An efficient k-means algorithm for the cluster head selection based on SAW and WPM

A wireless sensor network (WSN) offers the aggregation of data for the communication and processing in the exterior area or the base station. The main purpose of this study was to efficiently select the cluster heads (CHs) and carry out the synchronous data sink operation for the efficient energy and time utilization. An efficient approach based on the k-means algorithm for the cluster head selection has been proposed. It also includes simple additive weighting (SAW) and weighted product method (WPM) for the data sink operation priority by the decision performance ranking. In this approach, weights are assigned and pre-processed on the basis of the node operations or the attribute values. These values are used for clustering of the nodes. K-means have been applied for the clustering. The resultant data are then processed with the decision performance ranking methods. We have used SAW and WPM for the selection of CHs from the clusters. The variations in SAW and WPM results are minor and these approaches are efficient in providing the proper CHs selection from the obtained clusters. The result of the random selection priority scale also offers an energy efficient system. The proposed approach results in less delay in packet delivery and offers efficient energy consumption in contrast to the traditional method.