Denoising via Empirical Mode Decomposition

In this paper a signal denoising scheme based a multiresolution approach referred to as Empirical mode decomposition (EMD) [1] is presented. The denoising method is a fully data driven approach. Noisy signal is decomposed adaptively into intrinsic oscillatory components called Intrinsic mode functions (IMFs) using a decomposition algorithm algorithm called sifting process. The basic principle of the method is to reconstruct the signal with IMFs previously filtered or thresholded. The denoising method is applied to one real signal et to four simulated signals with different noise levels and the results compared to Wavelets, Averaging and Median methods. The effect of level noise value on the performances of the proposed denoising is analyzed. The study is limited to signals corrupted by additive white Gaussian random noise.