Pulse Active Bit (PAB) feature extractor for ECG biometric authentication

This paper presents a new feature extraction method known as Pulse Active Bit (PAB) implemented on electrocardiograph (ECG) signals for biometric authentication. This method is developed based on a simple amplitude comparison between two signals for feature extraction. A total of 486 ECGs from 113 subjects taken from the Physikalisch-Technische Bundesanstalt (PTB) database are used in this study. Biometric performance profile such as the area under ROC (AUR) and equal error rate (EER) are then used to evaluate the results. It is shown in this study that PAB outperforms conventional temporal feature extraction techniques.