Two-Layer Reinforcement Learning-Assisted Joint Beamforming and Trajectory Optimization for Multi-UAV Downlink Communications
Published in Submitted to IEEE Transactions on Wireless Communications, 2026
This work develops a two-layer learning framework for multi-UAV downlink communications. It combines topology-aware GNN beamforming with MAPPO-based cooperative trajectory planning under a centralized-training decentralized-execution paradigm.
| Links: arXiv | Code |
Recommended citation: R. Wang, E. M. A. Ghoura, O. A. M. Alhussein, et al. (2026). "Two-Layer Reinforcement Learning-Assisted Joint Beamforming and Trajectory Optimization for Multi-UAV Downlink Communications." Submitted to IEEE Transactions on Wireless Communications.
Download Paper