Clutter Suppression And Parameter Estimation Based On Atomic Norm

Airborne radar plays an extremely important role in the field of future war and current aviation safety.It plays the role of real-time monitoring of air and ground environment and target monitoring.The research on improving the clutter suppression and target detection ability of airborne radar echo is of great significance.Space time adaptive processing(STAP)technology can effectively filter the ground clutter of airborne radar and improve the ability of moving target detection through the joint processing of multi pulse and multi array elements.The space-time adaptive processing method based on sparse recovery is suitable for airborne radar clutter suppression and target detection in the absence of samples.This method depends on the construction of discrete space-time guidance vector dictionary,which has the problem of dictionary mismatch,which seriously affects the performance of clutter suppression and target detection.In this thesis,a space-time adaptive processing method for meshless sparse recovery based on atomic norm is proposed.Firstly,this thesis briefly describes the basic theory of sparse recovery,describes the airborne radar echo signal model used in this thesis,and analyzes the sparsity of clutter and target in angle doppler two-dimensional space-time spectrum,so as to pave the way for the subsequent clutter suppression and moving target parameter estimation algorithms.The existing space-time adaptive processing methods for sparse recovery need to build a discrete angle doppler two-dimensional space-time dictionary,which has the problem of dictionary mismatch,which affects the performance of sparse recovery.Due to the dictionary mismatch problem in the space-time adaptive processing method of sparse recovery,in terms of clutter suppression,the mismatch problem is that when the airborne radar is in a non-positive side view array,the clutter ridge is not on the grid points of the discrete two-dimensional dictionary.A meshless sparse recovery method based on atomic norm is proposed in this thesis,the sparse recovery of continuous space-time plane is realized by using the low rank matrix recovery theory,the dictionary mismatch problem in sparse recovery is overcome,and the high-resolution clutter space-time spectrum in the case of non-positive side view array is obtained,which effectively improves the clutter suppression performance.The simulation results show that the clutter suppression performance of this method is better than that of the existing dictionary discretization method in the case of non-positive side view array.In the part of moving target parameter estimation,the target is divided into uniform velocity target and uniform acceleration target based on whether the acceleration is zero.The mismatch problem is that the real doppler frequency and spatial frequency of the target have errors with the grid points of the constructed discrete space-time two-dimensional dictionary,and the small probability coincides with the grid points.In this thesis,an estimation method based on atomic norm meshless sparse recovery technology is proposed.This method makes use of the sparsity characteristics of target echo in angle doppler domain and realizes the meshless sparse recovery of target azimuth and velocity according to the low rank matrix recovery theory.The acceleration and velocity terms of uniformly accelerated target are separated by bilinear transformation,and then the parameters are estimated respectively,the estimation performance of moving target parameters is effectively improved.Simulation results show that the parameter estimation performance of the proposed method is better than that of the sparse restoration parameter estimation method with fixed dictionary grid.