Multi-objective quantization variation filtering based Bayesian prediction algorithm for WSN tracking and positioning

In order to further improve the wireless sensor network node's tracking accuracy and reduce energy consumption,a multi-objective crowding differential quantization variation filtering based Bayesian prediction algorithm for WSN tracking and positioning was proposed.Firstly,for the localization problem,the Bayesian quantization variation filtering method was used to predict the next position of target area,the quantized variation filtering method was used to select the appropriate participating nodes for positioning,for which the multi object parameter optimizing model of quantized variation filtering was designed.Secondly,aiming at the problem of optimization accuracy of traditional multi-objective optimization,a multi-objective differential evolution algorithm with the method of individual particle crowding distance sorting was proposed for the parameters optimization of quantization variation filtering,which implement the multi-objective parameters optimization.Finally,simulation results show that,the algorithm can effectively realize the tracking and locating of the target node,and also reduce the energy consumption.