An Innovative Framework for Human Biometric Identification Using ECG Signals

Biometric techniques, like fingerprint, iris, and facial recognition, provide ways to identify people based on their distinct physiological characteristics. An electrocardiogram can additionally function as a unique biometric signature, according to recent research. Empirical methods are the mainstay of current ECG-based identification algorithms for feature extraction. A different strategy is presented in this study, which makes use of a time-domain system of the ECG signal. Characterizing ECG signals is a good fit for auto-regressive integrated moving average (ARIMA) models, which are renowned for their reliable representation of periodic time series data. We describe a process for modeling electrocardiograms, taking features out of the resulting model representations, and using these features to identify specific people.