This repository contains the key function of the AFD based ECG denoising. The AFD based ECG denoising has been introduced in "Adaptive Fourier decomposition based ECG denoising". These codes are implemented by MATLAB. When using these codes, please cite our paper. The license follows CC BY-NC-ND 4.0.

The AFD decomposes a signal according to its energy distribution, thereby making this algorithm suitable for separating pure ECG signal and noise with overlapping frequency ranges but different energy distributions. A stop criterion for the iterative decomposition process in the AFD is calculated on the basis of the estimated signal-to-noise ratio (SNR) of the noisy signal.

The detailed program of the AFD can be found in Toolbox-for-Adaptive-Fourier-Decomposition with an online demo.

If you use this repository, please cite the related paper:

@article{wang_adaptive_2016,
	title = {Adaptive {Fourier} decomposition based {ECG} denoising},
	volume = {77},
	doi = {10.1016/j.compbiomed.2016.08.013},
	journal = {Computers in Biology and Medicine},
	author = {Wang, Ze and Wan, Feng and Wong, Chi Man and Zhang, Liming},
	year = {2016},
	pages = {195--205},
}