Wavelength-Spatial-Temporal Multiplexing Enabled High-Parallelism Optical Computing
We experimentally realize a wavelength-space-time multiplexed, high-parallelism optical tensor processor leveraging chromatic dispersion in a free-space diffraction grating and demonstrate batch DNN inference at 96.4% accuracy with optical energy consumption of 20 aJ/MAC.
Authors
- Chao Luan (ORCID: https://orcid.org/0000-0002-1458-7741)
- Dirk Englund (ORCID: https://orcid.org/0000-0002-1043-3489)
- Zaijun Chen (ORCID: https://orcid.org/0000-0003-1069-3122)
- Ryan Hamerly (ORCID: https://orcid.org/0000-0003-4976-2236)
- Ronald Davis
Institutions
- Cambridge Electronics (United States) (US)
- University of California, Berkeley (US)
Publication Details
- Published
- 2025-06-29
- DOI
- https://doi.org/10.23919/oecc/psc62146.2025.11110810
- Primary Topic
- Optical Network Technologies
- Type
- article
- Field-Weighted Citation Impact
- 0.00