Determining Location and Movement Pattern Using Anonymized WiFi Access Point BSSID

Abstract

Location tracking applications are usually based on exact GPS location information and non-anonymized WiFi Access Point (AP) information. It has been assumed that anonymizing WiFi AP information is an effective way to preserve location privacy of the connected users. In this work, our goal is to show that the privacy of connected users' movement pattern can still be compromised by a determined attacker even if the WiFi AP information is anonymized. We investigated the feasibility of tracking user's movement between locations from anonymized WiFi AP BSSIDs based on large Device Analyzer datasets. Our experiments show that the user's daily movement pattern can be identified with almost 83% accuracy.