Loading

Deep Neural Network-based Person Identification using ECG Signals
Rudresh T K1, Mallikarjun S H2, Shameem Banu L3

1Rudresh T. K., Lecturer, Department of Electronics and Communication Engineering, Government Polytechnic, Chamarajanagar (Karnataka), India.

2Mallikarjun S. H., Lecturer, Department of Electronics and Communication Engineering, Government Polytechnic, Kampli (Karnataka), India.

3Shameem Banu L, Lecturer, Department of Electronics and Communication Engineering, Government Polytechnic, Bellari (Karnataka), India.

Manuscript received on 21 July 2023 | Revised Manuscript received on 28 July 2023 | Manuscript Accepted on 15 August 2023 | Manuscript published on 30 August 2023 | PP: 14-21 | Volume-12 Issue-6, August 2023 | Retrieval Number: 100.1/ijeat.F42620812623 | DOI: 10.35940/ijeat.F4262.0812623

Open Access | Editorial and Publishing Policies | Cite | Zenodo | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: In recent times, biometrics is mainly utilized for the authentication or identification of a user for a vast civilian application. Many electronic systems have been proposed that employ distinct behavioural or physiological human signatures for automatically identifying or verifying users. Currently, Electrocardiogram (ECG)-oriented biometric systems are in the exploratory stage. The behaviour of the ECG signal is distinctive to every person. As ECG is an exclusive physiological signal present only in living people, it is utilised in new biometric systems for recognising individuals and counteracting fraud and forgery attacks. The majority of traditional techniques are limited by restrictions in several points of detection in the ECG signal. The contribution of this paper is the enhancement of the novel person identification model using ECG signals. Initially, the ECG signal collected from the three benchmark sources undergoes pre-processing, during which noise is removed using a low-pass filter (LPF) approach. Furthermore, the Empirical Mode Decomposition (EMD) is employed for decomposing the signal. As feature selection is a significant part of classification enhancement, Principal Component Analysis (PCA) is used as a practical feature extraction method that selects the most critical features from the signal. Finally, the adoption of a Deep Neural Network (DNN) is performed as a deep learning model that can identify the exact person from the given ECG signal. The effectiveness of the method is extensively validated on benchmark datasets, yielding the desired outcome.

Keywords: Deep Neural Network, ECG signals, PCA-based Feature Selection, Person Identification, Signal Decomposition.
Scope of the Article: Deep Learning