Research On Multi-Target Detection And Positioning Method Based On Distributed Radar Network

The multi-target detection and positioning system based on distributed radar network, usually lays radar transmitters and radar receivers on the ground in a wide range to monitor the target area in multiple perspectives, which has the advantage of high positioning accuracy, strong anti-interference ability and survival ability and especially can detect the stealth targets. Compared with the range-based multi-target positioning method, the projection imaging-based multi-target positioning method can avoid the problem of echo data association and solving the equation group. For the problems of large amount of computation and weak targets’ low positioning accuracy of the projection imaging-based multi-target positioning by distributed radar network, the multi-target positioning method based on distributed radar network is studied in this dissertation. The main work and innovation are listed as follows:1. The basic theory of multi-target positioning based on distributed radar network is studied. The time difference of arrival positioning method is briefly introduced and its shortcomings in the target positioning are analyzed. The multi-target positioning method based on projection imaging is studied. This dissertation introduces the geographical space projection multi-target positioning method, and three kinds of positioning signal processing models of the echo coherent-based accumulation, amplitude-based accumulation and statistical probability-based accumulation. Then the three kinds of models are compared by simulation. According to the spatial variability of the fuzzy region in the geographical space projection imaging, the basic principle and realization algorithm of bistatic space projection multi-target positioning method are introduced in detail. Finally, the dissertation briefly introduces the principle of GPU parallel computing programming.2. Radar echo data of stealth target is simulated based on electromagnetic simulation software FEKO to verify the effectiveness of the bistatic space projection multi-target of stealth positioning method. Firstly, the RCS distribution characteristics of stealth target are analyzed, then the radar echo data of stealth target is simulated and the influence of radar transmitter’s irradiation to stealth target from different directions on the bistatic RCS distribution is analyzed. Simulation results show that the bistatic space projection multi-target positioning method with large number of receivers can position the stealth target effectively.3. For the problem of intensive computation in bistatic space projection multi-target positioning method, the GPU parallelization of bistatic space projection multi-target positioning method is realized. The computation of bistatic space projection multi-target positioning method is analyzed. In view of the main factors that affect the computation, two parallel methods, one is based on the distributed radar network and the other based on the 3D space grid cell, are provided and the contrast analysis to the two methods are carried on. The more optimal parallel method which is based on 3D space grid cell is selected to realized. Simulation results show that the proposed parallelization method can greatly improve the operating efficiency of the positioning process, and with the increasing of monitoring scene, GPU parallel acceleration effect is more obvious.4. A multi-target positioning method based on multi-pulse mixture accumulation is proposed. The problem of low target positioning accuracy or difficulty to position for the traditional bistatic space projection multi-target positioning method under low signal to noise ratio is analyzed. The moving target radar echo model based on the distributed radar network is established. A multi-target positioning method based on multi-pulse mixture accumulation is proposed. The multi-pulse mixture accumulation requires multi-pulse signal coherent accumulation for a single receiver and then non coherent accumulation between the receivers. Simulation results show that in low signal to noise ratio conditions, compared with the traditional method, the multi-target positioning method based on multi-pulse mixture accumulation not only can better improve the positioning precision and can also effectively positioning the target with less receivers.