基于子频带黎曼距离的海杂波背景下目标检测

Target Detection in Sea Clutter Based on Subband Riemannian Distance

海杂波背景下的目标检测对军事和民用均具有重要意义,是雷达信号处理领域的研究热点与难点。传统广义似然比检测器在具有较强空间非均匀特性的海杂波中检测性能明显下降。文中提出基于子频带黎曼距离的检测器。首先,利用一组线性相位离散傅里叶变换调制滤波器将回波分解到不同子频带;随后,在每个子频带中,利用参考单元数据在Hermitian正定矩阵流形上的几何均值或几何中值估计杂波协方差矩阵,将待检测单元协方差矩阵与杂波协方差矩阵之间的黎曼距离作为检测统计量。基于仿真杂波数据和实测海杂波数据验证了所提方法可以有效提高海杂波背景下的目标检测性能。

Target detection in sea clutter is important for both military and civilian use. It is an important and challenging problem in the field of radar signal processing. Traditional generalized likelihood ratio test detector suffers performance degradation in nonhomogeneous sea clutter. In this paper, a detector based on subband Riemannian distance is proposed. First, the received signal is decomposed into multiple subbands using a linear phase discrete Fourier transform modulated filter bank. Then, in each subband, the geometric mean or geometric medium of the covariance matrices of references cells in Hermitian positive definite matrix manifold is used to estimate the clutter covariance matrix. The Riemannian distance between the covariance matrix of the cell under test and the clutter covariance matrix is used as the test statistics. Experiments based on simulated clutter and real sea clutter data have shown that the proposed detector effectively improves the performance of target detection in sea clutter.