Multi-target Positioning Method Research Based On Multi-station System

Multi-station system consists of transmitter and receiver that located in different positon. Compared with single station system, the multi-station system can obtain a higher information redundancy and a higher precision in target detecting, positioning and tracking after the data fusion. Target positioning method based on arrival time difference(TDOA) faces the complex data correlation when positon multi-target. Target positioning method based on imaging strategy has a large location error. To solve the problems exist in traditional multi-target positon method, a kind of Multi-target positioning method based on compressed sensing is proposed. In addition, to expand the area which can be located and detected range, the multi-station layout is optimized when the receivers' quantity is certain. In this thesis, the main work and innovation points are as follows.1. The traditional multi-target positioning method is introduced. Firstly, the target positioning method based on arrival time difference(TDOA) is introduced, the large amount of calculation and data association problem of this method are analyzed. Then, the multiple target positioning technology based on projection imaging is introduced. Even though the projection imaging target positioning method has solved the problem of data correlation, large amount of data calculation and slow speed of locating still exist. Finally, the target positioning method based on hierarchical projection imaging is introduced, it improves the target positioning speed greatly. The reasons that result in large positioning error and high positioning speed are analyzed respectively.2. The multi-target positioning method based on compressed sensing is proposed. Compressed sensing theory is applied in multi-target positioning, by using the phase information the sensing matrix is constructed, and the multi-target positioning model is established. The multi-target position is reconstructed through orthogonal matching pursuit(OMP) algorithm. This method can not only solve data association problem, but also improve the position precision by using the echo's phase information. Simulation results show that comparing with the target positioning method based on hierarchical projection imaging, this method can obtain a higher position precision.3. The multi-station layout is optimized. Under the condition that receivers' quantity is certain, the detection area optimization model is established, and the detection range is optimized. Simulation result show that the total area of all stations' detection range is improved after using the genetic algorithm. In order to improve the target area and expand multi-station detection range at the same time, the positioning area optimization model is established. Simulation result show that the positioning area is improved after using the genetic algorithm.4. The multi-target positioning method based on centralized-distributed layout of multi-station is researched. To solve the poor information redundancy and bad target position precision which caused by lack of receivers, centralized-distributed layout is constructed. Echo information fusion method is studied under this layout. Multiple observation signal and sensing matrix are stitched, and the multi-target positioning model which is suitable for centralized-distributed layout is established. The applicable condition is also given. Simulation result demonstrate that with the same number of stations, the location accuracy of the method proposed is higher than existing methods. Meanwhile, when the number of the receivers is deficiency in the concentrated area, this method still can get a high positioning precision.