Stepped-Frequency Radar Imaging Algorithm Based on Compression Sensing

Compressive sensing (CS) theory asserts that one can recover original signals from far fewer random samples under the condition of being sparse. CS theory is applied to high resolution imaging of vehicle-mounted stepped-frequency forward-looking ground-penetrating radar. This paper explores an approach of obtaining discrete scattering structure of the metal mine based on CS imaging and extracting geometry parameters to discriminate targets. Real data of vehicle-mounted stepped-frequency forward-looking ground-penetrating radar is processed. High resolution images of the metal mine with double-scattering structure are obtained. The feasibility of the method is tested through these images.