The Research On Target Localization And Tracking Of Wireless Sensor Networks

Wireless Sensor Networks(WSN) is one of the most influential and important technology in the 21st century. As the sensor network scale enlargement and improvement of technology with sensor's manufacturing process and energy, WSN is used in military and civilian. It can detect, locate and track targets. This paper is mainly about the applications of target tracking and localization in WSN.Firstly, the architecture of wireless sensor networks, the network structure and the application background are analyzed and discussed. Besides, the current achievements are also summarized. As a result of the relations between target localization and tracking, the localization of motion target is tracking.Then from the following three aspects, this paper discusses the target localization and tracking of WSN. Meanwhile, the improved algorithms are put forward and compared with previous methods in order to prove the superiority of these algorithms.(1) Based on the energy balance, I use three edges to locate the position of target. Because anyone in the past has proposed that he or she can choose such three nodes for target localization in the way of nodes'coverage, which has some malpractices. So this paper combines the coverage and energy balance to select three nodes for localization. The simulation proves that this new method can prolong further network lifetime and save energy.(2) This article will divide the area which is detected by many nodes into so many small grids in order to narrow the effective area in which the target locates. Besides, a correction value concept is also proposed in this paper. Through the weights which is the distance between the center of the effective grids coordinates and the true target position. At the same time, by modifying the weight exponential size and combining with the centroid localization method, the simulation shows that this improved method is so good.(3) The target tracking problem is also a nonlinear and non-Gauss noise problem and we can estimate the target position by using particle filter. So this paper will introduce the excellent weight selected particle filter particle filter to target localization. Because the excellent weight selected particle filter particle filter can well overcome the particle degeneracy problem and also can increase the diversity of particle. The simulation shows that this new method could have good target localization as long as we well select the reasonable particle number and well combine with the excellent weight selected particle filter in distributed dynamic clustering structure. Although this new method need to take much simulation time which isn't too long, which proves we can use this good method. So this new method has a certain value in the practice application. Besides, this method also saves more energy than centralized target tracking.In the end of this paper, the target localization and tracking algorithms are summarized. Besides, I also raise some problems which maybe lie in these researches.