Research on motion artifacts eliminating for wearable Electrocardiogram signal monitoring

Electrocardiogram(ECG) is the most effective and low cost tool in the diagnosis of heart diseases. The two main research topics about the ECG signal analysis are the noise reduction and the Q wave, R wave, S wave detection. In this paper, the Savitzky-Golay filter is used to removing baseline drift and motion artifacts in motion ECG signals. Here ECG signals are taken from the human body in the bend, walk and sit-stand condition. Then Q wave, R wave and S wave are detected in the filtered signals. We evaluate the advantages of Savitzky-Golay filter in the ECG motion artifact removal by comparing with Kalman filtering and wavelet decomposition. The comparison is done by evaluating different statistical parameters like time complexity and power spectrum density. Experimental results verify that Savitzky-Golay filter can be more effectively adapt to the changes of the ECG signal, more effectively remove the noise without much destroying the shape and the peak of the ECG waveform.