Cascaded Dual-Radar Fusion Detection with Loose-Threshold Preliminary Screening and False Alarm Suppression

Sea clutter exhibits complex, time-varying statistical characteristics due to variations in radar operational parameters and sea surface conditions, which severely degrade the false alarm control performance of conventional constant false alarm rate (CFAR) detectors. To achieve effective false alarm suppression without compromising detection performance in time-varying heterogeneous sea clutter, this paper proposes a dual-radar fusion detection scheme for two independent X-band staring radars deployed at different positions, featuring a cascaded architecture of signal-level fusion and false alarm discrimination. In the first stage, a relatively low threshold is adopted for signal-level fusion detection under the Rayleigh clutter assumption, thereby enhancing the target detection probability. Subsequently, a two-stage discriminator suppresses false alarms generated in this preliminary stage by exploiting the feature differences between targets and false alarms in pulse continuity and instantaneous frequency stability. Comparative evaluations on four datasets of real sea clutter measurements under sea state 3 demonstrate that the proposed scheme achieves superior detection performance and more robust false alarm control than existing multi-radar fusion detection methods, while exhibiting favourable adaptability to variations in wave height and wave direction.