Research On The Target Locating And Tracking Based On Wireless Sensor Network Spatial Meshing

Wireless sensor networks can meet special needs and work in poor condition, so it has become a hot area of research community in academia and industry.A large number of low-cost, small size, powerful wireless sensor network nodes are deployed randomly in poor condition and special area. For most applications in the WSN, the location of the incident is a essential part, so the target locating and tracking method in WSN environment increasingly become a hot topic of research.There are many research findings on target locating and tracking based on WSN but most of them have kinds of problems and are designed for two-dimensional scene. For WSN, these algorithms can’t be applied to all the requirements. Therefore, the researches in this thesis focus on the three-dimensional target locating and tracking method in WSN.Firstly, a brief introduction of relevant knowledge for WSN are given, and the development and significance of target locating and tracking are described, then the research status of object tracking in China and abroad are described. Then, a distributed localization algorithm based on WSN spatial meshing is proposed. In view of the current three-dimensional localization algorithm has some problems including location is not accurate enough, the calculation overhead is too high and power consumption is too large. Thus, a distributed three-dimensional localization mechanism for the environment of the WSN is proposed. The algorithm uses Cooperative Location-Sensing (CLS) to mesh grid and determine the target location by the estimated distance. It combines Gaussian fitting,signal sorting mechanism and Bounding-inbox algorithm which have effectively reduced signal interference. Moreover, it uses local meshing to reduce the grid voting overhead. Simulation results show that the algorithm is better than the existing three-dimensional localization algorithm in positioning accuracy and has a lower power consumption.Finally, for moving target, a further research on target tracking is given, and a tracking algorithm based on clustering idea is proposed. It uses a new spatial clustering structure, which includes two tier sleep scheduling mechanisms and unites cluster head mechanism. A spatial segmental linear fitting method is adopted to track the target, which has effectively reduced the network overhead and improved tracking efficiency. It also provides a scheduling strategy to wake up the sensor node to continue the tracking process, if the mobile target lost. Simulation results show that the algorithm has a good performance on tracking efficiency and power consumption.