One-Shot Learning for Modulation Format Identification in Evolving Optical Networks
Paper in proceeding, 2019
We report on the first successful application of one-shot machine learning scheme that identifies new modulation formats based on a single constellation diagram without re-training. 100% accuracy is achieved when expanding from 2 to 5 supported modulation formats.
Modulation
Machine learning
Machine vision
Fiber optics communications
Author
Carlos Natalino Da Silva
Chalmers, Electrical Engineering, Communication, Antennas and Optical Networks
Aleksejs Udalcovs
RISE Research Institutes of Sweden
Lena Wosinska
Chalmers, Electrical Engineering, Communication, Antennas and Optical Networks
Oskars Ozolins
RISE Research Institutes of Sweden
Marija Furdek Prekratic
Chalmers, Electrical Engineering, Communication, Antennas and Optical Networks
Optics InfoBase Conference Papers
10928081 (ISSN)
Vol. Part F131-IPRSN 2019 JW4A.2978-155752820-9 (ISBN)
Integrated Photonics Research, Silicon and Nanophotonics, IPRSN 2019
San Francisco, CA, USA,
San Francisco, CA, USA,
Subject Categories (SSIF 2011)
Telecommunications
Computer Systems