心电信号小波去噪的改进算法研究

发布时间：2018-02-28 10:34

本文关键词： 心电信号小波函数阈值阈值函数信噪比　出处：《中国生物医学工程学报》2017年01期 　论文类型：期刊论文

【摘要】：对心电信号(ECG)中的基线漂移、工频干扰和肌电干扰等噪声进行去除,在波形识别、医疗诊断和治疗等领域具有重要意义。提出用sym5小波函数对心电信号进行8层小波分解。根据有用信号强度在每一层平均分配而噪声强度随分解层数增加而减少的规律,将分解得到的每一层的小波细节系数设置不同的阈值,最后用所提出的新阈值函数进行小波阈值去噪。该阈值函数既能克服硬阈值函数在阈值附近不连续的缺点,又可弥补软阈值函数与原函数之间存在固定差值的不足。以MIT-BIH心电数据库中的101号文件作为原始数据,将整个数据文件进行平均分段,每段有1 200个数据点,对每段数据进行加噪仿真分析,结果表明所提出的去噪算法得到的去噪信号信噪比比硬阈值函数和软阈值函数分别提高2.31%和8.04%,从而证明所提算法的有效性。
[Abstract]:The baseline drift, power frequency interference and myoelectric interference in ECG signal are removed, and the waveform is recognized. The field of medical diagnosis and treatment is of great significance. An 8-layer wavelet decomposition of ECG signals using sym5 wavelet function is proposed. According to the rule that the intensity of useful signals is equally distributed in each layer and the noise intensity decreases with the increase of decomposition layers, The wavelet detail coefficients of each layer are set at different thresholds. Finally, the new threshold function is used to Denoise the wavelet threshold. The threshold function can not only overcome the disadvantage of discontinuity of the hard threshold function near the threshold. It can also make up for the deficiency of the fixed difference between the soft threshold function and the original function. Taking the 101 file in the MIT-BIH ECG database as the original data, the entire data file is divided into an average of 1,200 data points per segment. The simulation results show that the signal-to-noise ratio of the proposed denoising algorithm is 2.31% and 8.04 higher than that of the hard threshold function and the soft threshold function respectively, which proves the effectiveness of the proposed algorithm.
【作者单位】：莆田学院机电工程学院;清华大学医学院;
【基金】：福建省自然科学基金(2013J01251) 福建省高校现代精密测量与激光无损检测实验室项目(S20160404) 莆田学院支持项目(2015032)
【分类号】：R540.4;TN911.4

【参考文献】