As an effective feature extraction method in radar target high-resolution range profile(HRRP) recognition community, linear discriminant analysis(LDA) faces two main shortcomings which are referred to as small sample size problem and Gaussian-like assumption respectively. Therefore, a radar target recognition method based on nonparametric maximum margin criterion(NMMC) is proposed. Firstly, the non-stationary characters of HRRP are extracted by the auto-correlation wavelet transform, which are used as the target classification characteristics together with HRRPs. Then the classification characteristics are extracted by using the NMMC algorithm, and support vector machine(SVM) classier is used for target recognition. NMMC solves the small sample size problem and relaxes the requirement of Gaussian distribution assumption in LDA. Finally, simulation results based on a HRRP dataset of five aircraft models show the effectiveness of the proposed approach.