IEEE Transactions on Communications

Physical-Layer Secure Transmission for Semantic Communication Systems

Zijian Cao, Hua Zhang, Le Liang, Jipeng Gan, Shi Jin, Geoffrey Ye Li

Published Sep 16, 2025• 2 citations

Abstract

As a promising paradigm for the sixth-generation (6G) networks, task-oriented semantic communication significantly enhances transmission efficiency. However, it faces complex security challenges, particularly the risk of eavesdropping due to the open nature of wireless channels. To address this issue, we propose a secure semantic communication framework that integrates physical-layer secure beamforming (SBF) to safeguard semantic information from eavesdropping. Specifically, we design an SBF network to generate SBF vectors that focus signal beams on legitimate users to enhance signal power while directing designed artificial noise toward potential eavesdroppers to strengthen jamming. To further improve system adaptability across varying channel conditions, we introduce attention-based channel-aware modules that dynamically optimize the encoding, decoding, and beamforming processes based on perceived channel state information. Furthermore, task-oriented artificial noise is employed to degrade the task performance of eavesdroppers more effectively. Finally, a stepwise training strategy with task-specific loss functions is employed to jointly optimize the SBF and semantic modules, maximizing the performance gap in downstream tasks between legitimate users and eavesdroppers. The simulation results demonstrate that the proposed approach effectively maintains task performance for legitimate users while significantly suppressing eavesdroppers, outperforming conventional methods.

DOI: 10.1109/tcomm.2025.3610220

Related papers

Channel Capacity Constrained Secure Semantic Communications Against Eavesdropping

Yuxiang Liu, Xu Wang, Zheng Shi +3 more

With the breakthrough advancements in deep neural networks (DNNs) and natural language processing (NLP), semantic communication systems have emerged as a pivotal direction in the evolution of sixth-generation (6 G) networks. However, the broadcast nature of wireless channels inherently introduces user privacy risks, underscoring the urgency for robust security solutions. Existing secure semantic communication approaches predominantly focus on network-layer encryption, which often entails complex operations and carries inherent risks of key leakage. In contrast, physical layer security (PLS) harnesses the intrinsic time-varying nature and randomness of wireless channels, obviating the need for cumbersome cryptographic mechanisms. To address this critical challenge, this paper proposes a channel capacity-constrained secure semantic communication framework. The core design objective is to maximize channel capacity for legitimate channels while confining the ergodic secrecy capacity within a predefined security threshold. By developing a differentiable mutual information estimator, we establish a dual-channel capacity estimation model-encompassing both legitimate users and eavesdroppers-that is jointly optimized with the semantic communication system. Furthermore, we introduce a secure sentence similarity metric to quantitatively assess the security performance of semantic communication. Simulation results validate that despite a slight degradation in transmission accuracy, the proposed scheme achieves a significant enhancement in communication security.

International Conference on Wireless Communications and Signal Processing (WCSP)Oct 23, 20250 citations

Generative AI-Enabled Cooperative Jamming for Secure Semantic Communication

Xingyu Wang, Jiajia Huang, Shunpu Tang +3 more

Semantic communication (SemCom) has recently emerged as a promising approach to enhance communication efficiency by transmitting only important semantic information. In eavesdropping environments, the semantic information may be overheard by the eavesdropper even under poor channel conditions with a powerful semantic decoder. Some advanced secure communication deployments such as autoencoder-enhanced Sem-Com (AE-SemCom) and generative model-enhanced SemCom (GEN-SemCom) have been introduced to efficiently transmit the semantic information. However, these secure solutions require retraining the semantic encoder and decoder, which limits the flexibility and practical deployment. To address this issue, we propose a secure SemCom framework for AE-SemCom and GEN-SemCom without altering the parameters of the semantic encoder and decoder. Specifically, we propose to deploy a cooperative jammer to transmit optimized jamming signal to deteriorate the decoding ability of eavesdropper, while a residual refinement module (RRM) is used at the legitimate receiver to mitigate the impact of these jamming signal. Moreover, to achieve an efficient trade-off between transmission security and reliability, we then introduce a cooperative training strategy by formulating this optimization problem as a two-player game between the jammer and Bob. Finally, we conduct extensive simulations to evaluate the effectiveness of the proposed framework across face recognition dataset. In particular, with a jamming power of 0.3, the eavesdropper experiences up to a 12dB reduction in the peak signal-to-noise ratio (PSNR) of reconstructed images compared to the baselines without the proposed security framework while maintaining high-quality image reconstruction for legitimate users.

IEEE Transactions on CommunicationsJan 1, 20260 citations

Multi-Beam Symbol-Level Secure Communication for Hybrid Near- and Far-Field Communications

Wensheng Deng, Bin Qiu, Wenchi Cheng

This paper introduces a multi-beam secure communication scheme for hybrid near-field and far-field (hNF) scenarios, primarily designed to address the challenges faced by sixth-generation (6G) networks in simultaneously managing near-field and far-field communications. This method significantly reduces the signal quality at eavesdroppers(Eves) while ensuring high-quality reception for legitimate users, effectively enhancing communication security. At the transmitter, this study employs multi-beam symbol-level directional modulation to ensure secure and reliable transmission in a mixed eavesdropping environment. Given the uncertainty about eavesdropper information, the transmission beamforming vectors are specially designed to meet specific symbol-level constraints, thereby ensuring effective reception for legitimate users. Experimental and simulation results demonstrate the effectiveness of our approach in improving secrecy performance, reducing transmission power, and enhancing energy efficiency, offering a practical solution to the security challenges faced by future wireless networks.

International Conference on Ubiquitous Communication (Ucom)Jul 5, 20240 citations

Semantic Communication-Assisted Physical Layer Security Over Fading Wiretap Channels

Xidong Mu, Yuanwei Liu

A novel semantic communication (SC)-assisted secrecy transmission framework is proposed. In particular, the legitimate transmitter (Tx) sends the superimposed semantic and bit stream to the legitimate receiver (Rx), where the information may be eavesdropped by the malicious node (EVE). As the EVE merely has the conventional bit-oriented communication structure, the semantic signal acts as the type of beneficial information-bearing artificial noise (AN), which not only keeps strictly confidential to the EVE but also interferes with the EVE. The ergodic (equivalent) secrecy rate over fading wiretap channels is maximized by jointly optimizing the transmit power, semantic-bit power splitting ratio, and the successive interference cancellation decoding order at the Tx, subject to both the instantaneous peak and long-term average power constraints. To address this non-convex problem, both the optimal and suboptimal algorithms are developed by employing the Lagrangian dual method and the successive convex approximation method, respectively. Numerical results show that the proposed SC-assisted secrecy transmission scheme can significantly enhance the physical layer security compared to the baselines using the conventional bit-oriented communication and no-information-bearing AN. It also shows that the proposed suboptimal algorithm can achieve a near-optimal performance.

arXiv:2402.14581International Conference on CommunicationsJun 9, 202416 citations

Physical Layer Security Techniques for Wireless Communication Systems

Manish Nagpal, Sampa Chau Pattnaik, Paramjit Baxi +3 more

One of the most important issues for secure communication in the physical layer is developing physical layer security techniques to provide confidentiality and data integrity in wireless communication systems. The properties of the physical channel are used in these methodologies to safeguard communication from eavesdropping and some other security threats. This article presents physical layer security methods that can be implemented in wireless communication systems. Artificial Noise (AN) is one of the most popular techniques. The original signal is combined with a generated by the transmitter in such a way that eavesdroppers cannot easily decode the data. Cooperative Anti-jamming: Yet another strategy for jammer localization is to employ cooperative jamming, where multiple friendly users can add noise signals over the signal in an effort to confuse potential eavesdroppers. In addition, additional technologies such as multi-input multi-output beamforming and transmit power control can further improve the security of wireless communication systems. In this regard, strategies based on game theory have been proposed as a viable alternative to enhancing physical layer security. Viewing the interaction between the transmitter, receiver, and eavesdropper as a game allows us to reveal their optimal strategies and security levels.

International Conference on Automation and Computation (AUTOCOM)Mar 4, 20250 citations

Secure Transmission via Beamforming Optimization for NOMA Networks

Yang Cao, Nan Zhao, Yunfei Chen +4 more

Owing to the eminent performance gain, non-orthogonal multiple access (NOMA) has been regarded as a promising approach to achieve high throughput and low latency for the next-generation networks. Nevertheless, potential adversaries can still pose a threat to secure transmission in NOMA systems. Hence, in this article, secure transmission schemes based on beamforming optimization are designed to combat both internal and external eavesdropping for downlink multi-input single-output NOMA networks. Using the transmit beamforming design, we first propose two schemes to protect the privacy of a specific NOMA user in the presence of untrusted users. Then, the sum secrecy rate optimization problem is studied to combat the external eavesdropper through optimizing the precoding vectors, when the channel state information (CSI) is available. In addition, two secure beamforming strategies with the help of artificial noise and inter-user interference are proposed without eavesdropping CSI, respectively. Simulation results are presented to demonstrate the effectiveness of the proposed schemes, and some open issues are discussed as well.

IEEE Wireless CommunicationsAug 22, 201970 citations