Maritime target tracking from an airborne multi-channel radar

Detection and tracking of small maritime targets using high-resolution airborne radar is a difficult problem due to the non-stationary and non-Gaussian nature of sea clutter. This paper develops a target tracking algorithm for a scanning multi-channel radar operating in the maritime environment. The detection scheme uses space time adaptive processing (STAP) with a relatively high false alarm rate. While this improves the overall probability of detection, there are also a multitude of false (spurious) detections. As the radar scans, the target is illuminated over a range of azimuth angles, resulting in a large number of detections per target. A solution based on clustering the STAP output is proposed to create equivalent point target measurements, with the clustering output subsequently processed by a standard point-detection Bernoulli tracking filter. Through numerical simulations of high-fidelity airborne radar data, we demonstrate that the proposed approach achieves reliable tracking of targets characterised with low signal-to-interference ratios.