Enhancing Least Square Channel Estimation Using Deep Learning

Abdul Karim Gizzini, Marwa Chafii, Ahmad Nimr, Gerhard Fettweis

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Least square (LS) channel estimation employed in various communications systems suffers from performance degradation especially in low signal-to-noise ratio (SNR) regions. This is due to the noise enhancement in the LS estimation process. Minimum mean square error (MMSE) takes into consideration the noise effect and achieves better performance than LS with higher complexity. This paper proposes to correct the LS estimation error using deep learning (DL). Simulation results show that the proposed DL-based schemes perform better than both LS and MMSE channel estimation scheme, with less complexity than accurate MMSE.

Original languageEnglish (US)
Title of host publication2020 IEEE 91st Vehicular Technology Conference, VTC Spring 2020 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728152073
DOIs
StatePublished - May 2020
Event91st IEEE Vehicular Technology Conference, VTC Spring 2020 - Antwerp, Belgium
Duration: May 25 2020May 28 2020

Publication series

NameIEEE Vehicular Technology Conference
Volume2020-May
ISSN (Print)1550-2252

Conference

Conference91st IEEE Vehicular Technology Conference, VTC Spring 2020
Country/TerritoryBelgium
CityAntwerp
Period5/25/205/28/20

Keywords

  • Channel estimation
  • Channel estimation; Deep learning
  • DNN
  • DNN
  • Deep learning
  • LS
  • LS
  • MMSE
  • MMSE

ASJC Scopus subject areas

  • Computer Science Applications
  • Electrical and Electronic Engineering
  • Applied Mathematics

Fingerprint

Dive into the research topics of 'Enhancing Least Square Channel Estimation Using Deep Learning'. Together they form a unique fingerprint.

Cite this