DISTRIBUTED CLUSTERING-TASK SCHEDULING FOR WIRELESS SENSOR NETWORKS USING DYNAMIC HYPER ROUND POLICY
ABSTRACT
Prolonging the network life cycle is an essential requirement for many types of Wireless Sensor Network (WSN)applications. Dynamic clustering of sensors into groups is a popular strategy to maximize the network lifetime and increasescalability. In this strategy, to achieve the sensor nodes’ load balancing, with the aim of prolonging lifetime, network operationsare split into rounds, i.e. fixed time intervals. Clusters are configured for the current round and reconfigured for the next roundso that the costly role of the cluster head is rotated a the network nodes, i.e. Round-Based Policy (RBP). This loadbalancing approach potentially extends the network lifetime. However, the imposed overhead, due to the clustering in everyround, wastes network energy resources. This paper proposes a distributed energy-efficient scheme to cluster a WSN, i.e.Dynamic Hyper Round Policy (DHRP), which schedules clustering-task to extend the network lifetime and reduce energyconsumption. Although DHRP is applicable to any data gathering protocols that value energy efficiency, a Simple EnergyefficientDataCollecting(SEDC)protocolisalsopresentedtoevaluatetheusefulnessofDHRPandcalculatetheend-to-endenergyconsumption.ExperimentalresultsdemonstratethatSEDCwithDHRPismoreeffectivethantwowell-knownclusteringprotocols,HEEDandM-LEACH,forprolongingthenetworklifetimeandachievingenergyconservation.
EXISTING SYSTEM:
Clustering is a common strategy for prolonging the networklifetime and efficiently managing a WSN.Because an intelligent Cluster Head (CH) selection algorithmcan significantly lengthen the network lifetime, many clustering algorithms have been proposed in recentyears. Some algorithms select CHs without node energyawareness. Often a probabilistic functionis employed to select CHs; hence, a low overheadresults for the network. However, choosing the appropriateCHs to maximize the network lifetime is left by thesealgorithms. In contrast, some CH selection algorithmsapply the nodes’ energy to select the best CHs. However, a high message passing complexity is adrawback of the latter algorithms. Some proposed CHselection algorithms are distributed,they require partial knowledge of environment conditions.On the other hand, some algorithms need globalknowledge of network nodes to create the best clusters.These algorithms must be centralized. The centralizedapproache are not applicable to large-scaleWSNs because gathering all information at the sink nodeis a time- and energy-consuming task. One major problem with clustering approaches is howto distribute the high workload of CHs a the networknodes. For this purpose, clustering protocols can becategorized into static and dynamic methods. In the staticmethod, clusters are permanently formed (i.e. the shapeof the clusters are fixed during network operation).Therefore, a high amount of energy is spent due to themessage collisions, as some CHs may be located in closeproximity of each other. To achieve load balancing in dynamic methods the shape of the clusters is notfixed during the network life cycle and the CH role rotatesa all network nodes. This exchange of thenodes’ responsibility is performed periodically in eachstatic predefined round. Although periodic reclusteringachieves load balancing of the network nodes, it overloadsthe system. Besides, to justify reclustering in eachround, RBP assumes that all network nodes, especiallyCHs, consume a considerable amount of energy during around. Consequently, round-based clustering is onlycompatible with a continuous data delivery model and isnot suitable for other data delivery models, such as event-and query-driven ones. Therefore, there is a need formore flexible and energy-efficient clustering-task schedulingwhichremovesunnecessaryreclusterings,is compatiblewith all data delivery models, and employs an onlinedynamic clustering-task scheduling algorithm instead ofoffline static scheduling, RBP.
PROPOSED SYSTEM:
The significant contributionsofthepresentworkareasfollows: To achieve load balancing, DHRP schedules triggeringof the clustering-task only at the required times,i.e. this policy eliminates the unnecessary reclusteringsof the RBP. Therefore, by tuning the reclusteringtime dynamically, DHRP effectively controls the clusteringoverheadthroughoutthenetworklifetime. To enhance the performance, the clustering-task isscheduled by an online algorithm as opposed to theoffline algorithm (RBP) which does not consider thedynamism of the sensing region and network conditions As RBP predetermines the reclustering times, it is oftenappropriateforacontinuousdatadeliverymodel,inwhichnodesarecontinuouslymonitoringthesensingenvironment and, as a result, consuming energy.In contrast, DHRP considers the cluster heads’ residualenergy for scheduling clustering-task. Thus, thispolicy is adaptable to any data delivery model usedfor energy-efficient data collection to one or severalsink nodes, such as continuous and event-driven. Centralized approaches are not often scalable, eventhough scalability is a serious need for many WSNapplications. Therefore, in the current work, decisionmaking for clustering time is performed via a decentralizedapproach. In other words, the clustering operationisscheduledinadistributedfashion.This policy can be employed for scheduling the clustering-taskin many available data gathering protocols.In other words, DHRP is not only compatiblewith protocols such as LEAC,but it can improve their performance. Furthermore,the present study proposes a Simple Energy-efficientData Collecting (SEDC) protocol to evaluate DHRP’sperformance and calculate its energy consumption.This protocol focuses on achieving low clusteringmessage communication overhead, simple implementation(especially for continuous monitoring applications),and an overall low level of energy consumption.Hence, DHRP can better conserve energylevels using SEDC.
CONCLUSION
In this paper, the trade-off between load balancing improvementand clustering overhead reduction is addressedas a significant issue for prolonging the WSN lifecycle. Here, the main approach is to reduce the imposedclustering overhead via dynamic scheduling of the clustering-task.Therefore, an energy conservation policy,DHRP, is presented for the clustering-task scheduling in aWSN. Totally, when the sensor nodes’ energy consumptionin the setup phase, in comparison to that of thesteady phase, is considerable, DHRP may effectively decreaseclustering overhead, conserve energy, and extendnetwork lifetime. As DHRP determines the appropriatereclustering time by taking into account the residual energyof the CHs, this policy is suitable for most data deliverymodels,suchascontinuous,eventdriven,andquerydriven. Although DHRP may be applicable for anydata gathering protocol, a simple energy-efficient datacollecting protocol, SEDC, is also proposed to evaluatethe performance of DHRP. In addition, SEDC employsthe nodes’ remaining energy for CH selection, clusterformation, and route discovery processes. The provedtheorem and lemmas, the performed calculations, andexperimental results show the effectiveness of theDHRP’s findings on scalability, energy efficiency, andnetwork lifetime. Using DHRP, SEDC outperforms popularand well-known clustering protocols, HEED and MLEACH(which employ RBP), in terms of network lifetimeand energy efficiency. DHRP significantly improvesnetwork lifetime, but it cannot optimally achieve the HRlength due to its distributed nature. Therefore, presentinga centralized algorithm for small-scale WSNs to optimallydetermine the HR length remains as a future work
REFRENCE:
[1] E. Fadel, V. Gungor, L. Nassef, N. Akkari, M. A. Maik, S.Almasri, and I. F. Akyildiz, “A Survey on Wireless SensorNetworks for Smart Grid,” Computer Communications, 2015.
[2] L. Xuxun, “Atypical Hierarchical Routing Protocols for WirelessSensor Networks: A Review,” Sensors Journal, IEEE, vol. 15, no.10, pp. 5372-5383, 2015.
[3] K. Kredo, and P. Mohapatra, “Medium access control inwireless sensor networks,” Computer Networks, vol. 51, no. 4, pp.961-994, 2007.
[4] A. A. Abbasi, and M. Younis, “A survey on clusteringalgorithms for wireless sensor networks,” Computercommunications, vol. 30, no. 14, pp. 2826-2841, 2007.
[5] X. Zhu, L. Shen, and T.-S. Yum, “Hausdorff clustering andminimum energy routing for wireless sensor networks,”Vehicular Technology, IEEE Transactions on, vol. 58, no. 2, pp. 990997,2009.
[6] A. Chamam, and S. Pierre, “A distributed energy-efficientclustering protocol for wireless sensor networks,” Computers &electrical engineering, vol. 36, no. 2, pp. 303-312, 2010.
[7] N. Dimokas, D. Katsaros, and Y. Manolopoulos, “Energyefficientdistributed clustering in wireless sensor networks,”Journal of parallel and Distributed Computing, vol. 70, no. 4, pp.371-383, 2010.
[8] D. Wei, Y. Jin, S. Vural, K. Moessner, and R. Tafazolli, “Anenergy-efficient clustering solution for wireless sensornetworks,” Wireless Communications, IEEE Transactions on, vol.10, no. 11, pp. 3973-3983, 2011.
[9] A. Wang, D. Yang, and D. Sun, “A clustering algorithm based onenergy information and cluster heads expectation for wirelesssensor networks,” Computers & Electrical Engineering, vol. 38, no.3, pp. 662-671, 2012.