Analysis Of High-order Spectrum And Its Application In Seismic Exploration

Analysis of high-order spectrum is one of the most important methods in modern signal processing. Aimed at the problems of seismic wavelet estimation and identification of small fault in seismic data processing, based on utilizing the useful information of signal high-order spectrum, some research on theoretical data and practical application in seismic wavelet estimation and identification of small fault are studied in this thesis.On the seismic wavelet estimation, based on the fact that high-order spectrum contains the amplitude and phase spectrum of signal and using bispectrum of seismic, the phase spectrum and amplitude spectrum of wavelet is recovered and the wavelet is restructured. There is no any presumption of wavelet for this method. The processing results of synthetic data with different wavelet phase and real data show that the method is feasible and promising in improving the resolution of seismic data.On identification of small fault, based on the fact that the time difference of the reflected wavelet event is the main mark to identify small fault, time delay profile is drawn from time delay function of two adjacent traces of seismic data computed by fourth order cumulant, and it can show the distributed location of small fault. Through the time delay profile, the fault displacement curve can be obtained for fault quantitative interpretation. The processing results of synthetic data and real data show that this method is able to suppress noise and feasible in fault identification.