Comparisons of wavelet packet, lifting wavelet and stationary wavelet transform for de-noising ECG

In ECG de-noising, the goal is to recover the valid ECG from the undesired artifacts with the minimum distortion for presenting a signal, which allows easy visual interpretation and auto diagnosis. We adopt three de-noising algorithms based on wavelet packet (WPT), lifting wavelet (LWT) and stationary wavelet transform (SWT) for de-noising ECG signals, respectively. Through using simulated and real noisy ECG signals from MIT-BIH database, we evaluate the denoising results by means of SNR and visual inspection, and conclude that SWT algorithm is the suitable one for de-noising ECG signals.