ECG signals denoising method based on improved wavelet threshold algorithm

Abstract

Based on the understanding and analysis of the traditional soft and hard threshold denoising method and the existing improved threshold denoising method, this paper proposes a new threshold denoising method, and applies it in the problem of ECG signal denoising. The threshold function combines the soft and hard threshold, and the combining ratio and regulatory factors are computed automatically according to the wavelet coefficients, without the need of manual intervention. The thresholds are calculated separately based on the different scales after wavelet decomposition and adjusted using impact strength factors of the scales. The experimental results indicate that, the proposed algorithm can effectively filter out the noise of ECG signals, better retain the characteristic information of the ECG signal, has a higher signal-to-noise ratio, and achieve a better denoising effect.