Towards automated and proactive anomaly detection in fiber access networks

Communication networks are vital for society and network availability is therefore crucial. There is a huge potential in using network telemetry data and machine learning algorithms to proactively detect anomalies and remedy network problems before they affect the customers. In practice, however, there are many steps on the way to get there. In this presentation we would like to share the status of an ongoing research collaboration with the purpose of simplifying the operation and increase the availability of fiber access networks. The research project is part of a CELTIC_NEXT flagship research program (AI-NET) that has the overall target of accelerating the digital transformation in Europe by intelligent network automation. The project (“Palantir”) consists of two parts: A field trial in a Swedish municipality network where telemetry data from more than 500 access switches, connecting more than 12000 households, are collected and analyzed for anomalies using machine learning. (The