Reason - Resilience and security of geospatial data for critical infrastructures

Critical infrastructures are becoming increasingly dependent on accurate and continuous position, navigation, and timing (PNT) services provided by Global Navigation Satellite Systems (GNSS). PNT services are critical for, e.g., stock market, electricity transmission, banking and security information systems, building industry, logistics and transport (maritime and road transport as well as aviation), wireless communications, and rescue services. These critical services will not be available or they will need to rely on backup services if GNSS signals are unavailable in the area. This makes these services vulnerable when it comes to disruption in GNSS signals as a result of natural or intentional interference, or occurrence of unexpected GNSS constellation level problems. This calls for continuous monitoring of the GNSS signal quality so that any anomalies can be detected, isolated, and reported to authorities and a seamless shift to back-up solutions can be made. This study aims at improving the security of supply of the services that rely on GNSS-enabled PNT by the use of emerging Machine Learning methods (such as Deep Learning) for improved situational awareness in GNSS throughout Finland. The study is based on a GNSS-Finland monitoring platform, which uses the permanent GNSS reference network in Finland (FinnRef) to detect and localize the disruptions in the GNSS signals. Using the big data available from GNSS-Finland, Deep Learning (DL) methods will be developed to investigate possible trends in signal quality, and to detect or predict signal anomalies. This will provide an assessment of the continuity and forecast of critical failures in positioning and timing information and thus improve the resilience of critical PNTdependent services and operations in Finland. For the improved resilience of timing services, we also aim to explore solutions for cost-effective, fibre-optic time transfer to a large number of geographical locations as well as develop software-defined-radio-based technologies for monitoring low-frequency timing signals and other signals of opportunity. As a future effort, case studies in critical locations are planned in collaboration with end users, both for monitoring the GNSS signal quality and to explore the potential of using back-up timing services.