| Title |
ECG based Personal Authentication using Principal Component Analysis |
| Authors |
조주희(Cho, Ju-Hee) ; 조병준(Cho, Byeong-Jun) ; 이대종(Lee, Dae-Jong) ; 전명근(Chun, Myung-Geun) |
| DOI |
https://doi.org/10.5370/KIEEP.2017.66.4.258 |
| Keywords |
ECG ; PCA ; Personal authentication ; k-NN |
| Abstract |
The PCA(Principal Component Analysis) algorithm is widely used as a technique of expressing the eigenvectors of the covariance matrix that best represents the characteristics of the data and reducing the high dimensional vector to a low dimensional vector. In this paper, we have developed a personal authentication method based on ECG using principal component analysis. The proposed method showed excellent recognition performance of 98.2 [%] when it was experimented using electrocardiogram data obtained at weekly intervals. Therefore, it can be seen that it is useful for personal authentication by reducing the dimension without changing the information on the variability and the correlation set variable existing in the electrocardiogram data by using the principal component analysis technique. |