A Dual-cluster Heads Clustering Algorithm for Wireless Sensor Network Based on Improved Fuzzy C-means

Abstract

A dual-cluster heads clustering algorithm for wireless sensor network based on improved fuzzy C-means is proposed to address the limited energy and uneven node load issues in wireless sensor network. In the clustering phase, the proposed algorithm calculates the optimal number of clusters for current network nodes and uses the FCM algorithm optimized by the sparrow search algorithm to cluster nodes, solving the sensitivity problem of the FCM algorithm to initial clustering centers. In the cluster head selection phase, a dual-cluster heads mechanism is introduced to alleviate the energy load of a single cluster head, where the primary cluster head is responsible for data collection, fusion, and transmission, and the secondary cluster head is responsible for data relay. When selecting the primary and secondary cluster heads, two fitness functions are constructed based on the remaining energy of nodes, the communication distance within the cluster, and the distance from nodes to the base station. The primary and secondary cluster heads are selected according to different fitness functions. Simulation results show that the proposed algorithm can effectively balance network energy consumption and prolong the network lifetime.