End-to-End Learning in Optical Fiber Communications: Experimental Demonstration and Future Trends

Boris Karanov; Vinícius Oliari; Mathieu Chagnon; Gabriele Liga; Alex Alvarado; Vahid Aref; Domanic Lavery; Polina Bayvel; Laurent Schmalen

doi:10.1109/ECOC48923.2020.9333265

End-to-End Learning in Optical Fiber Communications: Experimental Demonstration and Future Trends

Boris Karanov, Vinícius Oliari, Mathieu Chagnon, Gabriele Liga, Alex Alvarado, Vahid Aref, Domanic Lavery, Polina Bayvel, Laurent Schmalen

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

9 Citations (Scopus)

261 Downloads (Pure)

Abstract

Fiber-optic auto-encoders are demonstrated on an intensity modulation/direct detection testbed, outperforming state-of-the-art signal processing. Algorithms for end-to-end optimization using experimentally collected data are discussed. The end-to-end learning framework is extended for performing optimization of the symbol distribution in probabilistically-shaped coherent systems.

Original language	English
Title of host publication	2020 European Conference on Optical Communications, ECOC 2020
Publisher	Institute of Electrical and Electronics Engineers
ISBN (Electronic)	9781728173610
DOIs	https://doi.org/10.1109/ECOC48923.2020.9333265
Publication status	Published - 4 Feb 2021
Event	46th European Conference on Optical Communications, ECOC 2020 - Brussels, Belgium, Virtual, Brussels, Belgium Duration: 6 Dec 2020 → 10 Dec 2020 Conference number: 46

Conference

Conference	46th European Conference on Optical Communications, ECOC 2020
Abbreviated title	ECOC 2020
Country/Territory	Belgium
City	Virtual, Brussels
Period	6/12/20 → 10/12/20

Bibliographical note

Publisher Copyright:
© 2020 IEEE.

Copyright:
Copyright 2021 Elsevier B.V., All rights reserved.

Funding

The work received funding from the EU’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie project COIN (676448), UK EPSRC TRANSNET grant EP/R035342/1, and the Netherlands Organisation for Scientific Research (NWO) via the VIDI Grant ICONIC (15685). The work of G. Liga is funded by the EU-ROTECH postdoc programme under the European Union’s Horizon 2020 research and innovation programme (Marie Skłodowska-Curie grant agreement No 754462). The work of A. Alvarado received funding from the European Research Council (ERC) under the EU’s Horizon 2020 research and innovation programme (757791).

Funders	Funder number
Marie Skłodowska‐Curie	676448
European Union's Horizon 2020 - Research and Innovation Framework Programme	754462, 757791
Engineering and Physical Sciences Research Council	EP/R035342/1
H2020 European Research Council
Nederlandse Organisatie voor Wetenschappelijk Onderzoek	15685

Access to Document

10.1109/ECOC48923.2020.9333265

End-to-End_Learning_Experimental_Future_TrendsFinal author version , 535 KBLicence: Unspecified

Cite this

Karanov, B., Oliari, V., Chagnon, M., Liga, G., Alvarado, A., Aref, V., Lavery, D., Bayvel, P., & Schmalen, L. (2021). End-to-End Learning in Optical Fiber Communications: Experimental Demonstration and Future Trends. In 2020 European Conference on Optical Communications, ECOC 2020 Article 9333265 Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ECOC48923.2020.9333265

@inproceedings{2ac330376cff40ae9e60f3b205c02005,

title = "End-to-End Learning in Optical Fiber Communications: Experimental Demonstration and Future Trends",

abstract = "Fiber-optic auto-encoders are demonstrated on an intensity modulation/direct detection testbed, outperforming state-of-the-art signal processing. Algorithms for end-to-end optimization using experimentally collected data are discussed. The end-to-end learning framework is extended for performing optimization of the symbol distribution in probabilistically-shaped coherent systems.",

author = "Boris Karanov and Vin{\'i}cius Oliari and Mathieu Chagnon and Gabriele Liga and Alex Alvarado and Vahid Aref and Domanic Lavery and Polina Bayvel and Laurent Schmalen",

note = " Publisher Copyright: {\textcopyright} 2020 IEEE. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.; 46th European Conference on Optical Communications, ECOC 2020, ECOC 2020 ; Conference date: 06-12-2020 Through 10-12-2020",

year = "2021",

month = feb,

day = "4",

doi = "10.1109/ECOC48923.2020.9333265",

language = "English",

booktitle = "2020 European Conference on Optical Communications, ECOC 2020",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

}

Karanov, B , Oliari, V, Chagnon, M, Liga, G , Alvarado, A, Aref, V, Lavery, D, Bayvel, P & Schmalen, L 2021, End-to-End Learning in Optical Fiber Communications: Experimental Demonstration and Future Trends. in 2020 European Conference on Optical Communications, ECOC 2020., 9333265, Institute of Electrical and Electronics Engineers, 46th European Conference on Optical Communications, ECOC 2020, Virtual, Brussels, Belgium, 6/12/20. https://doi.org/10.1109/ECOC48923.2020.9333265

End-to-End Learning in Optical Fiber Communications: Experimental Demonstration and Future Trends. / Karanov, Boris ; Oliari, Vinícius; Chagnon, Mathieu et al.
2020 European Conference on Optical Communications, ECOC 2020. Institute of Electrical and Electronics Engineers, 2021. 9333265.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - End-to-End Learning in Optical Fiber Communications

T2 - 46th European Conference on Optical Communications, ECOC 2020

AU - Karanov, Boris

AU - Oliari, Vinícius

AU - Chagnon, Mathieu

AU - Liga, Gabriele

AU - Alvarado, Alex

AU - Aref, Vahid

AU - Lavery, Domanic

AU - Bayvel, Polina

AU - Schmalen, Laurent

N1 - Conference code: 46

PY - 2021/2/4

Y1 - 2021/2/4

N2 - Fiber-optic auto-encoders are demonstrated on an intensity modulation/direct detection testbed, outperforming state-of-the-art signal processing. Algorithms for end-to-end optimization using experimentally collected data are discussed. The end-to-end learning framework is extended for performing optimization of the symbol distribution in probabilistically-shaped coherent systems.

AB - Fiber-optic auto-encoders are demonstrated on an intensity modulation/direct detection testbed, outperforming state-of-the-art signal processing. Algorithms for end-to-end optimization using experimentally collected data are discussed. The end-to-end learning framework is extended for performing optimization of the symbol distribution in probabilistically-shaped coherent systems.

UR - http://www.scopus.com/inward/record.url?scp=85101131453&partnerID=8YFLogxK

U2 - 10.1109/ECOC48923.2020.9333265

DO - 10.1109/ECOC48923.2020.9333265

M3 - Conference contribution

AN - SCOPUS:85101131453

BT - 2020 European Conference on Optical Communications, ECOC 2020

PB - Institute of Electrical and Electronics Engineers

Y2 - 6 December 2020 through 10 December 2020

ER -

End-to-End Learning in Optical Fiber Communications: Experimental Demonstration and Future Trends

Abstract

Conference

Bibliographical note

Funding

Access to Document

Other files and links

Fingerprint

FUN-NOTCH: Fundamentals of the Nonlinear Optical Channel

ICONIC: Increasing the Capacity of Optical Nonlinear Interfering Channels

Cite this

End-to-End Learning in Optical Fiber Communications: Experimental Demonstration and Future Trends

Abstract

Conference

Bibliographical note

Funding

Access to Document

Other files and links

Fingerprint

Projects

FUN-NOTCH: Fundamentals of the Nonlinear Optical Channel

ICONIC: Increasing the Capacity of Optical Nonlinear Interfering Channels

Cite this