A Decentralized Fuzzy C-Means Minimal Clustering Protocol for Energy Efficient Wireless Sensor Network

Abstract

A minimal clustering protocol based on Fuzzy-C means (FCM) is proposed for prolonging the lifetime of wireless sensor networks by balancing the node's energy consumption. The cluster formation of nodes is modeled as a fuzzy partition of sample space under an uneven distribution of the sensor nodes. The overall energy consumption of the networks depending on the minimal number of clusters is determined. A multicriteria objective function is proposed to select the CH. A decentralized method finds CH in each cluster itself reducing the energy consumption of the network. The efficiency of the proposed work over existing protocol are shown by various simulations.