Noise Removal from ECG Signal Using Various Filters

Abstract

Electrocardiogram signal is noninvasive technique that has great importance in the detection of cardiac abnormalities. The analysis of ECG signal is preferred for conveying information as it preserves the electrical performance of heart. ECG signals are very sensitive and characteristics of ECG signals get contaminated due to different types of noise. In many of the biomedical applications, for real time heart monitoring system, it is necessary to remove the noise from ECG recordings to achieve faithful signals for further processing. In this thesis, various filters are used results to reduce and remove the effect of noise to get refined signal. The power spectral density and average power are performance metrics used before and after filtration. Adaptive filter is used for noise cancellation of ECG signal. Adaptive filter function is based on error minimization between input signal which is noisy ECG signal and its reference input. There are many adaptive algorithms such as Least Mean Square (LMS), Recursive Least Square (RLS), and Normalized Least Mean Square (NLMS) etc.