Learning-based tracking of AoAs and AoDs in mmWave networks

Hossein S. Ghadikolaei, Hadi Ghauch, Carlo Fischione

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

Abstract

This paper considers a millimeter-wave communication system and proposes an efficient channel estimation scheme with a minimum number of pilots. We model the dynamics of the channel’s second-order statistics by a Markov process and develop a learning framework to obtain these dynamics from an unlabeled set of measured angles of arrival and departure. We then find the optimal precoding and combining vectors for pilot signals. Using these vectors, the transmitter and receiver will sequentially estimate the corresponding angles of departure and arrival, and then refine the pilot precoding and combining vectors to minimize the error of estimating the channel gains.

Original languageEnglish
Title of host publicationmmNets 2018 - Proceedings of the 2nd ACM Workshop on Millimeter Wave Networks and Sensing Systems, Co-located with MobiCom 2018
PublisherAssociation for Computing Machinery
Pages45-50
Number of pages6
ISBN (Electronic)9781450359283
DOIs
Publication statusPublished - 1 Oct 2018
Externally publishedYes
Event2nd ACM Workshop on Millimeter Wave Networks and Sensing Systems, mmNets 2018, , Co-located with MobiCom 2018 - New Delhi, India
Duration: 29 Oct 2018 → …

Publication series

NameProceedings of the Annual International Conference on Mobile Computing and Networking, MOBICOM

Conference

Conference2nd ACM Workshop on Millimeter Wave Networks and Sensing Systems, mmNets 2018, , Co-located with MobiCom 2018
Country/TerritoryIndia
CityNew Delhi
Period29/10/18 → …

Keywords

  • Channel estimation
  • Machine learning
  • Markov decision process
  • Millimeter-wave
  • Tracking

Fingerprint

Dive into the research topics of 'Learning-based tracking of AoAs and AoDs in mmWave networks'. Together they form a unique fingerprint.

Cite this