A novel dictionary-corrected sparse recovery stap algorithm based on prior knowledge
Zhiqi Gao, Caimei Zhao, Pingping Huang, Wei Xu, Weixian Tan · IET conference proceedings. · 2024
Sparse recovery space-time adaptive processing (SR-STAP) is prone to grid mismatch, which affects the clutter suppression performance. To cope with this problem, a dictionary correction sparse recovery STAP algorithm based on prior knowledge is proposed, which uses the principle of high power spectral value of clutter points. Firstly, the atoms matching the real clutt er points in the original dictionary are preliminarily selected by using the spectral value dimension reduction method. Then, around the selected atoms, the accurate clutter ridge is calculated by prior knowledge. Along parallel direction and vertical direction of the clutter ridge, space-time plane is evenly divided to form the local grid, and the optimal atoms to correct the original dictionary are searched in the local region. Finally, the modified clutter covariance matrix and the optimal STAP filtering weight are obtained. Simulation results show that, compared with the existing SR-STAP algorithm with dictionary-corrected, the proposed algorithm can find the grid points matching the real clutter ridge faster, effectively suppress the c lutter and improve the STAP performance.