Research On Clutter Modeling And Target Detection And Location Method Of Distributed UAV Radar

With the development of electronic countermeasure technology,the early warning radar faces more and more challenges.There's more and more difficult for single early warning radar to satisfy the needs of the battlefield.Compared with the traditional monostatic radar,the distributed radar system has many advantages such as wider detection range,higher detection accuracy and more flexible array arrangement,so it is of great significance to research on the distributed radar system.By analyzing distributed radar clutter,a simplified clutter model is established which was a condition for subsequent target detection and angle estimation.In order to solve the problem of mismatch between target steer vector and steer vector seted previously in matched filter,the method of using signal subspace for target detection is studied.Aiming to decrease the huge computation cost of the angle estimation by the maximum likelihood,the method of using the signal subspace is researched.Based on both of them,a method of joint target detection and angle estimation by using subspace is proposed.Aiming at the problem that the non-error signal subspace leads to the degradation of target detection and angle estimation performance in the presence of errors,a method to obtain the signal subspace from the clutter data is proposed.It can improve the target detection performance and angle estimation accuracy.The thesis is organized as follows:1.Clutter modeling and suppression methods of distributed radar is studied.According to the sending and receiving mode of distributed radar,the clutter modeling of distributed radar system can be divided into two modes: monostatic radar clutter modeling and bistatic radar clutter modeling.In the clutter modeling of bistatic radar,the method of acquiring bistatic clutter location by coordinate transformation is derived in detail.2.The method of using subspace detector to detect targets and estimate angles is researched.The performance of conventional target detection algorithm will decrease when the direction of target deviate from the center of the main lobe.In order to solve this problem,the performance of sum and difference beam subspace detector and eigensubspace detector is studied.Simulation results show that the subspace detector still has good detection performance when steer vector of target is mismatched.It is difficult for the conventional maximum likelihood angle estimation method to be used in practice because of its large computation.To solve this problem,this thesis researches the method of using signal subspaces combined with maximum likelihood angle estimation.By combining subspace detector and subspace angle estimation,a method which can reduce computational complexity is proposed.3.How to acquire signal subspace in the case of error is studied.In this thesis,a new method about how to obtain the signal subspace by using clutter data is studied.The subspace can be obtained by using the doppler filter to obtain the clutter data which has the corresponding spatial frequency based on the coupling relationship of spatial cone angle and the velocity cone angle.Simulation results show that the signal subspace obtained by the proposed method has better detection and angle estimation performance in the presence of errors.