This study proposes a wearable system that can measure electrocardiogram (ECG) signals reliably in an environment with high motion artifacts. This system employs a motion artifact extraction method based on a triple-axis accelerometer attached to each electrode independently to remove motion artifacts from ECG signals with high performance. Recursive Least Square (RLS) and Least Mean Square (LMS) algorithms remove extracted noise from the source signals, thereby obtaining a mean square error (MSE) of 0.0166 when using RLS and 0.0160 when using LMS. This means that the performance improved respectively by approximately 5.1% and 8.6% compared to that of the recently developed ECG monitoring system.
