Mobility detection through electromagnetic fingerprints in 5G networks

In recent years, our cities have become increasingly smart, thanks to the massive use of IoT (Internet-of-Things) sensors and smart devices owned by people, life in the city has improved considerably. One of the most popular and important studies concerns the monitoring of people flows with the aim to guarantee both better services to citizens, from the point of view of dimensioning transportation networks, bike lanes, and public routes. However, it is also significant for identifying and quantifying people in sensitive areas (e.g., for safety and security purposes, such as during large crowd gatherings) or in areas of transit. It is essential to introduce new technologies for tracking people in case of health emergencies, such as the one we have been experiencing for many months. In this thesis work, the analysis of people's mobility is studied in a restricted area between the Politecnico di Torino and Porta Susa train station. The study uses six commercial probe-detection sensors (APs scanners) installed in strategic positions to collect data from smart devices carried by passers-by. The scanners can collect WiFi and Bluetooth probe requests messages sent periodically by all the active devices searching for known APs or devices to connect with. The source MAC address of such messages (even if randomized and anonymized) allow inferring the mobility of each mobile device. By means of two NFV (Network functions virtualization) on the Edge Cloud, all the data are retrieved, stored, and visualized with some graphs and charts, relative to the detected MAC addresses, in a real-time dashboard. Thanks to the analysis of the data collected, it is possible to identify the most frequent patterns and types of mobility around the area of interest. Furthermore, it is relevant to see the effect that the lockdown, caused by COVID-19, has had on campus life but also in people's everyday travels. The platform's implementation took into account a future expansion on the number of sensors throughout the city; the platform adopts the MEC (Multi-Access Edge Computing) paradigm by which the computation is distributed across the mobile infrastructure, allowing to scale the approach and its accuracy.