Search

A hybrid machine learning (HML) model combining a-priori and a-posteriori knowledge is implemented and tested, which is shown to reduce the prediction error and training complexity, compared to an analytical or neural network learning model.

We propose and experimentally demonstrate an all-optical architecture for data center interconnect networks with reconfiguration times of a few seconds. Filtering and amplification transient effects have minimal impact on BER performance.

Optical transmission systems with high spectral efficiency require accurate quality of transmission estimation for optical channel provisioning. However, the wavelength-dependent gain effects of erbium-doped fiber amplifiers (EDFAs) complicate precise optical channel power prediction and low-margin operation. In this work, we examine supervised machine learning methods using multiple artificial neural networks (ANNs) to build models for gain spectra prediction of optical transmission line EDFAs under different operating conditions. Channel-loading configurations and channel input power spectra are used as an a posteriori knowledge data feature for model training. In a hybrid learning approach, estimated gain spectra calculated by an analytical model are added as an a priori input data feature to further improve the EDFA ANN model performance in terms of prediction accuracy, training time, and quantity of training data. Using these methods, the root mean square error and maximum absolute error of the predicted channel output power can be as low as 0.144 dB and 1.6 dB, respectively.

Recent advances in software and hardware greatly improve the multi-layer control and management of reconfigurable optical add-drop multiplexer (ROADM) systems facilitating wavelength switching. However, ensuring stable performance and reliable quality of transmission (QoT) remain difficult problems for dynamic operation. Optical power dynamics that arise from a variety of physical effects in the amplifiers and transmission fiber complicate the control and performance predictions in these systems.We present a deep-neural-network-based machine learning method to predict the power dynamics of a 90-channel ROADM system from data collection and training. We further show that the trained deep neural network can recommend wavelength assignments for wavelength switching with minimal power excursions.