In this paper, we developed the wrist motion-based authentication system using accelerometer on Android terminals. In the previous researches, hand moving of writing the letter in field was set to target, although our system is using the wrist motion to improve visual silence and availability. Our system requires the moving after the vibration to improve the accuracy of authentication. We got characteristics of motion in 8 subjects. As the result, we got 11.25% of FRR at 10.357% of FAR on change the threshold estimation value.