Study On Positioning And Tracking In Wireless Sensor Network

Target positioning and tracking are hot research issues in wireless sensor network (WSN). Usually, in the actual WSN monitoring fields, limited energy of sensor nodes and changing environment require that a cost-effective and strongly stable target monitoring system need to be designed. In this paper, considering the problems above, some effective positioning and tracking algorithms are proposed. Main research works are summarized as follows:The precondition of positioning is ranging, but ranging method based on the received signal strength indication (RSSI) is often affected by the node itself or the environment which can result in uncertain data in. Furthermore, complex formulas lead to long positioning time. So in this paper, we propose a positioning algorithm based on interval clustering. Firstly, collect and analyse RSSI data in the current environment, and use interval clustering method to make data more stable, then select good beacon nodes to improve positioning accuracy, finally establish standard RSSI-D (communication distance) reference sample space, which eliminates the process of complex calculation and correction, because we can look-up the table to get data. Simulation experiment results show that this algorithm has a positive effect on positioning speed and accuracy, and should have nice application prospects.Generally, parameters of traditional propagation model are initially set and fixed; obviously it’s unable to meet the needs of complex and changing positioning environment. Therefore, this paper proposes a FOA-based localization algorithm on optimizing parameters of WSN transmission model, taking use of the excellent optimization ability of FOA to adjust model parameters to the change of actual environment to improve positioning accuracy. Simulation results show that this is an effective and simple algorithm, reducing positioning errors and also can be applied in some high-precision field.Study on tracking of moving target is an advanced research based on target positioning. Most traditional tracking algorithms are based on ideal models such as uniform linear motion and uniformly accelerated linear motion, and then make continuous estimates and projections on target’s position with filtering methods. However, large numbers of moving targets in real-world follow no any law at all; as a result they are called the random walk model. So in this paper, a low complexity solution to resolve this kind of disorder movement model is proposed, the first step of this solution is dividing the WSN coverage area into many small average parts, the intersections of latitude and longitude are selected as interval clustering statistical points, then store these information in a data table, which can be used in the next step; the second step is tracking moving target through sequence matching. In short, this algorithm change positon tracking problem into a data sequence matching problem, effectively reducing the complexity of tracking, as well as providing a heuristic solution for tracking random moving targets. Numerous simulation experiments show that this scheme possess greater flexibility and universality while ensure tracking accuracy.