EUVIMEDEUVIMED entdecken
Crossref

Integrated WDM-compatible optical mode division multiplexing neural network accelerator

Ruoyu Yin, Huifu Xiao, Yongheng Jiang, Xu Han, Pu Zhang, Li Chen, Xudong Zhou, Mingrui Yuan, Guanghui Ren, Arnan Mitchell, Yonghui Tian

Optica · 20.12.2023 · 10 (12) · 1709

DOI: 10.1364/optica.500523 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Optica, online erschienen am 15.12.2023. Verfasst von Ruoyu Yin und 10 weiteren Autor:innen. Die Arbeit wurde bisher 82-mal zitiert (Crossref) und nennt 46 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Australian Research Council.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

On-chip photonic neural networks (PNN) are emerging as an attractive solution for artificial neural networks due to their high computing density, low energy consumption, and compact size. Matrix-vector multiplication (MVM) plays a key role in on-chip PNN, and can achieve high-speed multiply-accumulate operation. Most current schemes implement MVM by adopting wavelength division multiplexing technology to accumulate the power of different wavelengths together. This requires multiple laser sources. Additionally, both positive and negative domain MVM are inevitable for realizing precise PNNs, but because of the innate limitations of light, effective solutions to perform negative value computing are still inadequate. Here, we propose and demonstrate a PNN accelerator based on mode division multiplexing technology to reduce the use of multi-wavelength lasers. We show that it can satisfactorily tackle real-number-field computing (including positive and negative domains) based on a novel, to our knowledge, transformation mapping approach. As a proof-of-concept, we demonstrate a fabricated accelerator for image convolution and letter pattern detection, achieving a computing density of 1.37TOPS/mm 2 under the 22.38 Gbaud modulation rate.

Förderung

  • National Natural Science Foundation of China
  • Australian Research Council

Zitieren

APA

Yin, R., Xiao, H., Jiang, Y., Han, X., Zhang, P., Chen, L., Zhou, X., Yuan, M., Ren, G., Mitchell, A., & Tian, Y. (2023). Integrated WDM-compatible optical mode division multiplexing neural network accelerator. Optica, 10(12), 1709. https://doi.org/10.1364/optica.500523

Vancouver

Yin R, Xiao H, Jiang Y, Han X, Zhang P, Chen L, et al. Integrated WDM-compatible optical mode division multiplexing neural network accelerator. Optica. 2023;10(12):1709. doi:10.1364/optica.500523

BibTeX
@article{Yin2023,
  author = {Yin, Ruoyu and Xiao, Huifu and Jiang, Yongheng and Han, Xu and Zhang, Pu and Chen, Li and Zhou, Xudong and Yuan, Mingrui and Ren, Guanghui and Mitchell, Arnan and Tian, Yonghui},
  title = {Integrated WDM-compatible optical mode division multiplexing neural network accelerator},
  journal = {Optica},
  year = {2023},
  volume = {10},
  number = {12},
  pages = {1709},
  doi = {10.1364/optica.500523}
}
Kostenlos starten →