Flexible Raman Amplifier Optimization Based on Machine Learning-aided Physical Stimulated Raman Scattering Model

TL;DR

This paper introduces a machine learning-based approach to optimize Raman amplifiers by creating a differentiable model that enables gradient-based tuning of pump frequencies and powers for improved signal amplification over long distances.

Contribution

It develops a novel ML-aided differentiable Raman gain model enabling joint optimization of forward and backward pumps for enhanced amplifier performance.

Findings

Achieved gain flatness of less than 1 dB over 4 THz.

Successfully optimized a 250 km unrepeatered transmission link.

Validated the optimization results with numerical simulations.

Abstract

The problem of Raman amplifier optimization is studied. A differentiable interpolation function is obtained for the Raman gain coefficient using machine learning (ML), which allows for the gradient descent optimization of forward-propagating Raman pumps. Both the frequency and power of an arbitrary number of pumps in a forward pumping configuration are then optimized for an arbitrary data channel load and span length. The forward propagation model is combined with an experimentally-trained ML model of a backward-pumping Raman amplifier to jointly optimize the frequency and power of the forward amplifier's pumps and the powers of the backward amplifier's pumps. The joint forward and backward amplifier optimization is demonstrated for an unrepeatered transmission of 250 km. A gain flatness of $<$ 1~dB over 4 THz is achieved. The optimized amplifiers are validated using a numerical simulator.