Research On Airborne Multi-channel SAR Moving Target Indication And Motion Parameter Estimation Method

SAR moving target detection and parameter estimation technology can realize the functions of moving target imaging,detection,velocity measurement and positioning,which has important research significance in military,civil and other aspects.However,in the aspect of SAR moving target detection,most of the current detection methods,such as multi-channel VSAR method,dual channel DPCA,ATI method,are studied in the side view mode,which greatly limits the application scope.And aiming at the problem of the serious degradation of image quality and relocation accuracy of moving target in squint mode,the complexity of multi-channel image registration and phase error compensation,slow moving clutter suppression and so on,research on moving target detection and motion parameter estimation in squint mode is carried out.In the aspect of SAR multiple moving targets detection,aiming at the problem that multiple moving targets overlap in the same resolution unit in the image domain,which may lead to "velocity layover",the research on multiple moving targets separation and velocity estimation is carried out.The main contents and contributions of this paper are as follows:1.Aiming at the difficulty of traditional SAR moving target detection in squint mode,a DPCA-ATI(BP-DPCA-ATI)moving target detection method based on BP algorithm and a VSAR(BP-VSAR)moving target based on BP algorithm are proposed.First,a squint multi-channel SAR geometric model is established,and the position offset of the moving target in the SAR image based on BP algorithm is derived,which shows that the moving target has both azimuth offset and range offset in the squint mode,then combined with DPCA-ATI technology realizes moving target detection,radial velocity estimation and relocation in squint mode.Simulations verify the effectiveness of the BP-DPCA-ATI method in squint mode.However,the detection performance of this method deteriorates in a slow motion clutter environment.For this reason,a BP-VSAR moving target detection method is proposed,which realizes the moving target detection,radial velocity estimation and relocation in the slow motion clutter environment in the squint mode.Simulations verify the effectiveness of the method.Finally,it is compared with the BPDPCA-ATI method.The experimental results show that the BP-VSAR method can not only effectively suppress slow motion clutter,but also has higher velocity estimation accuracy and anti-noise performance is better.2.Aiming at the problem of "velocity layover" of moving targets,a method for separation and velocity estimation of multiple moving targets based on the Sparsity Bayesian Recovery via Iterative Minimum(SBRIM)algorithm is proposed.Firstly,the method of separation and velocity estimation of multiple moving targets based on spectrum estimation method is studied.The MVDR and MUSIC spectrum estimation methods are applied to multi-channel signal processing,which separates multiple overlaid moving targets and realizes high precision radial velocity estimation.Simulation experiments verify the effectiveness of the spectrum estimation method,but this method cannot effectively separate multiple overlaid moving targets when the signal-to-noise ratio(SNR)is low.In this regard,a SBRIM-based method for multiple moving targets separation and velocity estimation is proposed.This method constructs a linear observation model and corresponding measurement matrix by extracting the amplitude and phase information of the resolution unit where the multiple overlaid moving targets are located in the image domain.Utilizing the sparsity of the moving target in the velocity spectrum,separation and velocity estimation of multiple overlaid moving targets is realized through SBRIM sparse reconstruction.Simulation experiments verify that the proposed method can effectively solve the problem of "velocity layover".Finally,the simulation compares the proposed method,spectrum estimation method and traditional DFT method.The results show that the proposed method and spectrum estimation method can achieve higher velocity resolution,but the spectrum estimation method cannot effectively separate multiple overlaid moving targets at low signal-to-noise ratio(SNR ?10d B),and the proposed method(even in the case of SNR(28)0d B)can still achieve multiple overlaid moving targets separation and high precision radial velocity estimation,indicating that the proposed algorithm has better noise immunity.3.Aiming at the problem of moving targets separation and velocity estimation under non-uniform arrays,a moving target separation and velocity estimation method based on the Sparsity Bayesian Recovery via Iterative Minimum(SBRIM)algorithm under nonuniform arrays is proposed.Firstly,analyze the characteristics of multi-channel signals of multiple overlaid moving targets in a non-uniform antenna array,extract the corresponding amplitude and phase information,construct the observation model and the corresponding measurement matrix under the non-uniform arrays,and then implement the separation of moving target and velocity estimation by the SBRIM algorithm.Simulation experiments verify the effectiveness of the proposed method.Finally,a simulation comparison with the spectrum estimation method is carried out.The results show that the spectrum estimation method has completely failed under the non-uniform array,and the proposed method can effectively solve the problem of "velocity layover",realize the separation of multiple overlaid moving targets and high precision radial velocity estimation.