Enhancement of signal is highly essential to either remove or reduce the noise from it. More often we meet these difficulties due to nonlinearities in signal and due to its nonstationary nature. Fourier transform is a popular and efficient tool for this purpose along with wavelet transform. It may not work properly to its limitations, for the nonstationary type of signal it fails to prove its efficiency. Also it cannot offer the temporal and spatial information clearly. The empirical mode decomposition EMD of signal is an adaptive method and a powerful substitution to the Fourier and wavelet transform. This technique can be used for an effective way of analysis of the instantaneous frequency of signals. Though EMD has used by many researchers, it cannot be most popular due to its demerits in terms accurate mathematical model. Therefore the birth of variational mode decomposition VMD occurred as an alternative of EMD and can overcome the demerits of EMD. VMD decomposes the signal into discrete number of sub-signals (modes), where each mode has limited bandwidth in spectral domain. But the problem of this technique is how we can define the optimal number of modes where too large number of modes will lead to redundant VMD information, while too small number of modes will lead to mode mixing in the VMD results. In this paper we propose a combination of VMD and a correlation coefficients (CCs) to optimize the number of modes based on estimation of useful modes to reconstruct the original signal and determination of noisy modes to be removed. The robustness of VMD is evaluated on simulated signals under different parameters and the performance of the method for signal denoising is evaluated in terms of signal- to- noise ratio (SNR).
Key words: Variational mode decomposition, Empirical mode decomposition, correlation coefficients