S3: Side-Channel Attack on Stylus Pencil through Sensors
summary
System S3 infers what is written on an iPad by reading only background motion sensors and reconstructs strokes using a high-dimensional particle filter that tracks pencil tip location and 3D orientation, guided by an offline magnetic map built via wardriving. The attack requires only a benign background app to log magnetometer/accelerometer/gyroscope data and yields high accuracy in a 10-subject study: letters 93.9%, numbers 96%, shapes 97.9%, words 93.33%.
AI-written summary · about
related talks
06mListen to Your Fingers: User Authentication based on Geometry Biometrics of Touch Gesture
16mPerspective-Shift Attacks Against Optical Perception Sensors: A Novel Attack...
38mSide Channel Attacks: Lessons Learned Or Troubles Ahead?
24mMy Google Glass Sees Your Passwords!
21mInjected and Delivered: Fabricating Implicit Control over Actuation Systems by Spoofing Inertial Sensors
19mStay Cool! Understanding Thermal Attacks on Mobile-based User Authentication
30mThe Motion Sensor Western: The Good, the Bad, and the Ugly
34mA new categorization system for Side-channel attacks on mo...