119901Best119894=

119901119894119891 (119901119894) gt 119891 (119901Best

119894)

119901Best119894

otherwise(2)





V119894 (119905 + 1) = 120596V119894 (119905) + 11988811199031 (119901Best119894 minus V119894 (119905))

+ 11988821199032 (119892Bestminus V119894 (119905)) (4)




pos119894(119905 + 1) = pos

119894(119905) + V119894 (119905 + 1) (5)




pos1015840119894= pos

119894+ gaussian() (6)


119894and















30

45

60

100 200 300 400 500 600 700

Tim

e uni

ts

Number of nodes



30

45

60

100 200 300 400 500 600 700

Num

ber o

f act

ive n

odes

Number of nodes





0500

10001500200025003000350040004500

100 200 300 400 500 600 700

Num

ber o

f top

olog

yre

cons

truc

tions

Number of nodes








100600

11001600210026003100360041004600

100 200 300 400 500 600 700

Tim

e uni

ts (th

ousa

nds)

Number of nodes




5 Conclusions




Acknowledgments



References






























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation












establishment phase












119894 119890119894 119899119894) where pos



119894defines



gBest


Dynamic particle

s1

s2 s

3

s4

s5

s6

sM

p1

p2

p3

pN


|119875| le |119878| Each 119901119894= (V119894 pos119894 119901Best

119894 119892Best) where V

119894is a


119894is another








to zero




119894

119891 (119901119894) = 12057211003816100381610038161003816119873 (119901119894)1003816100381610038161003816 + 1205722 sum119901isin119873(119901119894)

119901119890 + 1205723119889119901 (1)




119894





119901Best119894=

119901119894119891 (119901119894) gt 119891 (119901Best

119894)

119901Best119894

otherwise(2)





V119894 (119905 + 1) = 120596V119894 (119905) + 11988811199031 (119901Best119894 minus V119894 (119905))

+ 11988821199032 (119892Bestminus V119894 (119905)) (4)




pos119894(119905 + 1) = pos

119894(119905) + V119894 (119905 + 1) (5)




pos1015840119894= pos

119894+ gaussian() (6)


119894and















30

45

60

100 200 300 400 500 600 700

Tim

e uni

ts

Number of nodes



30

45

60

100 200 300 400 500 600 700

Num

ber o

f act

ive n

odes

Number of nodes





0500

10001500200025003000350040004500

100 200 300 400 500 600 700

Num

ber o

f top

olog

yre

cons

truc

tions

Number of nodes








100600

11001600210026003100360041004600

100 200 300 400 500 600 700

Tim

e uni

ts (th

ousa

nds)

Number of nodes




5 Conclusions




Acknowledgments



References






























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation












119894





119901Best119894=

119901119894119891 (119901119894) gt 119891 (119901Best

119894)

119901Best119894

otherwise(2)





V119894 (119905 + 1) = 120596V119894 (119905) + 11988811199031 (119901Best119894 minus V119894 (119905))

+ 11988821199032 (119892Bestminus V119894 (119905)) (4)




pos119894(119905 + 1) = pos

119894(119905) + V119894 (119905 + 1) (5)




pos1015840119894= pos

119894+ gaussian() (6)


119894and















30

45

60

100 200 300 400 500 600 700

Tim

e uni

ts

Number of nodes



30

45

60

100 200 300 400 500 600 700

Num

ber o

f act

ive n

odes

Number of nodes





0500

10001500200025003000350040004500

100 200 300 400 500 600 700

Num

ber o

f top

olog

yre

cons

truc

tions

Number of nodes








100600

11001600210026003100360041004600

100 200 300 400 500 600 700

Tim

e uni

ts (th

ousa

nds)

Number of nodes




5 Conclusions




Acknowledgments



References






























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation














30

45

60

100 200 300 400 500 600 700

Tim

e uni

ts

Number of nodes



30

45

60

100 200 300 400 500 600 700

Num

ber o

f act

ive n

odes

Number of nodes





0500

10001500200025003000350040004500

100 200 300 400 500 600 700

Num

ber o

f top

olog

yre

cons

truc

tions

Number of nodes








100600

11001600210026003100360041004600

100 200 300 400 500 600 700

Tim

e uni

ts (th

ousa

nds)

Number of nodes




5 Conclusions




Acknowledgments



References






























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










100600

11001600210026003100360041004600

100 200 300 400 500 600 700

Tim

e uni

ts (th

ousa

nds)

Number of nodes




5 Conclusions




Acknowledgments



References






























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









research article energy-aware sink node...

Documents