Biometric Authentication using Hand Movement Information from Wrist-worn PPG Sensors

Recently, wrist-worn smart devices such as smart-watches are becoming popular, and there is a growing need for authentication systems on such devices. Many wrist-worn devices contain a photoplethysmographic (PPG) sensor with a purpose of measuring the user's heart rate. In this paper, in addition to using the PPG sensor to measure the heart rate, we use it to detect the user's hand movement. In our proposed method, we use not only the information of the hand movement, but also that of the stationary state to improve the robustness of authentication. We obtained the PPG signals from 15 subjects, whom in the experiment were asked to bend their wrist and put it back for several times, continuously. The results of the experiments revealed that a continuous wrist movement of three times achieved an equal error rate of 11.6 %, and a movement of nine times achieved an equal error rate of 8.8 %.