Research on the Human Identification Algorithm Using ECG Signal

Abstract

In the current research of ecg identification, the non-stationary characteristics of ecg waveform are not considered. In the process of power spectrum estimation, ecg waveform is regarded as Gaussian signal, which is greatly affected by Gaussian noise. In order to improve the accuracy of human identification, this paper proposes an identification algorithm of ecg signal based on higher order autocorrelation and non-fiducial point extraction. The proposed method is verified by ptb database. The results show that this method has the advantage of high recognition rate of ecg signal.