Performance Comparison of Modified LMS and RLS Algorithms in De-noising of ECG Signals

摘要

The electrocardiogram (ECG) is generally used for the diagnosis of cardiovascular diseases. In many of the biomedical applications it is necessary to remove the noise from ECG recordings. Several adaptive filter structures are proposed for noise cancellation. The main objective of our work is to develop an adaptive algorithm to remove the contaminating signal and to obtain original ECG data. A new, simple and efficient Least Mean Squares (LMS) based adaptive algorithm developed for optimal removing of interference in ECG signals is introduced. It uses a modified LMS (mLMS) algorithm to adjust filter weights according to non-stationary properties of processed signal. Furthermore, simulation studies shows that the modified LMS algorithm gives better performance compared to an existing RLS algorithm in de-noising of ECG signals.