Table of Content

Open AccessOpen Access


User Authentication System Based on Baseline‐corrected ECG for Biometrics

Gyu Ho Choia, Jae Hyo Jungb, Hae Min Moonc, Youn Tae Kimb, Sung Bum Panc

aDept. of Control and Measurement Engineering, Chosun Univ. Gwangju, South Korea;
bDept. of IT Convergence Engineering, Chosun Univ. Gwangju, South Korea;
cDept. of Electronics Engineering, Chosun Univ. Gwangju, South Korea

* Corresponding Author: Sung Bum Pan,

Intelligent Automation & Soft Computing 2019, 25(1), 193-204.


Recently, ECG-based user authentication technology, which is strong against forgery and falsification, has been actively studied compared to fingerprint and face authentication. It is impossible to measure the open ECG DB measured with expensive medical equipment in daily living, and the ECG measured with the developed device for easy ECG measurement has much noise. In this paper, we developed a device that easily measures the ECG for user authentication in everyday life, measured the ECG through the development equipment, adjusted the baseline correction of the measured ECG, extracted it from the adjusted ECG do. The proposed system includes the steps of obtaining ECGs, pre-processing ECG signals, segmenting the signals and extracting features thereof to evaluate user authentication performance. The ECG lead-Ⅰ was obtained from 100 subjects for 120 seconds while they were comfortably positioned. Although the maximum user authentication performance of the existing algorithm was EER of 1.82%, it was observed the maximum performance of the proposed user authentication system was 0.71% higher than the existing algorithm.


Cite This Article

G. H. Choi, . J. H. Jung, . H. M. Moon, . Y. T. Kim and . S. B. Pan, "User authentication system based on baseline‐corrected ecg for biometrics," Intelligent Automation & Soft Computing, vol. 25, no.1, pp. 193–204, 2019.

This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
  • 1183


  • 975


  • 0


Share Link

WeChat scan