Summary
International Conference on Emerging Technologies for Communications
2020
Session Number:B2
Session:
Number:B2-4
Machine-Learning Approach for Binary Classification of Signal Transmission Modes in Human Body Communication Channels
Ai-ichiro Sasaki, Kazuki Morita, Akinori Ban,
pp.-
Publication Date:2020/12/2
Online ISSN:2188-5079
PDF download
Summary:
Signal transmission paths in human body communication (HBC) systems are formed by capacitive coupling among human bodies, transceivers, and earth. When another person approaches the HBC system, the person is inevitably included in the system because the coupling between the person and the HBC system becomes strong. In this situation, eavesdropping and accidental data transmission can occur. To avoid these security problems, techniques for distinguishing two different HBC-channel modes must be developed. To address this problem, an attempt was made to predict the channel modes from received signal information by using an optical voltage sensor and machine learning. It was demonstrated that machine learning can predict correct modes with an accuracy of approximately 90%. Optical voltage sensors were found to be effective for correctly characterizing HBC channels.