Abstract
In this study, we propose BreathPass, a non-invasive authentication system that characterizes the chest/abdomen movement incurred by human breath to enable unlocking smart devices while wearing various types of face covers, clothing, in different postures, and dynamic status such as walking or running. To capture the breathing pattern, BreathPass uses speakers to emit ultrasound signals. The signals are reflected off the chest wall and abdomen and then back to the microphone, which records the reflected signals. The system then extracts the breathing pattern from the reflected signals, and further extracts fingerprints from the breathing pattern, and use these fingerprints to perform authentication. We carefully design a Deep Neural Network model and explore its capacity for feature abstraction in order to address the challenges associated with tiny position changes resulting in different breathing patterns and the extremely narrow bandwidth of breathing. We implement a prototype and conduct extensive experiments. BreathPass achieves an overall accuracy of 83%, a true positive rate of 73%, and a false positive rate of 5%, according to performance evaluation results.
Fan DANG
Assistant Professor
My research interests include industrial intelligence and edge computing.