Linear Attention for Joint Power Optimization and User-Centric Clustering in Cell-Free Networks

2025-12-19

AI Summary

Joint optimization of access point (AP) clustering and power allocation in user-centric cell-free massive MIMO remains challenging due to difficulties in simultaneously achieving dynamic topology adaptability, pilot contamination mitigation, and low computational complexity. Method: We propose the first end-to-end learning framework that relies solely on user/AP geometric coordinates—introducing a lightweight linear attention mechanism to model spatial user-AP interactions, integrating geometric coordinate encoding and pilot reuse constraint modules, and enabling joint clustering and power allocation without requiring channel state information (CSI). Contribution/Results: The method achieves O(N) complexity, supporting real-time scheduling for thousands of users. It attains near-optimal minimum spectral efficiency and achieves ≥98.3% optimality rate under dynamic scenarios, significantly enhancing robustness and scalability compared to existing approaches.

Technology Category

• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Optimal AP clustering and power allocation are critical in user-centric cell-free massive MIMO systems. Existing deep learning models lack flexibility to handle dynamic network configurations. Furthermore, many approaches overlook pilot contamination and suffer from high computational complexity. In this paper, we propose a lightweight transformer model that overcomes these limitations by jointly predicting AP clusters and powers solely from spatial coordinates of user devices and AP. Our model is architecture-agnostic to users load, handles both clustering and power allocation without channel estimation overhead, and eliminates pilot contamination by assigning users to AP within a pilot reuse constraint. We also incorporate a customized linear attention mechanism to capture user-AP interactions efficiently and enable linear scalability with respect to the number of users. Numerical results confirm the model's effectiveness in maximizing the minimum spectral efficiency and providing near-optimal performance while ensuring adaptability and scalability in dynamic scenarios.

Problem

• Optimizes AP clustering and power allocation in cell-free networks
• Addresses pilot contamination and computational complexity limitations
• Enables scalable, dynamic network adaptation without channel estimation

Innovation

• Lightweight transformer model predicts AP clusters and powers
• Architecture-agnostic to user load, no channel estimation overhead
• Linear attention enables scalability and captures user-AP interactions