基于小波变换的重力变化信号处理技术研究

针对重力敏感器中所测信号的特点，提出了小波变换的多分辨率分析方法。采用了强制去噪和阈值去噪两种方法，利用Daubechies小波函数，分别针对三种典型情况进行了分析。结果表明，对于信噪比非常小的重力变化信号，阈值去噪法相对于强制去噪法来说效果并没有明显提高，第9层下重建信号的逼近误差满足性能指标。文中还讨论了加入确定性干扰的情况下微弱信号的提取情况，结果表明，加入大的确定性干扰后，为了满足设计要求，小波分解的尺度有所增加，整体的滤波效果与未加入确定性干扰的情况基本相同。

Signal Processing for Gravity Variation Based on Wavelet Transform

In view of the characteristics of gravity variation signal, a multi-resolution analyses method, i.e. wavelet transform, is presented. By using compulsory decreasing noise and threshold decreasing noise, detailed analyses are demonstrated for three typical cases with Daubechies wavelet function. The simulation results show that the error is not improved with threshold decreasing noise method, because the SNR of gravity variation signal is too small. The result with Daubechies wavelet function in 9th scale can meet the required performance. The analysis with deterministic disturbance show that the scale of wavelet transform must be increased to satisfy the requirements. The filtering effect with and without deterministic disturbance is quite similar.