Resource Allocation and Scheduling in Non-coherent User-centric Cell-free MIMO

TL;DR

This paper proposes a novel resource allocation and user-scheduling framework for user-centric cell-free MIMO networks, achieving significant spectral efficiency gains over traditional schemes.

Contribution

It introduces an integrated algorithm combining block coordinate descent, fractional programming, and compressive sensing for user scheduling and beamforming in non-coherent, user-centric MIMO systems.

Findings

Achieves 8-10 times higher spectral efficiency than benchmark schemes.

Quantifies performance loss due to imperfect channel information.

Provides insights into pilot training overhead effects.

Abstract

We study the problem of user-scheduling and resource allocation in distributed multi-user, multiple-input multiple-output (MIMO) networks implementing user-centric clustering and non-coherent transmission. We formulate a weighted sum-rate maximization problem which can provide user proportional fairness. As in this setup, users can be served by many transmitters, user scheduling is particularly difficult. To solve this issue, we use block coordinate descent, fractional programming, and compressive sensing to construct an algorithm that performs user-scheduling and beamforming. Our results show that the proposed framework provides an 8- to 10-fold gain in the long-term user spectral efficiency compared to benchmark schemes such as round-robin scheduling. Furthermore, we quantify the performance loss due to imperfect channel state information and pilot training overhead using a defined area-based pilot-reuse factor.