Simplifying the Supervised Learning of Kerr Nonlinearity Compensation Algorithms by Data Augmentation

Vladislav Neskorniuk; Pedro Jorge Freire de Carvalho Souza; Antonio Napoli; Bernhard Spinnler; Wolfgang Schairer; Jaroslaw E. Prilepsky; Nelson Costa; Sergei K. Turitsyn

doi:10.1109/ECOC48923.2020.9333417

Simplifying the Supervised Learning of Kerr Nonlinearity Compensation Algorithms by Data Augmentation

Vladislav Neskorniuk^*, Pedro Jorge Freire de Carvalho Souza, Antonio Napoli, Bernhard Spinnler, Wolfgang Schairer, Jaroslaw E. Prilepsky, Nelson Costa, Sergei K. Turitsyn

^*Corresponding author for this work

Research output: Chapter in Book/Published conference output › Conference publication

Abstract

We propose a data augmentation technique to improve performance and decrease complexity of the supervised learning of nonlinearity compensation algorithms. We demonstrate both numerically and experimentally that the augmentation allows reducing the training dataset size up to 6 times while keeping the same post-compensation bit-error rate.

Original language	English
Title of host publication	2020 European Conference on Optical Communications, ECOC 2020
Publisher	IEEE
Number of pages	4
ISBN (Electronic)	978-1-7281-7361-0
ISBN (Print)	978-1-7281-7362-7
DOIs	https://doi.org/10.1109/ECOC48923.2020.9333417
Publication status	Published - 4 Feb 2021
Event	2020 European Conference on Optical Communications - Brussels, Belgium Duration: 6 Dec 2020 → 10 Dec 2020 https://ecoco2020.org/

Publication series

Name	2020 European Conference on Optical Communications, ECOC 2020

Conference

Conference	2020 European Conference on Optical Communications
Abbreviated title	ECOC 2020
Country/Territory	Belgium
City	Brussels
Period	6/12/20 → 10/12/20
Internet address	https://ecoco2020.org/

Bibliographical note

© 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

Access to Document

10.1109/ECOC48923.2020.9333417

Simplifying the Supervised Learning of Kerr Nonlinearity
© 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Submitted manuscript, 407 KB

Cite this

Neskorniuk, V., Freire de Carvalho Souza, P. J., Napoli, A., Spinnler, B., Schairer, W., Prilepsky, J. E., Costa, N., & Turitsyn, S. K. (2021). Simplifying the Supervised Learning of Kerr Nonlinearity Compensation Algorithms by Data Augmentation. In 2020 European Conference on Optical Communications, ECOC 2020 Article 9333417 (2020 European Conference on Optical Communications, ECOC 2020). IEEE. https://doi.org/10.1109/ECOC48923.2020.9333417

@inproceedings{d2ad042a3d7d47a78151358738d7c0f0,

title = "Simplifying the Supervised Learning of Kerr Nonlinearity Compensation Algorithms by Data Augmentation",

abstract = "We propose a data augmentation technique to improve performance and decrease complexity of the supervised learning of nonlinearity compensation algorithms. We demonstrate both numerically and experimentally that the augmentation allows reducing the training dataset size up to 6 times while keeping the same post-compensation bit-error rate.",

author = "Vladislav Neskorniuk and {Freire de Carvalho Souza}, {Pedro Jorge} and Antonio Napoli and Bernhard Spinnler and Wolfgang Schairer and Prilepsky, {Jaroslaw E.} and Nelson Costa and Turitsyn, {Sergei K.}",

note = "{\textcopyright} 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.; 2020 European Conference on Optical Communications, ECOC 2020 ; Conference date: 06-12-2020 Through 10-12-2020",

year = "2021",

month = feb,

day = "4",

doi = "10.1109/ECOC48923.2020.9333417",

language = "English",

isbn = "978-1-7281-7362-7",

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

publisher = "IEEE",

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

address = "United States",

url = "https://ecoco2020.org/",

}

Neskorniuk, V, Freire de Carvalho Souza, PJ, Napoli, A, Spinnler, B, Schairer, W, Prilepsky, JE, Costa, N & Turitsyn, SK 2021, Simplifying the Supervised Learning of Kerr Nonlinearity Compensation Algorithms by Data Augmentation. in 2020 European Conference on Optical Communications, ECOC 2020., 9333417, 2020 European Conference on Optical Communications, ECOC 2020, IEEE, 2020 European Conference on Optical Communications, Brussels, Belgium, 6/12/20. https://doi.org/10.1109/ECOC48923.2020.9333417

Simplifying the Supervised Learning of Kerr Nonlinearity Compensation Algorithms by Data Augmentation. / Neskorniuk, Vladislav; Freire de Carvalho Souza, Pedro Jorge; Napoli, Antonio et al.
2020 European Conference on Optical Communications, ECOC 2020. IEEE, 2021. 9333417 (2020 European Conference on Optical Communications, ECOC 2020).

Research output: Chapter in Book/Published conference output › Conference publication

TY - GEN

T1 - Simplifying the Supervised Learning of Kerr Nonlinearity Compensation Algorithms by Data Augmentation

AU - Neskorniuk, Vladislav

AU - Freire de Carvalho Souza, Pedro Jorge

AU - Napoli, Antonio

AU - Spinnler, Bernhard

AU - Schairer, Wolfgang

AU - Prilepsky, Jaroslaw E.

AU - Costa, Nelson

AU - Turitsyn, Sergei K.

N1 - © 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

PY - 2021/2/4

Y1 - 2021/2/4

N2 - We propose a data augmentation technique to improve performance and decrease complexity of the supervised learning of nonlinearity compensation algorithms. We demonstrate both numerically and experimentally that the augmentation allows reducing the training dataset size up to 6 times while keeping the same post-compensation bit-error rate.

AB - We propose a data augmentation technique to improve performance and decrease complexity of the supervised learning of nonlinearity compensation algorithms. We demonstrate both numerically and experimentally that the augmentation allows reducing the training dataset size up to 6 times while keeping the same post-compensation bit-error rate.

UR - https://ieeexplore.ieee.org/document/9333417

UR - http://www.scopus.com/inward/record.url?scp=85101058850&partnerID=8YFLogxK

U2 - 10.1109/ECOC48923.2020.9333417

DO - 10.1109/ECOC48923.2020.9333417

M3 - Conference publication

SN - 978-1-7281-7362-7

T3 - 2020 European Conference on Optical Communications, ECOC 2020

BT - 2020 European Conference on Optical Communications, ECOC 2020

PB - IEEE

T2 - 2020 European Conference on Optical Communications

Y2 - 6 December 2020 through 10 December 2020

ER -

Neskorniuk V, Freire de Carvalho Souza PJ, Napoli A, Spinnler B, Schairer W, Prilepsky JE et al. Simplifying the Supervised Learning of Kerr Nonlinearity Compensation Algorithms by Data Augmentation. In 2020 European Conference on Optical Communications, ECOC 2020. IEEE. 2021. 9333417. (2020 European Conference on Optical Communications, ECOC 2020). doi: 10.1109/ECOC48923.2020.9333417

Simplifying the Supervised Learning of Kerr Nonlinearity Compensation Algorithms by Data Augmentation

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Deep Learning Methods for Nonlinearity Mitigation in Coherent Fiber-Optic Communication Links

Machine Learning Techniques To Mitigate Nonlinear Impairments In Optical Fiber System

Cite this

Simplifying the Supervised Learning of Kerr Nonlinearity Compensation Algorithms by Data Augmentation

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Student theses

Deep Learning Methods for Nonlinearity Mitigation in Coherent Fiber-Optic Communication Links

Machine Learning Techniques To Mitigate Nonlinear Impairments In Optical Fiber System

Cite this