A Novel Biometric Identification Framework based on ECG Waveforms using k-mean Clustering Algorithm

Abstract

Recent advancements in computing and digital signal processing technologies have made automated identification of people based on their biological, physiological, or behavioral traits a feasible approach for access control. The wide variety of available technologies has also increased the number of traits and features that can be collected and used to more accurately identify people. Systems that use biological, physiological, or behavioral trait to grant access to resources are called biometric systems. In this paper, we present a biometric identification system based on the Electrocardiogram (ECG) signal. A novel hardware and software framework are proposed and investigated in this work. The proposed system is based on ECG amplitude features and a K-mean clustering for finding the significant features for the identification. The system is installed using an Arduino microcontroller and a Bluetooth signal transmitter is designed on electronic circuit and used for testing the system. Experimental results show significant success ratio and that the proposed system is reliable and feasible.