Paper Info
Title
Toward A Low Cost, High Performance Ecg Based Biometrics: A Preliminary Work
Abstract
An electrocardiogram (ECG) test is not only a very important tool in diagnosing heart diseases, but also it has been proved that the ECG shows unique patterns for biometric recognition. Hence, Individual can be identified once ECG signature is extracted. In this paper we prepare the field to design a low cost and high performance person recognition using mobile sensor. As the work can be divided into three parts, sensing signals from mobile device, simulating the recognition algorithm and implementing the solution in a mobile application, we focus in this paper on the second part. Indeed we develop a basic identification method with MATLAB fed by some ECG signals from Physionet database, which involves preprocessing, features extraction, dimensionality reduction and classification. Results are very encouraging and can highly be improved, first by enlarging database and second, by the improvement of each block in the system.
Year
DOI
Venue
2018
10.1109/SSD.2018.8570470
2018 15TH INTERNATIONAL MULTI-CONFERENCE ON SYSTEMS, SIGNALS AND DEVICES (SSD)
Keywords
DocType
ISSN
Biometrics, ECG, identification, Principal Components Analysis and Signal Processing
Conference
2474-0438
Citations 
PageRank 
References 
0
0.34
0
Authors
7
Name
Order
Citations
PageRank
Mabrouka Zemzemi100.34
Nesrine Keskes200.34
Rim Barioul300.68
Sameh Fakhfakh400.34
Kallel, I.K.543.87
Nabil Derbel610739.81
olfa kanoun76440.74