Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Acoustic Side-Channel Attacks on a Computer Mouse

Mauro Conti, Marin Duroyon, Gabriele Orazi, Gene Tsudik

TL;DR

This work investigates acoustic side-channel attacks on computer mice to determine whether mouse movements emit audio signatures that can be exploited. Through phased experiments, it demonstrates high accuracy for four-direction movements using a single microphone (approximately 98%), and sustained accuracy for twelve two-dimensional movements with stereo smartphone recordings (around 94%), along with 91% accuracy for detecting a window-close action. The approach combines MFCC features with ML classifiers (CNN and random forest) in both controlled and real-world scenarios, including multi-participant validation. These findings reveal a tangible security risk from mouse-based ASCAs and motivate mitigations and further exploration of multi-peripheral leakage, potentially in combination with keyboard ASCAs.

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.

Acoustic Side-Channel Attacks on a Computer Mouse

TL;DR

This work investigates acoustic side-channel attacks on computer mice to determine whether mouse movements emit audio signatures that can be exploited. Through phased experiments, it demonstrates high accuracy for four-direction movements using a single microphone (approximately 98%), and sustained accuracy for twelve two-dimensional movements with stereo smartphone recordings (around 94%), along with 91% accuracy for detecting a window-close action. The approach combines MFCC features with ML classifiers (CNN and random forest) in both controlled and real-world scenarios, including multi-participant validation. These findings reveal a tangible security risk from mouse-based ASCAs and motivate mitigations and further exploration of multi-peripheral leakage, potentially in combination with keyboard ASCAs.

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.
Paper Structure (19 sections, 10 figures, 5 tables)

This paper contains 19 sections, 10 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Research Gap:
  4. Methodology
  5. Mouse and Mouse Pad
  6. Audio Recording Methods
  7. Acoustic Analysis of Mouse Movements
  8. Single Microphone Analysis
  9. Audio Capture via Processing Software
  10. Continuous Audio Capture During Mouse Movements
  11. Limitation of single microphone approach
  12. Analyzing Proximity Through Sound Amplitude Variations
  13. Dual-Microphone Approach for Two-Dimensional Amplitude Analysis
  14. Real-world Implications and Security Risks
  15. Experiment with Other Participants
  16. ...and 4 more sections

Figures (10)

  • Figure 1: Recording phase of Left to Right movement. The user has to move the cursor from the green area to the red one.
  • Figure 2: Accuracy and loss graph for classifying four movement categories using Processing Record script.
  • Figure 3: Accuracy and loss graph for four area classification using background recording.
  • Figure 4: Circular recording method.
  • Figure 5: Audio waveforms for the mouse moving towards and away from the microphone.
  • ...and 5 more figures