On Out-of-Band Emissions of Quantized Precoding in Massive MU-MIMO-OFDM

TL;DR

This paper investigates the impact of low-resolution DACs and filtering techniques on out-of-band emissions in massive MU-MIMO-OFDM systems, proposing methods to reduce emissions while balancing error performance.

Contribution

It introduces a combined approach of analog filtering and digital predistortion to mitigate OOB emissions, analyzing the tradeoffs with Bussgang's theorem.

Findings

Analog filtering reduces OOB emissions significantly.

Digital predistortion degrades SINDR but lowers emissions.

Careful tuning balances emission reduction and error performance.

Abstract

We analyze out-of-band (OOB) emissions in the massive multi-user (MU) multiple-input multiple-output (MIMO) downlink. We focus on systems in which the base station (BS) is equipped with low-resolution digital-to-analog converters (DACs) and orthogonal frequency-division multiplexing (OFDM) is used to communicate to the user equipments (UEs) over frequency-selective channels. We demonstrate that analog filtering in combination with simple frequency-domain digital predistortion (DPD) at the BS enables a significant reduction of OOB emissions, but degrades the signal-to-interference-noise-and-distortion ratio (SINDR) at the UEs and increases the peak-to-average power ratio (PAR) at the BS. We use Bussgang's theorem to characterize the tradeoffs between OOB emissions, SINDR, and PAR, and to study the impact of analog filters and DPD on the error-rate performance of the massive MU-MIMO-OFDM downlink. Our results show that by carefully tuning the parameters of the analog filters, one can achieve a significant reduction in OOB emissions with only a moderate degradation of error-rate performance and PAR.