Machine learning-based EDFA Gain Model Generalizable to Multiple Physical Devices

We report a neural-network based erbium-doped fiber amplifier (EDFA) gain model built from experimental measurements. The model shows low gain-prediction error for both the same device used for training (MSE $\leq$ 0.04 dB$^2$) and different physical units of the same make (generalization MSE $\leq$...

Full description

Saved in:
Bibliographic Details
Main Authors Da Ros, Francesco, de Moura, Uiara Celine, Yankov, Metodi P
Format Journal Article
LanguageEnglish
Published 11.09.2020
Subjects
Computer Science - Learning
Physics - Optics
Online AccessGet full text

Cover

More Information
Summary:We report a neural-network based erbium-doped fiber amplifier (EDFA) gain model built from experimental measurements. The model shows low gain-prediction error for both the same device used for training (MSE $\leq$ 0.04 dB$^2$) and different physical units of the same make (generalization MSE $\leq$ 0.06 dB$^2$).
DOI:10.48550/arxiv.2009.05326