Implementation of EMD algorithm for ECG noise reduction

Abstract

In this paper, Empirical mode decomposition (EMD) algorithm is implemented on a digital signal processor (DSP) to eliminate artifacts from the electrocardiogram (ECG) signal. By extracting specified IMF components, power-line interference and baseline wander can be reduced. EMD algorithm is a novel technique which can decompose a signal into a series of intrinsic mode functions (IMFs), and use to eliminate the noise from biomedical signals. However, most of the publications are aimed at the off-line EMD-based signal processing through a personal computer. Therefore, it is valuable to implement the EMD algorithm on the DSP for the biomedical signal processing. In order to verify the features of the proposed system, emulations are firstly performed, and then the experimental results are presented using the fabricated platform.