非均匀噪声分布心电信号的奇异值小波消噪法

De-noising method for electrocardiograph of non-uniform noise distribution based on singular value decomposition and wavelet transform

针对一般消噪法对噪声非均匀分布心电信号消噪存在的不足,提出基于奇异值分解和小波阈值消噪相结合的消噪方法。该方法利用矩阵的奇异值分解将噪声非均匀分布的心电信号正交分解为噪声分布相对均匀的分量,在正交子空间中对每个分量进行小波阈值消噪,重构消噪后分量,得到消噪后的心电信号。研究结果表明:本方法有效地克服了因噪声分布不均匀而造成的小波阈值选择矛盾的缺点,有效地消除了大噪声区域的噪声,又完好保存小噪声区域的心电特征信息,且消噪后的信号与无噪信号之间的欧氏距离最小。

Aimed at the deficiency of common de-noising method for ECG（electrocardiograph）of non-uniform noise distribution,a de-noising method,which is based on singular value decomposition and wavelet transform,was proposed.ECG of non-uniform noise distribution was decomposed into several orthogonal-components,which had comparatively uniform noise,by singular value decomposition.Each component was de-noised by wavelet threshold de-noising method and de-noising ECG was obtained by rebuilding de-noised orthogonal-components.The results show that the proposed method can avoid the inconsistency of threshold choosing,which is caused by noise uneven distribution,the proposed method can de-noise thoroughly in big noise part and completely preserve the features of ECG in small noise part,and the Euclid distance between de-noised signal and non-noise signal is the least.