一种新的小波消噪阈值选取方法

A Novel Thresholding Algorithm of Wavelet De-noising

小波多尺度分解是一种有效的信号去噪方法。对于非平稳信号的消噪,主要是选取合适的小波及每层小波系数的阈值。介绍了阈值计算的SURE阈值、多分辨率SURE阈值和平移不变阈值,以及硬取阈值、软取阈值估计子,同时,提出了一种新的小波消噪阈值选取方法———半硬取阈值(SHT,SemiHardThresholding),并利用电子测量中的3σ准则和已有研究成果,提出了相关参数的确定原则;并与硬取阈值、软取阈值进行比较。仿真结果表明,消噪效果有明显的改观。

It is an effective method to reduce the noise in signals by wavelet multiscale decomposition. The keys of reducing noise in nonstationary signals are to select good mother wavelet and thresholds for each layer wavelet decomposition coefficients. Threshold calculation methods, such as SURE, Multiresolution SURE, and Translation-Invariant thresholding, are introduced. Besides, estimate arithmetic operaqtors, including hard-thresholding, soft-thresholding, are presented too. In addition, a new means of threshold selections, named as SemiHard Thresholding, for wavelet denoising is first introduced, and the principles of deciding corresponding parameters are put forward based on 3σ-rule applied in electronic measurement and relevant research achievement. Further more, the comparisons of it with hard-thresholding, soft-thresholding, are carried out. The simulations are conducted about the applications of them in digital signal denoising, and the results shows the denoising effects are improved.