Wavelet network with hybrid algorithm to linearize high power amplifiers

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

Abstract

This paper propose a linearizing scheme based on wavelet networks to reduce nonlinear distortion introduced by a high power amplifier over 256QAM signals. Parameters of the proposed linearizer are estimated by using a hybrid algorithm, namely least square and gradient descent. Computer simulation results confirm that once the 256QAM signals are amplified at an input back off level of 0 dB, there is a reduction of 29 dB spectrum re-growth. In addition proposed linearizing scheme has a low complexity and fast convergence.

Original languageEnglish
Title of host publicationAdvances in Artificial Life - 9th European Conference, ECAL 2007, Proceedings
PublisherSpringer Verlag
Pages1016-1023
Number of pages8
ISBN (Print)9783540749127
DOIs
StatePublished - 2007
Externally publishedYes
Event9th European Conference on Advance in Artificial Life, ECAL 2007 - Lisbon, Portugal
Duration: 10 Sep 200714 Sep 2007

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume4648 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference9th European Conference on Advance in Artificial Life, ECAL 2007
Country/TerritoryPortugal
CityLisbon
Period10/09/0714/09/07

Keywords

  • Digital modulation
  • Linearizing
  • Power amplifier
  • Wavelet network

Fingerprint

Dive into the research topics of 'Wavelet network with hybrid algorithm to linearize high power amplifiers'. Together they form a unique fingerprint.

Cite this