New K-means algorithm for clustering in wireless sensor networks
Walid Fakhet, Salim El Khediri, Adel Dallali, Abdennaceur Kachouri · 2017
Energy consumption is the main issue that researchers have dealt with to design any routing application. To design comprehensive reliable wireless sensors networks (WSNs), it is essential to consider node failure and energy constraint as inevitable phenomena. In this paper, a new method for optimal clustering is proposed, that accurately localizes sensors while minimizing their power consumption. It also splits the task of Cluster head between sensor nodes to send data to the base station (BS). The proposed scheme is based on K-means method to selected Cluster-Head. The selection of CH starts with the closest nodes to the centroid point. In the second step, we selected nodes with high power compared to its neighbors, after repeating this phenomenon until the energy of all the nodes is drained. This method can equilibrate the role of Cluster-Head between all the network sensors. Simulation results showed that our clustering mechanism seemed to effectively improve the network lifetime over Low Energy Adaptive Clustering Hierarchy (LEACH) and Energy efficient clustering algorithm for maximizing the lifetime of wireless sensor networks.