ECG Signal Denoising Using a Novel Solution to the Heat Equation Through Wavelet Transform

Abstract

This paper proposes a wavelet transformation solution to the Heat equation. This method, associated with wavelet obtained by taking derivative of the Gaussian kernel, has been applied to denoise ECG signals. The proposed method is tested by experimenting on different signals of the MIT-BIH Arrhythmia database corrupted by power line interference, electromyographic (EMG) noises and synthetic interference. It is observed that the proposed method gives appreciable results in the cases of EMG noise, power line interference and synthetic interference. The proposed method is compared with some of the recenetly reported methods. We found that the proposed method outperfoms the recently reported methods.