Enhanced shuffled shepherd algorithm for selecting cluster heads in wireless sensor networks
摘要
Clustering is considered as an efficient technique used to reduce and control the amount of consumed energy in wireless sensor networks. Optimized choice of cluster heads can significantly prolong the network lifetime. In our proposed approach for choosing cluster heads, the shuffled shepherd optimization algorithm (SSOA) is employed to determine the optimal location of each cluster head, which is enhanced by combining it with re-initialization of the worst agents using levy distribution, based on minimizing a specially designed fitness function. Our proposed fitness function is based on five factors which are the distance between the ordinary nodes to their cluster heads, the distance from the cluster heads to the base station, the ratio between remaining energy and consumed energy of each cluster head, node degree, and node centrality. Clusters are then formed using a weight function sensitive to CH energy and proximity. A comparison of our proposed scheme with its alternatives in the literature demonstrates that the proposed method reduces energy consumption by up to 26