Comparison between different wavelet transforms and thresholding techniques for ECG denoising

ECG Signal holds critical information regarding the physiology and pathological functions of heart. Quite often, they get corrupted by different type of noises, so to get proper information, it is essential to perform the denoising. This paper discusses the denoising of three different type of noises which generally corrupts the ECG signal viz. power line interference, baseline drift noise and EMG (wideband stochastic) noise. The denoising is performed over these noisy signals using five different thresholding techniques clubbed with ten different wavelet transform functions from the wavelet family. Finally, for each type of noise the optimized combination of thresholding technique and wavelet function is suggested.