Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/11527
Full metadata record
DC FieldValueLanguage
dc.contributor.authorSairul Safie-
dc.contributor.authorNur Haris-
dc.contributor.authorAzavitra Zainal-
dc.contributor.authorJohn Soraghan-
dc.contributor.authorLykourgos Petropoulakis-
dc.contributor.author(UniKL MITEC)-
dc.date.accessioned2015-11-16T23:16:05Z-
dc.date.available2015-11-16T23:16:05Z-
dc.date.issued2015-11-17-
dc.identifier.urihttp://localhost/xmlui/handle/123456789/11527-
dc.descriptionConference venue: Kuala Lumpuren_US
dc.description.abstractThis paper compares the authentication performance between three topological structures of the Pulse Active (PA) technique as applied to Electrocardiogram (ECG) biometrics. PA is a feature extraction method derived from the concept of Pulse Width Modulation (PWM). PA decomposes an ECG signal into a series of output pulses based on the linear dependency constant setting between the ECG and the modulated signal. In this work, three types of periodic signals namely triangular, sinusoidal and cosine waves are assigned as the modulation signal of the PA topology. The authentication performance of these periodic signals to generate output pulses and feature vector are analysed and compared. In this work, PA topology using a triangular function as the modulation signal outperforms the use of sinusoidal and cosine waves in PA-ECG based authentication system.en_US
dc.subjectPulse Activeen_US
dc.subjectFeature Extractionen_US
dc.subjectElctrocardiogram Biometricen_US
dc.subjectAuthenticationen_US
dc.titleComparison of Pulse Active (PA) Modulation Signal for Electrocardiogram (ECG) Authenticationen_US
dc.conference.name2015 IEEE International Conference on Signal and Image Processing (ICSIPA)en_US
dc.conference.year2015en_US
Appears in Collections:Conference Paper



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.