摘要
为了提高分析信号的信噪比,基于经验模态分解和自适应噪声抵消技术,提出了一种新的信号去噪方法。该方法首先对信号进行自适应噪声抵消,然后进行经验模态分解,得到不同尺度上的固有模态函数,再对不同尺度上的固有模态函数进行噪声属性判定,如果不是噪声则选用不同的滤波参数,进行自适应噪声抵消,最后对各尺度上噪声抵消后的信号进行重构,得到去噪后的信号。结果表明,该方法比基于最小均方误差准则的自适应噪声抵消方法更能有效地消除信号中的噪声。
In order to improve the signal noise ratio （SNR） of the analyzed signal, a kind of adaptive denoising scheme based on the empirical mode decomposition （EMD） and adaptive noise cancel （ANC） framework is presented. At first, the noisy signal is denoised by ANC, then it is decomposed adaptively into oscillatory components called imtrinsic mode functions （IMFs） by means of a process called sifting. Then the IMFs which are not noises, are processed separately by ANC. Finally, the filtered signals are synthesized to form the denoised signal. The results show that this method can eliminate the noise in the signal more effectively than the ANC method based on least mean square （LMS） criterion.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2008年第5期810-812,共3页
Systems Engineering and Electronics
基金
国家自然科学基金资助课题(60472108)
关键词
经验模态分解
自适应噪声抵消
最小均方算法
降噪
empirical mode decomposition
adaptive noise cancel
least mean square algorithm
noise reduction
