基于稀疏重构的机载雷达KA-STAP杂波抑制算法

机载雷达非均匀杂波环境下的空时自适应处理（STAP）算法会因杂波协方差矩阵估计不准导致其杂波抑制性能下降。传统知识辅助 STAP (KA-STAP)算法性能依赖于先验知识的准确程度以及配准精度，先验信息的失配可能会导致算法性能恶化。本文提出一种基于稀疏恢复技术构造杂波加噪声协方差矩阵的KA-STAP算法。该算法不依赖于先验信息，首先利用稀疏贝叶斯学习技术通过少量回波样本估计出稳健的辅助协方差矩阵，然后结合采样协方差矩阵进行空时处理。在小样本非均匀杂波场景下，该算法的输出性能优于传统KA-STAP算法。仿真结果表明了本文方法的有效性。

A KA-STAP Algorithm Based on Sparse Recovery for Airborne Radar Clutter Suppression

The clutter suppression effectiveness of space-time adaptive processing(STAP) in a heterogeneous clutter environment of airborne radar will be reduced due to the inaccurate estimation of the clutter covariance matrix. The performance of traditional knowledge-aided STAP (KA-STAP) algorithm depends on the accuracy and the matching precision of prior knowledge, therefore the mismatch of prior information can degrade the system performance. To mitigate this weakness, this paper proposes a STAP algorithm to reconstruct the covariance matrix by leveraging sparse recovery technique.Without prior information,this study uses a small amount of secondary data samples to estimate a robust auxiliary clutter and noise covariance matrix based on sparse Bayesian learning technique. The STAP result can be achieved by combining the sampling covariance matrix and the auxiliary covariance matrix. The performance of this method is better than that of traditional KA-STAP algorithm in a heterogeneous clutter environment with limited samples. Simulation results demonstrate the validity of the proposed method.