Recursive Neural Network With Phase-Normalization for Modeling and Linearization of RF Power Amplifiers

Arne Fischer-Bühner; Lauri Anttila; Manil Dev Gomony; Mikko Valkama

doi:10.1109/LMWT.2024.3393859

Recursive Neural Network With Phase-Normalization for Modeling and Linearization of RF Power Amplifiers

Arne Fischer-Bühner, Lauri Anttila, Manil Dev Gomony, Mikko Valkama

Electrical Engineering

Research output: Contribution to journal › Article › Scientific › peer-review

1 Citation (Scopus)

7 Downloads (Pure)

Abstract

This letter presents a novel phase-normalized recurrent neural network (PN-RNN) to linearize radio frequency (RF) power amplifiers (PAs) in high-bandwidth communication systems with significant memory effects. The proposed approach builds on proper phase alignment of the internal hidden variables in the recursive processing system. The provided RF measurement-based modeling and digital predistortion (DPD) results at 1.8 and 3.5 GHz demonstrate a significantly improved modeling capacity and predistortion ability when applying phase normalization, confirming the validity of the proposed approach.

Original language	English
Pages (from-to)	809-812
Number of pages	4
Journal	IEEE Microwave and Wireless Technology Letters
Volume	34
Issue number	6
DOIs	https://doi.org/10.1109/LMWT.2024.3393859
Publication status	Published - Jun 2024
Publication type	A1 Journal article-refereed

Keywords

Artificial neural networks
Baseband
Behavioral modeling
digital predistortion (DPD)
Logic gates
power amplifier (PA)
Predistortion
Radio frequency
recurrent neural network (RNN)
Training
Vectors

Publication forum classification

Publication forum level 2

ASJC Scopus subject areas

Electrical and Electronic Engineering
Condensed Matter Physics

Access to Document

10.1109/LMWT.2024.3393859Licence: CC BY

Recursive_Neural_NetworkFinal published version, 10.1 MBLicence: CC BY

https://urn.fi/URN:NBN:fi:tuni-202406117110Licence: CC BY

Cite this

@article{ff9b730f57fe4a73a1951905b259e3b3,

title = "Recursive Neural Network With Phase-Normalization for Modeling and Linearization of RF Power Amplifiers",

abstract = "This letter presents a novel phase-normalized recurrent neural network (PN-RNN) to linearize radio frequency (RF) power amplifiers (PAs) in high-bandwidth communication systems with significant memory effects. The proposed approach builds on proper phase alignment of the internal hidden variables in the recursive processing system. The provided RF measurement-based modeling and digital predistortion (DPD) results at 1.8 and 3.5 GHz demonstrate a significantly improved modeling capacity and predistortion ability when applying phase normalization, confirming the validity of the proposed approach.",

keywords = "Artificial neural networks, Baseband, Behavioral modeling, digital predistortion (DPD), Logic gates, power amplifier (PA), Predistortion, Radio frequency, recurrent neural network (RNN), Training, Vectors",

author = "Arne Fischer-B{\"u}hner and Lauri Anttila and Gomony, {Manil Dev} and Mikko Valkama",

note = "Publisher Copyright: Authors",

year = "2024",

month = jun,

doi = "10.1109/LMWT.2024.3393859",

language = "English",

volume = "34",

pages = "809--812",

number = "6",

}

TY - JOUR

T1 - Recursive Neural Network With Phase-Normalization for Modeling and Linearization of RF Power Amplifiers

AU - Fischer-Bühner, Arne

AU - Anttila, Lauri

AU - Gomony, Manil Dev

AU - Valkama, Mikko

N1 - Publisher Copyright: Authors

PY - 2024/6

Y1 - 2024/6

N2 - This letter presents a novel phase-normalized recurrent neural network (PN-RNN) to linearize radio frequency (RF) power amplifiers (PAs) in high-bandwidth communication systems with significant memory effects. The proposed approach builds on proper phase alignment of the internal hidden variables in the recursive processing system. The provided RF measurement-based modeling and digital predistortion (DPD) results at 1.8 and 3.5 GHz demonstrate a significantly improved modeling capacity and predistortion ability when applying phase normalization, confirming the validity of the proposed approach.

AB - This letter presents a novel phase-normalized recurrent neural network (PN-RNN) to linearize radio frequency (RF) power amplifiers (PAs) in high-bandwidth communication systems with significant memory effects. The proposed approach builds on proper phase alignment of the internal hidden variables in the recursive processing system. The provided RF measurement-based modeling and digital predistortion (DPD) results at 1.8 and 3.5 GHz demonstrate a significantly improved modeling capacity and predistortion ability when applying phase normalization, confirming the validity of the proposed approach.

KW - Artificial neural networks

KW - Baseband

KW - Behavioral modeling

KW - digital predistortion (DPD)

KW - Logic gates

KW - power amplifier (PA)

KW - Predistortion

KW - Radio frequency

KW - recurrent neural network (RNN)

KW - Training

KW - Vectors

U2 - 10.1109/LMWT.2024.3393859

DO - 10.1109/LMWT.2024.3393859

M3 - Article

AN - SCOPUS:85192184997

SN - 2771-957X

VL - 34

SP - 809

EP - 812

JO - IEEE Microwave and Wireless Technology Letters

JF - IEEE Microwave and Wireless Technology Letters

IS - 6

ER -

Recursive Neural Network With Phase-Normalization for Modeling and Linearization of RF Power Amplifiers

Abstract

Keywords

Publication forum classification

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this