Experimental Characterization of Raman Amplifier Optimization Through Inverse System Design

Optical communication systems are always evolving to support the need for ever-increasing transmission rates. This demand is supported by growth in complexity of which moving towards ultra-wideband and space-division multiplexing. Both directions will challenge design, modeling, optimization devices, subsystems, full systems. Amplification a key functionality this context, we recently demonstrated versatile machine learning framework designing modeling Raman amplifiers with arbitrary gains. In paper, perform thorough experimental characterization such framework. The applicability proposed approach, as well its ability accurately provide flat tilted gain-profiles, tested on several practical fiber types, showing errors below 0.5~dB. Moreover, channel power heavily employed further enhance rate, tolerance variations input signal spectral profile investigated. Results show that inverse design can highly accurate gain-profile adjustments different profiles even not considering information during training phase.