Optimization of clustering process for WSN with hybrid harmony search and K-means algorithm

As it is the fact that sensors deployed in the field are not accessible after deployment hence lifetime of the network is directly dependent on residual energy the sensors have. Conservation and efficient utilization of energy are very crucial in a sensor network. Clustering is one of the best-accepted strategies to efficiently utilize energy. Various approaches are proposed in the literature for clustering but meta-heuristic methods present promising results. When a hybrid version of two methods is derived to use best features of both, better results are anticipated. Two methods called Harmony search and K-means gives satisfactory results independently when applied to clustering process and we have made a hybrid version of these two methods to optimize the clustering process in this paper. Simulation is done with the NS-2 simulator for the above mentioned hybrid approach.