Graph-Neural Multi-Agent Coordination for Distributed Access-Point Selection in Cell-Free Massive MIMO

2026-02-20

AI Summary

This work addresses the access point selection (APS) problem in Cell-Free Massive MIMO systems, aiming to jointly optimize user spectral efficiency and network power consumption. The authors propose APS-GNN, a distributed multi-agent reinforcement learning framework based on graph neural networks (GNNs). In this approach, each AP–UE link is modeled as an agent that coordinates efficiently through local observation exchange and parameter sharing, while constrained reinforcement learning balances spectral efficiency and power consumption objectives. By innovatively integrating GNNs with multi-agent reinforcement learning, the method employs local reward and cost signals to enable scalable decision-making and leverages imitation learning for policy initialization to enhance training stability. Experimental results demonstrate that, under target spectral efficiency constraints, APS-GNN reduces the number of activated access points by 50–70% compared to heuristic and centralized MARL baselines, while achieving inference latency reductions of one to two orders of magnitude.

Technology Category

• Multiagent Systems: Multiagent Learning
• Search and Optimization: Learning to Search
• Machine Learning: Distributed Machine Learning & Federated Learning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Cell-free massive MIMO (CFmMIMO) systems require scalable and reliable distributed coordination mechanisms to operate under stringent communication and latency constraints. A central challenge is the Access Point Selection (APS) problem, which seeks to determine the subset of serving Access Points (APs) for each User Equipment (UE) that can satisfy UEs'Spectral Efficiency (SE) requirements while minimizing network power consumption. We introduce APS-GNN, a scalable distributed multi-agent learning framework that decomposes APS into agents operating at the granularity of individual AP-UE connections. Agents coordinate via local observation exchange over a novel Graph Neural Network (GNN) architecture and share parameters to reuse their knowledge and experience. APS-GNN adopts a constrained reinforcement learning approach to provide agents with explicit observability of APS'conflicting objectives, treating SE satisfaction as a cost and power reduction as a reward. Both signals are defined locally, facilitating effective credit assignment and scalable coordination in large networks. To further improve training stability and exploration efficiency, the policy is initialized via supervised imitation learning from a heuristic APS baseline. We develop a realistic CFmMIMO simulator and demonstrate that APS-GNN delivers the target SE while activating 50-70% fewer APs than heuristic and centralized Multi-agent Reinforcement Learning (MARL) baselines in different evaluation scenarios. Moreover, APS-GNN achieves one to two orders of magnitude lower inference latency than centralized MARL approaches due to its fully parallel and distributed execution. These results establish APS-GNN as a practical and scalable solution for APS in large-scale CFmMIMO networks.

Problem

• Cell-Free Massive MIMO
• Access Point Selection
• Spectral Efficiency
• Power Consumption
• Distributed Coordination

Innovation

• Graph Neural Network
• Multi-Agent Reinforcement Learning
• Access Point Selection
• Cell-Free Massive MIMO
• Constrained Reinforcement Learning