Azimuth Ambiguity Discrimination Using Doppler Spectrum of the Compressive Sensing-Based SAR Image with Downsampled PRF

A space-borne synthetic aperture radar (SAR) is widely used for maritime surveillance because it can observe a broad area. Although a low pulse repetition frequency (PRF) mode is generally selected to observe a wider area, it causes azimuth ambiguities that may lead to false detection and affect the accuracy of ship detection. In previous work, an azimuth ambiguity discrimination algorithm using the Doppler spectrum was presented. However, this algorithm did not eliminate false detection and presented challenges for ship detection performance. To achieve more precise detection, a novel algorithm using a compressive sensing (CS)-based SAR image is proposed. CS can reconstruct the ambiguity-suppressed SAR image, resulting in further improvement of azimuth ambiguity discrimination. The proposed algorithm was applied to a SAR image simulated by half-downsampling of the received signals. The evaluation shows that the proposed algorithm suppresses the power of azimuth ambiguity up to 15 dB and eliminates false detection. These findings demonstrate that the algorithm discriminates robustly and significantly improves ship detection performance.