Convolutional neural networks with multiple layers per span for nonlinearity mitigation in long-haul WDM transmission systems

Oleg Sidelnikov; Alexey Redyuk; Stylianos Sygletos; Mikhail Fedoruk; Sergei Turitsyn

doi:https://doi.org/10.1109/CLEO/Europe-EQEC52157.2021.9542317

Convolutional neural networks with multiple layers per span for nonlinearity mitigation in long-haul WDM transmission systems

Oleg Sidelnikov, Alexey Redyuk, Stylianos Sygletos, Mikhail Fedoruk, Sergei Turitsyn

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

Abstract

Overcoming fiber nonlinearity is one of the most challenging tasks in optical fiber links and it is a major limiting factor for extending their capacity. Digital backward propagation (DBP) method can be used to mitigate nonlinear transmission impairments [1] , but its complexity prevents any real-time implementation in these systems. On the other hand, it has been recently shown that deep neural networks can provide a good approximation of DBP at lower computational cost [2] . In this work, we continue the investigation of the proposed deep convolutional neural network (DCNN) [3] for long-haul WDM transmission systems. We study the effect of the number of neural network layers on the efficiency of nonlinear distortion compensation.

Original language	English
Title of host publication	The European Conference on Lasers and Electro-Optics, CLEO/Europe 2021
Publisher	The Optical Society
Number of pages	1
ISBN (Electronic)	9781557528209
DOIs	https://doi.org/10.1109/CLEO/Europe-EQEC52157.2021.9542317
Publication status	Published - 30 Sept 2021
Event	2021 European Conference on Lasers and Electro-Optics, CLEO/Europe 2021 - Virtual, Online, Germany Duration: 21 Jun 2021 → 25 Jun 2021

Publication series

Name	Optics InfoBase Conference Papers

Conference

Conference	2021 European Conference on Lasers and Electro-Optics, CLEO/Europe 2021
Country/Territory	Germany
City	Virtual, Online
Period	21/06/21 → 25/06/21

Bibliographical note

Funding Information:
The work was supported by the Russian Science Foundation (Grant No. 17-72-30006).

Access to Document

https://doi.org/10.1109/CLEO/Europe-EQEC52157.2021.9542317

Cite this

Sidelnikov, O., Redyuk, A., Sygletos, S., Fedoruk, M., & Turitsyn, S. (2021). Convolutional neural networks with multiple layers per span for nonlinearity mitigation in long-haul WDM transmission systems. In The European Conference on Lasers and Electro-Optics, CLEO/Europe 2021 (Optics InfoBase Conference Papers). The Optical Society. https://doi.org/10.1109/CLEO/Europe-EQEC52157.2021.9542317

@inproceedings{0b7dd2e63b704ea0a675b3de301721a9,

title = "Convolutional neural networks with multiple layers per span for nonlinearity mitigation in long-haul WDM transmission systems",

abstract = "Overcoming fiber nonlinearity is one of the most challenging tasks in optical fiber links and it is a major limiting factor for extending their capacity. Digital backward propagation (DBP) method can be used to mitigate nonlinear transmission impairments [1] , but its complexity prevents any real-time implementation in these systems. On the other hand, it has been recently shown that deep neural networks can provide a good approximation of DBP at lower computational cost [2] . In this work, we continue the investigation of the proposed deep convolutional neural network (DCNN) [3] for long-haul WDM transmission systems. We study the effect of the number of neural network layers on the efficiency of nonlinear distortion compensation.",

author = "Oleg Sidelnikov and Alexey Redyuk and Stylianos Sygletos and Mikhail Fedoruk and Sergei Turitsyn",

note = "Funding Information: The work was supported by the Russian Science Foundation (Grant No. 17-72-30006).; 2021 European Conference on Lasers and Electro-Optics, CLEO/Europe 2021 ; Conference date: 21-06-2021 Through 25-06-2021",

year = "2021",

month = sep,

day = "30",

doi = "https://doi.org/10.1109/CLEO/Europe-EQEC52157.2021.9542317",

language = "English",

series = "Optics InfoBase Conference Papers",

publisher = "The Optical Society",

booktitle = "The European Conference on Lasers and Electro-Optics, CLEO/Europe 2021",

address = "United States",

}

Sidelnikov, O, Redyuk, A, Sygletos, S, Fedoruk, M & Turitsyn, S 2021, Convolutional neural networks with multiple layers per span for nonlinearity mitigation in long-haul WDM transmission systems. in The European Conference on Lasers and Electro-Optics, CLEO/Europe 2021. Optics InfoBase Conference Papers, The Optical Society, 2021 European Conference on Lasers and Electro-Optics, CLEO/Europe 2021, Virtual, Online, Germany, 21/06/21. https://doi.org/10.1109/CLEO/Europe-EQEC52157.2021.9542317

Convolutional neural networks with multiple layers per span for nonlinearity mitigation in long-haul WDM transmission systems. / Sidelnikov, Oleg; Redyuk, Alexey; Sygletos, Stylianos et al.
The European Conference on Lasers and Electro-Optics, CLEO/Europe 2021. The Optical Society, 2021. (Optics InfoBase Conference Papers).

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

TY - GEN

T1 - Convolutional neural networks with multiple layers per span for nonlinearity mitigation in long-haul WDM transmission systems

AU - Sidelnikov, Oleg

AU - Redyuk, Alexey

AU - Sygletos, Stylianos

AU - Fedoruk, Mikhail

AU - Turitsyn, Sergei

N1 - Funding Information: The work was supported by the Russian Science Foundation (Grant No. 17-72-30006).

PY - 2021/9/30

Y1 - 2021/9/30

N2 - Overcoming fiber nonlinearity is one of the most challenging tasks in optical fiber links and it is a major limiting factor for extending their capacity. Digital backward propagation (DBP) method can be used to mitigate nonlinear transmission impairments [1] , but its complexity prevents any real-time implementation in these systems. On the other hand, it has been recently shown that deep neural networks can provide a good approximation of DBP at lower computational cost [2] . In this work, we continue the investigation of the proposed deep convolutional neural network (DCNN) [3] for long-haul WDM transmission systems. We study the effect of the number of neural network layers on the efficiency of nonlinear distortion compensation.

AB - Overcoming fiber nonlinearity is one of the most challenging tasks in optical fiber links and it is a major limiting factor for extending their capacity. Digital backward propagation (DBP) method can be used to mitigate nonlinear transmission impairments [1] , but its complexity prevents any real-time implementation in these systems. On the other hand, it has been recently shown that deep neural networks can provide a good approximation of DBP at lower computational cost [2] . In this work, we continue the investigation of the proposed deep convolutional neural network (DCNN) [3] for long-haul WDM transmission systems. We study the effect of the number of neural network layers on the efficiency of nonlinear distortion compensation.

UR - http://www.scopus.com/inward/record.url?scp=85120433388&partnerID=8YFLogxK

UR - https://ieeexplore.ieee.org/document/9542317

UR - https://opg.optica.org/abstract.cfm?uri=CLEO_Europe-2021-ci_p_6

U2 - https://doi.org/10.1109/CLEO/Europe-EQEC52157.2021.9542317

DO - https://doi.org/10.1109/CLEO/Europe-EQEC52157.2021.9542317

M3 - Conference publication

AN - SCOPUS:85120433388

T3 - Optics InfoBase Conference Papers

BT - The European Conference on Lasers and Electro-Optics, CLEO/Europe 2021

PB - The Optical Society

T2 - 2021 European Conference on Lasers and Electro-Optics, CLEO/Europe 2021

Y2 - 21 June 2021 through 25 June 2021

ER -

Sidelnikov O, Redyuk A, Sygletos S, Fedoruk M, Turitsyn S. Convolutional neural networks with multiple layers per span for nonlinearity mitigation in long-haul WDM transmission systems. In The European Conference on Lasers and Electro-Optics, CLEO/Europe 2021. The Optical Society. 2021. (Optics InfoBase Conference Papers). doi: https://doi.org/10.1109/CLEO/Europe-EQEC52157.2021.9542317

Convolutional neural networks with multiple layers per span for nonlinearity mitigation in long-haul WDM transmission systems

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this