Mehran SOLTANIĀ (Technical University of Denmark)
Spatial and spectral power evolution design using machine learning-enabled Raman amplification
Abstract
Raman amplification is one of the key technologies in improving the performance of fiber optic communication systems. In this talk, I will present our machine learning framework on Raman amplifier design for shaping the signal power evolution over the frequency and fiber distance. The proposed framework adjusts the Raman pump power values to obtain the desired two-dimensional (2D) profiles using a convolutional neural network (CNN) followed by the differential evolution (DE) technique. The CNN learns the mapping between the 2D profiles and their corresponding pump power values. Nonetheless, its performance is not accurate for designing 2D profiles of practical interest, such as a 2D flat or a 2D symmetric (with respect to the middle point in fiber distance). To adjust the pump power values more accurately, the DE fine-tunes the power values initialized by the CNN to design the proposed 2D profile with a lower cost value. The results assert the very good performance of the proposed CNN-assisted DE framework utilized in designing 2D flat and symmetric power profiles defined over the whole C-band. Furthermore, the proposed DE with the CNN initialization provides higher accuracy with lower variance compared to the randomly initialized DE optimization.