Graph Neural Networks for Resource Allocation in Interference-limited Multi-Channel Wireless Networks with QoS Constraints

2025-09-08

AI Summary

To address scalability and theoretical reliability challenges in QoS-constrained resource allocation for interference-limited multi-channel wireless networks, this paper proposes JCPGNN-M: a joint power and subcarrier allocation framework integrating Graph Neural Networks (GNNs) with Lagrangian primal-dual optimization, leveraging eWMMSE structure as a structural prior. The method rigorously enforces minimum rate constraints and guarantees convergence to a stable point—overcoming the lack of theoretical guarantees inherent in purely data-driven approaches. Experiments demonstrate that JCPGNN-M matches eWMMSE in spectral efficiency while achieving significantly faster inference. Moreover, it exhibits strong generalization across varying network scales and robustness under imperfect channel state information (CSI).

Technology Category

• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Search and Optimization: Mixed Discrete/Continuous Search

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: Experiences and lessons learnt from Web-based algorithms and system deployments
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

Meeting minimum data rate constraints is a significant challenge in wireless communication systems, particularly as network complexity grows. Traditional deep learning approaches often address these constraints by incorporating penalty terms into the loss function and tuning hyperparameters empirically. However, this heuristic treatment offers no theoretical convergence guarantees and frequently fails to satisfy QoS requirements in practical scenarios. Building upon the structure of the WMMSE algorithm, we first extend it to a multi-channel setting with QoS constraints, resulting in the enhanced WMMSE (eWMMSE) algorithm, which is provably convergent to a locally optimal solution when the problem is feasible. To further reduce computational complexity and improve scalability, we develop a GNN-based algorithm, JCPGNN-M, capable of supporting simultaneous multi-channel allocation per user. To overcome the limitations of traditional deep learning methods, we propose a principled framework that integrates GNN with a Lagrangian-based primal-dual optimization method. By training the GNN within the Lagrangian framework, we ensure satisfaction of QoS constraints and convergence to a stationary point. Extensive simulations demonstrate that JCPGNN-M matches the performance of eWMMSE while offering significant gains in inference speed, generalization to larger networks, and robustness under imperfect channel state information. This work presents a scalable and theoretically grounded solution for constrained resource allocation in future wireless networks.

Problem

• Meeting minimum data rate QoS constraints in wireless networks
• Reducing computational complexity for multi-channel resource allocation
• Ensuring theoretical convergence and constraint satisfaction in deep learning

Innovation

• GNN-based multi-channel allocation per user
• Lagrangian primal-dual optimization for QoS constraints
• Theoretically convergent framework with GNN integration