Abstract: ECG signal in the process of collecting is always mixed with a variety of noise, the basic theory and method of wavelet transform can be used to remove the noise. After the study of wavelet multi-resolution theory and Donoho’s soft and hard threshold de-noising method, an improved threshold function method is proposed. Improved threshold function can overcome the?defects of soft and hard threshold method. MIT-BIH ECG database is used for verification and different methods de-noising results are analyzed. The experimental results show that the improved threshold method can effectively remove noise, and gives better SNR performance than the soft and hard threshold method.

Key words: electrocardiogram（ECG）, wavelet transform, improved threshold, de-noise

摘要： 心电信号（ECG）在采集的过程中总是参杂着各种噪声，可利用小波变换基本原理和方法进行去噪处理。对小波多分辨率理论进行研究后，在分析Donoho的软、硬阈值去噪法的基础上，提出一种改进阈值函数量化方法，改进阈值函数能克服软、硬阈值存在的缺陷，运用MIT-BIH心电数据库进行验证，并对ECG 信号用不同方法去噪后的结果进行了分析比较。实验表明，改进阈值方法可以有效地去除不同噪声干扰，在信噪比指标上也明显优于常用的软、硬阈值去噪方法。

关键词: 心电信号（ECG）, 小波变换, 改进阈值, 去噪