The paper discusses the principle of zero-phase wavelet extracted from Ground Penetrating Radar (GPR) record at the base of coefficient of reflection being white noise and used to remove the random noise in the GPR signal with cross correlation filter.

Application of correlation filter in processing of microseismic data;

Five predict methods have been used like step by step regression analysis, neural network, correlation filter, cokriging and nonparametric regression analysis.

The principle and effection of correlation filter are also discussed in.

By means of stepwise regression analysis, neural network, correlation filtering, CoKrige, and nonparametric regression, this article realizes the prediction of porosity distribution from seismic attributes and logging data.

On the basis of analyzing and summarizing existing methods for prediction of reservoir parameters, the paper comprehensively utilizes geologic and geophysical information and uses the methods for planar distribution prediction of porosity parameters such as gradual linear regression, correlation filtering, neural network, co-Kriging and non-parameter regression through m.

The simulated signal analysis showed that the bearing defect characteristic frequency could be extracted effectively by means of Laplace wavelet correlation filtering.

According to the information characteristics of the dynamic imbalance signal and the definition of correlation function,the principles and algorithm of relevant filter with automatic tracking is given.