Classification of Radar Signals with Convolutional Neural Networks

In this paper, we propose a method to classify radar signals according to the jamming techniques by applying the machine learning to the parameter data extracted from the received radar signals. In the present army, the radar signal is classified according to the type of threats by refening to the library composed of radar signal parameters mostly built by prior investigations. Since radar technology is continuously evolving and diversifying, however, the library based method can not properly classify the signals for new threats which are not in the existing libraries, thus limiting the choice of appropriate jamming techniques. Therefore, it is necessary to classify the signals so that the optimal jamming technique can be selected by using only the parameter data of the radar signal. In this paper, we propose a method based on machine learning to cope with new threat signals of radars. The method classifies the radar signals according to the jamming method with convolutional neural networks, and does not refer to the preexisting library.