小波分析及其去噪性能研究

Study on the wavelet transform and the de-noising property

小波变换是一种具有一定时间和频率分辨率的分析方法。为了寻找更有效的去噪方法,对连续小波变换(CWT)、离散序列小波变换(DTWT)和正交小波变换(QWT)作了理论分析。结果表明:CWT对噪声具有良好的抑制作用;DTWT在通过域值处理的同时,也考虑了因噪声造成的某些信号细节的丢失;而QWT则能有效地提取淹没在噪声中的微弱信号。与传统的傅里叶变换(FT)相比较,具有显著的优点,值得关注。

The wavelet transform is a kind of multi-resolution method on finite time and frequency. In order to seek out de-noising method,this paper studied the continous wavelet transform (CWT),discrete wavelet transform (DTWT) and orthenormal wavelet transform (QWT) in theory and practice. The results showed that CWT has a good suppression function for noise. When it is going through the Valve Processing, DTWT also consider useless noise to produce some losses of particular signal; But QWT can effectively pick up the weak signal submerged in the noise. Compared with the Fourier transform, the wavelet has marked ad vantages, hence worthy of note.