PrintListener: Uncovering the Vulnerability of Fingerprint Authentication via the Finger Friction Sound

TL;DR

This work identifies a new acoustic-side-channel vulnerability in fingerprint authentication by showing that finger-friction sounds from screen swiping, captured via ordinary social apps, can reveal fingerprint pattern features and enable targeted PatternMasterPrint dictionary attacks. The authors implement PrintListener, a three-stage pipeline that preprocesses audio, maps fingerprint patterns through wide/deep feature fusion, and synthesizes PatternMasterPrints (independent, sequential, and synthetic) via a random-restart hill-climbing strategy focused on a central region of the fingerprint. Empirical results across multiple datasets, devices, and environments demonstrate substantial attack potency, including wASR improvements over baseline MasterPrint methods, with up to 27.9% wASR for partial fingerprints at FAR = 0.01% within five attempts. The findings highlight practical risks for real-world deployments and suggest mitigations such as limiting audio sampling rates and employing noise-suppressing or user-behavior-aware defenses. Overall, the work underscores the need to reassess biometric security in the presence of covert acoustic leakage channels and to design robust countermeasures against such side-channel attacks.

Abstract

Fingerprint authentication has been extensively employed in contemporary identity verification systems owing to its rapidity and cost-effectiveness. Due to its widespread use, fingerprint leakage may cause sensitive information theft, enormous economic and personnel losses, and even a potential compromise of national security. As a fingerprint that can coincidentally match a specific proportion of the overall fingerprint population, MasterPrint rings the alarm bells for the security of fingerprint authentication. In this paper, we propose a new side-channel attack on the minutiae-based Automatic Fingerprint Identification System (AFIS), called PrintListener, which leverages users' fingertip swiping actions on the screen to extract fingerprint pattern features (the first-level features) and synthesizes a stronger targeted PatternMasterPrint with potential second-level features. The attack scenario of PrintListener is extensive and covert. It only needs to record users' fingertip friction sound and can be launched by leveraging a large number of social media platforms. Extensive experimental results in realworld scenarios show that Printlistener can significantly improve the attack potency of MasterPrint.

Figures (16)

• Figure 1: Attack scenario of PrintListener.
• Figure 2: Variations of $FAR$ with threshold settings in D0 (mixed patterns), D1 (whorl), D2 (left loop), and D3 (right loop) Datasets.
• Figure 3: Propagation paths of frictional sound waves.
• Figure 4: 2D optical graph and corresponding 3D topographic map (including ridge and valley lines) of the left loop, whorl, and right loop. (The 3D topographic areas of greater pressure are marked as red. The dashed lines indicate the overall direction of ridges lines, the triangles represent the singular points of different patterns).
• Figure 5: A proof of concept for two-dimensional mapping of frictional sound characteristics of (a) different fingerprint patterns; (b) blurred fingers.