Research on denoising algorithm for ECG signals

In the course of collecting ECG signal data, electromyography interference, baseline drift and 50 Hz power line interference will be introduced inevitably, which often makes it difficult to identify the characteristics of ECG signals to some degree by using conventional identification methods. Median filter, a nonlinear signal filter with simple operation and high speed, is used to remove the low-frequencies noises in Electrocardiogram signals, such as baseline drift. Because the dyadic wavelet of WTS is a set of band-pass filters having different frequency bands in every scale, the wavelet transformation was selected to decompose the original signals. An interference-eliminated ECG signal was formed by reconstruction from the changed coefficients of wavelet. A simulation experiment is adopted to make it sure how to determine the self-adaptive threshold selections, the proper decomposition order and wavelet functions. The method was tested by using both ECG signals from MIT/BIH database and ECG signals generated via computer simulation. The results show that the algorithm can suppress the main noises existing in ECG signals efficiently and can satisfy the requirements of clinical analysis and diagnosis on ECG waveforms.