Citation

Kim S, Lee K, You K. Sensors (Basel) 2020; 20(7): e1879.

Affiliation

Department of Electrical and Computer Engineering, Sungkyunkwan University, Suwon 16419, Korea.

Copyright

(Copyright © 2020, MDPI: Multidisciplinary Digital Publishing Institute)

DOI

10.3390/s20071879

PMID

32231075

Abstract

The discrimination between earthquakes and explosions is a serious issue in seismic signal analysis. This paper proposes a seismic discrimination method using support vector machine (SVM), wherein the amplitudes of the P-wave and the S-wave of the seismic signals are selected as feature vectors. Furthermore, to improve the seismic discrimination performance using a heterodyne laser interferometer for seismic wave detection, the Hough transform is applied as a compensation method for the periodic nonlinearity error caused by the frequency-mixing in the laser interferometric seismometer. In the testing procedure, different kernel functions of SVM are used to discriminate between earthquakes and explosions. The outstanding performance of a laser interferometer and Hough transform method for precision seismic measurement and nonlinearity error compensation is confirmed through some experiments using a linear vibration stage. In addition, the effectiveness of the proposed discrimination method using a heterodyne laser interferometer is verified through a receiver operating characteristic curve and other performance indices obtained from practical experiments.

Language: en

Keywords

Hough transform; heterodyne laser interferometer; measurement accuracy; seismic discrimination; support vector machine