A large scale photonic matrix processor enabled by charge accumulation
dc.contributor.author | Brückerhoff-Plückelmann, F | |
dc.contributor.author | Bente, I | |
dc.contributor.author | Wendland, D | |
dc.contributor.author | Feldmann, J | |
dc.contributor.author | Wright, CD | |
dc.contributor.author | Bhaskaran, H | |
dc.contributor.author | Pernice, W | |
dc.date.accessioned | 2022-11-17T10:27:39Z | |
dc.date.issued | 2022-10-28 | |
dc.date.updated | 2022-11-16T17:23:04Z | |
dc.description.abstract | Integrated neuromorphic photonic circuits aim to power complex artificial neural networks (ANNs) in an energy and time efficient way by exploiting the large bandwidth and the low loss of photonic structures. However, scaling photonic circuits to match the requirements of modern ANNs still remains challenging. In this perspective, we give an overview over the usual sizes of matrices processed in ANNs and compare them with the capability of existing photonic matrix processors. To address shortcomings of existing architectures, we propose a time multiplexed matrix processing scheme which virtually increases the size of a physical photonic crossbar array without requiring any additional electrical post-processing. We investigate the underlying process of time multiplexed incoherent optical accumulation and achieve accumulation accuracy of 98.9% with 1 ns pulses. Assuming state of the art active components and a reasonable crossbar array size, this processor architecture would enable matrix vector multiplications with 16,000 × 64 matrices all optically on an estimated area of 51.2 mm2, while performing more than 110 trillion multiply and accumulate operations per second. | en_GB |
dc.description.sponsorship | Deutsche Forschungsgemeinschaft | en_GB |
dc.description.sponsorship | European Commission | en_GB |
dc.description.sponsorship | Bundesministerium für Bildung und Forschung | en_GB |
dc.identifier.citation | Published online 28 october 2022 | en_GB |
dc.identifier.doi | https://doi.org/10.1515/nanoph-2022-0441 | |
dc.identifier.grantnumber | CRC 1459 | en_GB |
dc.identifier.grantnumber | 101017237 | en_GB |
dc.identifier.grantnumber | 101046878 | en_GB |
dc.identifier.grantnumber | 724707 | en_GB |
dc.identifier.grantnumber | 899598 | en_GB |
dc.identifier.grantnumber | HYPHONE | en_GB |
dc.identifier.uri | http://hdl.handle.net/10871/131794 | |
dc.identifier | ORCID: 0000-0003-4087-7467 (Wright, C David) | |
dc.language.iso | en | en_GB |
dc.publisher | De Gruyter | en_GB |
dc.rights | © 2022 the author(s), published by De Gruyter, Berlin/Boston. Open access. This work is licensed under the Creative Commons Attribution 4.0 International License. | en_GB |
dc.subject | matrix vector multiplication | en_GB |
dc.subject | photonic computing | en_GB |
dc.subject | time-multiplexing | en_GB |
dc.title | A large scale photonic matrix processor enabled by charge accumulation | en_GB |
dc.type | Article | en_GB |
dc.date.available | 2022-11-17T10:27:39Z | |
dc.identifier.issn | 2192-8606 | |
dc.description | This is the final version. Available on open access from De Gruyter via the DOI in this record | en_GB |
dc.identifier.eissn | 2192-8614 | |
dc.identifier.journal | Nanophotonics | en_GB |
dc.relation.ispartof | Nanophotonics, 0(0) | |
dc.rights.uri | https://creativecommons.org/licenses/by/4.0/ | en_GB |
dcterms.dateAccepted | 2022-10-17 | |
rioxxterms.version | VoR | en_GB |
rioxxterms.licenseref.startdate | 2022-10-28 | |
rioxxterms.type | Journal Article/Review | en_GB |
refterms.dateFCD | 2022-11-17T10:23:06Z | |
refterms.versionFCD | VoR | |
refterms.dateFOA | 2022-11-17T10:27:42Z | |
refterms.panel | B | en_GB |
refterms.dateFirstOnline | 2022-10-28 |
Files in this item
This item appears in the following Collection(s)
Except where otherwise noted, this item's licence is described as © 2022 the author(s), published by De Gruyter, Berlin/Boston. Open access. This work is licensed under the Creative Commons Attribution 4.0 International License.