Mobile apps have profoundly reshaped modern lifestyles in different aspects. Several concerns are naturally raised about the privacy risk of mobile apps. Despite the prevalence of encrypted communication, app fingerprinting (AF) attacks still pose a serious threat to users’ online privacy. However, existing AF attacks are usually hampered by four challenging issues, namely i) hidden destination, ii) invisible boundary, iii) app multiplexing, and iv) open-world recognition, when they are applied to wireless traffic. None of existing AF attacks can address all these challenges. In this paper, we advance a novel AF attack, dubbed PACKETPRINT, to recognize user activities associated with the app of interest from encrypted wireless traffic and tackle the above challenges by proposing two novel models, i.e., sequential XGBoost and hierarchical bag-of- words model. We conduct extensive experiments to evaluate the proposed attack in a series of challenging scenarios, including i) open-world setting, ii) packet loss and network congestion, iii) simultaneous use of different apps, and iv) cross-dataset recognition. The experimental results show that PACKETPRINT can accurately recognize user activities associated with the apps of interest. It achieves the average F1-score 0.884 for open-world app recognition and the average F1-score 0.959 for in-app user action recognition.

View More Papers

An In-Depth Analysis on Adoption of Attack Mitigations in...

Ruotong Yu (Stevens Institute of Technology, University of Utah), Yuchen Zhang, Shan Huang (Stevens Institute of Technology)

Read More

V-Range: Enabling Secure Ranging in 5G Wireless Networks

Mridula Singh (CISPA - Helmholtz Center for Information Security), Marc Roeschlin (ETH Zurich), Aanjhan Ranganathan (Northeastern University), Srdjan Capkun (ETH...

Read More

Too Afraid to Drive: Systematic Discovery of Semantic DoS...

Ziwen Wan (University of California, Irvine), Junjie Shen (University of California, Irvine), Jalen Chuang (University of California, Irvine), Xin Xia...

Read More

Interpretable Federated Transformer Log Learning for Cloud Threat Forensics

Gonzalo De La Torre Parra (University of the Incarnate Word, TX, USA), Luis Selvera (Secure AI and Autonomy Lab, The...

Read More