Flexible Raman Amplifier Optimization Based on Machine Learning-aided Physical Stimulated Raman Scattering Model
This work addresses the problem of improving signal quality and efficiency in optical communication systems, representing an incremental advancement in amplifier optimization.
The paper tackled the optimization of Raman amplifiers by using a machine learning-aided physical model to jointly optimize forward and backward pumping configurations, achieving a gain flatness of less than 1 dB over 4 THz for an unrepeatered 250 km transmission.
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.