摘要
提出一种基于离散平稳小波的改进自适应降噪方法。首先,利用离散平稳小波的冗余特性,解决离散二进小波变换降噪方法在奇异点存在振荡效应的问题;其次,针对传统小波阈值降噪算法忽略尺度系数噪声影响的不足,利用噪声强度估计各分解层阈值,对尺度和小波系数同时进行自适应降噪;最后,将此方法应用于不同信噪比下典型信号的降噪对比试验。仿真结果表明:该方法在消噪的定性和定量指标上,整体优于传统离散二进小波方法,消噪效果改善明显。
An improved self-adaptive de-noising method based on discrete stationary wavelet transform was given.The problem of the singular-points oscillation effects,was solved by the redundant nature of the discrete stationary wavelet transform.Aiming to the shortage of ignoring the noise influence in approximation coefficients,both detail and approximation coefficients were self-adaptive denoised using the thresholds,which were estimated by the noise intensity accordingly.At last,the method was applied to the typical signals de-noisong experiments in different SNR.The simulation result shows that in qualitative and quantitative indicators,the de-noising effect of the method is better than the traditional discrete binary wavelet transform.
出处
《仪表技术与传感器》
CSCD
北大核心
2012年第11期166-168,171,共4页
Instrument Technique and Sensor
关键词
离散平稳小波变换
阈值降噪
自适应
尺度系数
discrete stationary wavelet transform
threshold de-noising
self-adaptive method
approximation coefficients
