Compatibility of mother wavelet functions with the electrocardiographic signal

Electrocardiographic ECG gives the electrical activity of the heart captured over time by an external electrode attached to the skin. The proper processing of ECG signal and its accurate detection is very much essential since it determines the condition of the heart. ECG consists of various noise such as baseline wander, power line interference and muscle noise etc. The wavelet transform is a useful tool to remove the noises. Mother wavelet has become an effective tool to describe the nonstationary signal such as the repeated patterns of the recorded ECG signals. Selection of appropriate mother wavelet for the ECG signal is a challenging task. In this paper, one-hundred and ten potential mother wavelet functions (Daubechies, Coiflets, Biorthogonal, Reverse Biorthogonal, Discrete Meyer and Symlets) are investigated to choose a compatible mother wavelet function with the ECG signal by considering the minimum mean square error (MSE), larger signal-to-noise (SNR) ratio and minimum percentage RMS difference (PRD). The values showed that the performance of Biorthogonal (bior3.9) for denoising is best out of the other mother wavelet functions.