A Two-Step Space-Time Adaptive Matrix Information Geometry Detector for Multichannel Radar

The multi-channel radars are widely used in the circumference perception of moving platforms. Suffering from the limited reference samples and heterogeneous clutter, the detection performance degrades severely in strong clutter environments, especially in weak target detection. To address this problem, a space-time adaptive detector which is based on the matrix information geometry is proposed to improve the performance of clutter suppression and weak target detection. First, utilizing the eigenvalue decomposition to decompose the clutter into the subspace spanned by steering vectors, and then deriving the detection formula in this subspace with eliminating the clutter components whose spectrum peak is inconsistent to the target's. Secondly, aimed at improving the detection performance of weak targets, a whitening step is added to the proposed detector to achieve a more accurate estimate of the target's reflection coefficient. The proposed method vectorized the matrix information methods to achieve better detection performance and computational efficiency. Experiment results based on the mountaintop data, phased array radar data and IPIX data demonstrate the performance advantages of the proposed method. The results show that the proposed method obtains better performance in comparison with the state-of-the-art methods.