Research Article
Electromagnetic Signal Interference Based on Convolutional Autoencoder
@INPROCEEDINGS{10.1007/978-3-031-23902-1_18, author={Kaiyuan Zhao and Sa Xiao and Xiangyu Wu and Yang Wang and Xian Cheng}, title={Electromagnetic Signal Interference Based on Convolutional Autoencoder}, proceedings={Mobile Multimedia Communications. 15th EAI International Conference, MobiMedia 2022, Virtual Event, July 22-24, 2022, Proceedings}, proceedings_a={MOBIMEDIA}, year={2023}, month={2}, keywords={Intelligent interference Convolutional autoencoder Signal to interference ratio}, doi={10.1007/978-3-031-23902-1_18} }- Kaiyuan Zhao
Sa Xiao
Xiangyu Wu
Yang Wang
Xian Cheng
Year: 2023
Electromagnetic Signal Interference Based on Convolutional Autoencoder
MOBIMEDIA
Springer
DOI: 10.1007/978-3-031-23902-1_18
Abstract
At present, electromagnetic interference methods are mainly divided into traditional interference methods and intelligent interference methods. Traditional interference is currently dominated by barrage interference. Intelligent interference solves the shortcomings of barrage interference by sending out fixed-frequency and directional targeted interference waveform. However, most of the current intelligent interference methods require prior information and cannot deal with highly dynamic electromagnetic environments. Therefore, this study introduces an intelligent interference method without prior information. This study is based on a convolutional autoencoder model, which is used to extract high-order features of disturbed communication signal waveform without prior information. By covering some indistinct features and using a deconvolution network to generate similar signals to generate the best interference waveform, this method has an ideal bit error rate. The target signal is reconstructed by a convolutional autoencoder, and the optimal interference waveform is generated in the network by covering the high-order features of the input signal. Finally, the simulation is carried out using the method in this paper. In the BPSK communication system, a bit error rate of 48.7% can be achieved with a low signal-to-noise ratio. In practical engineering, the interference method in this paper can also realize covert jamming, which greatly improves the safety of jammer itself.
