Smart_Auth: User Identity Authentication Based on Smartphone Motion Sensors
Zhengjie Wang, Naisheng Zhou, Fang Chen, Xiaoxue Feng, Fei Liu, Yinjing Guo, Da Chen · 2021
Nowadays, the smartphone saves a lot of the private information of users. The traditional authentication methods, such as passwords or patterns, have some limitations, so it is not enough for legitimate users to protect personal privacy. In this paper, we propose a user identity authentication system for smartphone users called Smart_Auth. Multiple motion sensors of the smartphone are used to collect data, and a deep learning network is employed to realize smartphone user identity authentication. We invite 11 users to participate experiment. The legitimate user performs circle actions, and the illegitimate user performs random actions. The sampling frequency of smartphone motion sensors is 200Hz, and about 7260 samples are collected. The system achieves a 99.18% accuracy of legitimate users authentication and prevents 98.18% of illegitimate users from intruding into the smartphone.