Mobile apps have become ubiquitous, but their network traffic is often encrypted. Researchers at arXiv introduce MAppPrint, a method for app fingerprinting (AF) that analyzes patterns in encrypted traffic to infer which apps are in use. The paper describes a technique based on traffic behavior correlation graphs, which capture relationships between traffic flows to identify apps without decrypting the data.

The abstract highlights two key design goals: robustness and real-time performance. By correlating traffic behaviors rather than relying on simple packet statistics, MAppPrint aims to withstand variations in network conditions and app versions. The method is positioned as a practical AF attack that works despite encryption, raising privacy concerns for mobile users.

Since the abstract does not include experimental results, the actual accuracy or speed of MAppPrint is not detailed here. The contribution lies in the graph-based approach, which offers a new angle on a known problem. The paper is available on arXiv under identifier 2610.22148.