High-performance networking is essential for Unmanned Aerial Vehicle (UAV) swarms to accomplish complex, coordinated missions. A central challenge in UAV swarm networking is managing concurrent multi-hop transmissions, where traditional protocols often struggle due to routing path conflicts and co-channel interference. To address this, we propose a novel multi-agent reinforcement learning (MARL)-based cross-layer transmission framework that maximizes system throughput by jointly optimizing network-layer routing, link-layer resource allocation, and UAV trajectories. We decouple this complex joint optimization problem and solve it with a routing-prioritized iterative scheme. For the routing sub-problem, an MARL approach is designed for agents to collaboratively plan concurrent routing paths. The non-convex resource allocation and trajectory sub-problems are handled using successive convex approximation (SCA). Experimental results demonstrate that our proposed framework significantly outperforms existing benchmarks in system throughput, end-to-end delay, and packet delivery ratio.
Yang Shen, Bing Li, Rongqing Zhang· IEEE Transactions on Wireles...· 0 citations
Low Earth Orbit (LEO) constellations are required to process increasing volumes of heterogeneous tasks from ground networks. Intermittent inter-satellite links, heterogeneous onboard resources, and time-varying traffic loads make collaborative task offloading difficult for static or reactive strategies. Although multi-agent reinforcement learning (MARL) provides an adaptive solution, existing model-free MARL methods often suffer from slow convergence, insufficient foresight, and limited robustness in dynamic satellite environments. To address these challenges, this paper proposes a World Model-Based Multi-Agent Proximal Policy Optimization (WM-MAPPO) framework for space computing power networks. The offloading problem is formulated as a partially observable multi-agent decision-making process, where LEO satellites make decentralized decisions under incomplete local observations. A predictive world model learns latent transition dynamics of network states and provides future context for proactive planning. Meanwhile, a Transformer-based policy architecture captures inter-agent dependencies and supports cooperative scheduling under centralized training and decentralized execution (CTDE). Simulation results show that WM-MAPPO achieves higher task completion ratios, lower average latency, improved energy efficiency, and stronger robustness than model-free MARL baselines, heuristic methods, Lyapunov-based scheduling, and MINLP-inspired optimization.
Yuqi Cong, Zhiwei Wei, Jiarui Chen et al.· IEEE Transactions on Cogniti...· 0 citations