Abstract
Conference Title: 2017 International Conference on Electrical and Computing Technologies and Applications (ICECTA) Conference Start Date: 2017, Nov. 21 Conference End Date: 2017, Nov. 23 Conference Location: Ras Al Khaimah, United Arab Emirates Authentication using biological peculiarities, e.g. voice, retina, fingerprint, and face is a key solution in security issues. Recently, there has been a growing interest in the development of authentication algorithms based on the use of cardiovascular signals measured by electrocardiogram (ECG). In this paper, we develop an authentication algorithm based on ECG signals. In particular, we use eight features extracted from the ECG signal as inputs to a linear classifier where it has been found that each individual has a distinguished ECG feature and thus can be used for individual identification and authentication. We use the Linear Discrimination Analysis (LDA) to classify the ECG signals based on the value of the eight extracted features and identify the person from the data base in whom the input ECG signal belongs to. The effectiveness of proposed algorithm is demonstrated using actual ECG signals, and the performance is compared with the state-of-the-art techniques available in literature.