Evaluation of single-channel electrocardiograph from the wrist inorder to confirm the identity using amplitude and frequencydomain features

In this study, in order to investigate the possibility of using theelectrocardiographic signal for identification purposes, which plays animportant role in the field of security and privacy, a device was designedto record the signal from the wrist. Also, data was collected from ۱۲healthy volunteers in the age range of ۲۵ ± ۵ years using the mentioneddevice. After sorting, the data was refined by Butterworth filter and ۱۷features were extracted in the time and frequency domain from the ۹۶obtained data. The raw data were windowed in one-second windowswithout overlap. Then the feature extraction results were presented to adecision tree classifier. Based on the obtained classification, the datawere separated with ۹۸.۹۶ percent accuracy, which indicates the highpotential of this signal to be used as a person identifier in identityrecognition systems. It seems that it is necessary to conduct more studiesin order to use this technology in real industry projects.

Keywords:

Electrocardiography , signal processing , identity recognition , featureextraction , decision tree