End-to-End Deep Learning of Long-Haul Coherent Optical Fiber Communications via Regular Perturbation Model

Vladislav Neskorniuk; Andrea  Carnio; Vinod Bajaj; Domenico Marsella; Sergei K. Turitsyn; Jaroslaw E. Prilepsky; Vahid Aref

End-to-End Deep Learning of Long-Haul Coherent Optical Fiber Communications via Regular Perturbation Model

Vladislav Neskorniuk^*, Andrea Carnio, Vinod Bajaj, Domenico Marsella, Sergei K. Turitsyn, Jaroslaw E. Prilepsky, Vahid Aref

^*Corresponding author for this work

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

Abstract

We present a novel end-to-end autoencoder-based learning for coherent optical communications using a "parallelizable" perturbative channel model. We jointly optimized constellation shaping and nonlinear pre-emphasis achieving mutual information gain of 0.18 bits/sym./pol. simulating 64 GBd dual-polarization single-channel transmission over 30x80 km G.652 SMF link with EDFAs.

Original language	English
Title of host publication	ECOC Exhibition 2021
Number of pages	4
Publication status	Published - 26 Jul 2021
Event	ECOC 2021 - Bordeaux, France Duration: 13 Sept 2021 → 15 Sept 2021

Conference

Conference	ECOC 2021
Country/Territory	France
City	Bordeaux
Period	13/09/21 → 15/09/21

Bibliographical note

© 2021 The Authors.

Funding: This project has received funding from
EU Horizon 2020 program under the Marie Skłodowska-Curie
grant agreement No. 766115 (FONTE). JEP is supported by
Leverhulme Project RPG-2018-063. SKT acknowledges the
support of EPSRC project TRANSNET.

Access to Document

https://arxiv.org/abs/2107.12320Licence: CC BY 3.0

Deep Learning Methods for Nonlinearity Mitigation in Coherent Fiber-Optic Communication Links
Neskorniuk, V. (Author), Turitsyn, S. (Supervisor) & Prylepskiy, Y. (Supervisor), Nov 2022
Student thesis: Doctoral Thesis › Doctor of Philosophy

File

Cite this

@inproceedings{ed644cea3af94fc8aa02966f4af285a9,

title = "End-to-End Deep Learning of Long-Haul Coherent Optical Fiber Communications via Regular Perturbation Model",

abstract = "We present a novel end-to-end autoencoder-based learning for coherent optical communications using a {"}parallelizable{"} perturbative channel model. We jointly optimized constellation shaping and nonlinear pre-emphasis achieving mutual information gain of 0.18 bits/sym./pol. simulating 64 GBd dual-polarization single-channel transmission over 30x80 km G.652 SMF link with EDFAs.",

author = "Vladislav Neskorniuk and Andrea Carnio and Vinod Bajaj and Domenico Marsella and Turitsyn, {Sergei K.} and Prilepsky, {Jaroslaw E.} and Vahid Aref",

note = "{\textcopyright} 2021 The Authors. Funding: This project has received funding from EU Horizon 2020 program under the Marie Sk{\l}odowska-Curie grant agreement No. 766115 (FONTE). JEP is supported by Leverhulme Project RPG-2018-063. SKT acknowledges the support of EPSRC project TRANSNET.; ECOC 2021 ; Conference date: 13-09-2021 Through 15-09-2021",

year = "2021",

month = jul,

day = "26",

language = "English",

booktitle = "ECOC Exhibition 2021",

}

TY - GEN

T1 - End-to-End Deep Learning of Long-Haul Coherent Optical Fiber Communications via Regular Perturbation Model

AU - Neskorniuk, Vladislav

AU - Carnio, Andrea

AU - Bajaj, Vinod

AU - Marsella, Domenico

AU - Turitsyn, Sergei K.

AU - Prilepsky, Jaroslaw E.

AU - Aref, Vahid

N1 - © 2021 The Authors. Funding: This project has received funding from EU Horizon 2020 program under the Marie Skłodowska-Curie grant agreement No. 766115 (FONTE). JEP is supported by Leverhulme Project RPG-2018-063. SKT acknowledges the support of EPSRC project TRANSNET.

PY - 2021/7/26

Y1 - 2021/7/26

N2 - We present a novel end-to-end autoencoder-based learning for coherent optical communications using a "parallelizable" perturbative channel model. We jointly optimized constellation shaping and nonlinear pre-emphasis achieving mutual information gain of 0.18 bits/sym./pol. simulating 64 GBd dual-polarization single-channel transmission over 30x80 km G.652 SMF link with EDFAs.

AB - We present a novel end-to-end autoencoder-based learning for coherent optical communications using a "parallelizable" perturbative channel model. We jointly optimized constellation shaping and nonlinear pre-emphasis achieving mutual information gain of 0.18 bits/sym./pol. simulating 64 GBd dual-polarization single-channel transmission over 30x80 km G.652 SMF link with EDFAs.

M3 - Conference publication

BT - ECOC Exhibition 2021

T2 - ECOC 2021

Y2 - 13 September 2021 through 15 September 2021

ER -

End-to-End Deep Learning of Long-Haul Coherent Optical Fiber Communications via Regular Perturbation Model

Abstract

Conference

Bibliographical note

Access to Document

Fingerprint

Student theses

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

Cite this