基于稀疏贝叶斯学习的字典失配杂波空时谱估计方法

杂波谱稀疏恢复空时自适应处理（STAP）是一种有效减少杂波样本数需求的机载雷达杂波抑制方法。然而，空时平面被离散地划分为若干个网格点来构建空时导向矢量字典，当字典在失配时，杂波脊不能准确落在预先离散化的网格点上，稀疏恢复STAP性能严重下降。提出了一种基于稀疏贝叶斯学习的字典失配杂波空时谱估计方法，首先利用二维泰勒级数建立空时动态字典模型，然后将字典失配误差作为待估超参数构建贝叶斯稀疏恢复模型，并利用失配误差估计值对空时导向矢量字典进行修正，最后利用修正后的空时导向矢量字典重构杂波协方差矩阵，进而计算杂波空时谱。实验证明，该方法能够有效提高字典失配情况下的杂波谱稀疏恢复精度，杂波抑制性能优于已有字典预先离散化的稀疏贝叶斯学习STAP方法。

Sparse Bayesian learning method for eliminating dictionary mismatch in clutter space-time spectrum estimation

Space-Time Adaptive Processing (STAP) via clutter spectrum sparse recovery can reduce the requirement for clutter sample size, and suppress the clutter of airborne radar effectively using limited training samples. The space-time plane is discretized into some grid points, and the space-time steering vector dictionary with discretized grid points is designed. However, the clutter ridge is not located exactly on the pre-discretized grid points in presence of dictionary mismatch. The dictionary mismatch effect degrades significantly the performance of STAP via sparse recovery. In this paper, a sparse Bayesian learning method for eliminating dictionary mismatch in clutter spectrum estimation is proposed. A dynamic dictionary is established by two-dimensional Taylor's series, and the dictionary mismatch biases are considered as parameters to be estimated by the sparse Bayesian learning method. The space-time steering dictionary is compensated by the estimated mismatch biases. The clutter-plus-noise covariance matrix is reconstructed with the compensated dictionary, and then the clutter space-time spectrum is calculated finally. Numerical results show that the proposed method can obtain better accuracy in clutter spectrum sparse recovery, and provide better performance of clutter suppression in comparison with the STAP using the sparse Bayesian learning method with the pre-discretized dictionary.