Background
The widespread use of personal mobile devices, such as tablets and smartphones, has opened new possibilities for collecting data on individuals’ movements within cities while maintaining their anonymity. Extensive research has explored how personal mobile devices can serve as tools for measuring human presence. To ensure privacy, the collected data must be either anonymous or pseudo-anonymous, leading to a preference for management data.
A popular method involves analysing probe requests, which are Wi-Fi protocol messages sent by mobile devices as they search for access points. These messages contain media access control (MAC) addresses, which were once unique identifiers. However, to protect user privacy, companies like Google, Apple, and Microsoft have introduced algorithms that randomize MAC addresses, causing them to change frequently and unpredictably.
Our work aims to address the issue of fingerprinting Wi-Fi devices through the analysis of management messages, providing an alternative to outdated methods reliant on MAC addresses. By identifying messages from the same device, it becomes possible to count the number of devices in an area, track their duration of stay, and correlate these metrics with the people carrying them.
Datasets
In this context, we developed datasets containing traces of individually analysed devices on three non-overlapping channels, which can be used to study device behaviour or to generate synthetic traces with multiple devices.
Here are the links to our datasets:
Here is the software used to create our datasets:
CONFRONT – Challenge ON wifi FRame fingerprinting for people cOunting aNd Tracking (2024)
The CONFRONT challenge took place in 2024 as part of the MOST project. The challenge focused on analysing Wi-Fi management frames to identify patterns that could be used for people counting and tracking in an indoor environment. Participants were provided with Wi-Fi sniffed data, and their task was to develop algorithms capable of extracting useful insights from this data, with a particular emphasis on accuracy and efficiency. The competition aimed to foster collaboration and innovation in the area of Wi-Fi-based device fingerprinting, attracting participants interested in the practical applications of network traffic analysis. Some of the participants presented their work in the special session “Challenges and advanced technologies for crowd mobility monitoring: a 5G/6G approach” at the EuCNC & 6G Summit 2024 conference.
Contacts
lucia.pintor[at]unica.it
Funding
Ministero dello Sviluppo Economico – Cagliari Digital Lab – G27F22000040008
Ministero dell’Università e della Ricerca – Sustainable Mobility Center (MOST) – 00000023