Abstract:
Automatic identification of individuals using biometric features is an area that has gained
high importance nowadays. This study presents a novel approach for biometric identification
through ECG signal using hybridization of different features and Radial Basis Function Neural
Network (RBF-NN). Three different features, namely ARIMA, Wavelet Entropy, and Sample
Entropy, are extracted from an ECG dataset. The features are then fed to an RBF-NN to identify
different individuals. In the past, these features were used individually for person identification.
This paper presents an approach for person identification by hybridization of the features
mentioned above. The proposed approach shows promising results with an accuracy of 99.50% to
identify 55 individuals correctly
