A Simple, Efficient and Innovative Biometric Human Identification Using Weighted Thresholding and KNN

Biometric methods for identifying human beings are in demand for various real-time monitoring and security purposes. However, the methods used for the identification have their own restrictions and setbacks. Consider the iris and fingerprint scanners, wherein the real-time monitoring is restricted because of portability. Unlike these, ECG biometric is a suitable tool that can be used for security and is portable. ECG has properties of being a biometric because of the unique features that can be extracted from the fiducial points (P, Q, R, S, T), most notably the wrist, leading to simple real-time wearable product development. Here, a detailed procedure for implementing ECG as a biometric has been mentioned with an accuracy of 98% for the standard 24 hour MIT-BIH arrhythmia database.