End-to-end Learning for GMI Optimized Geometric Constellation Shape

Rasmus Thomas Jones; Metodi Plamenov Yankov; Darko Zibar

doi:10.1049/cp.2019.0886

End-to-end Learning for GMI Optimized Geometric Constellation Shape

Rasmus Thomas Jones, Metodi Plamenov Yankov, Darko Zibar

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

53 Downloads (Pure)

Abstract

Autoencoder-based geometric shaping is proposed that includes optimizing bit mappings. Up to 0.2 bits/QAM symbol gain in GMI is achieved for a variety of data rates and in the presence of transceiver impairments. The gains can be harvested with standard binary FEC at no cost w.r.t. conventional BICM.

Original language	English
Title of host publication	Proceedings of 45th European Conference on Optical Communication
Publisher	Institution of Engineering and Technology
Publication date	2019
Pages	560-564
ISBN (Print)	978-1-83953-185-9
DOIs	https://doi.org/10.1049/cp.2019.0886
Publication status	Published - 2019
Event	45th European Conference on Optical Communication - Royal Dublin Showground, Dublin, Ireland Duration: 22 Sep 2019 → 26 Sep 2019 Conference number: 45 http://www.ecoc2019.org

Conference

Conference	45th European Conference on Optical Communication
Number	45
Location	Royal Dublin Showground
Country/Territory	Ireland
City	Dublin
Period	22/09/2019 → 26/09/2019
Internet address	http://www.ecoc2019.org

Series	45th European Conference on Optical Communication (ecoc 2019)

Keywords

Geometric constellation shaping
Generalized mutual information
Fiber channel
End-to-end learning

Access to Document

10.1049/cp.2019.0886

FulltextSubmitted manuscript, 286 KB

Cite this

@inproceedings{40ad407657204b9780d0918a492548a0,

title = "End-to-end Learning for GMI Optimized Geometric Constellation Shape",

abstract = "Autoencoder-based geometric shaping is proposed that includes optimizing bit mappings. Up to 0.2 bits/QAM symbol gain in GMI is achieved for a variety of data rates and in the presence of transceiver impairments. The gains can be harvested with standard binary FEC at no cost w.r.t. conventional BICM.",

keywords = "Geometric constellation shaping, Generalized mutual information, Fiber channel, End-to-end learning",

author = "Jones, {Rasmus Thomas} and Yankov, {Metodi Plamenov} and Darko Zibar",

year = "2019",

doi = "10.1049/cp.2019.0886",

language = "English",

isbn = "978-1-83953-185-9",

series = "45th European Conference on Optical Communication (ecoc 2019)",

pages = "560--564",

booktitle = "Proceedings of 45th European Conference on Optical Communication",

publisher = "Institution of Engineering and Technology",

note = "45th European Conference on Optical Communication, ECOC 2019 ; Conference date: 22-09-2019 Through 26-09-2019",

url = "http://www.ecoc2019.org",

}

Jones, RT , Yankov, MP & Zibar, D 2019, End-to-end Learning for GMI Optimized Geometric Constellation Shape. in Proceedings of 45th European Conference on Optical Communication. Institution of Engineering and Technology, 45th European Conference on Optical Communication (ecoc 2019), pp. 560-564, 45th European Conference on Optical Communication, Dublin, Ireland, 22/09/2019. https://doi.org/10.1049/cp.2019.0886

End-to-end Learning for GMI Optimized Geometric Constellation Shape. / Jones, Rasmus Thomas ; Yankov, Metodi Plamenov ; Zibar, Darko.

Proceedings of 45th European Conference on Optical Communication. Institution of Engineering and Technology, 2019. p. 560-564 (45th European Conference on Optical Communication (ecoc 2019)).

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

TY - GEN

T1 - End-to-end Learning for GMI Optimized Geometric Constellation Shape

AU - Jones, Rasmus Thomas

AU - Yankov, Metodi Plamenov

AU - Zibar, Darko

N1 - Conference code: 45

PY - 2019

Y1 - 2019

N2 - Autoencoder-based geometric shaping is proposed that includes optimizing bit mappings. Up to 0.2 bits/QAM symbol gain in GMI is achieved for a variety of data rates and in the presence of transceiver impairments. The gains can be harvested with standard binary FEC at no cost w.r.t. conventional BICM.

AB - Autoencoder-based geometric shaping is proposed that includes optimizing bit mappings. Up to 0.2 bits/QAM symbol gain in GMI is achieved for a variety of data rates and in the presence of transceiver impairments. The gains can be harvested with standard binary FEC at no cost w.r.t. conventional BICM.

KW - Geometric constellation shaping

KW - Generalized mutual information

KW - Fiber channel

KW - End-to-end learning

U2 - 10.1049/cp.2019.0886

DO - 10.1049/cp.2019.0886

M3 - Article in proceedings

SN - 978-1-83953-185-9

T3 - 45th European Conference on Optical Communication (ecoc 2019)

SP - 560

EP - 564

BT - Proceedings of 45th European Conference on Optical Communication

PB - Institution of Engineering and Technology

T2 - 45th European Conference on Optical Communication

Y2 - 22 September 2019 through 26 September 2019

ER -

End-to-end Learning for GMI Optimized Geometric Constellation Shape

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this