Abstract: Radar High Range Resolution Profile (HRRP) plays an important role in target detection and recognition. For the properties of radar scattering response at high frequencies, a novel radar high-resolution imaging method using sparse signal representation is proposed in this paper. Based on ideal point-scattering model, an overcomplete basis is constructed, and an algorithm via smoothed l0 norm is also introduced to obtain radar HRRP. Experimental results indicate that good HRRP can be reconstructed even under the condition of data sampling rate decreasing or observation data missing, which shows that the proposed method is an effective radar imaging method.