Timely detection of Seismic waves in Ground motion data using Improved S-Transform

Abstract

In Seismology, exact recognition of arrival times of seismic wave has been a key challenge for a geologist. The arrival time of body waves provides significant information about the hypocenter location, source systems and tomography analysis. In this paper, Linear Gaussian kernel window method is incorporated with S-transform for time frequency decomposition of seismic signals. It progressively controls the window width to increase the concentration of energy and reduce the smearing effect in time-frequency region and minimize the spreading of wave. The IST is applied for time identification body wave components called Primary and Secondary waves. The accelerograph of famous earthquakes of San Francisco in 2014 and California in 2012 regions have been used for the analysis. The superiority of the results obtained from IST is compared with benchmark techniques continuous Wavelet Transform (CWT), Discrete Wavelet Transform (DWT) and S-Transform (ST) in terms of time detection.