Leveraging Machine Learning for Millimeter Wave Beamforming in Beyond 5G Networks

Basem M. Elhalawany, Sherief Hashima, Kohei Hatano, Kaishun Wu, Ehab Mahmoud Mohamed

Research output: Contribution to journalArticlepeer-review

13 Citations (Scopus)


Millimeter wave (mmWave) communication has attracted considerable attention as a key technology for the next-generation wireless communications thanks to its exceptional advantages. MmWave leads the way to achieve a high transmission quality with directed narrow beams from source to multiple destinations by adopting different antenna beamforming (BF) techniques, which have a pivotal role in establishing and maintaining robust links. However, realizing such BF gains in practice requires overcoming several challenges, such as severe signal deterioration, hardware constraints, and design complexity. The elevated complexity of configuring mmWave BF vectors encourages researchers to leverage relevant machine learning (ML) techniques for better BF configurations deployment in 5G and beyond. In this article, we summarize mmWave BF strategies employed for future wireless networks. Then, we provide a comprehensive overview of ML techniques plus its applications and promising contributions toward efficient mmWave BF deployment. Furthermore, we discuss mmWave BF's future research directions and challenges. Finally, we discuss a single and concurrent mmWave BF case study by applying multiarmed bandit to confirm the superiority of ML-based methods over conventional ones.

Original languageEnglish
Pages (from-to)1739-1750
Number of pages12
JournalIEEE Systems Journal
Issue number2
Publication statusPublished - Jun 1 2022

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Information Systems
  • Computer Science Applications
  • Computer Networks and Communications
  • Electrical and Electronic Engineering


Dive into the research topics of 'Leveraging Machine Learning for Millimeter Wave Beamforming in Beyond 5G Networks'. Together they form a unique fingerprint.

Cite this