Multi-Agent Deep Reinforcement Learning for Collaborative UAV Relay Networks under Jamming Atatcks

2025-12-09

AI Summary

To address the challenge of jointly optimizing throughput, collision avoidance, and interference resilience in UAV swarm relay communications under adversarial jamming, this paper proposes a multi-agent deep reinforcement learning (MARL)-based cooperative optimization method. Leveraging a centralized training with decentralized execution (CTDE) framework, we design a global-state-driven centralized critic and local-observation-driven decentralized policies, enabling agents to autonomously evolve jamming-resilient coordination without explicit programming. The approach dynamically balances jamming suppression and link maintenance, exhibiting emergent resilience. Experimental results demonstrate approximately 50% improvement in aggregate throughput, near-zero collision rate, and substantial performance gains over heuristic baselines under complex, dynamic jamming scenarios.

Technology Category

• Multiagent Systems: Adversarial Agents
• Search and Optimization: Adversarial Search
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

The deployment of Unmanned Aerial Vehicle (UAV) swarms as dynamic communication relays is critical for next-generation tactical networks. However, operating in contested environments requires solving a complex trade-off, including maximizing system throughput while ensuring collision avoidance and resilience against adversarial jamming. Existing heuristic-based approaches often struggle to find effective solutions due to the dynamic and multi-objective nature of this problem. This paper formulates this challenge as a cooperative Multi-Agent Reinforcement Learning (MARL) problem, solved using the Centralized Training with Decentralized Execution (CTDE) framework. Our approach employs a centralized critic that uses global state information to guide decentralized actors which operate using only local observations. Simulation results show that our proposed framework significantly outperforms heuristic baselines, increasing the total system throughput by approximately 50% while simultaneously achieving a near-zero collision rate. A key finding is that the agents develop an emergent anti-jamming strategy without explicit programming. They learn to intelligently position themselves to balance the trade-off between mitigating interference from jammers and maintaining effective communication links with ground users.

Problem

• Develops MARL for UAV relay networks under jamming attacks
• Balances throughput, collision avoidance, and anti-jamming resilience
• Uses CTDE to enable emergent anti-jamming strategies

Innovation

• Multi-Agent Reinforcement Learning with CTDE framework
• Centralized critic guides decentralized actors using local observations
• Agents learn emergent anti-jamming strategy without explicit programming