MAC Address De-Anonymisation

TL;DR

This paper investigates the potential to de-anonymize users by linking WiFi probe request MAC addresses to individual identities through algorithm development and a distributed detection system, addressing privacy concerns.

Contribution

It introduces an identification algorithm for MAC addresses and demonstrates a distributed detection system, advancing privacy research and device tracking capabilities.

Findings

MAC address identification is feasible with simulated data

Distributed probe request detection can be implemented in real-world settings

The system effectively links MAC addresses to device locations

Abstract

This project is an exploration into analysing WiFi probe requests, a management frame described as part of the IEEE 802.11 protocol which publicly broadcasts the senders MAC address. The intention was to collect these probe requests to use as a basis to link people's information to the MAC of their device. This would enable people to be identified in future by the presence of their device near a probe request detector. Due to data protection and privacy issues preventing access to real data of people's names and locations, this project was divided into two parts. Firstly, an identification algorithm was developed and tested on simulated data sets of MAC addresses and names, to prove MAC address identification is possible. And secondly, a distributed system of probe request detectors coupled with a centralised MAC address database was developed to demonstrate that these simulated MAC addresses are obtainable in the real world. The capturing software was initially developed for Unix systems and uses a Django-powered web server to store data from multiple capturing devices. Python was used to model and test the identification algorithm.