In this paper, we propose the amplitude modulation (AM) bandwidth and the frequency modulation (FM) bandwidth for far-field point target model and near-field multi-point target model, and analyze the statistical differences of their distributions. The result shows that there exists a significance difference between the distribution of AM (resp. FM) bandwidth under the model of ideal far-field point-target and near-field multi-point target. The two bandwidths, namely AM bandwidth and FM bandwidth, have been used as an input to support vector machine (SVM) for classifying far-field deceptive jamming signal and the near-field body target signal. The simulation results show the effectiveness of the proposed method for classification.