Publication

Abstract : Clustering is one of the main aspects in wireless sensor networks and K-means algorithm is the most efficient way to cluster analysis. K-means algorithm forms the group of clusters which based upon random selection of initial centroids. If initial centroids are selected based on the distribution of clusters in a particular area, then obtain the best set of cluster. This manuscript proposes a method depending the rectangular area division algorithm to find the best initial centroid which forms a more precise cluster is considerably to reduce the computing period. The analysis and experimental outcomes show that the proposed rectangular area division technique can increase the cluster’s accuracy as well as lessen the computational time of K-means algorithm.