基于小波分析的信号滤波方法研究

Signal Filtering Based on Wavelet Transform

论述了小波分析的基本原理，研究了应用小波变换进行信号滤波的方法，通过正交小波包对信号的分解，把频率成分复杂的信号分解到互不重叠的频带，根据需要删除某些频带的信号（噪声），然后用小波包重构算法对信号进行重构，可实现对信号的滤波．给出了应用小波包变换对振动信号进行滤波的实例，结果表明该方法可有效地滤除信号中的确定性噪声和随机噪声，可提取出淹没在信号中的非常微弱的特征信息，与传统的信号滤波方法相比具有明显优点．

Signal filtering based on wavelet decomposition and reconstruction was presented． Firstly， the signal was decomposed Into different channels by wavelet transform． Then the signals of some channels were deleted and the signals of other channels were reconstructed， so the noise in the original signal was filtered．example of signal filtering using wavelet packet an was provided． The result demonstrates that the new method has excellent features in signal filtering and weak signal extracting．