Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 20 min ago
PreprintCC BY-NC-SA 4.0arXiv

Deep Reinforcement Learning for Blind Passive Beamforming in RIS-Assisted MIMO Systems

  • Oumaïma Bounhar
  • Thomas Chêne
  • Ghaya Rekaya-Ben Othman
  • Oussama Damen
Open at sourcePDFDOICite

Abstract

Reconfigurable Intelligent Surfaces (RIS) are a promising solution for improving millimeter-wave (mmWave) communications by controlling signal propagation. Yet, selecting the optimal RIS configuration remains challenging due to the high-dimensional search space and signaling overhead. We propose a Deep Reinforcement Learning (DRL) framework for blind passive beamforming, where a Deep Q-Learning (DQL) agent learns an adaptive policy to select near-optimal RIS configurations while sounding only few codewords, without channel state information. Results show that our approach outperforms Q-Learning (QL) and conventional baselines in convergence, performance, and overhead reduction.

Related

  1. Microwave Linear Analog Computer (MiLAC) for Simultaneous Active and Passive Beamforming
    Matteo Nerini, Bruno Clerckx
    arXivPreprintCC BY 4.0Full textGeneral mathematicsarXiv
  2. Data-Assimilation-Assisted Reinforcement Learning for Power Grid Control under Load Uncertainty
    Ruoyu Hu, Feng Bao, Zezhong Zhang +2
    arXivPreprintCC BY 4.0Full textOptimisation & controlarXiv
  3. Deep Reinforcement Learning for BD-RIS-Assisted Multi-UAV RSMA Communications
    Aamer Mohamed Huroon, Ali A. Nasir, Daniel B. da Costa
    arXivPreprintarXiv licenceFull textGeneral mathematicsarXiv
  4. NeuroMAS: Multi-Agent Systems as Neural Networks with Joint Reinforcement Learning
    Haoran Lu, Luyang Fang, Wenxuan Zhong +1
    arXivPreprintarXiv licenceFull textStatisticsarXiv
  5. Multi-Agent Reinforcement Learning for Movable Antenna-aided Cell-Free Massive MIMO Systems
    Bokai Xu, Jiayi Zhang, Shuaifei Chen +4
    arXivPreprintarXiv licenceFull textGeneral mathematicsarXiv

Cite

BibTeXRIS
BibTeX
@misc{bounhar2026deep,
  title = {{Deep Reinforcement Learning for Blind Passive Beamforming in RIS-Assisted MIMO Systems}},
  author = {Oumaïma Bounhar and Thomas Chêne and Ghaya Rekaya-Ben Othman and Oussama Damen},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2610.39827},
  eprint = {2610.39827},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.39827},
  note = {CC BY-NC-SA 4.0},
}
RIS
TY  - UNPB
TI  - Deep Reinforcement Learning for Blind Passive Beamforming in RIS-Assisted MIMO Systems
AU  - Oumaïma Bounhar
AU  - Thomas Chêne
AU  - Ghaya Rekaya-Ben Othman
AU  - Oussama Damen
PY  - 2026
DA  - 2026/10/29
PB  - arXiv
DO  - 10.48550/arxiv.2610.39827
UR  - https://arxiv.org/abs/2610.39827
AB  - Reconfigurable Intelligent Surfaces (RIS) are a promising solution for improving millimeter-wave (mmWave) communications by controlling signal propagation. Yet, selecting the optimal RIS configuration remains challenging due to the high-dimensional search space and signaling overhead. We propose a Deep Reinforcement Learning (DRL) framework for blind passive beamforming, where a Deep Q-Learning (DQL) agent learns an adaptive policy to select near-optimal RIS configurations while sounding only few codewords, without channel state information. Results show that our approach outperforms Q-Learning (QL) and conventional baselines in convergence, performance, and overhead reduction.
ER  -