Design of a digital twin for NB-IoT satellite constellations

The deployment of Low Earth Orbit satellite constellations is challenging due to the dynamic nature of these networks, intermittent connections, and power limitations. To better understand and anticipate the system’s behaviour under new conditions, Digital Twins (DTs) has been studied as a valuable tool for simulation and analysis. DTs enable satellite operators to perform real-monitoring, anomaly detection, and forecasting studies of software updates or orbital changes. The present work sets forth a DT architecture that integrates real-time satellite data with AI-powered forecasting. This facilitates real-time, continuous monitoring of the constellation performance, enabling performance forecasting and the detection of anomalies. To conclude, different forecasting models are presented to overcome the discontinuity and interruption on satellite communications, as well as the results obtained that illustrate the errors of the different forecasting models. Presenting a framework for a DTs architecture dedicated to manage any satellite constellation, while the Distributed system Simulator can model satellite contacts, aiming to enhance global connectivity and efficiency.