WristAuthen: A Dynamic Time Wrapping Approach for User Authentication by Hand-Interaction through Wrist-Worn Devices

摘要: The growing trend of using wearable devices for context-aware computing and pervasive sensing systems has raised its potentials quick reliable authentication techniques. Since personal writing habitats differ from each other, it is possible to realize user through writing. This great significance as sensible information easily collected by these devices. paper presents a novel system wrist-worn analyzing the interaction behavior with users, which both accurate efficient future usage. key feature our approach lies in much more effective Savitzky-Golay filter Dynamic Time Wrapping method obtain fine-grained metrics authentication. These new are relatively unique person independent platform. Analyses conducted on wristband-interaction data 50 users diversity gender, age, height. Extensive experimental results show that proposed can identify timely manner, false-negative rate 1.78\%, false-positive 6.7\%, Area Under ROC Curve 0.983 . Additional examination robustness various mimic attacks, tolerance training data, comparisons further analyze applicability.