Experimental Prediction and Design of Ultra-Wideband Raman Amplifiers Using Neural Networks

Abstract

A machine learning method for Raman gain prediction and multi-pump broadband amplifier design is experimentally demonstrated over a 100 nm-wide optical bandwidth. We show high accuracy and ultra-fast prediction of arbitrary gain profile over a 100 km-long SSMF span.

Identifiers

DOI

10.1364/ofc.2020.w1k.3

Citations and references