QoT Estimation for Light-path Provisioning in Un-Seen Optical Networks using Machine Learning

Abstract

We propose the use of machine-learning based regression model to predict the quality of transmission (QoT) of an un-established lightpath (LP) in an un-seen network prior to its actual deployment, based on telemetry data of already established LPs of different network. This advance prediction of the QoT of un-established LP in an un-seen network has a promising factor not only for the optimal designing of this network but also enables the possibility to automatically deploy the LPs with a minimum margin in a reliable manner. The QoT metric of the LPs are defined by the Generalized Signal-to-Noise Ratio (GSNR) which includes the effect of both Amplified Spontaneous Emission (ASE) noise and Non-Linear Interference (NLI) accumulation. In the response of present simulation scenario, the real field telemetry data is mimicked by using a well reliable and tested network simulation tool GNPy. Using the generated data set, a machine-learning technique is applied, demonstrating the GSNR prediction of an un-established LP in an unrevealed network with maximum error of 0.53 dB.