An Enhanced Machine Learning Based Biometric Authentication System Using RR-Interval Framed Electrocardiograms
share
The disadvantages of traditional authentication systems include the risks of forgetfulness, loss, and theft. Hence, biometric authentication is rapidly replacing traditional authentication methods and is becoming an everyday part of life. The electrocardiogram (ECG) is recently introduced and an ECG-based authentication system suitable for security checks. The proposed authentication system supports investigators studying ECG-based biometric authentication techniques to reshape input data by slicing based on RR-Interval and to define the Overall Performance (OP) measure which is the combined performance metric of multiple authentication measures and this new performance measure is newly proposed in this paper. We evaluated the performance of the proposed system using a confusion matrix and this authentication system could be performed up to the 95 percent accuracy with the compact data analysis. We are also applying the Amang ECG (amgecg) toolbox in MATLAB to investigate the MSE-based the upper range control limit (UCL) that directly affects three authentication performances: the accuracy, the number of accepted samples and the OP. Using this approach, we found that the OP is optimized by using a UCL of 0.0028 which indicates 61 accepted samples out of 70 and 95 percent accuracy of the proposed authentication system.
READ FULL TEXT