Skip to main navigation Skip to search Skip to main content

Digital Predistortion Enhancement by Convolutional Neural Network for Probabilistic Shaped Discrete Multi-Tone Signal Transmission in Passive Optical Network

  • Qi Zhou
  • , Rui Zhang
  • , Shuyi Shen
  • , Chin Wei Hsu
  • , Shuang Yao
  • , Shang Jen Su
  • , Gee Kung Chang
  • Georgia Institute of Technology

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

3 Scopus citations

Abstract

We experimentally demonstrate a 68.2-Gb/s net data-rate probabilistic shaped discrete multi-tone transmission in passive optical network with 11G-class devices. The convolutional-neural-network strengthens digital predistortion performance with 1.1-dB improvement of system sensitivity over linear pre-equalization.

Original languageEnglish
Title of host publication2021 Optical Fiber Communications Conference and Exhibition, OFC 2021 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781943580866
StatePublished - Jun 2021
Event2021 Optical Fiber Communications Conference and Exhibition, OFC 2021 - San Francisco, United States
Duration: Jun 6 2021Jun 11 2021

Publication series

Name2021 Optical Fiber Communications Conference and Exhibition, OFC 2021 - Proceedings

Conference

Conference2021 Optical Fiber Communications Conference and Exhibition, OFC 2021
Country/TerritoryUnited States
CitySan Francisco
Period06/6/2106/11/21

Fingerprint

Dive into the research topics of 'Digital Predistortion Enhancement by Convolutional Neural Network for Probabilistic Shaped Discrete Multi-Tone Signal Transmission in Passive Optical Network'. Together they form a unique fingerprint.

Cite this