An Optimization-Based User Scheduling Framework for mmWave Massive MU-MIMO Systems

TL;DR

This paper introduces a flexible optimization-based user scheduling framework for mmWave massive MU-MIMO systems that improves performance over existing methods by effectively selecting user groups using realistic channel models.

Contribution

It presents a novel, adaptable scheduling framework that solves a nonconvex optimization problem with forward-backward splitting, supporting various cost functions and resource constraints.

Findings

Outperforms existing scheduling methods in simulations

Closely approaches the performance of exhaustive search

Demonstrates effectiveness with realistic mmWave channel models

Abstract

We propose a novel user equipment (UE) scheduling framework for millimeter-wave (mmWave) massive multiuser (MU) multiple-input multiple-output (MIMO) wireless systems. Our framework determines (sub)sets of UEs that should transmit simultaneously in a given time slot by approximately solving a nonconvex optimization problem using forward-backward splitting. Our UE scheduling framework is flexible in the sense that it (i) supports a variety of cost functions, including post-equalization mean square error and sum rate, and (ii) enables precise control over the minimum and maximum number of resources the UEs should occupy. We demonstrate the efficacy of our framework using realistic mmWave channel vectors generated with a commercial ray-tracer. We show that our UE scheduler outperforms a range of existing scheduling methods and closely approaches the performance of an exhaustive search.