DNN-Enabled Multi-User Beamforming for Throughput Maximization under Adjustable Fairness

2026-02-17

AI Summary

This work addresses the non-convex, multi-objective optimization challenge in wireless communication systems that arises from balancing user fairness against aggregate throughput, a problem whose complexity escalates with network scale. The authors propose an unsupervised learning approach based on the Wireless Transformer (WiT), which integrates fairness constraints into an end-to-end deep learning framework via Lagrange multipliers. By coupling this architecture with a dual ascent algorithm, the method automatically tunes the multipliers to maximize throughput under controllable fairness guarantees. Notably, the approach operates without labeled data and efficiently approximates the Pareto frontier, enabling flexible trade-offs between fairness and system performance in multi-user scenarios. Experimental results demonstrate significant improvements over existing state-of-the-art solutions.

Technology Category

• Search and Optimization: Non-convex Optimization
• Machine Learning: Optimization
• Computer Vision: Learning & Optimization for CV

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Fairness, privacy, and diversity in economic environments

Abstract

Ensuring user fairness in wireless communications is a fundamental challenge, as balancing the trade-off between fairness and sum rate leads to a non-convex, multi-objective optimization whose complexity grows with network scale. To alleviate this conflict, we propose an optimization-based unsupervised learning approach based on the wireless transformer (WiT) architecture that learns from channel state information (CSI) features. We reformulate the trade-off by combining the sum rate and fairness objectives through a Lagrangian multiplier, which is updated automatically via a dual-ascent algorithm. This mechanism allows for a controllable fairness constraint while simultaneously maximizing the sum rate, effectively realizing a trace on the Pareto front between two conflicting objectives. Our findings show that the proposed approach offers a flexible solution for managing the trade-off optimization under prescribed fairness.

Problem

• fairness
• throughput maximization
• multi-user beamforming
• non-convex optimization
• Pareto front

Innovation

• wireless transformer
• unsupervised learning
• multi-user beamforming
• fairness-throughput trade-off
• Pareto optimization