Deep Learning-Based Power Control for Uplink Cell-Free Massive MIMO Systems

TL;DR

This paper introduces an unsupervised deep learning framework for power control in uplink cell-free massive MIMO systems, achieving high spectral efficiency with low training complexity.

Contribution

It proposes a novel unsupervised deep neural network approach for power control that does not require known optimal solutions, reducing training complexity.

Findings

Outperforms traditional optimization methods in spectral efficiency.

Effective for max-min, max-sum-rate, and max-product optimizations.

Low computational complexity and fast training.

Abstract

In this paper, a general framework for deep learning-based power control methods for max-min, max-product and max-sum-rate optimization in uplink cell-free massive multiple-input multiple-output (CF mMIMO) systems is proposed. Instead of using supervised learning, the proposed method relies on unsupervised learning, in which optimal power allocations are not required to be known, and thus has low training complexity. More specifically, a deep neural network (DNN) is trained to learn the map between fading coefficients and power coefficients within short time and with low computational complexity. It is interesting to note that the spectral efficiency of CF mMIMO systems with the proposed method outperforms previous optimization methods for max-min optimization and fits well for both max-sum-rate and max-product optimizations.