Deployment algorithm based on dynamic multi-populations particle swarm optimization for wireless sensor networks

Lei Hong

School of Information Technology, Jinling Institute of Technology, Nanjing 211169, China

Aiming at improving coverage rate and reducing coverage holes of wireless sensor networks, this paper proposes a deployment algorithm based on dynamic multi-populations particle swarm optimization. K-Means clustering algorithm is employed to divide the network into several sub-populations dynamically, which could weaken particles on the pursuit of local optima, realize the improvement of basic PSO (Particle Swarm Optimization) algorithm, and solve the “premature” problem of basic PSO algorithm effectively. In addition, it also accelerates the convergence of the algorithm. Simulation results show that this deployment algorithm can improve the network coverage rate effectively. Comparing with the conventional particle swarm optimization algorithm, its coverage rate is increased by 3.66%.