A Quiet Place: An In-Depth Study of Mobile Public-to-Private Attacks

Being carried everywhere by end-users, smartphones constantly gather user data, some of which are private (e.g., geolocations) and others public (e.g., accelerometer readings). To protect against data leakage attacks, most mobile platforms implement permission schemes that restrict access only to private rather than public data. Ironically, such unprotected public data can be craftily exploited by attackers, resulting in an emerging threat of leaking users’ private information, which we refer to as a Mobile Public-to-Private (MP2P) attack. However, for this particular attack, the current surveys lack details and need more focus on its distinct features, while the proposed attack methods have become increasingly similar. To better understand the MP2P attack, we conducted an in-depth study of academic examples spanning over a decade (2012-2024). Specifically, we first investigate and categorize public resources that can be exploited to leak private information. Then, we systematically explore the standard workflow and methodologies of MP2P attacks. Further, we propose a new MP2P attack method within a typical user scenario, taking flight. Based on our experiments with real flight trips, our method successfully infers passengers’ flight numbers, departure locations, destinations, and routes. Having deeply analyzed MP2P attacks and unveiled a new attack method, this paper can assist security engineers in devising countermeasures against this attack.