LLM-Aided Joint Secrecy Precoding and Trajectory for RSMA-Based Heterogeneous UAV Networks
Researchers have developed a novel approach for optimizing secure communications in drone networks using Large Language Models (LLMs). This method jointly addresses secrecy, energy efficiency, and trajectory planning for multiple drones serving ground terminals while evading eavesdroppers. The proposed LLM-guided multi-agent reinforcement learning technique, LLM-HeMARL, incorporates LLM-generated expert policies to enable energy-aware and security-driven flight paths for the drones. AI
IMPACT This research could lead to more secure and efficient drone communication systems for various applications.