Behavioral Modeling of Power Amplifiers with Modern Machine Learning Techniques

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

2 Citations (Scopus)
165 Downloads (Pure)

Abstract

In this study, modern machine learning (ML) methods are proposed to predict the dynamic non-linear behavior of wideband RF power amplifiers (PAs). Neural networks, k-nearest neighbor, and several tree-based ML algorithms are first adapted to handle complex-valued signals and then applied to the PA modeling problem. Their modeling performance is evaluated with measured data from two base station PAs. Gradient boosting is seen to outperform the other ML approaches and to give comparable performance to the generalized memory polynomial (GMP) reference model in terms of both the normalized mean squared error (NMSE) and adjacent channel error power ratio (ACEPR). This is the first study in the open literature to consider modern ML approaches, besides neural networks, for PA behavioral modeling.
Original languageEnglish
Title of host publication2019 IEEE MTT-S International Microwave Conference on 5G Hardware and Systems (IMC-5G)
PublisherIEEE
Number of pages3
ISBN (Electronic)978-1-7281-3143-6
ISBN (Print)978-1-7281-3142-9
DOIs
Publication statusPublished - 2019
Publication typeA4 Article in a conference publication
EventIEEE MTT-S International Microwave Conference on 5G Hardware and System Technologies - Atlanta, United States
Duration: 15 Aug 201916 Aug 2019

Conference

ConferenceIEEE MTT-S International Microwave Conference on 5G Hardware and System Technologies
Country/TerritoryUnited States
CityAtlanta
Period15/08/1916/08/19

Keywords

  • Behavioral modeling
  • Power amplifiers
  • Generalized memory polynomial
  • Machine learning
  • Neural networks
  • Gradient boosting
  • Tree-based approaches
  • Decision tree

Publication forum classification

  • Publication forum level 1

Fingerprint

Dive into the research topics of 'Behavioral Modeling of Power Amplifiers with Modern Machine Learning Techniques'. Together they form a unique fingerprint.

Cite this