Frequency-domain digital predistortion for Massive MU-MIMO-OFDM Downlink

Yibo Wu; Ulf Gustavsson; Mikko Valkama; Alexandre Graell I Amat; Henk Wymeersch

doi:10.1109/GLOBECOM48099.2022.10001382

Frequency-domain digital predistortion for Massive MU-MIMO-OFDM Downlink

Yibo Wu
, Ulf Gustavsson
, Mikko Valkama
, Alexandre Graell I Amat
, Henk Wymeersch

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

2 Citations (Scopus)

25 Downloads (Pure)

Abstract

Digital predistortion (DPD) is a method commonly used to compensate for the nonlinear effects of power amplifiers (sPAs). However, the computational complexity of most DPD algorithms becomes an issue in the downlink of massive multi-user (MU) multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM), where potentially up to several hundreds of PAs in the base station (BS) require linearization. In this paper, we propose a convolutional neural network (CNN)-based DPD in the frequency domain, taking place before the precoding, where the dimensionality of the signal space depends on the number of users, instead of the number of BS antennas. Simulation results on generalized memory polynomial (GMP)-based PAs show that the proposed CNN-based DPD can lead to very large complexity savings as the number of BS antenna increases at the expense of a small increase in power to achieve the same symbol error rate (SER).

Original language	English
Title of host publication	2022 IEEE Global Communications Conference, GLOBECOM 2022 - Proceedings
Publisher	IEEE
Pages	579-584
Number of pages	6
ISBN (Electronic)	9781665435406
ISBN (Print)	9781665435413
DOIs	https://doi.org/10.1109/GLOBECOM48099.2022.10001382
Publication status	Published - 2022
Publication type	A4 Article in conference proceedings
Event	IEEE Global Communications Conference - Virtual, Online, Brazil Duration: 4 Dec 2022 → 8 Dec 2022

Conference

Conference	IEEE Global Communications Conference
Country/Territory	Brazil
City	Virtual, Online
Period	4/12/22 → 8/12/22

Funding

This work was supported by the Swedish Foundation for Strategic Research (SSF), grant no. ID19-0021. The authors would like to thank Fan Jiang at Chalmers University of Technology for fruitful discussions.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Publication forum classification

Publication forum level 1

ASJC Scopus subject areas

Artificial Intelligence
Computer Networks and Communications
Hardware and Architecture
Signal Processing
Renewable Energy, Sustainability and the Environment
Safety, Risk, Reliability and Quality

Access to Document

10.1109/GLOBECOM48099.2022.10001382

Frequency_domain_DPD
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Accepted author manuscript, 468 KBLicence: Unspecified

https://urn.fi/URN:NBN:fi:tuni-202302152410Licence: Unspecified

Cite this

@inproceedings{e1df95dd20f94a1aa8c2f9bc3344de34,

title = "Frequency-domain digital predistortion for Massive MU-MIMO-OFDM Downlink",

abstract = "Digital predistortion (DPD) is a method commonly used to compensate for the nonlinear effects of power amplifiers (sPAs). However, the computational complexity of most DPD algorithms becomes an issue in the downlink of massive multi-user (MU) multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM), where potentially up to several hundreds of PAs in the base station (BS) require linearization. In this paper, we propose a convolutional neural network (CNN)-based DPD in the frequency domain, taking place before the precoding, where the dimensionality of the signal space depends on the number of users, instead of the number of BS antennas. Simulation results on generalized memory polynomial (GMP)-based PAs show that the proposed CNN-based DPD can lead to very large complexity savings as the number of BS antenna increases at the expense of a small increase in power to achieve the same symbol error rate (SER).",

author = "Yibo Wu and Ulf Gustavsson and Mikko Valkama and \{I Amat\}, \{Alexandre Graell\} and Henk Wymeersch",

note = "Funding Information: This work was supported by the Swedish Foundation for Strategic Research (SSF), grant no. ID19-0021. The authors would like to thank Fan Jiang at Chalmers University of Technology for fruitful discussions. Publisher Copyright: {\textcopyright} 2022 IEEE.; IEEE Global Communications Conference ; Conference date: 04-12-2022 Through 08-12-2022",

year = "2022",

doi = "10.1109/GLOBECOM48099.2022.10001382",

language = "English",

isbn = "9781665435413",

pages = "579--584",

booktitle = "2022 IEEE Global Communications Conference, GLOBECOM 2022 - Proceedings",

publisher = "IEEE",

address = "United States",

}

TY - GEN

T1 - Frequency-domain digital predistortion for Massive MU-MIMO-OFDM Downlink

AU - Wu, Yibo

AU - Gustavsson, Ulf

AU - Valkama, Mikko

AU - I Amat, Alexandre Graell

AU - Wymeersch, Henk

N1 - Funding Information: This work was supported by the Swedish Foundation for Strategic Research (SSF), grant no. ID19-0021. The authors would like to thank Fan Jiang at Chalmers University of Technology for fruitful discussions. Publisher Copyright: © 2022 IEEE.

PY - 2022

Y1 - 2022

N2 - Digital predistortion (DPD) is a method commonly used to compensate for the nonlinear effects of power amplifiers (sPAs). However, the computational complexity of most DPD algorithms becomes an issue in the downlink of massive multi-user (MU) multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM), where potentially up to several hundreds of PAs in the base station (BS) require linearization. In this paper, we propose a convolutional neural network (CNN)-based DPD in the frequency domain, taking place before the precoding, where the dimensionality of the signal space depends on the number of users, instead of the number of BS antennas. Simulation results on generalized memory polynomial (GMP)-based PAs show that the proposed CNN-based DPD can lead to very large complexity savings as the number of BS antenna increases at the expense of a small increase in power to achieve the same symbol error rate (SER).

AB - Digital predistortion (DPD) is a method commonly used to compensate for the nonlinear effects of power amplifiers (sPAs). However, the computational complexity of most DPD algorithms becomes an issue in the downlink of massive multi-user (MU) multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM), where potentially up to several hundreds of PAs in the base station (BS) require linearization. In this paper, we propose a convolutional neural network (CNN)-based DPD in the frequency domain, taking place before the precoding, where the dimensionality of the signal space depends on the number of users, instead of the number of BS antennas. Simulation results on generalized memory polynomial (GMP)-based PAs show that the proposed CNN-based DPD can lead to very large complexity savings as the number of BS antenna increases at the expense of a small increase in power to achieve the same symbol error rate (SER).

U2 - 10.1109/GLOBECOM48099.2022.10001382

DO - 10.1109/GLOBECOM48099.2022.10001382

M3 - Conference contribution

AN - SCOPUS:85146933193

SN - 9781665435413

SP - 579

EP - 584

BT - 2022 IEEE Global Communications Conference, GLOBECOM 2022 - Proceedings

PB - IEEE

T2 - IEEE Global Communications Conference

Y2 - 4 December 2022 through 8 December 2022

ER -

Frequency-domain digital predistortion for Massive MU-MIMO-OFDM Downlink

Abstract

Conference

Funding

UN SDGs

Publication forum classification

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this