An Architecture for Detection of Anomalies in Deterministic Time within Real-Time Communication Networks.
Christian Maier, Jia Lei Du, Stefan Fahrthofer, Peter Dorfinger (2021): An Architecture for Detection of Anomalies in Deterministic Time within Real-Time Communication Networks. In: CORETA 2021: Advances on Core Technologies and Applications.
Anomaly detection is a classical and important topic within the domain of communication networks. For critical applications, the time it takes to detect an anomaly and to respond to it is an important metric of an anomaly detection system. To address this, we propose an end-to-end real-time anomaly detection architecture. In this architecture, the collection and transmission of the required data, the analysis of this data in a machine learning model and the subsequent reaction when an anomaly is detected, are carried out in deterministic time. We study two different use cases where this architecture may be applied in the future and we investigate a demo implementation of one of these use cases as a proof of concept for the proposed architecture. This contributes to the future application of machine learning for anomaly detection in deterministic time in time critical application areas.