Removing the power line interference from ECG signal using Kalman least mean square filter

Abstract

Electrocardiogram (ECG) is the graph obtained from the electrical activity of the heart and is used to examine function of heart and for diagnosis cardiac problems. The voltage level of ECG signal ranges from as low as 0.5 to 5mV and frequency components fall into the range of 0.05 to 100Hz. Various noises like Power Line Interference (PLI), baseline wander and noise due to muscles contraction and relaxation get added into the recorded ECG signal. Out of them, PLI is considered to be the major source of noise. PLI may be stationary or non stationary. Non adaptive filters are employed to remove stationary power line interference and adaptive cancellers are used to handle non-stationary power line interference. In this paper adaptive LMS algorithm along with Kalman filtering became KLMS has been investigated to eliminate PLI from contaminated ECG signal. The results of KLMS filter has been compared with the LMS filter before and after filtering the ECG signal. Results shows MSE equal to 0.6314 and 0.0409 before filtering for LMS and KLMS filter respectively. MSE equal to 0.1983 and 0.0245 after filtering for LMS and KLMS filter respectively. Also, PSNR before filtering is 50.1279 and 62.0168 for LMS and KLMS filter respectively. PSNR after filtering is 55.1586 and 64.2352 for LMS and KLMS respectively. The filters algorithm has designed using MATLAB software and tested on ECG signal contaminated with power line frequencies.