Barrage Jamming Detection and Classification Based on Convolutional Neural Network for Synthetic Aperture Radar

Abstract

Suppression technology of barrage jamming is an important approach to ensure the normal operation of the synthetic aperture radar (SAR) system. The detection and classification of jamming is a necessary procedure in this technology. Unsuitable thresholds set in the traditional methods may reduce the detection accuracy. In order to avoid it, this paper proposes a new method of barrage jamming detection and classification for SAR based on convolutional neural network (CNN). The signal model is constructed based on the statistical characteristics of the SAR echo signal. Based on this, a data set containing echo signals and interference signals is generated by simulation. Finally, the convolution neural network VGG16 is used to detect whether the signals in the dataset is contaminated by barrage jamming and identify the type of the interference. The experiment result illustrates that the VGG16 network trained by the frequency domain signals can effectively detect and classify the jamming signals.