基于小波变换的心音信号降噪方法

Noise reduction for heart sound based on wavelet transform

摘要:

为了弥补传统阈值函数在消噪过程中存在的不足，得到高信噪比的心音信号更好地进行心音分析，本文提出一种新的阈值函数。该函数通过灵活调节参数a和m的大小，更好地对染噪心音信号小波分解的每一层高频系数进行阈值量化。仿真中，应用传统的软、硬阈值函数及新阈值函数分别对大量的标准心音信号进行消噪处理，并对消噪效果进行了比较分析，同时将新阈值函数应用到实测心音信号消噪中。结果表明，新阈值函数能有效地消除噪声和保留心音信号的特征，具有较强的实用价值。

Abstract:

To compensate for the deficiencies of traditional threshold functions and obtain the heart sound with high SNR for better analysis，a novel threshold function was proposed in this study. The function quantified the detail coefficients of each level of noisy heart sounds effectively through adjusting the values of the parameters a and m. In the simulation, the traditional soft and hard threshold functions and the new threshold function were used in noise reduction for many cases of standard heart sounds respectively and the comparative analysis of the effect of noise reduction was made. The new threshold function was applied to noise reduction of the measured heart sounds as well. The results indicated that the new threshold function could better reserve the characteristics of heart sounds, eliminate noise effectively, and was of strong practical value.