|
| 优化VMD联合改进小波阈值的心电信号去噪方法 |
| Optimized VMD Combined with Improved Wavelet Threshold for ECG Signal Denoising |
| 投稿时间：2025-08-07 修订日期：2025-12-01 |
| DOI： |
| 中文关键词: 心电信号去噪 变分模态分解 麻雀搜索算法 小波阈值 |
| 英文关键词: electrocardiogram signal denoising variational modal decomposition sparrow search algorithm wavelet threshold |
| 基金项目: |
|
| 摘要点击次数: 53 |
| 全文下载次数: 0 |
| 中文摘要: |
| 心电图(ECG)信号采集时易受到噪声干扰,对后续研究产生影响。变分模态分解(VMD)算法广泛应用于ECG信号去噪,但算法依赖于参数设置。针对此问题,文中提出一种改进麻雀搜索算法(ISSA)优化VMD算法参数,并联合改进小波阈值的ECG信号去噪方法。在麻雀搜索算法(SSA)中添加四种策略进行改进,并利用该算法优化VMD算法的参数。优化后对ECG信号进行VMD分解,得到K个模态分量。对噪声主导的分量利用改进的小波阈值算法进行去噪,然后与信号主导的分量叠加,构成去噪后的ECG信号。为验证算法的去噪性能,文中选择含模拟噪声的ECG信号和实测的ECG信号进行去噪实验。实验结果表明,文中所提算法的去噪能力强于EMD、EEMD、CEEMDAN和VMD结合经验小波变换算法,去噪性能指标也优于这些算法。 |
| 英文摘要: |
| Electrocardiogram (ECG) signal acquisition is susceptible to noise interference, which has an impact on subsequent studies. The variational modal decomposition (VMD) algorithm is widely used for ECG signal denoising, but the algorithm depends on parameter settings. To address this problem, an ECG signal denoising method with improved sparrow search algorithm (ISSA) to optimise the parameters of the VMD algorithm and joint improved wavelet thresholding is proposed in the paper. Four strategies are added to the Sparrow Search Algorithm (SSA) for improvement and the algorithm is used to optimise the parameters of the VMD algorithm. The optimised VMD decomposition of the ECG signal is performed to obtain K modal components. The noise-dominated components are denoised using the improved wavelet thresholding algorithm and then superimposed with the signal-dominated components to form the denoised ECG signal. In order to verify the denoising performance of the algorithm, the ECG signal containing simulated noise and the measured ECG signal are selected for denoising experiments in the paper. The experimental results show that the denoising ability of the algorithm proposed in the paper is stronger than that of EMD, EEMD, CEEMDAN and VMD combined with empirical wavelet transform algorithms, and the denoising performance index is also better than these algorithms. |
|
View Fulltext
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
