Machine Learning for PIN Side-Channel Attacks Based on Smartphone Motion Sensors

Motion sensors are integrated into all mobile devices, providing useful information for a variety of purposes. However, these sensor data can be read by any application and website accessed through a browser, without requiring security permissions. In this paper, we show that information about smartphone movements can lead to the identification of a Personal Identification Number (PIN) typed by the user. To reduce the amount of sniffed data, we use an event-driven approach, where motion sensors are sampled only when a key is pressed. The acquired data are used to train a Machine Learning (ML) algorithm for the classification of the keystrokes in a supervised manner. We also consider that users insert the same PIN each time authentication is required, leading to further side-channel information available to the attacker. Numerical results show the feasibility of PIN cyber-attacks based on motion sensors, with no restrictions on the PIN length and on the possible digit combinations. For example, 4-digit PINs are correctly recognized at the first attempt with an accuracy of 37%, and in five attempts with an accuracy of 63%.

Automated summary

Motion sensors in mobile devices can be accessed by any application or website without security permissions, allowing for the identification of PIN numbers typed by the user. The study uses an event-driven approach to reduce data sniffing, training a supervised machine learning algorithm to classify keystrokes. Numerical results show the feasibility of motion sensor-based PIN cyber-attacks, with high accuracy rates even for 4-digit PIN numbers.