Target Tracking Based On Particle Filtering In Wireless Sensor Network

Wireless Sensor Network(WSN), which integrates the technologies of sensor, Micro-electro-mechanism system, wireless communication and distributed computing, is a new mode of computing and a hot spot information technology which influences the lifestyle of human beings profoundly after Internet in 21st century.Based on the characteristics of low cost, low power, small size, high redundancy and multi-function, it is more convenient, more reliable and more effective for WSN to apply to target tracking than traditional system. The main purpose of WSN target tracking is to fix the location, the number, the velocity, the direction of targets and so on, its research aims to cost the low power consumption of WSN and obtain the high accuracy of target tracking. In the WSN target tracking, the models of the target tracking are non-linear and non-Gaussian, but Kalman filters and traditional methods of data fusion can not solve non-linear and non-Gaussian problems effectively where Particle filter performs admirably.A particle set, which is randomly sampled from probability function and has corresponding weights, is introduced to approach the posterior distribution. Therefore it can handle non-linear and non-Gaussian problems without any limits.For the sake of better effect,we apply filter algorithms to target tracking in WSN, In respect to above idea, following works are done:Firstly, this project has studied the key technologies about target tracking in WSN, Secondly, we develop the independent modular application software and hardware platform for WSN, named target tracking system in terms of dynamic location. The platform takes a supporting role for the research and application.We realize integrated programming based on LabVIEW and MATLAB, and this realization definitely provides a way to solve some targets tracking algorithms in the tracking system.Thirdly, we introduce extended kalman filter(EKF) theory based on the analysis of non-linear filter. Since the disadvantage of EKF target tracking algorithm, particle filter(PF) has been presented in this dissertation. After making the comparison of PF with LSQ and EKF through the MATLAB simulation, PF algorithm turns out to possess better performance in WSN target tracking.