Automatic modulation classification using principle composition analysis based features selection

Afan Ali, Fan Yangyu

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

8 Scopus citations

Abstract

Automatic Modulation Classification (AMC) plays an important role in both military and civilian applications. Feature based AMC is used in this paper. Principle Component Analysis (PCA) is employed to reduce dimensions of the feature vector. Two classifiers mainly k-nearest neighbor (KNN) and Support Vector Machine (SVM) are used to investigate the correct classification rate against different SNRs for test signals. Experiments are conducted using data trained at two different SNRs of 15dB and 3dB respectively. Results show that KNN classifier shows better results when data is trained at high SNRs. However, both the classifiers show almost same performance when data is trained at low SNR.

Original languageEnglish
Title of host publicationProceedings of Computing Conference 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages294-296
Number of pages3
ISBN (Electronic)9781509054435
DOIs
StatePublished - 8 Jan 2018
Event2017 SAI Computing Conference 2017 - London, United Kingdom
Duration: 18 Jul 201720 Jul 2017

Publication series

NameProceedings of Computing Conference 2017
Volume2018-January

Conference

Conference2017 SAI Computing Conference 2017
Country/TerritoryUnited Kingdom
CityLondon
Period18/07/1720/07/17

Keywords

  • Automatic Modulation Classification
  • classifier
  • Feature based
  • k-Nearest Neighbor
  • Principle Component Analysis
  • Support Vector Machine

Fingerprint

Dive into the research topics of 'Automatic modulation classification using principle composition analysis based features selection'. Together they form a unique fingerprint.

Cite this