View of STATE OF ART ON CLUSTERING APPROACHES OF WIRELESS SENSOR NETWORK

(1)

Vol.03, Issue 09, Conference (IC-RASEM) Special Issue 01, September 2018 Available Online: www.ajeee.co.in/index.php/AJEEE

STATE OF ART ON CLUSTERING APPROACHES OF WIRELESS SENSOR NETWORK

Satyaprakash Shikari Research Scholar

Elcetronics & Communication, SIRT, SAGE UNIVERSITY, Indore satyshikari@gmail.com

Prof. Pallavi Pahadiya Asst. Professor

Electronics & Communication Department, SIRT, SAGE UNIVERSITY, Indore Pallavi_19_2000@yahoo.com

Abstract –Wireless sensor network consists of large number of sensor nodes or devices that are distributed spatially and communicates using radio signals. These sensor nodes are usually placed in the geographical area to gather information from the surroundings. Once the sensor nodes are deployed they are typically unapproachable to the humans. Sensor nodes in WSN sense the data from the environment and send it to the BS or sink node where this data is analysed.

Sensor nodes in WSN are powered with a battery. Sensor nodes consume the battery power mainly in the tasks like data transmission, data reception and sensing. Sometimes it is impractical to replace a battery in WSN because humans can’t reach. Therefore once energy or computational resources are consumed, immediate recovery of these resources is a complex task so it is necessary to make use of battery power efficiently to increase the lifetime of the sensor nodes that will also increase the lifetime of the whole network. Based on the need to optimize energy consumption, new research topics appear, such as energy collection and consumption optimization. This paper is framed within these fields, trying to study the state of the art of the theoretical models on clustering approaches of wireless sensor network.

Keywords – ADC, CH, GA, WSN.

I. INTRODUCTION

WSN consist of large number of sensor nodes that are deployed in the environment and are powered by battery and replacing the battery of each and every node in the sensor network is impractical so it is necessary to make use of this limited energy efficiently therefore an energy efficient algorithm has to be designed [1]. The nodes in the WSN are distributed spatially and to send the sensed data to the base station or sink node needs multi-hop communication when the base station or sink is not in range so the number of intermediate nodes are required to send this sensed data to the desired destination which consumes more energy of the network because all the intermediate nodes forward the data coming from their neighbour and to do this job they consume energy and this motivate us to reduce the number of hops and intermediate nodes taking part in transmission of data by using clustering algorithm in which the nodes are grouped into clusters and each node in cluster send data only to their concerned cluster head which is then aggregate all the data

coming from the member nodes and send it to the sink or base station [2-3].

II. ARCHITECTURE OF SENSOR NODE IN

WSN In WSN the main components of sensor nodes are sensors, microcontroller, transceiver, power source and ADC as shown in Figure 1.

Sensors: sensor is a device which generates a measurable response to a change in physical conditions like temperature, sound and pressure etc.

Sensors produce a continuous analog signal which is then digitized by analog- to-digital converter and after digitization the signal is sent to the microcontroller for processing. A sensor node in WSN has one or more than one sensor equipped in it [4].

Analog-to-Digital Converter: Analog-to- digital converter (ADC) is used to convert analog signal produced by the sensors into a digital form so it can be easily processed by the microcontroller.

Generally the output of the ADC is two’s compliment binary digit which is

(2)

proportional to the input. The output of the ADC is in digital form so it has only two states 0 and 1. Sometimes to convert the analog signal into a current or voltage value transducers are used. ADC performs to operations on the input signal first is Sampling and hold and the second is Quantizing and encoding [5-6].

Microcontroller: A microcontroller is used to control all the processes and functions performed by the sensor node components. The main reason of using microcontroller is its flexibility to connect other devices and low cost although there are many alternatives which are used in place of microcontroller for example a general purpose microprocessor, Digital Signal Processor, Field Programmable Gate Array (FPGA) and many more.

Microcontrollers consume less energy as compared to the other alternatives [7].

Transceiver: A transceiver is used to transmit and receive data from the base station or from the other nodes or devices.

In WSN sensor nodes uses ISM band.

WSN uses radio frequency for wireless transmission. It uses license free frequencies i.e. 433, 868, 915 MHz and 2.4 GHz in which 2.4 GHz is most popular. The transceiver operates in different states which are transmitting state, receiving state, idle state and sleep.

In transmission mode the transceiver is used as transmitter which transmit the data to the sink or base station while in receive mode the transceiver acts as receiver which receives the data from the other nodes or base station. In idle mode the transceiver is not transmitting or receiving anything but in this mode it is ready to receive although consumes energy which is same as the energy consumed in receiving mode so it is better to put the transceiver in sleep mode when not in use in sleep mode important parts of transceiver are switched off to reduce energy consumption although it requires some amount of energy to change the state of transceiver from sleeping mode to transmitting mode. In sleeping mode the transceiver is not able to receive anything immediately.

Power Source: In WSN a sensor node is placed somewhere in the environment where it is out of reach for humans hence to provide power to sensor nodes a battery is used due to this reason the sensor nodes have limited amount of power and it is impractical to change battery of the sensor nodes so to reduce power consumption of the sensor node some energy efficient algorithms are used.

Battery used in the sensor may be chargeable or non-chargeable although many sensor nodes use a solar energy to charge the batteries [8].

Figure 1: Sensor node architecture III. LITERATURE REVIEW

N R Wankhade and D N Choudhari, in their paper “Advanced Energy Efficient Routing Protocol for Clustered Wireless

Sensor Network: Survey” [9], Due to the attributes that separate sensor organize from contemporary correspondence and wireless Ad-hoc Network the directing in Power Source

External Memory

Sensor Analog-to-

Digital Converter Microcontroller Transceiver

(3)

sensor arrange is a testing one. To send the huge number of sensor nodes it isn't conceivable to construct a worldwide tending to conspire so the great IP-based conventions can't be utilized for sensor arrange. Furthermore, to a particular sink or goal node the sensor organize applications require detected information spill out of numerous locales or sources which is in opposition to run of the mill correspondence arrange. Thirdly, as numerous sensors can create comparable information inside close to a wonder the produced information movement contains excess so to enhance vitality and transmission capacity utilize steering conventions must utilize it. Fourthly, sensor nodes require legitimate asset administration since they are very obliged in regards to transmission control, process limit, on-board vitality, and capacity.

You-Chiun Wang, Wen-Chih Peng and Yu-Chee Tseng, in their paper

"Vitality Balanced Dispatch of Mobile Sensors in a Hybrid Wireless Sensor Network" [10], considered a hybrid wireless sensor coordinate with static and versatile nodes. Checking the environment and announcing the occasions happening in the focused on region are finished by Static sensors. The general life expectancy is amplified in light of the fact that they plan the versatile sensors voyaging ways in a vitality adjusted way and demonstrates that it has been a NP-finish issue. To plan versatile sensors voyaging ways they proposed an incorporated and a circulated heuristics which permits subjective number of portable sensors and occasion areas in each round having a vitality adjusted idea as a top priority.

The incorporated heuristic attempts to limit the vitality required in moving portable sensors while keeping their vitality utilization adjusted.

Gerard Chalhoub and Michel Misson, in their paper "Group tree Based Energy Efficient Protocol for Wireless Sensor Networks" [11], proposed a period division approach which enhances the general system execution, spares vitality and empowers nature of administration as far as ensured access to the medium.

By synchronizing nodes action utilizing tree based topology the time division is accomplished. Crash free signal proliferation along the group tree is

ensured by synchronization period.

Keeping in mind the end goal to enhance the vitality productivity of the system and its execution they propose an information gathering period. At long last, to enhance promote progressively the system execution and assurance and end-to-end defer they include transfer time interims between facilitators.

Concerning a cluster tree arrangement the general evaluated vitality utilization is diminished in their outcomes likewise there is increment of 20% to 40%

in the got outlines and by and large the normal number of impacts is partitioned by 2.

Rathna R. and Sivasubramanian A., in their paper “Improving Energy Efficiency in Wireless Sensor Networks through Scheduling and Routing” [12], they proposed an ecological monitoring use of wireless sensor network. The more consideration is given to the two zones i.e.

routing and clustering. Their work is an endeavor to lessen the power utilization of the sensor nodes, by focusing on the radio. The Cluster based rest/wake-up booking strategy is turns out to be productive subsequent to testing in a reenacted WSN. For reproduction they utilize Network Simulator (NS2).

Minimum Spanning Tree (MST) and Shortest Hop path Tree (SPT) shapes the tree structures and by applying clustering systems to these structures WSN is mimicked. In MST the bunching of nodes gives best outcomes. The scheduling algorithm proposed in it depends on TDMA totally. Diminishing various circumstances a node needs to wake up amid a schedule opening to be in dynamic mode helps in decreasing vitality utilization. Their primary design is to utilize vitality proficiently.

Mohammad Soleimani,

Mohammad Ghasemzadeh and Mehdi Agha Sarram, in their paper “A New Cluster Based routing Protocol for Prolonging Network Lifetime in Wireless Sensor Network” [13], introduced a routing protocol in light of group with considering the low energy utilization in wireless sensor network. The energy stack among the sensor nodes is adjusted by this convention. The nodes which are near each other have more cover and cause a misuse of energy by creating redundant information since they sense similar information from condition. In the

(4)

proposed convention, a specific number of nodes are indicated in each round; the nodes which have no less than one neighbouring node at a separation not as much as the limit. At that point, to rest mode the nodes with their neighbours have been picked. Likewise, by incorporating the separation of nodes from the base station the energy unevenness among sensor nodes is diminished. This technique increment the lifetime of the system and can counteract squandering of energy.

Sanjay Waware, Nisha Sarwade and Pallavi Gangurde, in their paper “A Review of Power Efficient Hierarchical Routing Protocols in Wireless Sensor Networks” [14], introduced a portion of the power-efficient hierarchical routing protocols for wireless sensor network. In a various leveled engineering, to process and send the data the nodes with higher vitality can be utilized while to play out the detecting assignment in the focused on region the nodes with low vitality can be utilized. The general framework adaptability, vitality effectiveness and lifetime of the system will be enhanced by making clusters and allotting unique undertakings to cluster heads. To diminish the quantity of transmitted messages to the BS information collection and information combination has been done to decrease the vitality utilization inside the clusterin progressive steering having two layers in which one is utilized to choose the cluster head and other is utilized for routing.

Sunita Rani and Tarun Gulati, in their paper “An Improved PEGASIS Protocol to Enhance Energy Utilization in Wireless Sensor Network” [15], In WSN, each sensor has certain energy to screen the physical or natural condition, for example, sound, vibration, temperature at various areas. The wireless sensor nodes are conveyed into the system. The conventions assume a critical part in limiting the postponement while offering high energy productivity and increment organize lifetime. One of such convention is PEGASIS which depends on the chain structure, each chain have one group head, The vast measure of energy is devoured by bunch heads and the seasons of each round expanding. The benefit of PEGASIS is node sends information to its nearest neighbour, it increases the lifetime of the system by

sparing the battery energy for WSN. The proposed work is to choose the following neighbouring node dependably.

Samia A. Ali and Shreen k. Refaay, in their paper “Chain-Chain Based Routing Protocol” [16], presented a chain- chain based energy efficient routing protocol. It accomplishes both diminished energy utilization and decreased postponement. Utilizing covetous calculation the CCBRP convention partitions a WSN into various chains and keeps running in two stages. In the principal stage, the information is transmitted to the chain pioneer in each chain by the sensor nodes in parallel. In the second stage, all chain pioneer nodes frame a chain and information is sent to the pioneer node picked arbitrarily by the chain pioneer nodes than information combination is performed by the haphazardly picked node and after this intertwined information is sent to the BS or sink node. The deferral of the proposed CCBRP is same as LEACH and CCM however 75% not as much as that of PEGASIS.

Lathies Bhasker, in his paper

“Genetically Derived Secure Custer-based Data Aggregation in Wireless Sensor Networks” [17], presented a genetically optimized cluster head election in WSN which Initially, in light of the node availability the CHs were chosen, which goes about as an information aggregator.

At that point, utilizing the GA the clustering procedure was executed. An encryption strategy is utilized that offers realness, privacy and uprightness when a cluster needs to transmit the information to aggregator or a clusterhead. The proposed method diminishes the energy utilization, guaranteed information security and decreased the transmission overhead.

Stephanie Lindsey and Cauligi Raghavendra, in their paper “Data Gathering Algorithms in Sensor Networks Using Energy Metrics” [18], presented three new protocols for wireless sensor network. PEGASIS is one of these protocols it is a greedy chain protocol that is near optimal for sensor network having data-gathering problems. By minimizing the distance, eliminating the overhead of dynamic cluster formation, limiting the number of transmissions and receptions among all the nodes and using only single transmission to the BS per round

(5)

PEGASIS out-performs LEACH. The simulation shows that when 1%, 25%, 50%, and 100% of nodes die for different topologies and network sizes the PEGASIS performs better than LEACH by about 100% to 200%. Compared to LEACH-C, which doesn’t have the cluster formation over-head in each round, the improvements will be slightly lesser. As the size of the network increases PEGASIS shows an even further improvements.

Ouadoudi Zytoune and Driss Aboutajdine, in their paper “A Lifetime Extension Protocol for Data Gathering in Wireless Sensor Networks” [19], presented a new algorithm in WSN for gathering.

This is based on chain formation using greedy algorithm. Its primary focus is to distribute the energy load equitably over the whole network nodes. To avoid dying of nodes quickly, the role of leader node is better distributed over nodes based on their energies required to transmit to the sink. Thus, all the network nodes would have the same lifetime and then as result, the lifetime of the network will increase.

In the proposed algorithm the transmission energy over the whole network nodes is balanced correctly which increase the network lifetime. The result of simulation shows that as compared to the well-known protocol for chaining in wireless sensor network this technique provides improvement.

Razieh Sheikhpour, Sam Jabbehdari and Ahmad khademzadeh, in their paper “A Cluster-Chain based Routing Protocol for Balancing Energy Consumption in Wireless Sensor Networks” [20], Presents an Energy Efficient Cluster-Chain based Protocol (ECCP) for wireless sensor network that aims at maximizing stability period, network lifetime, and balancing the energy consumption among sensor nodes.

Using multiple metrics sensor nodes are organized into cluster in ECCP and within the cluster a chain is constructed among the sensor nodes so they can receive data from previous neighbours and transmits to a next neighbour. In ECCP chain based data transmission mechanism is also used to send data packets to base station from the cluster heads. ECCP offers an advantage of small transmit distance by chaining the nodes in each cluster and by conserving their energy ECCP help them to be operational for longer time period.

The result of simulation shows that in

term of network lifetime, stability and instability period and balancing the energy consumption among the sensor nodes ECCP is more efficient as compared to CBRP, LEACH and PEGASIS.

Cong Wang and Cuirong Wang, in their paper “A Concentric Data Aggregation Model in Wireless Sensor Network” [21], proposed a concentric data aggregation model. The main purpose of this model is to consider the base station location, and divide the whole WSN into several hierarchical and concentric zones refer to the base station’s location, each zone is also divided into some areas and nodes are organized as PEGASIS in every area. Sensor nodes collects the data which goes through proper areas belong to different level zones towards the base station and in each hop that data is aggregated. The aggregated data is transmitted to the base station by the head on the last area which is the nearest node of the base station. The simulation results shows that in transmission delay and energy efficiency it performs better it also avoids data, gathering latency and reduce energy consumption.

IV. CONCLUSION

With the technological advancements wireless sensor networks found enormous applications in various fields such as rescue operations, military fields etc.

these networks are usually deployed in hostile areas or adverse environmental conditions where human involvement is not possible. Furthermore these sensor nodes are usually operated by battery which is normally not easy to replace.

Hence energy efficiency is of prime need in wireless sensor networks. Clustering is the key technique for diminishing energy dissipation in the network and enhancing the network stability. Nodes location may be far away from BS so direct communication is not feasible due to limited battery as this requires high energy. Many clustering protocols are designed in this regard. LEACH is first basic algorithm which relies on this procedure of clusterhead election. Such clustering algorithm relies on a random number generated by each sensor node in every round for the process of clusterhead election. If the generated random number is less than predefined threshold value then the current node will be elected as clusterhead node for current round.

(6)

Vol.03, Issue 09, Conference (IC-RASEM) Special Issue 01, September 2018 Available Online: www.ajeee.co.in/index.php/AJEEE REFERENCE

1. M.A. Matin and M.M. Islam,“Over view of Wireless SensorNetwork”, In Tech, pp. 03- 24, 2012.

2. Harshwinder Singh and Navpreet Kaur,“Energy Efficiency Techniques for Wire less Sensor Networks: A Review”, International Journal of Innovative Researchin Computer and Communication Engineering, Volume2,Issue5,pp. 4138-4142, May2014.

3. Rajesh Chaudhary and Sonia Vatta,

“Review Paperon Energy-Efficient Protocolsin Wireless SensorN etworks”, International organization of Scientific Research Journal of Engineering (IOSRJEN), Volume 4, Issue2, pp.01-07, February2014.

4. SubhajitPal,DebnathBhattacharyya,Geeta mS.TomarandTai-

hoonKim,“WirelessSensorNetworksandits RoutingProtocols:AComparativeStudy”,Int ernationalConferenceonComputationalInt elligenceandCommunicationNetworks IEEE,pp.314-319, November 2010.

5. LathiesBhasker,“GeneticallyDerivedSecur eCluster-

BasedDataAggregationinWireless SensorNetworks”,IETInformationSecurity, Volume8,Issue 1,pp.01–07,May2013.

6. SanjeevKumarGupta,NeerajjainandPoona mSinha,“ClusteringProtocolsinWirelessSe nsorNetworks:ASurvey”,InternationalJour nalofAppliedInformation

Systems(IJAIS),Volume5, No. 2,pp.41-50, January2013.

7. MohiniKumrawatandManojDhawan,“Surv eyonClusteringAlgorithmsofWirelessSenso rNetwork”,InternationalJournalofComput erScienceandInformation Technologies (IJCSIT),Volume6,No.3,pp. 2046-2049, 2015.

8. NoritakaShigei,HiromiMiyajima,HirokiMor ishitaandMichiharuMaeda,“Centralizedan dDistributedClusteringMethodsforEnergy EfficientWirelessSensorNetworks”,Proceed ingsoftheInternationalMultiConferenceof

Engineers and Computer

Scientists,HongKong, Volume1, March 2009.

9. N.R.WankhadeandD.N.Choudhari,“Advan cedEnergyEfficientRoutingProtocolforClus teredWirelessSensorNetwork:Survey”,Inte rnationalJournalofEmergingTechnologiesi nComputationalandAppliedSciences(IJET CAS),Volume 9,No.3,pp. 237-242, June- August 2014.

10. You-ChiunWang,Wen-ChihPengandYu- CheeTseng,“Energy-

BalancedDispatchofMobileSensorsinaHyb ridWirelessSensorNetwork”,IEEETransact ionsonParallelandDistributedSystems,Vol ume21,No.12,December 2010.

11. GerardChalhoubandMichelMisson,“Clust er-

treeBasedEnergyEfficientProtocolforWirel essSensorNetworks”,InternationalConfere nceonNetworking, Sensing andControl (ICNSC)IEEE,pp. 664-669, April2010.

12. RathnaR.andSivasubramanianA.,“Improvi ngEnergyEfficiencyinWirelessSensorNetw orksThroughSchedulingandRouting”,Inter nationalJournalOfAdvancedSmart

SensorNetwork Systems(IJASSN),Volume 2,No.1,pp.21-27,January2012.

13. MohammadSoleimani,MohammadGhase mzadehandMehdiAghaSarram,“ANewClus terBasedroutingProtocolforProlongingNet workLifetimeinWirelessSensorNetworks”, Middle-

EastJournalofScientificResearch,Volume7 , No. 6,pp. 884-890, 2011.

14. SanjayWaware,NishaSarwadeandPallaviG angurde,“AReviewofPowerEfficientHierarc hicalRoutingProtocolsinWirelessSensorNe tworks”,InternationalJournalofEngineerin gResearchandApplications(IJERA),Volume 2,Issue 2,pp.1096-1102, March-April 2012.

15. SunitaRaniandTarunGulati,“AnImproved PEGASISProtocoltoEnhanceEnergyUtiliza

tion in

WirelessSensorNetwork”,InternationalJou rnalof

AdvancedResearchinComputerSciencean dSoftwareEngineering,Volume2,Issue 11,pp.448-253,November2012.

16. SamiaA.AliandShreenK.Refaay,“Chain- ChainBasedRoutingProtocol”,Internationa lJournalofComputerScienceIssues(IJCSI), Volume8,Issue3,No. 2,pp. 105-112, May2011.

17. LathiesBhasker,“GeneticallyDerivedSecur eCluster-

BasedDataAggregationinWireless SensorNetworks”,IETInformationSecurity, Volume8,Issue 1,pp.01–07,May2013.

18. StephanieLindsey,CauligiRaghavendraan dKrishnaM.Sivalingam,“DataGatheringAlg orithmsinSensorNetworksUsingEnergyMe trics”,IEEETransactionsonParallelandDist ributedSystems,Volume13,No.9,pp.924- 935, September 2002.

19. OuadoudiZytouneandDrissAboutajdine,“A LifetimeExtensionProtocolforDataGatheri nginWirelessSensorNetworks”,Internation alJournalofInnovationandAppliedStudies, Volume4,No.3,pp.477-

482,November2013.

20. RaziehSheikhpour,SamJabbehdariandAh madkhademzadeh,“ACluster-

ChainbasedRoutingProtocolforBalancingE nergyConsumptioninWirelessSensorNetw orks”,InternationalJournalofMultimediaa ndUbiquitousEngineering,Volume7,No.

2,pp. 01-16,April 2012.

21. CongWangandCuirongWang,“AConcentri cDataAggregationModelinWirelessSensor Network”,ProgressInElectromagneticsRes earchSymposium, Beijing, China,pp.436- 441, March 2009.