Acoustic Side-Channel Attacks on a Computer Mouse

2025-05-05

International Conference on Detection of intrusions and malware, and vulnerability assessment

AI Summary

This work presents the first systematic study of acoustic side-channel attacks (ASCAs) emanating from mouse operations, demonstrating their feasibility for inferring user interaction behavior. Leveraging smartphone microphones, we capture faint acoustic emissions generated by mouse clicks, scrolling, and movement. Using CNN and RNN models, we extract discriminative features and classify these signals under controlled conditions—achieving 97% accuracy across four fundamental mouse actions. With six participants, we extend classification to twelve distinct mouse actions with 94% accuracy. Furthermore, in realistic settings, we successfully detect full-screen window closure—a privacy-sensitive operation—with 91% accuracy. This research transcends prior acoustic attack paradigms focused on keyboards, establishing mice as viable targets for acoustic side-channel analysis. It is the first to empirically validate that mouse-generated acoustic emissions pose a tangible security threat, thereby broadening the scope of hardware-based side-channel vulnerabilities and providing critical insights for designing robust mitigation strategies.

Technology Category

• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Machine Learning: Privacy
• Data Mining & Knowledge Management: Knowledge Acquisition from the Web

Application Category

• Security and Privacy: Large-scale security measurements
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Web Mining and Content Analysis: Web measurements

Abstract

Acoustic Side-Channel Attacks (ASCAs) extract sensitive information by using audio emitted from a computing devices and their peripherals. Attacks targeting keyboards are popular and have been explored in the literature. However, similar attacks targeting other human interface peripherals, such as computer mice, are under-explored. To this end, this paper considers security leakage via acoustic signals emanating from normal mouse usage. We first confirm feasibility of such attacks by showing a proof-of-concept attack that classifies four mouse movements with 97% accuracy in a controlled environment. We then evolve the attack towards discerning twelve unique mouse movements using a smartphone to record the experiment. Using Machine Learning (ML) techniques, the model is trained on an experiment with six participants to be generalizable and discern among twelve movements with 94% accuracy. In addition, we experiment with an attack that detects a user action of closing a full-screen window on a laptop. Achieving an accuracy of 91%, this experiment highlights exploiting audio leakage from computer mouse movements in a realistic scenario.

Problem

• Investigates acoustic side-channel attacks on computer mice
• Classifies mouse movements with high accuracy using ML
• Detects user actions via audio leakage from mice

Innovation

• Classifies mouse movements using acoustic signals
• Employs smartphone-recorded audio for movement detection
• Uses ML to generalize mouse movement classification