Nonlinear equalization method based on Gaussian mixture model clustering algorithm for a coherent optical OFDM communication system

Tian Wang, Qinghua Tian, Qi Zhang, Yongjun Wang, Feng Tian, Xishuo Wang, Leijing Yang, Yunxiao Zu, Xiangjun Xin

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this paper, an unsupervised clustering algorithm based on the Gaussian Mixture Model (UCGMM algorithm) for the coherent optical OFDM communication system is proposed to determine the constellation diagram. The purpose of nonlinear equalization of communication systems is achieved. In a back to back transmission system, compared to the Kmeans algorithm and the without any clustering algorithm, the UCGMM algorithm can obtain gains of approximately 0.6dB and 2dB respectively. For the cases of simulation in optical fiber transmission, the transmission distance of UCGMM algorithm is extended by 45km relative to the K-means algorithm, and 75km relative to without any clustering algorithm. In both cases, the effectiveness of the proposed UCGMM algorithm in nonlinear equilibrium is proved.

Original languageEnglish
Title of host publicationAOPC 2020
Subtitle of host publicationOptical Information and Network
EditorsJie Zhang, Songnian Fu, He Zhang
PublisherSPIE
ISBN (Electronic)9781510639607
DOIs
Publication statusPublished - 2020
Externally publishedYes
Event2020 Applied Optics and Photonics China: Optical Information and Network, AOPC 2020 - Beijing, China
Duration: 30 Nov 20202 Dec 2020

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume11569
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

Conference2020 Applied Optics and Photonics China: Optical Information and Network, AOPC 2020
Country/TerritoryChina
CityBeijing
Period30/11/202/12/20

Keywords

  • Clustering Algorithm
  • Coherent Optical OFDM
  • Fiber Nonlinearity Equalizer
  • Gaussian Mixture Model (GMM)

Fingerprint

Dive into the research topics of 'Nonlinear equalization method based on Gaussian mixture model clustering algorithm for a coherent optical OFDM communication system'. Together they form a unique fingerprint.

Cite this