小波变换在信号降噪中的应用研究

Applied Research on Wavelet Transform in Signal De-noise

分析了小波变换的基本原理,采用小波变换和傅里叶变换方法,降噪处理了noisdopp信号并对比了处理结果,结果表明,小波阈值降噪在降噪的同时,能很好保存有用信号中的尖峰部分和突变部分.计算机仿真了实时采集的语音信号,结果表明,小波变换能从强噪声背景中提取有用信号,并能保留信号的大部分能量,且与原始信号有较好的相似性,能极大提高信号的信噪比,具有一定的工程应用价值.

The basic principle of wavelet transform is analyzed. Using the wavelet transform and FFT, noisdopp signal is de-noised, and the results are contrasted. Results show that when the wavelet threshold de-noising it can well retain the peak portion and mutation of useful signal. The speech signal collected in real-time is simulated by computer. Results show that the wavelet transform can extract useful signal from the strong background noise and retains most of the energy of signals, while having better similarity with the original signal. The wavelet transform improves the signal-to-noise ratio greatly and has important value of engineering application in signal de-noising.