End-to-end Learning of a Constellation Shape Robust to Variations in SNR
  and Laser Linewidth

Ognjen Jovanovic; Metodi P. Yankov; Francesco Da Ros; Darko Zibar

End-to-end Learning of a Constellation Shape Robust to Variations in SNR and Laser Linewidth

Ognjen Jovanovic, Metodi P. Yankov, Francesco Da Ros, Darko Zibar

Centre of Excellence for Silicon Photonics for Optical Communications

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

103 Downloads (Pure)

Abstract

We propose an autoencoder-based geometric shaping that learns a constellation robust to SNR and laser linewidth estimation errors. This constellation maintains shaping gain in mutual information (up to 0.3 bits/symbol) with respect to QAM over various SNR and laser linewidth values.

Original language	English
Title of host publication	Proceedings of European Conference on Optical Communications 2021
Publisher	IEEE
Publication date	2021
ISBN (Print)	978-1-6654-3868-1
Publication status	Published - 2021
Event	European Conference on Optical Communications 2021 - Parc des expositions, Bordeaux, France Duration: 13 Sept 2021 → 16 Sept 2021 https://www.ecoc2021.org/

Conference

Conference	European Conference on Optical Communications 2021
Location	Parc des expositions
Country/Territory	France
City	Bordeaux
Period	13/09/2021 → 16/09/2021
Internet address	https://www.ecoc2021.org/

Access to Document

FulltextAccepted author manuscript, 653 KB

OpenUrl availability

Full text

Cite this

@inproceedings{1e4bbedd639647b38d1b068c6761858e,

title = "End-to-end Learning of a Constellation Shape Robust to Variations in SNR and Laser Linewidth",

abstract = "We propose an autoencoder-based geometric shaping that learns a constellation robust to SNR and laser linewidth estimation errors. This constellation maintains shaping gain in mutual information (up to 0.3 bits/symbol) with respect to QAM over various SNR and laser linewidth values.",

author = "Ognjen Jovanovic and Yankov, {Metodi P.} and {Da Ros}, Francesco and Darko Zibar",

year = "2021",

language = "English",

isbn = "978-1-6654-3868-1",

booktitle = "Proceedings of European Conference on Optical Communications 2021",

publisher = "IEEE",

address = "United States",

note = "European Conference on Optical Communications 2021, ECOC 2021 ; Conference date: 13-09-2021 Through 16-09-2021",

url = "https://www.ecoc2021.org/",

}

TY - GEN

T1 - End-to-end Learning of a Constellation Shape Robust to Variations in SNR and Laser Linewidth

AU - Jovanovic, Ognjen

AU - Yankov, Metodi P.

AU - Da Ros, Francesco

AU - Zibar, Darko

PY - 2021

Y1 - 2021

N2 - We propose an autoencoder-based geometric shaping that learns a constellation robust to SNR and laser linewidth estimation errors. This constellation maintains shaping gain in mutual information (up to 0.3 bits/symbol) with respect to QAM over various SNR and laser linewidth values.

AB - We propose an autoencoder-based geometric shaping that learns a constellation robust to SNR and laser linewidth estimation errors. This constellation maintains shaping gain in mutual information (up to 0.3 bits/symbol) with respect to QAM over various SNR and laser linewidth values.

M3 - Article in proceedings

SN - 978-1-6654-3868-1

BT - Proceedings of European Conference on Optical Communications 2021

PB - IEEE

T2 - European Conference on Optical Communications 2021

Y2 - 13 September 2021 through 16 September 2021

ER -