Design of Hybridized Wireless Sensor Network using K-Means Clustering and Genetic Algorithm

Abstract

Prolonging the network lifetime in wireless sensor networks (WSNs), Clustering has been recognized has one of the significant methods in achieving this, It entails grouping of sensor nodes into clusters and electing cluster heads (CHs) for all the clusters. CH’s accept data from relevant cluster’s nodes and forward the aggregate data to base station. A main challenge in WSNs is the selection of appropriate cluster heads. This work proposes a system that is efficient, scalable and load balanced. The proposed scheme combines two known algorithms namely k-means clustering and genetic algorithms based on the weaknesses identified in the two. The simulated data is obtained through the enhancement of clustering by the cluster head (base station) that helps in locating the nearest node that is important in the data transfer instead of transferring to a node that is not necessary, thereby wasting time and resources. The obtained simulation results indicate that this approach is efficient and last longer in elongating the battery life time than the conventional method by 60%.