DeepFP: Deep-Unfolded Fractional Programming for MIMO Beamforming

2026-01-06

IEEE Transactions on Communications

AI Summary

This work addresses the high computational complexity and step-size selection challenges in weighted sum-rate beamforming for multi-cell MIMO wireless networks by proposing a novel approach that integrates deep unfolding networks with an improved fractional programming framework, termed FastFP. For the first time, a deep unfolding architecture is incorporated into the FastFP framework to enable adaptive step-size optimization, thereby eliminating the need for large-scale matrix inversions and manual tuning of Lagrange multipliers. The proposed method preserves theoretical convergence guarantees while significantly enhancing data-driven efficiency. Experimental results demonstrate that, compared to learning-based approaches built upon the WMMSE algorithm, the proposed scheme substantially reduces computational overhead without compromising beamforming performance.

Technology Category

• Search and Optimization: Learning to Search
• Machine Learning: Optimization
• Natural Language Processing: Learning & Optimization for NLP

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

This work proposes a mixed learning-based and optimization-based approach to the weighted-sum-rates beamforming problem in a multiple-input multiple-output (MIMO) wireless network. The conventional methods, i.e., the fractional programming (FP) method and the weighted minimum mean square error (WMMSE) algorithm, can be computationally demanding for two reasons: (i) they require inverting a sequence of matrices whose sizes are proportional to the number of antennas; (ii) they require tuning a set of Lagrange multipliers to account for the power constraints. The recently proposed method called the reduced WMMSE addresses the above two issues for a single cell. In contrast, for the multicell case, another recent method called the FastFP eliminates the large matrix inversion and the Lagrange multipliers by using an improved FP technique, but the update stepsize in the FastFP can be difficult to decide. As such, we propose integrating the deep unfolding network into the FastFP for the stepsize optimization. Numerical experiments show that the proposed method is much more efficient than the learning method based on the WMMSE algorithm.

Problem

• MIMO beamforming
• weighted-sum-rate
• fractional programming
• multicell networks
• computational complexity

Innovation

• Deep Unfolding
• Fractional Programming
• MIMO Beamforming
• Stepsize Optimization
• FastFP