Application Of Higher-Order Statistics In The Analysis Of Digital Earthquake Precursory Monitoring Data

The study of digital earthquake precursory monitoring data are the basis of earthquake prediction. Seismic precursory network have collected a large amount of observational data as seismic monitoring technologies become more and more mature. How to more effectively extract seismic anomaly information is a pressing problem. Higher-order statistics has become a new signal analysis tool. When some signals were analyzed by correlation function or power spectrum, the satisfactory results have been obtained. However, these issue can be soluted by using the higher-order statistics . At present, there is no theory and applied research which have been applied to the study of digital earthquake precursory monitoring data .Higher-order statistics methods have been applied to digital earthquake precursory monitoring data of Shandong Province. The relationship between seismic anomaly information and earthquake have been analyzed. The anomaly information extraction method show that some mutations or gradients of seismic precursory signals can be effectively identified.Compared with traditional methods, its resolution is higher. It can suppress noise, highlight the unusual features of information. It can make up the defects of the first order, second-order statistical information in the phase information. The lowest point of results corresponds to the beginning of the abnormal information. The lowest point is considered the warning point to determine abnormal. It should be able to bring must assist to earthquake prediction.The relationship between solid tidal abnormal and earthquake have been studied. The results show that accompanied by earthquake preparation solid tides would produce a certain abnormal changes. In tidal observing system which is sensitiver and stabler, the abnormal change can be effectively identified by higher-order statistics. The lowest point of results has guiding significance to extract seismic abnormal information. It shows that this method is very effective.Higher-order statistics method in the digital earthquake precursor research should have a very good development prospects,it should be able to provide some help to earthquake prediction.