Assessment and comparison of different coherent techniques for detection of small maritime radar targets

The CP-140 Aurora is the mainstay patrol aircraft of the Canadian Air Force. The CP-140 radar's ability to detect small targets such as periscopes is vital to its surveillance role. As stealth technology improves, further advances must be made in detection processing to maintain surveillance capabilities.; This thesis examines small target detection algorithms through analysis of data collected in a variety of sea states at multiple ranges. Various coherent processing techniques are evaluated by applying them to real radar data with embedded simulated targets. The targets embedded in the data serve as a gauge to quantify the performance of the different signal processing techniques. Spotlight Synthetic Aperture Radar (SSAR) data collected under three different conditions of sea state are used. These conditions consist of Cross-swell, Down-swell and Up-swell flight profiles. The process of embedding targets properly accounts for aircraft and target velocity headings as well as range migration, azimuth migration and Doppler shifts. The coherent processing methods compared in order to assess their relative performance (ie. ability to detect a target a set target) are: (1) Choi-William Transform, (2) Short Time Fourier Transform, and (3) Wigner-Ville Transform.; The performance of the various techniques are analyzed and compared against different Probability of False Alarm (PFA) rates using Receiver Operating Characteristic curves. It is found that the detection performance of the techniques is dependant upon the aircraft orientation respective to the sea. The Wigner-Ville Transform performed best in the cross-swell and down-swell profiles and the Short Time Fourier Transform is better in the up-swell profile. The obtained results should be useful to improve detection performance of small targets in clutter plus noise.