多层降噪自动编码器的心电信号去噪算法

ECG Signal Denoising Algorithm for Multilayer Noise Reduction Automatic Encoder

传统的心电信号(ECG)去噪算法在去除线性的、平稳的ECG信号噪声时效果显著,但是在面对非线性、非平稳的ECG信号噪声时去噪效果不理想。为了提高心电信号算法的去噪能力运用了一种基于降噪自动编码器的ECG去噪算法。降噪自动编码器(DAE)具有噪声鲁棒性的特点,可以在信号受到污染的情况下尽可能地恢复数据的原始状态。为了进一步提升降噪自动编码器算法的去噪效果用多个降噪自动编码器堆叠形成深度神经网络对心电信号进行降噪处理。通过实验结果表明:多层降噪自动编码器(SDAE)算法相较于DAE算法和传统的心电信号去噪算法,SDAE算法对非线性、非平稳的信号噪声具有更好的降噪效果,而且保留了原始心电信号绝大部分的细节信息,对噪声具有较强的抗干扰能力,满足了心电信号的去噪要求。

The traditional ECG denoising algorithm has a significant effect in removing linear and stable ECG signal noise,but it is not ideal in the face of non-linear and non-stationary ECG signal noise.In order to improve the de-noising ability of ECG algorithm,an ECG de-noising algorithm based on automatic de-noising encoder is used.The noise reduction automatic encoder(DAE)has the characteristics of noise robustness,which can recover the original state of data as much as possible when the signal is polluted.In order to further improve the denoising effect of the algorithm,a deep neural network is formed by stacking multiple automatic coders to denoise the ECG signal.The experimental results show that compared with DAE algorithm and traditional ECG signal denoising algorithm,sdae algorithm has a better denoising effect on non-linear and non-stationary signal noise,and retains most of the details of the original ECG signal,it has a strong anti-interference ability to noise,and meets the requirements of ECG signal denoising.