GPU-Based implementation of pruned artificial neural networks for digital predistortion linearization of wideband power amplifiers

This paper presents a feature selection technique based on l1 regularization to select the most relevant weights of artificial neural networks (ANNs) for digital predistortion (DPD) linearization of wideband radio-frequency (RF) power amplifiers (PAs). The proposed pruning method is applied to the f...

Descripción completa

Detalles Bibliográficos
Autores: Li, Wantao|||0000-0002-2634-6742, Criado Simón, Raúl, Thompson, William, Montoro López, Gabriel|||0000-0002-1328-4175, Chuang, Kevin, Gilabert Pinal, Pere Lluís|||0000-0001-6183-6977
Tipo de recurso: artículo
Fecha de publicación:2025
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/445544
Acceso en línea:https://hdl.handle.net/2117/445544
https://dx.doi.org/10.1109/JMW.2025.3560420
Access Level:acceso abierto
Palabra clave:Radio frequency
Graphics processing units
Linearity
Artificial neural networks
Peak to average power ratio
Throughput
Predistortion
Real-time systems
Wideband
Power generation
Neural network
Power amplifier
Digital predistortion
GPU-based implementation
Predistortion linearization
Output power
Hidden layer
Processing unit
Graphics processing unit
Power efficiency
Relevant weight
Mean power output
Pruning strategy
Peak-to-average power ratio
Model performance
Behavioral model
Parallelization
Iterative learning control
Normalized mean square error
Throughput performance
Artificial neural network model
Orthogonal frequency division multiplexing
Reference signal
Digital signal processing
Parameter vector
Multiple blocks
Load-modulated balanced amplifier
Model-order reduction
id ES_3cdba21f915e8987e1ee6a9b97c8d253
oai_identifier_str oai:upcommons.upc.edu:2117/445544
network_acronym_str ES
network_name_str España
repository_id_str
spelling GPU-Based implementation of pruned artificial neural networks for digital predistortion linearization of wideband power amplifiersLi, Wantao|||0000-0002-2634-6742Criado Simón, RaúlThompson, WilliamMontoro López, Gabriel|||0000-0002-1328-4175Chuang, KevinGilabert Pinal, Pere Lluís|||0000-0001-6183-6977Radio frequencyGraphics processing unitsLinearityArtificial neural networksPeak to average power ratioThroughputPredistortionReal-time systemsWidebandPower generationNeural networkPower amplifierDigital predistortionGPU-based implementationPredistortion linearizationOutput powerHidden layerProcessing unitGraphics processing unitPower efficiencyRelevant weightMean power outputPruning strategyPeak-to-average power ratioModel performanceBehavioral modelParallelizationIterative learning controlNormalized mean square errorThroughput performanceArtificial neural network modelOrthogonal frequency division multiplexingReference signalDigital signal processingParameter vectorMultiple blocksLoad-modulated balanced amplifierModel-order reductionThis paper presents a feature selection technique based on l1 regularization to select the most relevant weights of artificial neural networks (ANNs) for digital predistortion (DPD) linearization of wideband radio-frequency (RF) power amplifiers (PAs). The proposed pruning method is applied to the first hidden layer of a feed-forward real-valued time-delay neural network, commonly used for DPD purposes. In addition, this paper presents the ANN-based DPD implementation using a graphic processing unit (GPU) with compute unified device architecture (CUDA) units. Thanks to the proposed pruning strategy, it is possible to reduce the ANN complexity significantly, thereby achieving a higher data throughput with the GPU-based implementation. The trade-off among RF performance metrics, number of model parameters and throughput of the GPU implementation is evaluated considering the linearization of a high-efficiency pseudo-Doherty load modulated balanced amplifier (LMBA). The linearized PA operating at an RF frequency of 2 GHz delivers a mean output power of 40 dBm with approximately 50% power efficiency when excited with 5G new radio (NR) signals with up to 200 MHz bandwidth and an 8 dB peak-to-average power ratio (PAPR). The real-time GPU implementation of the ANN-based DPD can meet the linearity specifications with a throughput circa 1 GSa/s.This work was supported in part by MCIN/AEI/10.13039/50110001103 under Project PID2020-113832RB-C21; in part by MICIU/AEI/10.13039/501100011033/FEDER, UE, under project PID2023-146245OB-C21; in part by the Government of Catalonia; and in part by European Social Fund under Grant 2021-FI-B-137Peer ReviewedInstitute of Electrical and Electronics Engineers (IEEE)20252025-05-0120252025-11-05journal articlehttp://purl.org/coar/resource_type/c_6501VoRhttp://purl.org/coar/version/c_970fb48d4fbd8a85info:eu-repo/semantics/articleapplication/pdfhttps://hdl.handle.net/2117/445544https://dx.doi.org/10.1109/JMW.2025.3560420reponame:UPCommons. Portal del coneixement obert de la UPCinstname:Universitat Politècnica de Catalunya (UPC)Inglésengopen accesshttp://purl.org/coar/access_right/c_abf2Attribution 4.0 Internationalhttp://creativecommons.org/licenses/by/4.0/info:eu-repo/semantics/openAccessoai:upcommons.upc.edu:2117/4455442026-05-27T15:37:01Z
dc.title.none.fl_str_mv GPU-Based implementation of pruned artificial neural networks for digital predistortion linearization of wideband power amplifiers
title GPU-Based implementation of pruned artificial neural networks for digital predistortion linearization of wideband power amplifiers
spellingShingle GPU-Based implementation of pruned artificial neural networks for digital predistortion linearization of wideband power amplifiers
Li, Wantao|||0000-0002-2634-6742
Radio frequency
Graphics processing units
Linearity
Artificial neural networks
Peak to average power ratio
Throughput
Predistortion
Real-time systems
Wideband
Power generation
Neural network
Power amplifier
Digital predistortion
GPU-based implementation
Predistortion linearization
Output power
Hidden layer
Processing unit
Graphics processing unit
Power efficiency
Relevant weight
Mean power output
Pruning strategy
Peak-to-average power ratio
Model performance
Behavioral model
Parallelization
Iterative learning control
Normalized mean square error
Throughput performance
Artificial neural network model
Orthogonal frequency division multiplexing
Reference signal
Digital signal processing
Parameter vector
Multiple blocks
Load-modulated balanced amplifier
Model-order reduction
title_short GPU-Based implementation of pruned artificial neural networks for digital predistortion linearization of wideband power amplifiers
title_full GPU-Based implementation of pruned artificial neural networks for digital predistortion linearization of wideband power amplifiers
title_fullStr GPU-Based implementation of pruned artificial neural networks for digital predistortion linearization of wideband power amplifiers
title_full_unstemmed GPU-Based implementation of pruned artificial neural networks for digital predistortion linearization of wideband power amplifiers
title_sort GPU-Based implementation of pruned artificial neural networks for digital predistortion linearization of wideband power amplifiers
dc.creator.none.fl_str_mv Li, Wantao|||0000-0002-2634-6742
Criado Simón, Raúl
Thompson, William
Montoro López, Gabriel|||0000-0002-1328-4175
Chuang, Kevin
Gilabert Pinal, Pere Lluís|||0000-0001-6183-6977
author Li, Wantao|||0000-0002-2634-6742
author_facet Li, Wantao|||0000-0002-2634-6742
Criado Simón, Raúl
Thompson, William
Montoro López, Gabriel|||0000-0002-1328-4175
Chuang, Kevin
Gilabert Pinal, Pere Lluís|||0000-0001-6183-6977
author_role author
author2 Criado Simón, Raúl
Thompson, William
Montoro López, Gabriel|||0000-0002-1328-4175
Chuang, Kevin
Gilabert Pinal, Pere Lluís|||0000-0001-6183-6977
author2_role author
author
author
author
author
dc.subject.none.fl_str_mv Radio frequency
Graphics processing units
Linearity
Artificial neural networks
Peak to average power ratio
Throughput
Predistortion
Real-time systems
Wideband
Power generation
Neural network
Power amplifier
Digital predistortion
GPU-based implementation
Predistortion linearization
Output power
Hidden layer
Processing unit
Graphics processing unit
Power efficiency
Relevant weight
Mean power output
Pruning strategy
Peak-to-average power ratio
Model performance
Behavioral model
Parallelization
Iterative learning control
Normalized mean square error
Throughput performance
Artificial neural network model
Orthogonal frequency division multiplexing
Reference signal
Digital signal processing
Parameter vector
Multiple blocks
Load-modulated balanced amplifier
Model-order reduction
topic Radio frequency
Graphics processing units
Linearity
Artificial neural networks
Peak to average power ratio
Throughput
Predistortion
Real-time systems
Wideband
Power generation
Neural network
Power amplifier
Digital predistortion
GPU-based implementation
Predistortion linearization
Output power
Hidden layer
Processing unit
Graphics processing unit
Power efficiency
Relevant weight
Mean power output
Pruning strategy
Peak-to-average power ratio
Model performance
Behavioral model
Parallelization
Iterative learning control
Normalized mean square error
Throughput performance
Artificial neural network model
Orthogonal frequency division multiplexing
Reference signal
Digital signal processing
Parameter vector
Multiple blocks
Load-modulated balanced amplifier
Model-order reduction
description This paper presents a feature selection technique based on l1 regularization to select the most relevant weights of artificial neural networks (ANNs) for digital predistortion (DPD) linearization of wideband radio-frequency (RF) power amplifiers (PAs). The proposed pruning method is applied to the first hidden layer of a feed-forward real-valued time-delay neural network, commonly used for DPD purposes. In addition, this paper presents the ANN-based DPD implementation using a graphic processing unit (GPU) with compute unified device architecture (CUDA) units. Thanks to the proposed pruning strategy, it is possible to reduce the ANN complexity significantly, thereby achieving a higher data throughput with the GPU-based implementation. The trade-off among RF performance metrics, number of model parameters and throughput of the GPU implementation is evaluated considering the linearization of a high-efficiency pseudo-Doherty load modulated balanced amplifier (LMBA). The linearized PA operating at an RF frequency of 2 GHz delivers a mean output power of 40 dBm with approximately 50% power efficiency when excited with 5G new radio (NR) signals with up to 200 MHz bandwidth and an 8 dB peak-to-average power ratio (PAPR). The real-time GPU implementation of the ANN-based DPD can meet the linearity specifications with a throughput circa 1 GSa/s.
publishDate 2025
dc.date.none.fl_str_mv 2025
2025-05-01
2025
2025-11-05
dc.type.none.fl_str_mv journal article
http://purl.org/coar/resource_type/c_6501
VoR
http://purl.org/coar/version/c_970fb48d4fbd8a85
dc.type.openaire.fl_str_mv info:eu-repo/semantics/article
format article
dc.identifier.none.fl_str_mv https://hdl.handle.net/2117/445544
https://dx.doi.org/10.1109/JMW.2025.3560420
url https://hdl.handle.net/2117/445544
https://dx.doi.org/10.1109/JMW.2025.3560420
dc.language.none.fl_str_mv Inglés
eng
language_invalid_str_mv Inglés
language eng
dc.rights.none.fl_str_mv open access
http://purl.org/coar/access_right/c_abf2
Attribution 4.0 International
http://creativecommons.org/licenses/by/4.0/
dc.rights.openaire.fl_str_mv info:eu-repo/semantics/openAccess
rights_invalid_str_mv open access
http://purl.org/coar/access_right/c_abf2
Attribution 4.0 International
http://creativecommons.org/licenses/by/4.0/
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Institute of Electrical and Electronics Engineers (IEEE)
publisher.none.fl_str_mv Institute of Electrical and Electronics Engineers (IEEE)
dc.source.none.fl_str_mv reponame:UPCommons. Portal del coneixement obert de la UPC
instname:Universitat Politècnica de Catalunya (UPC)
instname_str Universitat Politècnica de Catalunya (UPC)
reponame_str UPCommons. Portal del coneixement obert de la UPC
collection UPCommons. Portal del coneixement obert de la UPC
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869406403211296768
score 15,198674

Similares en LA Referencia