Summary
International Workshop on Smart Info-Media Systems in Asia
2021
Session Number:RS3
Session:
Number:RS3-2
A Study on Anomalous Signal Detection Using Convolutional Neural Network for ELF Band Electromagnetic Wave
Hideo Kohatsu, Akitoshi Itai, Ichi Takumi, Hiroshi Yasukawa,
pp.128-132
Publication Date:2021/9/20
Online ISSN:2188-5079
PDF download
Summary:
The electromagnetic (EM) wave radiated from the earth’s crust is known as important information to understand the precursor phenomena of great earthquakes. We record the EM wave in an extremely low frequency (ELF) band. The recorded data includes various noises from the thunder radiation in near fields and tropics, artificial radiation, and so on. The detection and extraction of anomalous signals related to the precursor of an earthquake is an important task. Some supervised classification approaches are proposed to detect anomalous signals from the ELF EM wave signal. However, the analysis of anomalous signals is not discussed enough. If the analysis of anomalous signal is progressed, the performance of supervised classification is improved. In this paper, we discuss a possibility of the anomalous signal detection and analysis using a convolutional neaural network (CNN) and waveform of EM wave signals.