A hierarchical hybrid of genetic algorithm and particle swarm optimization for distributed clustering in large-scale wireless sensor networks

Abstract

In clustering wireless sensor networks (WSNs), the collected data from sensor nodes can be aggregated by cluster head instead of being sent directly to the base station. So distributed clustering in WSNs can eliminate the redundant data generated by nodes, and thus save the energy consumption by avoiding the unnecessary redundant data transmission. In this paper, a hierarchical hybrid of genetic algorithm (GA) and particle swarm optimization (PSO) is proposed for distributed clustering in large-scale WSNs. All sensor nodes can be organized into two levels of logic architecture. The bottom layer is made up of independent subgroups, and GA is used for global search. The optimum individuals of all sub groups constitute the elite group at the upper level, and the evolution of the PSO algorithm is mainly responsible for the local search of the elite individual. Then the design solution of distributed clustering can be obtained and the convergence speed of the algorithm will be accelerated effectively. Simulation results demonstrate that the proposed approach effectively reduces the energy consumed and it helps to increase the network lifetime.