Ultra-low Crosstalk Multiplexer for Neuromorphic Photonic Data Processing

  • Frank Brückerhoff-Plückelmann
  • , Johannes Feldmann
  • , Helge Gehring
  • , Wen Zhou
  • , C. David Wright
  • , Harish Bhaskaran
  • , Wolfram Pernice

Research output: Contribution to journalConference articlepeer-review

Abstract

Neuromorphic architectures aim for powering artificial intelligence algorithms energy efficiently. We fabricate a neuromorphic photonic circuit for matrix vector multiplications using phase change materials and integrate wavelength multiplexers with a crosstalk below -41 dB on-chip.

Original languageEnglish
Article numberJTh3A.51
JournalOptics InfoBase Conference Papers
StatePublished - 2022
Externally publishedYes
EventCLEO: Science and Innovations, S and I 2022 - San Jose, United States
Duration: 15 May 202220 May 2022

Fingerprint

Dive into the research topics of 'Ultra-low Crosstalk Multiplexer for Neuromorphic Photonic Data Processing'. Together they form a unique fingerprint.

Cite this