EUVIMEDEUVIMED entdecken
Crossref

Optical Reflecting Intelligent Surface-Assisted Secure Visible Light Communication: An Inverse Pre-Reflection Model-Driven Deep Reinforcement Learning Approach

Sicong Liu, Hong Zhang, Linqi Song, Zhu Han

IEEE Transactions on Wireless Communications · 2026 · 25 · 19162-19178

DOI: 10.1109/twc.2026.3708130 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Wireless Communications, 2026. Verfasst von Sicong Liu und 3 weiteren Autor:innen. Die Arbeit nennt 48 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Basic and Applied Basic Research Foundation of Guangdong Province, Shenzhen Science and Technology Program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Visible light communication (VLC) is emerging as a promising technology for next-generation networks, attracting considerable attention due to its potential for high-speed and secure indoor communication. To cope with the open visible light channel exposed to wiretapping, integrating optical reflecting intelligent surface (ORIS) into VLC systems enables dynamic manipulation of signal reflection paths, thereby altering channel states for legitimate users or eavesdroppers to enhance communication security. Consequently, this paper investigates ORIS-assisted multi-user secure VLC scenarios. Specifically, a deep reinforcement learning (DRL) based scheme is proposed, aiming to maximize the system utility in terms of the secrecy rate, bit error rate, and energy consumption by jointly optimizing the beamforming pattern of the LEDs and the orientations of the ORIS. To overcome the challenges of traditional data-driven learning in high-dimensional continuous spaces and time-varying complex environments, an inverse pre-reflection model-driven DRL scheme is devised. It can effectively address the sparse reward issue in the optimization of the ORIS mirror orientation, providing a more precise guidance for the learning and exploration of the DRL agent. Furthermore, to better balance the quality-of-service among multiple users and try to avoid service outage, an outage-aware DRL-based approach is proposed to reduce the outage probability. Simulation results demonstrate that the proposed model-driven DRL-based scheme can effectively solve the problem of multiuser anti-eavesdropping VLC in complex environments, and achieve superior performance over some typical baseline schemes.

Förderung

  • National Natural Science Foundation of China
  • Basic and Applied Basic Research Foundation of Guangdong Province
  • Shenzhen Science and Technology Program
  • 2025 Software Key RDI Project, IIT Department of Fujian Province, China
  • Research Grants Council of Hong Kong, SAR
  • NSF
  • Japan Science and Technology Agency Adopting Sustainable Partnerships for Innovative Research Ecosystem
  • University of Houston

Zitieren

APA

Liu, S., Zhang, H., Song, L., & Han, Z. (2026). Optical Reflecting Intelligent Surface-Assisted Secure Visible Light Communication: An Inverse Pre-Reflection Model-Driven Deep Reinforcement Learning Approach. IEEE Transactions on Wireless Communications, 25, 19162-19178. https://doi.org/10.1109/twc.2026.3708130

Vancouver

Liu S, Zhang H, Song L, Han Z. Optical Reflecting Intelligent Surface-Assisted Secure Visible Light Communication: An Inverse Pre-Reflection Model-Driven Deep Reinforcement Learning Approach. IEEE Transactions on Wireless Communications. 2026;25:19162-19178. doi:10.1109/twc.2026.3708130

BibTeX
@article{Liu2026,
  author = {Liu, Sicong and Zhang, Hong and Song, Linqi and Han, Zhu},
  title = {Optical Reflecting Intelligent Surface-Assisted Secure Visible Light Communication: An Inverse Pre-Reflection Model-Driven Deep Reinforcement Learning Approach},
  journal = {IEEE Transactions on Wireless Communications},
  year = {2026},
  volume = {25},
  pages = {19162--19178},
  doi = {10.1109/twc.2026.3708130}
}
Kostenlos starten →