Artificial neural network equalizers for PAM-4 using DML-on-silicon

Nikolaos Panteleimon Diamantopoulos, Takuro Fujii, Hidetaka Nishi, Koji Takeda, Takaaki Kakitsuka, Shinji Matsuo

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Nonlinear equalizers (NLEs) based on artificial neural networks (ANNs) are studied for short-reach 4-level pulse amplitude modulation (PAM-4) transmissions [1,2] utilizing directly-modulated membrane lasers fabricated on silicon substrates (DMLs-on-Si) [3,4]. Using 28-GBaud signals over 2-km of single-mode fiber (SMF), we compare two ANN-NLEs with a previously developed reduced-complexity Volterra-NLE (VNLE) [5].

Original languageEnglish
Title of host publicationThe European Conference on Lasers and Electro-Optics, CLEO_Europe_2019
PublisherOSA - The Optical Society
ISBN (Electronic)9781557528209
Publication statusPublished - 2019
Externally publishedYes
EventThe European Conference on Lasers and Electro-Optics, CLEO_Europe_2019 - Munich, Germany
Duration: 2019 Jun 232019 Jun 27

Publication series

NameOptics InfoBase Conference Papers
VolumePart F140-CLEO_Europe 2019

Conference

ConferenceThe European Conference on Lasers and Electro-Optics, CLEO_Europe_2019
CountryGermany
CityMunich
Period19/6/2319/6/27

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Mechanics of Materials

Fingerprint Dive into the research topics of 'Artificial neural network equalizers for PAM-4 using DML-on-silicon'. Together they form a unique fingerprint.

Cite this