Association-Aware GNN for Precoder Learning in Cell-Free Systems

2026-03-13

AI Summary

This work addresses the challenge of precoder design in cell-free systems, where dynamic user equipment (UE)–access point (AP) association patterns and channel state information must be jointly modeled—a coupling that traditional methods struggle to handle efficiently. To this end, the authors propose an Association-Aware Graph Neural Network (AAGNN), which explicitly embeds UE–AP association states into the graph neural architecture for the first time. By incorporating permutation equivariance to reduce training complexity and integrating attention mechanisms to enhance generalization, the proposed method achieves superior performance with significantly lower training and inference overhead. Experimental results demonstrate that AAGNN outperforms existing learning-based baselines, delivering notable improvements in both precoding efficacy and cross-scenario generalization capability.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Multiagent Systems: Adversarial Agents

Application Category

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

Abstract

Deep learning has been widely recognized as a promising approach for optimizing multi-user multi-antenna precoders in traditional cellular systems. However, a critical distinction between cell-free and cellular systems lies in the flexibility of user equipment (UE)-access point (AP) associations. Consequently, the optimal precoder depends not only on channel state information but also on the dynamic UE-AP association status. In this paper, we propose an association-aware graph neural network (AAGNN) that explicitly incorporates association status into the precoding design. We leverage the permutation equivariance properties of the cell-free precoding policy to reduce the training complexity of AAGNN and employ an attention mechanism to enhance its generalization performance. Simulation results demonstrate that the proposed AAGNN outperforms baseline learning methods in both learning performance and generalization capabilities while maintaining low training and inference complexity.

Problem

• cell-free systems
• precoder learning
• UE-AP association
• graph neural network
• channel state information

Innovation

• association-aware
• graph neural network
• cell-free systems
• precoder learning
• permutation equivariance