Power Evolution Prediction and Optimization in a Multi-span System Based on Component-wise System Modeling

TL;DR

This paper presents a machine learning approach to predict and optimize power profiles in multi-span optical communication systems using component-wise system modeling.

Contribution

It introduces a novel combination of ML-based gain modeling and differentiable fiber models for power prediction and optimization in multi-span systems.

Findings

Accurate power profile predictions in multi-span systems.

Effective optimization of power distribution for system performance.

Demonstrated applicability to real optical communication setups.

Abstract

Cascades of a machine learning-based EDFA gain model trained on a single physical device and a fully differentiable stimulated Raman scattering fiber model are used to predict and optimize the power profile at the output of an experimental multi-span fully-loaded C-band optical communication system.