Secure Cooperative Semantic Communication with Swin Transformer and Jamming

Hung Ngo Luu Tan, Toan Van Nguyen, Phuong Luu Vo

Abstract: Semantic communication is becoming increasingly popular for wireless image transmission due to its superior communication efficiency. However, current semantic systems remain vulnerable to eavesdropping and often overlook security measures at the physical layer. To address this issue, this paper presents a cooperative semantic communication system with a Swin Transformer design and physical jamming signals that enhance the security of semantic communication systems. Specifically, the designed Swin Transformer utilizes shifted windows for representation computation, enabling it to efficiently extract and transmit semantic features while reducing the risk of privacy leakage. Artificial noise at the physical layer is employed at the source and relay nodes to degrade the wiretapping ability of the eavesdropper. Simulation results show that, under the artificial noise strategy, the legitimate destination achieves better Euclidean distance, structural similarity index measure (SSIM), and peak signal-to-noise ratio (PSNR) than the eavesdropper, demonstrating the effectiveness of our proposed security strategy. Moreover, the cooperative semantic system with the designed Swin Transformer and jamming strategy outperforms separated source-and-channel coding models in terms of Euclidean Distance, SSIM, and PSNR, while preserving the confidentiality of semantic information during wireless transmission.

External IDs:dblp:conf/iccci/TanNV25