Cell-Free Massive MIMO under a Non-Linear Power Amplifier Consumption Model

TL;DR

This paper proposes a novel approach to minimize power amplifier consumption in Cell-Free Massive MIMO systems, accounting for non-linear power models, leading to significant energy savings and scalable solutions.

Contribution

It introduces a penalty-method-based optimization that reduces amplifier power consumption considering non-linear models, improving scalability and promoting sparsity in power allocation.

Findings

Up to 24% reduction in power consumption.

Scalable solution with better network size handling.

Promotes sparsity in power allocation.

Abstract

Existing works on Cell-Free Massive MIMO primarily focus on optimising system throughput and energy efficiency under high-traffic scenarios with only a limited focus on variable user demand as required by higher network layers. Additionally, existing works only minimise the transmitted power instead of the consumed power at the power amplifier. This work introduces a penalty-method-based approach to minimise the amplifier's power consumption while scaling much better with network size than current solutions and promoting sparsity in the power allocated to each access point. Furthermore, we demonstrate substantial reductions in power consumption (up to 24%) by considering the non-linear power consumption.