Ultra-Reliable Risk-Aggregated Sum Rate Maximization via Model-Aided Deep Learning

2025-09-30

AI Summary

This paper addresses the joint optimization of weighted sum rate (WSR) maximization and ultra-high reliability of user rates in multi-input single-output (MISO) downlink networks—a challenge exacerbated by tail-rate degradation and deep rate outages under channel uncertainty. Method: We propose a conditional value-at-risk (CVaR)-aware optimization framework, establishing a rigorous equivalence between the CVaR-WSR problem and a risk-averse weighted mean squared error (MSE) formulation. This yields a differentiable, unrolled optimization paradigm. Furthermore, we design an α-Robust Graph Neural Network (αRGNN) that learns robust beamforming policies in a model-assisted manner. Contribution/Results: The approach effectively suppresses tail-rate deterioration and completely eliminates deep rate outages for individual users. It maintains competitive ergodic rate performance while achieving 99.999%-level ultra-reliable communication—constituting the first solution that jointly ensures theoretical rigor and deep learning deployability for risk-sensitive wireless resource allocation.

Technology Category

• Computer Vision: Learning & Optimization for CV
• Intelligent Robots: Learning & Optimization for ROB
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

We consider the problem of maximizing weighted sum rate in a multiple-input single-output (MISO) downlink wireless network with emphasis on user rate reliability. We introduce a novel risk-aggregated formulation of the complex WSR maximization problem, which utilizes the Conditional Value-at-Risk (CVaR) as a functional for enforcing rate (ultra)-reliability over channel fading uncertainty/risk. We establish a WMMSE-like equivalence between the proposed precoding problem and a weighted risk-averse MSE problem, enabling us to design a tailored unfolded graph neural network (GNN) policy function approximation (PFA), named α-Robust Graph Neural Network (αRGNN), trained to maximize lower-tail (CVaR) rates resulting from adverse wireless channel realizations (e.g., deep fading, attenuation). We empirically demonstrate that a trained αRGNN fully eliminates per user deep rate fades, and substantially and optimally reduces statistical user rate variability while retaining adequate ergodic performance.

Problem

• Maximizing weighted sum rate in MISO downlink networks
• Enforcing ultra-reliable user rates over channel fading uncertainty
• Reducing statistical user rate variability while maintaining ergodic performance

Innovation

• Risk-aggregated formulation using Conditional Value-at-Risk
• Unfolded graph neural network for precoding optimization
• Alpha-Robust GNN maximizes lower-tail rates reliably