Can Hardware Distortion Correlation be Neglected When Analyzing Uplink SE in Massive MIMO?

TL;DR

This paper investigates the impact of hardware distortion correlation on uplink spectral efficiency in Massive MIMO systems, finding that the correlation can often be neglected when there are many users, simplifying analysis.

Contribution

It analytically and numerically demonstrates conditions under which hardware distortion correlation can be ignored in Massive MIMO uplink SE analysis.

Findings

Correlation can be neglected with five or more users in Rayleigh fading.

Distortion correlation has minimal impact on SE in typical Massive MIMO scenarios.

Analytical proofs support the approximation's validity in multi-user settings.

Abstract

This paper analyzes how the distortion created by hardware impairments in a multiple-antenna base station affects the uplink spectral efficiency (SE), with focus on Massive MIMO. The distortion is correlated across the antennas, but has been often approximated as uncorrelated to facilitate (tractable) SE analysis. To determine when this approximation is accurate, basic properties of the distortion correlation are first uncovered. Then, we focus on third-order non-linearities and prove analytically and numerically that the correlation can be neglected in the SE analysis when there are many users. In i.i.d. Rayleigh fading with equal signal-to-noise ratios, this occurs when having five users.