2021
DOI: 10.3390/s21051568
|View full text |Cite
|
Sign up to set email alerts
|

Efficiently Updating ECG-Based Biometric Authentication Based on Incremental Learning

Abstract: Recently, the interest in biometric authentication based on electrocardiograms (ECGs) has increased. Nevertheless, the ECG signal of a person may vary according to factors such as the emotional or physical state, thus hindering authentication. We propose an adaptive ECG-based authentication method that performs incremental learning to identify ECG signals from a subject under a variety of measurement conditions. An incremental support vector machine (SVM) is adopted for authentication implementing incremental … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
5
0

Year Published

2021
2021
2023
2023

Publication Types

Select...
8
1

Relationship

1
8

Authors

Journals

citations
Cited by 13 publications
(7 citation statements)
references
References 65 publications
(56 reference statements)
0
5
0
Order By: Relevance
“…Table 3 shows the number of beat data generated by each subject for training and evaluating the proposed model. During the training process, the synthetic minority oversampling technique (SMOTE) [ 83 ], which is an oversampling method for the data augmentation, was applied to improve performance during the training process [ 45 ]. It is the method of generating a new sample using the distance between selected samples within the same group by applying the K-nearest neighbor (KNN) [ 84 ] algorithm.…”
Section: Methodsmentioning
confidence: 99%
See 1 more Smart Citation
“…Table 3 shows the number of beat data generated by each subject for training and evaluating the proposed model. During the training process, the synthetic minority oversampling technique (SMOTE) [ 83 ], which is an oversampling method for the data augmentation, was applied to improve performance during the training process [ 45 ]. It is the method of generating a new sample using the distance between selected samples within the same group by applying the K-nearest neighbor (KNN) [ 84 ] algorithm.…”
Section: Methodsmentioning
confidence: 99%
“…However, authentication with these ECGs has some limitations. Violent activities such as exercise may change the ECG features [ 43 ]; drugs such as caffeine may change the ECG features [ 19 ]; emotional changes may cause difficulties in ECG-based authentication [ 44 ]; the heart rate may change every day [ 45 ]. In this paper, experimental data were created to design robust models for problems caused by ECG that vary daily among these challenges.…”
Section: Introductionmentioning
confidence: 99%
“…Binary Convolutional Neural Network (BCNN), Support Vector Machine (SVM), Gaussian Naive Bayes (GaussianNB) and K-Nearest Neighbour (3-NN) classifiers were explored for thigh ECGbased identification. These classifiers were chosen because they have already been addressed in the literature as a method for identifying individuals using ECG signals [18][19][20].…”
Section: Classificationmentioning
confidence: 99%
“…This shortcoming makes it difficult to evaluate whether a method tested in resting can be applied in real life. For example, Kim et al [44] designed and tested a system using single-channel ECG signals measured from 11 males for 10 min over six days. However, they only performed experimental validation using data measured in the resting state and did not consider signals measured in real-life conditions.…”
Section: Ecg Based Authentication System Including Various Real-life Conditionsmentioning
confidence: 99%