ECG signals for human identification based on fiducial and non-fiducial approaches

Biometric systems are mostly used for human identification and authentication. Recent developments have shown that ECG human identification can be used as a powerful tool as it gives more reliable and accurate results. In this paper, a proposed human identification system based on ECG as a biometric is presented with different feature extraction methods. Different feature extraction methods such as Daubechies wavelet ('db3', 'db8' and 'db10'), Symlets wavelet 'sym7' and Biorthogonal wavelet 'bior2.6' are exploited in this work. A combination of radial basis function (RBF) neural network and Backpropagation (BP) neural network is used as a classifier. The proposed system gives an identification rate of 98.41% with Daubechies wavelet 'db8'. In addition, the identification rate for Daubechies wavelet ('db3'and 'db10'), Symlets wavelet 'sym7' and Biorthogonal wavelet 'bior2.6' increases when R-R intervals are added as fiducial features of the non-fiducial features.