Litcius/Paper detail

Unsupervised Deep Learning for Power Control of Cell-Free Massive MIMO Systems

Yongshun Zhang, Jiayi Zhang, Stefano Buzzi, Huahua Xiao, Bo Ai

2023IEEE Transactions on Vehicular Technology23 citationsDOI

Abstract

The problem of power control for the uplink and the downlink in cell-free massive multiple-input multiple-output (CF mMIMO) systems is considered in this paper. In order to achieve a balance between the conflicting requirements of good performance levels and of low computational complexity, we employ unsupervised learning based on deep learning (DL). The proposed power control strategies can work using as input the large scale fading coefficients only and are capable to obtain very satisfactory performance levels, as compared to conventional, highly complex, optimization methods and to heuristic methodologies. Moreover, they can be used also in a user centric system wherein each mobile station is served by a subset of the active access points.

Topics & Concepts

Telecommunications linkMIMOFadingComputer sciencePower controlHeuristicPower (physics)Artificial intelligenceAlgorithmComputer networkDecoding methodsPhysicsChannel (broadcasting)Quantum mechanicsAdvanced MIMO Systems OptimizationCooperative Communication and Network CodingAdvanced Wireless Communication Technologies
Unsupervised Deep Learning for Power Control of Cell-Free Massive MIMO Systems | Litcius