Abstract
Electrocardiogram (ECG) signals are time series data that are acquired by time change. A problem with these signals is that comparison data that have the same size as the registration data must be acquired every time. A network model of an auxiliary classifier based generative adversarial neural network that is capable of generating synthetic ECG signals is proposed to resolve the data size inconsistency problem. After constructing comparison data with various combinations of the real and generated synthetic ECG signal cycles, a user recognition experiment was performed by applying them to an ensemble network of parallel structure. Recognition performance of 98.5% was demonstrated when five cycles of real ECG signals were used. Moreover, 98.7% and 97% accuracies were provided when the first cycle of synthetic ECG signals and the fourth cycle of real ECG signals were repetitively used as the last cycle, respectively, in addition to the four cycles of real ECG. When two cycles of synthetic ECG signals were used with three cycles of real ECG signals, 97.2% accuracy was shown. When the last third cycle was repeatedly used with the three cycles of real ECG signals, the accuracy was 96%, which was 1.2% lower than the performance obtained while using the synthetic ECG. Therefore, even if the size of the registration data and that of the comparison data are not consistent, the generated synthetic ECG signals can be applied to a real life environment, because a high recognition performance is demonstrated when they are applied to an ensemble network of parallel structure.
Highlights
Among the biometric signals investigated in studies of user recognition methods, electrocardiogram (ECG) signals are bio-signals that are produced autonomously and show unique characteristics of individuals according to such factors as the heart’s position, size, and structure, as well as age and gender [1]
In conventional user recognition studies using ECG signals, experiments are performed by constructing comparison data with the same size as the registration data in the initial experimental environment
After constructing comparison data with various combinations of the real ECG signal cycles and the generated synthetic ECG signal cycles, the user recognition experiment was conducted by applying them to an ensemble network of parallel structure
Summary
Among the biometric signals investigated in studies of user recognition methods, electrocardiogram (ECG) signals are bio-signals that are produced autonomously and show unique characteristics of individuals according to such factors as the heart’s position, size, and structure, as well as age and gender [1]. If the ECG signal acquired in response to the user’s condition change is used as comparison data, not the ECG signal acquired in the same environment, the cause of user recognition performance deterioration occurs To solve this problem, studies have been conducted for various data schemes using neural networks. In Section the class synthetic data gendivided, the structure of ACGAN the discriminator model is designed as the a CNN modelnetwork that repeats eration model using the proposed in this study and ensemble of the convolution of a structure is not deep when compared with that of the parallel structure operation for user recognition arethat described. This work is organized as follows: In Section 2, previous studies for generative adversarial-neural-network-based synthetic data generation are analyzed and conventional user recognition methods which use the ECG signal. Sec of 13 tion 5 provides the conclusion of this work
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
