New estimation algorithm for epicenter location of low frequency seismograms

Monitoring low frequency and long period ground motions is important to investigate the motion mechanism under deep ground and helps to understand earthquake occurrence, where the signal processing techniques and numerical optimization have played a fundamental role in detecting seismic information from the seismograms. In this paper an extended spectral analysis based algorithm is presented to estimate the travel time difference of seismic waves for epicenter location. The spectra of the seismograms are calculated through a fast algorithm, and the difference of travel time between a reference position and the seismographic stations is given by an optimization algorithm in the frequency domain. Furthermore, a GA based estimation algorithm is used for epicenter location. The new approach is more efficient than the existing time domain methods, and can compensate the discretization error caused by low sampling rate. Its performance is demonstrated through some numerical examples, and it is shown that the proposed algorithm is applicable to the analysis of low frequency seismograms.