Branch Identification in Passive Optical Networks using Machine Learning
A machine learning approach for improving monitoring in passive optical networks with almost equidistant branches is proposed and experimentally validated. It achieves a high diagnostic accuracy of 98.7 localization error of 0.5m
READ FULL TEXT