A Comparison of the Denoising Performance Using Capon Time-Frequency and Empirical Wavelet Transform Applied on Biomedical Signal

Abstract

The empirical wavelet transforms and Capon time-frequency applications are used in this work. EMG and EEG are non-invasive ways to measure muscle activity and the electrical activity of the brain. The visual analyses of EMG and EEG are significantly impacted by these noises. The EWT's effectiveness in eliminating the sounds has been assessed by using several common metrics between the clean original signal and the filter's signal output. According to the results, the EWT has performed better than other denoising algorithms in terms of locating the different parts of the aberrant biomedical signal data. The EMG signal is non-stationary. Hence, employing time-frequency methods is unavoidable. Capon is the preferred parametric time-frequency technique. The Capon technique outperforms other non-parametric time-frequency algorithms mentioned in the scientific literature in terms of resolution and removing any interference terms. This study explains why the use of EWT and Capon approaches together is beneficial for biomedicine.