Privacy-Aware Crowd Monitoring and WiFi Traffic Emulation for Effective Crisis Management

Estimating the number of people in a given area, denoted as “people counting” process, plays a vital role in crisis management and disaster response, enabling accurate monitoring of crowd dynamics and facilitating effective decision-making. In this work, we focus on WiFi fingerprint technique which exploits the MAC address of the mobile devices as proxy for people counting. Due to the European GDPR regulation and the strict actions undertaken by the major smart-devices vendors to enhance users' privacy (e.g., MAC randomization), most of the techniques investigated in the past must be redesigned and rethought. Here, we propose an ad-hoc WiFi traffic generator, tailored to emulate a realistic behaviour of the WiFi cards and to provide the ground truth for the counting algorithms. Furthermore, we propose a technique for crowd monitoring that leverages Bloom filters to guarantee a formal “deniability” property, which preserves users' privacy. Our solution is also compatible with trajectory-based crowd monitoring.