Space-time Adaptive Signal Processing Based On Block Sparse Recovery

Space-time adaptive processing(STAP)technology utilizes spatial and temporal information to suppress clutter effectively,which can reduce the number of the samples required combined with sparse recovery(SR).However,when the number of samples is seriously insufficient,the clutter recovered by SR-STAP is far from the true clutter spectrum.Considering the block sparsity of the clutter in the Angle-Doppler domain,the block sparse recovery algorithm can be applied to STAP to improve the clutter suppression performance.In this paper,block sparse recovery algorithm is applied to STAP.When the number of samples is seriously insufficient,the reconstruction accuracy and clutter suppression performance can be greatly improved with calculation time accepted.The main work is organized as follows:1.The principle of STAP,the characteristics and metrics of the clutter spectrum and the methods of SR are introduced.Two classical methods,OMP and SLO,are chosen to apply to STAP,and then the shortcoming of performance is analyzed.2.The principle of block sparse recovery is introduced,which is combined with STAP technology innovatively based on the analysis of the block sparse structure of STAP in the Angle-Doppler domain.The steps of block sparse recovery combining STAP approach are innovatively proposed and studied in detail.3.OMP is extended to BOMP,when which applied to STAP the recovered clutter is different from true one,due to the block boundary of STAP clutter unknowed.Therefore,a modified BOMP is proposed to improve clutter suppression performance,which re-choose the optimal atomic block judging by given threshold.The new method with BOMP is simulated in Matlab to show the effectiveness with both simulation data and real data from MountainTop program.4.SLO is extended to BSLO to be applied to STAP.BOMP and BSLO combing STAP are compared with OMP and SLO together under the same conditions.The simulation results of both simulation data and real data from MountainTop program show the effectiveness of block sparse recovery combining STAP.