EUVIMEDEUVIMED entdecken
Crossref

Baseband-Free End-to-End Communication System Based on Diffractive Deep Neural Network

Xiaokun Teng, Wankai Tang, Xiao Li, Shi Jin

IEEE Transactions on Wireless Communications · 2026 · 25 · 7649-7664

DOI: 10.1109/twc.2025.3632886 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Wireless Communications, 2026. Verfasst von Xiaokun Teng und 3 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 42 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Key Technologies Research and Development Program of Jiangsu, Postgraduate Research and Practice Innovation Program of Jiangsu-Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Diffractive deep neural network (D2NN), also known as reconfigurable intelligent metasurface based deep neural networks (Rb-DNNs) or stacked intelligent metasurfaces (SIMs) in wireless communications, has emerged as a promising signal processing paradigm that enables computing-by-propagation. However, existing architectures are limited to implementing specific functions such as precoding and combining, while still relying on digital baseband modules for other essential tasks like modulation and detection. In this work, we propose a baseband-free end-to-end (BBF-E2E) wireless communication system where modulation, beamforming, and detection are jointly realized through the propagation of electromagnetic (EM) waves. The BBF-E2E system employs D2NNs at both the transmitter and the receiver, forming an autoencoder architecture optimized within a complex-valued neural network (CVNN) framework. The transmission coefficients of each metasurface layer are trained using the mini-batch stochastic gradient descent (SGD) to minimize the cross-entropy loss. To reduce computational complexity during diffraction calculation, the angular spectrum method (ASM) is adopted over the Rayleigh–Sommerfeld formula. Extensive simulations demonstrate that BBF-E2E achieves robust symbol transmission under various channel conditions with significantly reduced hardware requirements. In particular, the proposed system matches the performance of a conventional multi-antenna system with 81 RF chains while requiring only a single RF chain and 1024 metasurface elements. These results highlight the potential of this wave-domain neural computing paradigm to replace digital baseband modules in future wireless transceivers.

Förderung

  • National Natural Science Foundation of China
  • Key Technologies Research and Development Program of Jiangsu
  • Postgraduate Research and Practice Innovation Program of Jiangsu-Province
  • Start-Up Research Fund of Southeast University

Zitieren

APA

Teng, X., Tang, W., Li, X., & Jin, S. (2026). Baseband-Free End-to-End Communication System Based on Diffractive Deep Neural Network. IEEE Transactions on Wireless Communications, 25, 7649-7664. https://doi.org/10.1109/twc.2025.3632886

Vancouver

Teng X, Tang W, Li X, Jin S. Baseband-Free End-to-End Communication System Based on Diffractive Deep Neural Network. IEEE Transactions on Wireless Communications. 2026;25:7649-7664. doi:10.1109/twc.2025.3632886

BibTeX
@article{Teng2026,
  author = {Teng, Xiaokun and Tang, Wankai and Li, Xiao and Jin, Shi},
  title = {Baseband-Free End-to-End Communication System Based on Diffractive Deep Neural Network},
  journal = {IEEE Transactions on Wireless Communications},
  year = {2026},
  volume = {25},
  pages = {7649--7664},
  doi = {10.1109/twc.2025.3632886}
}
Kostenlos starten →