Empirical Mode Decomposition-based Subtraction Techniques for 50 Hz Interference Reduction from Electrocardiogram

摘要

In general, most of the Biomedical signals such as Electrocardiogram (ECG), Electroencephalogram, and Electro-oculogram are nonstationary signals, suffers from different interferences like power line interference (PU) and with other biomedical signals that gets added with it. Analysis of these signals means the extraction of useful information from the signal, and in this paper it is carried out using a new nonlinear and nonstationary data analysis method called Empirical Mode Decomposition (EMD). The key feature of this method is that it can decompose the signal into different IMFs and makes the analysis simple. Compared with other tools like Fourier analysis and wavelet methods, EMD is purely a data-driven and adaptive technique. Thus, it is well suited to analyze nonstationary signals like biosignals. This paper foregrounds an EMD-based, two-weight adaptive filter structure to reduce the PU in ECG signals. Two methodologies are studied based on EMD and the simulations are carried out in a MATLAB environment. The denoised signals are visually impressive and the methodologies are well suited for real-time implementation.