Estimation Algorithm for Low Frequency Seismic Motions Based on Extended Semblance Function

. Monitoring low frequency and long period ground motions is very important for earthquake detection and alarm systems, and signal processing techniques help to investigate the mechanism of deep ground motions and earthquake occurrence by detecting effective seismic information from the seismograms. In this paper a frequency domain algorithm is presented to estimate the travel time difference of seismic waves with multi-path interferences caused by the refraction effect. The spectra of the seismograms are calculated through the fast algorithm, and the multipath parameters as well as the difference of travel time between a reference position and the seismographic stations are given by an optimization algorithm in the frequency domain. The new approach can work under conditions of refraction interference, and improve the estimation performance by using an extended semblance function. Its effectiveness is demonstrated through some numerical examples, and it is shown that the proposed algorithm is applicable to the analysis of low frequency seismograms.