Bayesian optimization based mixed-word-length quantization scheme for fiber nonlinearity compensation in coherent optical communication systems

Wenkai Wan; Aiying Yang; Peng Guo; Zhe Zhao

doi:10.1109/OECC54135.2024.10975392

Bayesian optimization based mixed-word-length quantization scheme for fiber nonlinearity compensation in coherent optical communication systems

Wenkai Wan, Aiying Yang^*, Peng Guo, Zhe Zhao

^*Corresponding author for this work

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

Abstract

We propose a mixed-word-length quantization scheme assisted by Bayesian optimization for fiber nonlinearity compensation (NLC). The experimental results of an 80Gbps 1800-km single channel (SC) 16-QAM coherent optical communication system show that our proposed method reduces the hardware complexity by 58.2% compared with the uniform quantization counterpart without degrading Q-factor.

Original language	English
Title of host publication	Proceedings of the 29th Opto-Electronics and Communications Conference, OECC 2024
Publisher	Institute of Electrical and Electronics Engineers Inc.
Edition	2024
ISBN (Electronic)	9798350308396
DOIs	https://doi.org/10.1109/OECC54135.2024.10975392
Publication status	Published - 2024
Externally published	Yes
Event	29th Opto-Electronics and Communications Conference, OECC 2024 - Melbourne, Australia Duration: 30 Jun 2024 → 4 Jul 2024

Conference

Conference	29th Opto-Electronics and Communications Conference, OECC 2024
Country/Territory	Australia
City	Melbourne
Period	30/06/24 → 4/07/24

Keywords

Bayesian optimization
nonlinearity compensation
perturbation-based theory
quantization

Access to Document

10.1109/OECC54135.2024.10975392

Cite this

Wan, W., Yang, A., Guo, P., & Zhao, Z. (2024). Bayesian optimization based mixed-word-length quantization scheme for fiber nonlinearity compensation in coherent optical communication systems. In Proceedings of the 29th Opto-Electronics and Communications Conference, OECC 2024 (2024 ed.). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/OECC54135.2024.10975392

@inproceedings{065bb091d9f245dcb4b87ee5073d9494,

title = "Bayesian optimization based mixed-word-length quantization scheme for fiber nonlinearity compensation in coherent optical communication systems",

abstract = "We propose a mixed-word-length quantization scheme assisted by Bayesian optimization for fiber nonlinearity compensation (NLC). The experimental results of an 80Gbps 1800-km single channel (SC) 16-QAM coherent optical communication system show that our proposed method reduces the hardware complexity by 58.2\% compared with the uniform quantization counterpart without degrading Q-factor.",

keywords = "Bayesian optimization, nonlinearity compensation, perturbation-based theory, quantization",

author = "Wenkai Wan and Aiying Yang and Peng Guo and Zhe Zhao",

note = "Publisher Copyright: {\textcopyright} 2024 IEEE.; 29th Opto-Electronics and Communications Conference, OECC 2024 ; Conference date: 30-06-2024 Through 04-07-2024",

year = "2024",

doi = "10.1109/OECC54135.2024.10975392",

language = "English",

booktitle = "Proceedings of the 29th Opto-Electronics and Communications Conference, OECC 2024",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

edition = "2024",

}

Wan, W, Yang, A , Guo, P & Zhao, Z 2024, Bayesian optimization based mixed-word-length quantization scheme for fiber nonlinearity compensation in coherent optical communication systems. in Proceedings of the 29th Opto-Electronics and Communications Conference, OECC 2024. 2024 edn, Institute of Electrical and Electronics Engineers Inc., 29th Opto-Electronics and Communications Conference, OECC 2024, Melbourne, Australia, 30/06/24. https://doi.org/10.1109/OECC54135.2024.10975392

Bayesian optimization based mixed-word-length quantization scheme for fiber nonlinearity compensation in coherent optical communication systems. / Wan, Wenkai; Yang, Aiying ; Guo, Peng et al.
Proceedings of the 29th Opto-Electronics and Communications Conference, OECC 2024. 2024. ed. Institute of Electrical and Electronics Engineers Inc., 2024.

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

TY - GEN

T1 - Bayesian optimization based mixed-word-length quantization scheme for fiber nonlinearity compensation in coherent optical communication systems

AU - Wan, Wenkai

AU - Yang, Aiying

AU - Guo, Peng

AU - Zhao, Zhe

PY - 2024

Y1 - 2024

N2 - We propose a mixed-word-length quantization scheme assisted by Bayesian optimization for fiber nonlinearity compensation (NLC). The experimental results of an 80Gbps 1800-km single channel (SC) 16-QAM coherent optical communication system show that our proposed method reduces the hardware complexity by 58.2% compared with the uniform quantization counterpart without degrading Q-factor.

AB - We propose a mixed-word-length quantization scheme assisted by Bayesian optimization for fiber nonlinearity compensation (NLC). The experimental results of an 80Gbps 1800-km single channel (SC) 16-QAM coherent optical communication system show that our proposed method reduces the hardware complexity by 58.2% compared with the uniform quantization counterpart without degrading Q-factor.

KW - Bayesian optimization

KW - nonlinearity compensation

KW - perturbation-based theory

KW - quantization

UR - http://www.scopus.com/inward/record.url?scp=105005757217&partnerID=8YFLogxK

U2 - 10.1109/OECC54135.2024.10975392

DO - 10.1109/OECC54135.2024.10975392

M3 - Conference contribution

AN - SCOPUS:105005757217

BT - Proceedings of the 29th Opto-Electronics and Communications Conference, OECC 2024

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 29th Opto-Electronics and Communications Conference, OECC 2024

Y2 - 30 June 2024 through 4 July 2024

ER -

Bayesian optimization based mixed-word-length quantization scheme for fiber nonlinearity compensation in coherent optical communication systems

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this