Recognition of radar deception jamming based on convolutional neural network

Since the radar deception jamming interferes with the radar in complicated electromagnetic environments, recognizing the deception jamming is an intractable problem. Various improved jamming strategies have been proposed, while the jamming technology is updated continuously. In this study, a new radar jamming detection method based on the convolutional neural network (CNN) is presented. The method firstly uses the nearest-neighbour interpolation (NNI) and short-time Fourier transform (STFT) to analyse the signal within the time-frequency (TF) range, to obtain the great adaptability on the different condition; then, it uses CNN to detect whether the signal is contaminated by radar jamming and determine the type of jamming. The simulation results show that this method can not only identify transmitting deceptive jamming and velocity deception jamming effectively, but also has good anti-noise performance and robustness. Compared with the recognition method based on time-frequency distribution, the recognition accuracy is 15% higher than the conventional detection model.