Application of Support Vector Machines to Pulse Repetition Interval Modulation Recognition

The preprocessing of the pulse repetition interval (PRI) train is essential to the PRI modulation recognition of radar emitter signals when intelligent recognition methods are adopted. In this paper, a feature extraction method is proposed to deal with the PRI train to decrease the dimension of classifier inputs and to improve the robustness of recognition. Also, neural networks and support vector machines are adopted to design classifiers to identify the PRI types automatically. Experimental results show that the proposed method achieves lower error recognition rate and stronger capability of noise-suppression than the method proposed by Noone