Research on electronic jamming identification based on CNN

In this paper, we apply the idea of deep learning to radar interference recognition. Firstly we introduced history and concept of Convolutional Neural Network, and summarized the method of interference recognition. Secondly, the structure of improved LeNet CNN is described, considering the character of radar echo wave. Thirdly, 7 kinds of jamming are introduced. Fourthly, several important parameters of net structure such as kernel size and batch size are adjusted to achieve best performance, through measured interference radar echo. The accuracy rate reaches up to 99.7%. Finally, we summarize advantages of the method proposed in this paper.