OSNR prediction for optical links via learned noise figures

We predict the per-channel OSNR of optical links with up to 23 EDFAs via a machine learning model based on learned noise figures from experimental data. For a 20 span link, the error margin to cover 99% of cases is less than 0.35 dB.

Saved in:
Bibliographic Details
Published in2021 European Conference on Optical Communication (ECOC) pp. 1 - 4
Main Authors Kamel, Sarah, Hafermann, Hartmut, Le Gac, Dylan, Dos Santos, Ludovic, Kegl, Balazs, Frignac, Yann, Charlet, Gabriel
Format Conference Proceeding
LanguageEnglish
Published IEEE 13.09.2021
Subjects
Artificial neural networks
Europe
Machine learning
Noise figure
Predictive models
Training
Uncertainty
Online AccessGet full text

Cover

More Information
Summary:We predict the per-channel OSNR of optical links with up to 23 EDFAs via a machine learning model based on learned noise figures from experimental data. For a 20 span link, the error margin to cover 99% of cases is less than 0.35 dB.
DOI:10.1109/ECOC52684.2021.9605932