Effective Identity Management on Mobile Devices Using Multi-Sensor Measurements
Feng, Tao 1985-
MetadataShow full item record
Due to the dramatic increase in popularity of mobile devices in the past decade, sensitive user information is stored and accessed on these devices every day. Securing sensitive data stored and accessed from mobile devices, makes user-identity management a problem of paramount importance. The tension between security and usability renders the task of user-identity verification on mobile devices challenging. Meanwhile, an appropriate identity management approach is missing since most existing technologies for user-identity verification are either one-shot user verification or only work in restricted controlled environments. To solve the aforementioned problems, we investigated and sought approaches from the sensor data generated by human-mobile interactions. The data are collected from the on-board sensors, including voice data from microphone, acceleration data from accelerometer, angular acceleration data from gyroscope, magnetic force data from magnetometer, and multi-touch gesture input data from touchscreen. We studied the feasibility of extracting biometric and behaviour features from the on-board sensor data and how to efficiently employ the features extracted to perform user-identity verification on the smartphone device. Based on the experimental results of the single-sensor modalities, we further investigated how to integrate them with hardware such as fingerprint and Trust Zone to practically fulfill a usable identity management system for both local application and remote services control. User studies and on-device testing sessions were held for privacy and usability evaluation.