MAGRPO: Accelerated MARL Training for Fluid Antenna-Assisted Wireless Network Optimization

2026-04-19

AI Summary

This study addresses the joint non-convex optimization of antenna placement, beamforming, and power allocation in fluid-antenna-assisted wireless networks without inter-base-station coordination. The problem is formulated as a decentralized partially observable Markov decision process (Dec-POMDP), and a multi-agent group-relative policy optimization algorithm (MAGRPO) is proposed under the centralized training with decentralized execution (CTDE) paradigm. By replacing the conventional critic network with a group-relative advantage estimator and leveraging parameter sharing, MAGRPO reduces computational complexity by nearly 50% and derives an upper bound on the variance of cumulative rewards. Experimental results demonstrate that fluid-antenna networks achieve several-fold improvements in sum rate over fixed-antenna systems, and MAGRPO attains performance comparable to MAPPO while reducing training time by 30%–40%.

Technology Category

• Multiagent Systems: Multiagent Learning
• Planning, Routing, and Scheduling: Planning with Markov Models (MDPs, POMDPs)
• Search and Optimization: Non-convex Optimization

Application Category

• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Fluid antenna system (FAS) becomes a promising paradigm for next-generation wireless networks, which enables position-flexible antenna elements that can dynamically adjust to more favorable channel conditions. However, the optimization of fluid antenna (FA) positions, beamforming, and power allocation in FA-assisted wireless networks is challenging, due to the non-convexity and the lack of base station (BS) coordination. In this paper, we first formulate this challenging optimization problem as a decentralized partially observable Markov decision process, and then propose a multi-agent group relative policy optimization (MAGRPO) algorithm under the centralized training decentralized execution (CTDE) paradigm. Compared with multi-agent proximal policy optimization (MAPPO), MAGRPO replaces the critic network with group relative advantage estimation. This design reduces computational complexity by nearly half under parameter sharing. Furthermore, we derive a variance upper bound of the cumulative reward, which scales with network parameters, e.g., the number of BSs, users, and FAs. Simulation results show that compared with wireless networks with fixed antenna positions, FA-assisted wireless networks achieve multiple-fold sum-rate enhancement. Moreover, the proposed MAGRPO attains sum-rates comparable to those of MAPPO in testing, while reducing training time by $30\% \sim 40\%$.

Problem

• fluid antenna system
• wireless network optimization
• multi-agent reinforcement learning
• non-convex optimization
• base station coordination

Innovation

• Fluid Antenna System
• Multi-Agent Reinforcement Learning
• Group Relative Policy Optimization
• Decentralized Optimization
• Centralized Training Decentralized Execution