Skip to content

Passive crowd sensing with personal communication technologies

Abstract

Personal wireless communication technologies such as Wi-Fi, Bluetooth, and cellular systems such as 4G and 5G have come to play a central role in contemporary society. These technologies enable users to establish seamless connectivity, facilitate communication, and exchange data across increasingly complex and interconnected digital infrastructures. Despite their ubiquity, the underlying technical mechanisms that support such connectivity remain largely opaque to end users. At the physical layer, radio waves enable all forms of wireless communication by carrying EM energy through the atmosphere and surrounding environment. Without them, no wireless transmission of information would be possible. Owing to their pervasive integration into everyday life, wireless communication services are often implicitly assumed to be continuously available and reliable. Beyond their conventional role in data transmission, radio signals can also be exploited for environmental sensing. By analysing the interaction between transmitted signals and the surrounding environment, it becomes possible to extract information about objects, motion, and spatial behaviour. This sensing capability effectively transforms wireless infrastructures into distributed sensing systems, thereby enabling a new class of applications that extend well beyond traditional communication paradigms. Current people-counting systems often rely on methods that involve privacy-intrusive data collection, such as visual or identity-related information from individuals within a monitored area. Radio-based sensing approaches are, in principle, not exempt from such concerns. However, extracting meaningful information from radio signals typically requires complex signal processing techniques, particularly when decoding the data embedded within these signals. Importantly, such decoding is only necessary if the objective is to access the transmitted information itself. In this dissertation, we propose an alternative, non-intrusive approach. We hypothesise that crowd presence and density can be inferred by analysing the aggregate energy of ambient radio signals generated by wireless devices. Specifically, we assume that higher levels of detected radio signal energy correlate with a greater number of individuals, while lower energy levels indicate fewer individuals. This approach deliberately avoids inspection of the encoded data within the signals, focusing instead on measuring signal power over time as a proxy for crowd estimation. To accomplish this, a measurement framework is developed based on the acquisition and analysis of RF signals. In contrast to inspecting the payload or content of transmitted data packets, this work emphasises the analysis of the EM energy emitted by radio waves. Specifically, the captured signals are examined in both the time and frequency domains to extract relevant characteristics and patterns.

View source

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.