Clock Synchronization Algorithm Based on Particle Swarm Optimization with Natural Selection

In order to improve the clock synchronization accuracy of wireless sensor network (WSN) and avoid the synchronization error caused by the unreliable data and the change of the network topology structure in the synchronization process, a simulation model of particle swarm optimization (PSO) algorithm based on natural selection was established. Firstly, using Kalman filtering algorithm to locally filter the measurement data of multiple sensors, the redundant information in the bottom of the measurement data could be removed so that the accuracy of measurement data is improved. Then, particle swarm optimization algorithm based on natural selection was used to set up a data fusion model and calculate the optimal fused estimation. Finally,the simulation results of the fusion model of PSO algorithm based on natural selection show that the fusion algorithm can effectively realize the fusion of five node clock in the networks, and improve the fusion accuracy of time offset and frequency offset of node crystal oscillating, producing a virtual clock of high stability. The concentration ranges of fusion accuracy of time offset and frequency offset are 10-5 and 10-7, respectively. Compared with traditional adaptive fusion algorithms, the concentration ranges of fusion accuracy of time offset and frequency offset are improved by one order of magnitude.