小波去噪在超声多普勒信号处理中的应用

Wavelet denoising in doppler signal processing applications

在超声多普勒流量计的研制过程中,由于多普勒信号属于微弱信号,超声波回波信号的去噪处理这一步是十分必要的。通常采用小波阈值去噪的方法对此信号进行预处理,本文所采用的方法是在小波阈值去噪方法的基础上,利用一种改进的阈值函数,此阈值函数克服了硬阈值函数不连续的不足,继承了软阈值函数的连续性特点,并且解决了软阈值函数中存在的小波系数与分解小波系数之间存在着恒定偏差的缺陷,与此同时,它具有软硬阈值函数不可比拟的灵活性。对实验所获得的超声波回波信号进行去噪,仿真结果表明,采用新的阈值函数去噪能有效抑制在信号奇异点附近产生的Pseudo-Gibbs现象,并且从信噪比增益和均方误差意义上分析看出改进型阈值函数对于超声回波信号的去噪效果均优于传统的软、硬阈值方法。

In the development process of the doppler flowmeter, because of doppler signal belongs to the weak signal, denoising of ultrasonic echo signal is very important step. Usually it used wavelet threshold denoising to this signal pretreatment, The used method of this paper is based on the wavelet threshold denoising, that used a kind of improved threshold function, that overcome the shortcomings of the hard threshold for the discontinuous function, inherited the soft threshold function characteristics of continuity. It also solves the shortcoming that there is an invariable dispersion between the estimated wavelet coefficients and the decomposed wavelet coefficients of the soft-thresholding method. Meanwhile, It has incomparable flexibility compared to the soft and hard thresholding function. Denoising to the obtained ultrasonic echo signal, experiment result indicate, the new threshold function effectively restrain the Pseudo-Gibbs phenomenon in signal singularities from the nearby, At the same time, the improved threshold function is superior to traditional method of the soft and hard thresholding for denoising of ultrasonic echo signal by analyzing the signal-to-noise ratio （SNR） gain and the mean square error.