ECG signal based human identification method using features in temporal and wavelet domains

This paper presents an effective method for human identification using temporal and wavelet domain features extracted from electrocardiogram (ECG) signal. Instead of directly using the ECG data of a person as the feature, first, it is shown that a few number of reflection coefficients extracted from the autocorrelation function of the data can efficiently perform the recognition task. Next, the discrete wavelet transform (DWT) coefficients are utilized as features, which offer even a better recognition performance. A combination of these two features is found to demonstrate a high within class compactness and between class separation. A pre-processing scheme is incorporated prior to feature extraction to reduce the effect of different noises and artefacts. In the recognition phase, a linear discriminant based classifier is employed, where the two features are jointly utilized. The proposed human identification method has been tested on a standard ECG database and high recognition accuracy is achieved with a low feature dimension.