Date of Award
11-20-2015
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Computer Science
First Advisor
Mengjun Xie
Abstract
Wrist worn smart devices are increasingly popular. As those devices collect sensitive personal information and often work as a companion component of smart phones, thus being treated trustworthy, appropriate user authentication mechanisms are necessary to prevent illegitimate access to those devices. However, the small form and function-based usage of wearable devices also pose a big challenge to authentication, which should be user friendly and unobtrusive. We propose MotionAuth, a behavioral biometric based authentication method, which collects user’s behavioral biometrics through a wrist worn smart device for building a user’s profile, and applies the profile in user validation. We implement MotionAuth using Android smart watch and test its effectiveness with data collected in a user study involving 30 users and four simple, natural gestures. MotionAuth is shown to achieve high accuracy (as low as 2.6% EER value) with average EER values lower than 5% using a histogram based verification method.
Recommended Citation
Yang, Junshuang, "MotionAuth: Motion-Based Authentication for Wrist Worn Smart Devices" (2015). Theses and Dissertations. 647.
https://research.ualr.edu/etd/647
