Low-altitude UAV Friendly-Jamming for Satellite-Maritime Communications via Generative AI-enabled Deep Reinforcement Learning

TL;DR

The paper tackles securing LEO satellite–maritime communications against eavesdropping by deploying a low-altitude UAV as a friendly jammer. It formulates a dynamic, long-horizon multi-objective optimization problem balancing secrecy rate and UAV energy under spectrum-sharing and interference constraints, then reformulates it as an MDP. To solve the NP-hard problem efficiently in dynamic seas, it introduces TransSAC, a transformer-enhanced soft actor-critic algorithm with an MAB-based weight optimization to capture temporal dependencies and diversify objective trade-offs. Simulations show TransSAC achieving near-optimal secrecy rates with reduced UAV energy compared to baselines, and reveal suitable constraint settings for stable performance. The approach provides a scalable, real-time capable framework for secure satellite–maritime links with potential extensions to multi-UAV deployments and predictive eavesdropper positioning.

Abstract

Low Earth orbit (LEO) satellites can be used to assist maritime wireless communications for wide-area data transmission. However, the extensive coverage of LEO satellites, combined with the openness of channels, can cause the communication process to suffer from security risks. This paper presents a LEO satellite-maritime communication system assisted by low-altitude unmanned aerial vehicle (UAV) friendly-jamming to ensure data security at the physical layer. Since such a system requires balancing the conflicting performance metrics of secrecy rate and energy consumption of the UAV to meet evolving scenario demands, we formulate a secure satellite-maritime communication multi-objective optimization problem (SSMCMOP). In order to solve the dynamic and long-term optimization problem, we reformulate it into a Markov decision process. We then propose a transformer-enhanced soft actor-critic (TransSAC) algorithm, which is a generative artificial intelligence-enabled deep reinforcement learning approach to solve the reformulated problem, thus capturing strong temporal correlations and diversely exploring weights. Simulation results demonstrate that the TransSAC algorithm outperforms comparative approaches and algorithms, maximizing the secrecy rate while effectively minimizing the energy consumption of the UAV. Moreover, the results identify more suitable constraints for the system.

Figures (11)

• Figure 1: An LEO satellite-maritime communication system assisted by low-altitude UAV.
• Figure 2: LEO satellite orbit model.
• Figure 3: Vessel coordinate systems.
• Figure 4: The framework of the proposed TransSAC algorithm for solving the SSMCMOP, where a transformer-enhanced learning strategy and an MAB-based weight optimization scheme are integrated into the network to capture global dependencies and explore weights diversely.
• Figure 5: Average reward of different hyper-parameters of the TransSAC algorithm.