Radar Active Jamming Recognition based on Recurrence Plot And Convolutional Neural Network

Considering the problems of low accuracy and poor robustness of traditional active jamming algorithms, this paper proposes an algorithm that can intelligently recognize the types of active jamming. In this paper we develop an intelligent recognition method based on recurrence plot and convolutional neural network(CNN). Firstly the algorithm realizes the graphical representation of radar active jamming based on the recurrence plot , and then uses CNN for learning, training, recognition and classification. Simulation shows that for eight types of active jamming such as interrupted sampling repeater jamming, the algorithm proposed in this paper can achieve a correct recognition probability of more than 99%, and is significantly better than traditional recognition methods based on manual feature extraction in terms of accuracy and robustness.