基于CS-MVDR的多目标方位估计新方法

A new method of multi-target DOA estimation with Sub-Nyquist spatial-temporal sampling based on CS-MVDR

针对空时欠采样条件下多目标方位估计存在的问题,将压缩感知理论引入到最小方差无畸变响应(MVDR)方法中进行多目标方位估计,通过空间角度网格划分形式实现信号在空时域的稀疏性表示,利用空间谱估计恢复重构空间稀疏向量,从而实现空间上多信号入射角的方位估计。该方法不仅运算量低,在低信噪比、低阵元采样数、高估计精度和稳健性等方面的优势更为突出。

In view of the existing problems of the multi-target DOA estimation under the condition of the sub-Nyquist spatial-temporal sampling,the compressed sensing(CS)theory is introduced to the Minimum Variance Distortionless Response(MVDR)method for the multi-target DOA estimation.The sparse representation of signals in the spatial-temporal domain is realized by meshing in spatial angles.The spatial sparse vector is recovered and reconstructed through spatial spectrum estimation to realize the DOA estimation of multi-signal incident angles.The method is superior in terms of low SNR,low array element sampling number,high estimation accuracy and robustness with a small amount of computation.