Filtering of the Airborne Gravity Data by Wavelet Thresholding

The principle of the wavelet threshold filtering is introduced briefly.Based on the airborne gravity real data,the effectiveness of using wavelet thresholding in filtering airborne gravity disturbance is checked.Different wavelet,different thresholding techniques and different decomposition level are tested.Results show that the db8,db10 and coif5 are the three wavelets suit for airborne gravimetry.If the highest decomposition level of the wavelets is seven or eight and in each level a different but suitable threshold is used,the accuracy of 3~5×10-5ms-2 can be reached in comparison to the ground reference.This accuracy is almost same as the approximation introduced by the Butterworth lowpass filter with 0.005 Hz cutoff frequency.