Machine learning techniques applied to system characterization and equalization

Darko Zibar, Jakob Thrane, Jesper Wass, Rasmus Thomas Jones, Molly Piels, Christian Schaeffer

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review


Linear signal processing algorithms are effective in combating linear fibre channel impairments. We demonstrate the ability of machine learning algorithms to combat nonlinear fibre channel impairments and perform parameter extraction from directly detected signals.
Original languageEnglish
Title of host publicationProceedings of 2016 Optical Fiber Communications Conference and Exhibition
Number of pages3
PublisherOptical Society of America (OSA)
Publication date2016
ISBN (Print)9781943580071
Publication statusPublished - 2016
Event2016 Optical Fiber Communication Conference and Exhibition - Anaheim Convention Center, Anaheim, United States
Duration: 20 Mar 201624 Mar 2016


Conference2016 Optical Fiber Communication Conference and Exhibition
LocationAnaheim Convention Center
Country/TerritoryUnited States

Bibliographical note

From the session: DSP for Coherent Systems (Tu3K)


  • signal processing
  • equalisers
  • learning (artificial intelligence)
  • optical fibre communication
  • parameter extraction
  • system characterization
  • system equalization
  • linear signal processing algorithms
  • machine learning algorithms
  • nonlinear fibre channel
  • Optical noise
  • Signal to noise ratio
  • Modulation
  • Phase noise
  • Estimation
  • Nonlinear optics
  • Optical polarization
  • Optical communication
  • Communication channel equalisation and identification
  • Signal processing and detection
  • Knowledge engineering techniques
  • Digital signal processing


Dive into the research topics of 'Machine learning techniques applied to system characterization and equalization'. Together they form a unique fingerprint.

Cite this