Abstract
Recently, user recognition methods to authenticate personal identity has attracted significant attention especially with increased availability of various internet of things (IoT) services through fifth-generation technology (5G) based mobile devices. The EMG signals generated inside the body with unique individual characteristics are being studied as a part of next-generation user recognition methods. However, there is a limitation when applying EMG signals to user recognition systems as the same operation needs to be repeated while maintaining a constant strength of muscle over time. Hence, it is necessary to conduct research on multidimensional feature transformation that includes changes in frequency features over time. In this paper, we propose a user recognition system that applies EMG signals to the short-time fourier transform (STFT), and converts the signals into EMG spectrogram images while adjusting the time-frequency resolution to extract multidimensional features. The proposed system is composed of a data pre-processing and normalization process, spectrogram image conversion process, and final classification process. The experimental results revealed that the proposed EMG spectrogram image-based user recognition system has a 95.4% accuracy performance, which is 13% higher than the EMG signal-based system. Such a user recognition accuracy improvement was achieved by using multidimensional features, in the time-frequency domain.
Highlights
In recent years, the ability to access various internet of things (IoT) services using mobile devices has become possible with the commercialized of 5G services, and such services have led to major changes in the interaction with the external environment which is not limited to only communication between users. 5G based mobile environment allows users to conveniently access services at anytime from anywhere on a realtime basis, and the structure of mobile networks has changed dramatically through information and communication technologies (ICT) technology [1]
It consists of a total of 40 subjects, and the data was organized using three hand gestures, such as Fig. 9, which were performed within the palm range, except for actions that were large in order to recognize hand gestures using hand gestures as ciphers
Experiments showed 82.1% user recognition accuracy when used the convolutional neural networks (CNN) designed in this paper composed of time information like the method of Hao et al and 86.4% user recognition accuracy was shown when used the artificial neural networks (ANN) designed in this paper composed of time and frequency feature like the method of Shin et al [15] and the user recognition accuracy obtained by converting the EMG signals into spectrogram images that can be analyzed in a multidimensional time-frequency domain was 95.4%
Summary
The ability to access various IoT services using mobile devices has become possible with the commercialized of 5G services, and such services have led to major changes in the interaction with the external environment which is not limited to only communication between users. 5G based mobile environment allows users to conveniently access services at anytime from anywhere on a realtime basis, and the structure of mobile networks has changed dramatically through information and communication technologies (ICT) technology [1]. In order to address such problems of biometrics information, there has been attention on a user recognition technology using bio-signals that are unique to each individual and generated inside the body. Studies are actively being conducted to improve the performance of user recognition technologies using ECG signals that contain unique features such as electrophysiological properties of the heart, the heart’s position and size, and the physical conditions. There is a problem of having low recognition accuracy To resolve this problem, this paper proposes an EMG signal-based user recognition system that extracts features in two-dimensional (2D) from a spectrogram image that can be analyzed multidimensionally in the time-frequency domain. In the proposed method of this paper, the short-time fourier transform (STFT) is applied to extract 2D features from EMG signals in the time-frequency domain, and the extracted features are converted to EMG spectrogram images.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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 callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
