Muscle Noise Filtering in ECG Signal

Abstract

This work deals with muscle noise filtering in ECG signals using wiener filtration and optimization of Wiener filter numerical parameters. The optimization was performed by using the exhaustive search method, which belongs to the brute force methods and seeks the global solution. The Criterial function in the optimization process was the average SNR of filtered signals. The testing was performed on 20 random signals from the CSE database. The testing has shown that the greatest difference between the input SNR and the average output SNR was at the minimum input SNR.