Automation of switching in COADM using machine learning algorithm

Abstract

In this paper a Machine Learning (ML) algorithm has been proposed based on application in field of Optical Network, where in it makes use of large data set to learn, train the switching nodes and predicts the traffic in the network. Configurable Optical Add-Drop Multiplexer (COADM) are used as the switching nodes. Once prediction is done, the traffic at the node is directed to the next node automatically. This improves the performance in terms of efficiency and reduces the delay in the network due to automation.