Noise suppression using lifting-based wavelet transform and level-dependent threshold estimator

We propose an effective technique for the denoising of electrocardiogram (ECG) signals corrupted by nonstationary noise. The technique is based on a lifting scheme providing a frame for the construction of a second generation wavelet transform and level-dependent threshold estimator. Wavelet coefficients of ECG signals are obtained by a lifting-based DB6 filter. To obtain ECG corrupted by nonstationary noise, white noise is considered. Performance of the proposed method is evaluated by means of signal-to-noise ratio (SNR) and visual inspection. The results obtained by our proposed algorithm are compared to results obtained by median filtering. The comparison shows the superior denoising performance of our proposed method over that of median filtering.