摘要
处理低信噪比的测量信号的关键是去除噪声,小波阈值消噪法是较为实用的一种处理方法。分析了Donoho的去噪阈值函数的优劣,基于sigmoid函数构造了一个新的阈值函数,该函数克服了硬阈值函数的分段性及软阈值函数的恒定偏差及导数不连续的弱点。对低信噪比的3种典型信号进行对比去噪仿真实验,结果表明该函数的去噪效果优于Donoho的阈值函数去噪以及传统的快速傅里叶变换去噪。
The key to the processing of measuring signals with lower SNR is denoising. A practical method is wavelet threshold denoising. A new threshold denoising function is given based on the sigmoid function. The new function avoids to be discontinuous as hard threshold function and has no constant deviation of soft threshold function, The derivative of this new function is continuous. The denosing function of Donoho is analyzed. The simulation results show that the noises in three typical signals are denoised by the new threshold function more effectively than that by Donoho and FFT denoising.
出处
《计量学报》
EI
CSCD
北大核心
2007年第2期163-166,共4页
Acta Metrologica Sinica
关键词
计量学
小波分析
阈值去噪
信噪比
Metrology
Wavelet analysis
Threshold denoise
Signal- to- noise ratio （ SNR ）
