A radar target recognition method based on wavelet power spectrum and power offset

Radar high resolution range profile (HRRP) is sensitive to the target aspect and highly overlapped in feature space between different targets, thereby hybrid features are suitable for representing the target´s property. In this paper, the quadratic spline wavelet with compact support properties was used to extract the energy spectrum features of HRRP by multi-resolution decomposition, and the power offset was defined to characterize the Structural characteristics of aircraft under different target-aspect. Then the joint features which combine the energy spectrum features and the power offset of HRRP were selected in the kernel space. The simulation results based on support vector machine (SVM) classifier show that the proposed method is effective.