Blind nonlinearity equalization by machine learning based clustering for single- and multi-channel coherent optical OFDM

Elias Giacoumidis, Amir Matin, Jinlong Wei, Nick J. Doran, Liam P. Barry, Xu Wang

Research output: Contribution to journalArticle

Abstract

Fiber-induced intra- and inter-channel nonlinearities are experimentally tackled using blind nonlinear equalization (NLE) by unsupervised machine learning based clustering (MLC) in ∼46-Gb/s single-channel and ∼20-Gb/s (middle-channel) multi-channel coherent multi-carrier signals (OFDM-based). To that end we introduce, for the first time, Hierarchical and Fuzzy-Logic C-means (FLC) based clustering in optical communications. It is shown that among the two proposed MLC algorithms, FLC reveals the highest performance at optimum launched optical powers (LOPs), while at very high LOPs Hierarchical can compensate more effectively nonlinearities only for low-level modulation formats. FLC also outperforms K-means, Fast-Newton support vector machines, supervised artificial neural networks and a NLE with deterministic Volterra analysis, when employing BPSK and QPSK. In particular, for the middle channel of a QPSK WDM coherent optical OFDM system at optimum -5 dBm of LOP and 3200 km of transmission, FLC outperforms Volterra-NLE by 2.5 dB in Q-factor. However, for a 16-quadrature amplitude modulated single-channel system at 2000 km, the performance benefit of FLC over IVSTF reduces to ∼0.4 dB at a LOP of 2 dBm (optimum). Even when using novel sophisticated clustering designs in 16 clusters, no more than additional ∼0.3 dB Q-factor enhancement is observed. Finally, in contrast to the deterministic Volterra-NLE, MLC algorithms can partially tackle the stochastic parametric noise amplification.
Original languageEnglish
Pages (from-to)721-727
JournalJournal of Lightwave Technology
Volume36
Issue number3
Early online date1 Dec 2017
DOIs
Publication statusPublished - 1 Dec 2017

Fingerprint

machine learning
nonlinearity
logic
quadrature phase shift keying
Q factors
binary phase shift keying
quadratures
newton
format
optical communication
modulation
fibers
augmentation

Bibliographical note

© 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

Funding: Partnership Resource Project of Quantum Communications Hub (EPSRC), EU Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 713567, SFI CONNECT Research Centre and Sterlite Techn. Ltd. We thank S. T. Le (Nokia-Bells labs) and M. E. McCarthy (Oclaro) for their support

Keywords

  • Algorithm design and analysis
  • clustering
  • Clustering algorithms
  • coherent detection
  • coherent optical OFDM
  • Machine learning
  • nonlinearity mitigation
  • OFDM
  • Optical fiber amplifiers
  • Optical fiber networks
  • Phase shift keying

Cite this

Giacoumidis, Elias ; Matin, Amir ; Wei, Jinlong ; Doran, Nick J. ; Barry, Liam P. ; Wang, Xu. / Blind nonlinearity equalization by machine learning based clustering for single- and multi-channel coherent optical OFDM. In: Journal of Lightwave Technology. 2017 ; Vol. 36, No. 3. pp. 721-727.
@article{1fa7bb3db82543008922a25244cdd565,
title = "Blind nonlinearity equalization by machine learning based clustering for single- and multi-channel coherent optical OFDM",
abstract = "Fiber-induced intra- and inter-channel nonlinearities are experimentally tackled using blind nonlinear equalization (NLE) by unsupervised machine learning based clustering (MLC) in ∼46-Gb/s single-channel and ∼20-Gb/s (middle-channel) multi-channel coherent multi-carrier signals (OFDM-based). To that end we introduce, for the first time, Hierarchical and Fuzzy-Logic C-means (FLC) based clustering in optical communications. It is shown that among the two proposed MLC algorithms, FLC reveals the highest performance at optimum launched optical powers (LOPs), while at very high LOPs Hierarchical can compensate more effectively nonlinearities only for low-level modulation formats. FLC also outperforms K-means, Fast-Newton support vector machines, supervised artificial neural networks and a NLE with deterministic Volterra analysis, when employing BPSK and QPSK. In particular, for the middle channel of a QPSK WDM coherent optical OFDM system at optimum -5 dBm of LOP and 3200 km of transmission, FLC outperforms Volterra-NLE by 2.5 dB in Q-factor. However, for a 16-quadrature amplitude modulated single-channel system at 2000 km, the performance benefit of FLC over IVSTF reduces to ∼0.4 dB at a LOP of 2 dBm (optimum). Even when using novel sophisticated clustering designs in 16 clusters, no more than additional ∼0.3 dB Q-factor enhancement is observed. Finally, in contrast to the deterministic Volterra-NLE, MLC algorithms can partially tackle the stochastic parametric noise amplification.",
keywords = "Algorithm design and analysis, clustering, Clustering algorithms, coherent detection, coherent optical OFDM, Machine learning, nonlinearity mitigation, OFDM, Optical fiber amplifiers, Optical fiber networks, Phase shift keying",
author = "Elias Giacoumidis and Amir Matin and Jinlong Wei and Doran, {Nick J.} and Barry, {Liam P.} and Xu Wang",
note = "{\circledC} 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. Funding: Partnership Resource Project of Quantum Communications Hub (EPSRC), EU Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 713567, SFI CONNECT Research Centre and Sterlite Techn. Ltd. We thank S. T. Le (Nokia-Bells labs) and M. E. McCarthy (Oclaro) for their support",
year = "2017",
month = "12",
day = "1",
doi = "10.1109/JLT.2017.2778883",
language = "English",
volume = "36",
pages = "721--727",
journal = "Journal of Lightwave Technology",
issn = "0733-8724",
publisher = "IEEE",
number = "3",

}

Blind nonlinearity equalization by machine learning based clustering for single- and multi-channel coherent optical OFDM. / Giacoumidis, Elias; Matin, Amir; Wei, Jinlong; Doran, Nick J.; Barry, Liam P.; Wang, Xu.

In: Journal of Lightwave Technology, Vol. 36, No. 3, 01.12.2017, p. 721-727.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Blind nonlinearity equalization by machine learning based clustering for single- and multi-channel coherent optical OFDM

AU - Giacoumidis, Elias

AU - Matin, Amir

AU - Wei, Jinlong

AU - Doran, Nick J.

AU - Barry, Liam P.

AU - Wang, Xu

N1 - © 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. Funding: Partnership Resource Project of Quantum Communications Hub (EPSRC), EU Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 713567, SFI CONNECT Research Centre and Sterlite Techn. Ltd. We thank S. T. Le (Nokia-Bells labs) and M. E. McCarthy (Oclaro) for their support

PY - 2017/12/1

Y1 - 2017/12/1

N2 - Fiber-induced intra- and inter-channel nonlinearities are experimentally tackled using blind nonlinear equalization (NLE) by unsupervised machine learning based clustering (MLC) in ∼46-Gb/s single-channel and ∼20-Gb/s (middle-channel) multi-channel coherent multi-carrier signals (OFDM-based). To that end we introduce, for the first time, Hierarchical and Fuzzy-Logic C-means (FLC) based clustering in optical communications. It is shown that among the two proposed MLC algorithms, FLC reveals the highest performance at optimum launched optical powers (LOPs), while at very high LOPs Hierarchical can compensate more effectively nonlinearities only for low-level modulation formats. FLC also outperforms K-means, Fast-Newton support vector machines, supervised artificial neural networks and a NLE with deterministic Volterra analysis, when employing BPSK and QPSK. In particular, for the middle channel of a QPSK WDM coherent optical OFDM system at optimum -5 dBm of LOP and 3200 km of transmission, FLC outperforms Volterra-NLE by 2.5 dB in Q-factor. However, for a 16-quadrature amplitude modulated single-channel system at 2000 km, the performance benefit of FLC over IVSTF reduces to ∼0.4 dB at a LOP of 2 dBm (optimum). Even when using novel sophisticated clustering designs in 16 clusters, no more than additional ∼0.3 dB Q-factor enhancement is observed. Finally, in contrast to the deterministic Volterra-NLE, MLC algorithms can partially tackle the stochastic parametric noise amplification.

AB - Fiber-induced intra- and inter-channel nonlinearities are experimentally tackled using blind nonlinear equalization (NLE) by unsupervised machine learning based clustering (MLC) in ∼46-Gb/s single-channel and ∼20-Gb/s (middle-channel) multi-channel coherent multi-carrier signals (OFDM-based). To that end we introduce, for the first time, Hierarchical and Fuzzy-Logic C-means (FLC) based clustering in optical communications. It is shown that among the two proposed MLC algorithms, FLC reveals the highest performance at optimum launched optical powers (LOPs), while at very high LOPs Hierarchical can compensate more effectively nonlinearities only for low-level modulation formats. FLC also outperforms K-means, Fast-Newton support vector machines, supervised artificial neural networks and a NLE with deterministic Volterra analysis, when employing BPSK and QPSK. In particular, for the middle channel of a QPSK WDM coherent optical OFDM system at optimum -5 dBm of LOP and 3200 km of transmission, FLC outperforms Volterra-NLE by 2.5 dB in Q-factor. However, for a 16-quadrature amplitude modulated single-channel system at 2000 km, the performance benefit of FLC over IVSTF reduces to ∼0.4 dB at a LOP of 2 dBm (optimum). Even when using novel sophisticated clustering designs in 16 clusters, no more than additional ∼0.3 dB Q-factor enhancement is observed. Finally, in contrast to the deterministic Volterra-NLE, MLC algorithms can partially tackle the stochastic parametric noise amplification.

KW - Algorithm design and analysis

KW - clustering

KW - Clustering algorithms

KW - coherent detection

KW - coherent optical OFDM

KW - Machine learning

KW - nonlinearity mitigation

KW - OFDM

KW - Optical fiber amplifiers

KW - Optical fiber networks

KW - Phase shift keying

UR - http://www.scopus.com/inward/record.url?scp=85037653067&partnerID=8YFLogxK

U2 - 10.1109/JLT.2017.2778883

DO - 10.1109/JLT.2017.2778883

M3 - Article

AN - SCOPUS:85037653067

VL - 36

SP - 721

EP - 727

JO - Journal of Lightwave Technology

JF - Journal of Lightwave Technology

SN - 0733-8724

IS - 3

ER -