The article presents a cluster-based technique to detect and locate eavesdropping events in optical line systems. This method uses optical performance monitoring (OPM) data collected at the receiver to detect such events. To locate these events, in-line OPM data is used. The application of machine learning in this context helps maintain transmission quality while monitoring. The method finds a balance between supervised learning, which requires large amounts of labeled data, and unsupervised learning, which is more suitable for anomaly detection as it does not require prior knowledge of abnormal events.
Publication date: 28 Sep 2023
Project Page: Not provided
Paper: https://arxiv.org/pdf/2309.14541
