The network clustering optimization algorithm of wireless sensor network with energy replenishment

Herein, we propose an improved cluster-formation algorithm of binary particle swarm optimization (SOL-BPSO). Through the establishment of mathematic models, network clustering problems are transformed into combination optimization ones. Additionally, the search space of particles is constructed as the binary N-dimensional space that works as cluster heads. The search speed of particles decides the probability that nodes become cluster heads. Meanwhile, a fitness function is designed under the situation of energy replenishment. Parameter adjustment increases the energy efficiency of nodes and maintains the loading balance of clustering, which makes the nodes whose energy harvesting ability is high become cluster heads with a priority. Finally, optimal cluster heads will be decided through multiple iterations, which stands for the accomplishment of clustering and saves cluster-head energy and balances node loading. The simulation results show that SOL-BPSO algorithm effectively balances the energy consumption of nodes and extends the life cycle of network while energy replenishment is considered. Therefore, the network life cycle of SOL-BPSO algorithm is 12.5% longer than that of PHC algorithm.