Active Jamming Signal Recognition based on Residual Neural Network*

Abstract

The problem how to effectively identify the type of active jamming signal has important practical significance for the accurate perception of radar anti-jamming system. Therefore, a radar active jamming identification method based on fractional Fourier transform and residual neural network is proposed. Before the jamming signal pattern recognition, the time-frequency structure model of the signal is established, and the influence factors such as radar technical system and jamming signal processing cycle are comprehensively considered. The constraints of the time-frequency analysis kernel function and processing on the jamming signal type, the cross term of the composite modulation signal and the effectiveness of the distorted signal characteristics are analyzed. Then, according to the requirements of availability and recognition rate of subsequent signal classifiers, the recognition model based on residual neural network (RESNET) is used to solve the problem. The simulation results show that the recognition effect of multiple active jamming patterns under different interference to signal ratio is higher than 90%, which verifies the effectiveness and rationality of the method.