Cluster head selection strategy of WSN based on binary multi-objective adaptive fish migration optimization algorithm

In deployed wireless sensor networks (WSNs), how to efficiently transmit information collected by sensor nodes with limited energy is a challenging problem. An appropriate cluster head selection strategy can efficiently solve this problem, but there are many factors to be considered, such as energy consumption, the coverage of cluster head nodes, and the number of cluster head nodes. Each factor has a profound impact on the performance of wireless sensor networks, and there are conflicts among them. In order to solve the conflict of multiple factors and obtain the optimal selection strategy of the cluster head node, this paper proposes a Binary Multi-Objective Adaptive Fish Migration Optimization (BMAFMO) algorithm. The algorithm introduces the Pareto optimal solution storage strategy to improve the global search ability of the optimization algorithm and transform the continuous solution into a binary solution according to the sigmoid transformation function to solve the problem of cluster head node selection. The new algorithm was comprehensively tested using eight test problems and four test metrics. At the same time, the reliability of the algorithm is tested by rank sum test. The test results show that the BMAFMO algorithm obtained the best results in 78.13% test problems compared with other algorithms. Finally, the BAMFMO algorithm is applied to solve the cluster head selection problem of WSN and the simulation results show the novel algorithm has better optimization ability than other heuristic algorithms.