EUVIMEDEUVIMED entdecken
Crossref

Deep-Learning-Assisted Orbital Angular Momentum Mode Recovery Under Atmospheric Turbulence for Free-Space Optical Communication

Wenmin Ren, Hao Zhang, Xingyu Mao, Dongye Xu, Zehui Lu, Wei Lin, Shaoxiang Duan, Bo Liu

Journal of Lightwave Technology · 15.09.2025 · 43 (18) · 8647-8656

DOI: 10.1109/jlt.2025.3591694 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Lightwave Technology, erschienen am 15.09.2025. Verfasst von Wenmin Ren und 7 weiteren Autor:innen. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 34 Literaturangaben. Gefördert von: National Natural Science Foundation of China, National Key R&D Program of China, Natural Science Foundation of Tianjin Municipality.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Free-space optical communication systems are susceptible to atmospheric turbulence, resulting in the degradation of beam quality at the receiver. Therefore, the rapid and accurate identification of orbital angular momentum (OAM) modes is crucial for vortex-beam-based optical communication. In this work, an OAM mode recovery approach is proposed and experimentally demonstrated through a tabletop experiment for a distance of 2 meter based on convolutional neural networks (CNNs). Three fusion frameworks, namely Res-UNet, Dense- UNet and Inception-UNet, are designed to recover the distorted vortex beam (VB) pattern. 16 superimposed OAM modes are input into the model as datasets for autonomous learning. Both of the simulation and experimental results show that the Pearson correlation coefficients (PCCs) of the 16 OAM modes propagating through a strong turbulent environment of $C_n^2$ = 10−13 m−2/3 are beyond 98% with good recovery quality. In addition, the recovery qualities of the three proposed fusion frameworks based on 6 types of datasets prove to be better than those of the pure UNet and generative adversarial network. For the OAM-SK-FSOC system, the PCCs of the 10 images restored using our proposed model reach more than 90% with excellent communication performance. Ultimately, to extend the tabletop condition to a real-world scenario, we have experimentally demonstrated the applicability of communication through a 120-meter turbulent environment and investigated the effective methods to enhance OAM communication performance through complex and long-range atmospheric turbulent channels. The proposed CNN-based mode recovery approach paves a new way toward long-distance and low-crosstalk vortex-beam-based communication through atmospheric turbulence channels.

Förderung

  • National Natural Science Foundation of China
  • National Key R&D Program of China
  • Natural Science Foundation of Tianjin Municipality
  • Tianjin Development Program for Innovation and Entrepreneurship
  • Fundamental Research Funds for the Central Universities
  • Nankai University

Zitieren

APA

Ren, W., Zhang, H., Mao, X., Xu, D., Lu, Z., Lin, W., Duan, S., & Liu, B. (2025). Deep-Learning-Assisted Orbital Angular Momentum Mode Recovery Under Atmospheric Turbulence for Free-Space Optical Communication. Journal of Lightwave Technology, 43(18), 8647-8656. https://doi.org/10.1109/jlt.2025.3591694

Vancouver

Ren W, Zhang H, Mao X, Xu D, Lu Z, Lin W, et al. Deep-Learning-Assisted Orbital Angular Momentum Mode Recovery Under Atmospheric Turbulence for Free-Space Optical Communication. Journal of Lightwave Technology. 2025;43(18):8647-8656. doi:10.1109/jlt.2025.3591694

BibTeX
@article{Ren2025,
  author = {Ren, Wenmin and Zhang, Hao and Mao, Xingyu and Xu, Dongye and Lu, Zehui and Lin, Wei and Duan, Shaoxiang and Liu, Bo},
  title = {Deep-Learning-Assisted Orbital Angular Momentum Mode Recovery Under Atmospheric Turbulence for Free-Space Optical Communication},
  journal = {Journal of Lightwave Technology},
  year = {2025},
  volume = {43},
  number = {18},
  pages = {8647--8656},
  doi = {10.1109/jlt.2025.3591694}
}
Kostenlos starten →