Data Fusion and Machine Learning for Innovative GNSS Science Use Cases

The volume of data produced worldwide is growing rapidly, from 33 zettabytes in 2018 to an expected 175 zettabytes in 2025 Furthermore, the way in which data are stored and processed will change dramatically over the coming 5 years. Today 80% of the processing and analysis of data takes place in centralised computing facilities, and 20% in smart connected objects, such as cars, home appliance, manufacturing robots and computing facilities close to the user ('edge computing'). By 2025 these proportions are likely to be inverted. In the GNSS space segment, according to current development plans, over 120 GNSS satellites (including European Galileo satellites) will provide, already this decade, continuous data, in several frequencies, without interruption and on a permanent basis. This unique opportunity for science has been recognised by the European Space Agency (ESA) with the creation of the Navigation Science Office, which leverages on GNSS infrastructure to deliver innovative solutions across Earth Science, Space Science, Metrology and Fundamental Physics domains. At the core of this initiative, the GNSS Science Support Centre (GSSC) combines Big Data and Machine learning (ML) technologies to extract knowledge and discover patterns between GNSS-related inputs and outputs given the sheer volume of data. In this work, we introduce key GNSS Science Use Cases, providing a detailed view of GSSC on-going initiatives regarding troposphere and ionosphere characterisation through ML science pipelines to exploit a unique, publicly available repository of multifaceted GNSS data and products.