Title: ECG modelling using wavelet networks: application to biometrics

Authors: Samer Chantaf, Amine Nait-Ali, Patrick Karasinski, Mohamad Khalil

Addresses: Laboratoire Images, Signaux et Systemes Intelligents, LiSSi, EA. 3956, Universite Paris-Est Creteil (UPEC), 61, Avenue du general de Gaulle, Creteil 94010, France. ' Laboratoire Images, Signaux et Systemes Intelligents, LiSSi, EA. 3956, Universite Paris-Est Creteil (UPEC), 61, Avenue du general de Gaulle, Creteil 94010, France. ' Laboratoire Images, Signaux et Systemes Intelligents, LiSSi, EA. 3956, Universite Paris-Est Creteil (UPEC), 61, Avenue du general de Gaulle, Creteil 94010, France. ' Azm Center for Research in Biotechnology and its Applications, Doctoral School for Sciences and Technology, Lebanese University, El Mitein Street, B.P 210, Tripoli, Lebanon

Abstract: This paper deals with human identification using normal ECGs. Precisely, we would like to highlight how one can achieve human identification by considering only the most significant parameters extracted from a model. In this work, parameters are extracted by modelling the ECG using wavelet networks. The radial basis neural network method is then used to classify these parameters. Thus, a useful analysis is performed to evaluate the robustness of the identification. For each recording condition, the proposed technique has been evaluated on a set of ECG signals corresponding to normal subjects. Consequently, very encouraging results have been obtained.

Keywords: ECG; electrocardiograms; biometrics; wavelet networks; RBF neural networks; classification; human identification; modelling.

DOI: 10.1504/IJBM.2010.033388

International Journal of Biometrics, 2010 Vol.2 No.3, pp.236 - 249

Published online: 01 Jun 2010 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article