Weighted Feature-Based Spatial Reconstruction Suppression Method for Sea Clutter

To address the poor detection performance of airborne radar for sea surface targets under complex conditions， this study presents an innovative spatiotemporal signal model of sea clutter for an airborne multi-subarray radar system. Building upon traditional techniques like MTI， MTD， and frequency-stepped noncoherent processing， we analyze the spatiotemporal characteristics of sea clutter and propose a statistical covariance matrix decomposition method to extract the spatial distribution features of signal components. By leveraging spatial differences within Doppler channels， we achieve effective clutter suppression. To reduce the computational burden of spatiotemporal joint processing， we introduce a joint Doppler-spatially adaptive STAP framework. Our proposed method employs weight-based features for spatial reconstruction and suppression， innovatively utilizing angle and energy factors from the feature subspace to enable adaptive subspace reconstruction. This approach mitigates energy loss in target detection caused by conventional adaptive filters. The echo signals are segmented into range intervals for localized processing， and joint vectors are constructed using adjacent Doppler and subarray channels. Environmental perception is realized through a sample selection strategy， facilitating weak target detection in high sea states. Airborne flight tests were conducted to validate the system， obtaining echo data for large and small targets under high sea states and at large grazing angles. Comparative analysis demonstrates that our method outperforms existing spatiotemporal techniques in adaptability to target and clutter environments， achieving superior sea clutter suppression and enhanced target detection performance in complex sea surface conditions. The proposed method significantly improves the detection probability of weak targets and is readily applicable to airborne multichannel systems for moving sea surface target detection.