Seismic events modeling via FM/sup m/let transform

Author

Fan, Wanchun ; Qiu, Hongmao ; Fan, Yanfang ; Sun, Yu ; Wang, Haijun

Author_Institution

Dept. of Autom. Control, Xi´´an Jiaotong Univ., China

Volume

1

fYear

2003

fDate

14-17 Dec. 2003

Firstpage

696

Abstract

A novel method is proposed to extract the time-frequency effective features of the seismic ground motion with both the linear and nonlinear time-variant structures. The extraction is based on FM/sup m/let transform that is potentially suitable for the characterization of nonstationary processes. The FM/sup m/let atoms are dilated and translated windowed exponential frequency modulated functions with Gaussian envelope. The features are defined by the parameters of FM/sup m/let atoms and the weighting coefficients by matching the signal local natures successively. The FM/sup m/let based signal representations naturally lead to highly compact solutions to a plethora of stationary and nonstationary scenarios. Accordingly, the time-dependent evolutionary random process model of underground nuclear explosions is obtained by their mutual parametric features.

Keywords

frequency modulation; geophysical signal processing; nuclear explosions; random processes; seismology; signal representation; time-frequency analysis; transforms; FM/sup m/let transform; Gaussian envelope; compact solutions; evolutionary random process model; frequency modulated function; mutual parametric features; nonstationary process characterization; seismic events modeling; seismic ground motion; signal representations; time-dependent random process model; time-frequency effective feature; time-variant structure; underground nuclear explosions; Chirp; Density functional theory; Explosions; Feature extraction; Frequency modulation; Random processes; Signal representations; Spectral analysis; Sun; Time frequency analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks and Signal Processing, 2003. Proceedings of the 2003 International Conference on

Conference_Location

Nanjing

Print_ISBN

0-7803-7702-8

Type

conf

DOI

10.1109/ICNNSP.2003.1279370

Filename

1279370