Search

In order to separate effectively small deformation from GPS signal containing the error, Based on multiwavelet theory, the paper makes DGHM for GPS deformation monitoring signal processing, because the DGHM biorthogonal wavelet is with short support, orthogonality, symmetry ( antisymmetry ) and two approximation and other characteristics, in the same length of filtering, multi-wavelet method is more superior than traditional wavelet method, improves the signal to noise ratio, obtains higher accuracy, confirms correct and practical in the actual problem, the algorithm has opened a new way for deformation monitoring signal processing

In the deformation monitor data, the noise exists inevitably. According to the different dissemination characteristic of the signal and the noise in the different scale, Wavelet modular maximum de-nosing algorithm combines the method of threshold de-noising with scale trading theory by Witkin, and picks out modular maximum value of the noise, makes use of Modular Maximum value of the signal to recover the coefficients of different scale efficiently, until recovers the signal. Contrast to the traditional threshold value and translation invariance, the processing result shows de-nosing method based on wavelet modular maximum can pick out the noise in the deformation data, protect the original information, have a better effect on de-noising, enforce the signal to noise ratio,Wavelet modular maximum de-nosing algorithm is a effective method in the deformation monitor data processing.

With wavelet technology used more widely in GPS deformation analysis, the paper will talk multi-wavelet (the second generation wavelet) theory used for GPS deformation monitoring data analysis. Due to multi-wavelet has a short branch set, orthogonal, symmetric (antisymmetric) and two order approximation characteristics, in the same filter length, using multiple wavelet on deformation monitoring signal can obtain better denoising effect than traditional wavelet. The paper studies signal adopting different preprocessing methods, makes a study of the selection problem in multi-wavelet preprocessing method. The GPS deformation monitoring signal is disposed using different multi-wavelet which adopts optima preprocessing method, and the paper makes a comparison to the conventional odd wavelet. The result confirms: multi-wavelet is more superiority than conventional wavelet, which decreases RMSE, advances SNR, obtains higher analytic precision, conforms the validity and practicability in physical problem, and offers a new road for deformation monitoring signal process