Unmanned aerial vehicle (UAV) swarms have become a promising solution to enhance wireless communication in complicated environments. In this paper, we study a UAV swarmassisted secure communication system, where multiple UAVs cooperatively construct an aerial virtual antenna array (AVAA) to deliver confidential information to ground users under the threat of an aerial eavesdropper. we seek to maximize the secrecy energy efficiency (SEE) through the joint optimization of UAV positions and excitation current weights. To handle this highly non-convex problem, we cast it as a Markov decision process (MDP) and propose an energy-penalty based TD3 (EP-TD3) algorithm. In particular, the sum secrecy rate is adopted as the main reward term, while energy consumption and constraint violations are incorporated as penalty terms to guide the learning process toward a better balance of secrecy enhancement and propulsion energy expenditure. Numerical simulation results reveal that the proposed EP-TD3 achieves better overall performance compared with benchmark schemes regarding average sum secrecy rate, total energy consumption, and average SEE.
Chun-Jia Tang, Zhihong Lu, Zhiyu Huang et al.· IEEE International Conferenc...· 0 citations
Accurate channel state information (CSI) prediction is essential for mitigating channel aging and feedback delay in communication systems. This letter proposes an enhanced masked autoencoder (MAE) framework for CSI prediction. Specifically, singular value decomposition (SVD) is first applied to CSI reconstruction and noise suppression, preserving dominant signal components while reducing noise interference. Then, a time–frequency hopping sampling strategy is designed to refine the MAE random masking mechanism, improving uniform subcarrier coverage over the time–frequency grid and enabling the model to learn representative channel features. Furthermore, a multi-scale aligned fusion mechanism is employed to aggregate information across resolutions for capturing diverse multipath dynamics with varying time–frequency scales. These three modules act complementarily on input enhancement, observation coverage, and multi-scale representation learning. Experimental results demonstrate that the proposed model achieves higher prediction accuracy while maintaining an improved accuracy–complexity tradeoff compared with baselines. Our code is publicly available at https://github.com/OpenCommAI/Enhanced_MAE
