Radar Deception Jamming Recognition Based on GCN and Transformer

As the variety of radar jamming increases, there is a significant threat to radar systems. Anti-jamming techniques are crucial for addressing this major threat, with jamming recognition being the initial step. In this paper, we propose a multi-channel radar deception jamming recognition model based on graph convolution networks and transformer (GCN-T). Firstly, the short-time Fourier transform (STFT) is performed on the jamming signal to obtain the time-frequency distribution, which is modeled as an undirected graph. Secondly, GCN is used to acquire the spatial information in the data, transformer is used to capture the global features in the data. Finally, a fully connected layer is used to obtain the classification results. Extensive experiments on various jamming datasets and comparisons with existing methods are included to demonstrate the effectiveness of the proposed recognition model, and the recognition accuracy can reach 97.27 % to 99.14 %.