Application Of Optimized Variational Mode Decomposition Algorithm In Signal Denoising

Signal denoising is to remove the noise contained in the signal by technical means so as to get a cleaner signal,which has a very wide range of applications in the intelligent terminal,medical field,military field and aerospace field.However,in our daily life,due to the limitation of equipment,environmental influence and uncontrollable transmission process in various fields,a cleaner signal cannot be obtained.Therefore,how to remove the noise contained in the signal has become an urgent problem to be solved.Because signals are divided into speech and image signals,for these two signals,In this paper,the optimized Variational Mode decompo-sition(VMD)algorithm is combined with wavelet threshold,and the optimized 2D-VMD algorithm is combined with wavelet threshold to solve the problem of noise in the signal.The specific research contents are as follows:(1)For speech signals,because the speech signals are unstable and discontinuous,it is difficult to remove the noise in the speech signals with ordinary methods.Therefore,this paper proposes an optimized denoising method combining variational mode decomposition and wavelet threshold.The concrete idea of this method is that the grey Wolf algorithm is used to optimize the variational mode decomposition algorithm,and then the input speech signal is processed by the optimized variational mode decomposition algorithm,and the processed speech signal is decomposed into finite intrinsic mode components.The correlative coefficient is used to select the noise-containing modes,and the new threshold wavelet transform is used to process these modes.Finally,the processed mode and the effective mode are reconstructed so that the denoised speech signal is obtained.In this paper,10 d B Gaussian white noise is added into the simulation signal and the real speech signal respectively,and then they are used in the experiment.The experimental results show that the proposed method can solve the above problems better than other algorithms,and the resulting graph is closer to the original graph,with the highest SNR and the lowest MSE,which can better remove the noise.(2)For image signal,it is very complicated and difficult to decompose the image signal because the image is a two-dimensional signal,so this paper proposes an optimal two-dimensional variational mode decomposition and wavelet threshold to realize image signal denoising method.The concrete idea of this method is to first optimize 2D-VMD by using the optimization algorithm of Grey Wolf algorithm,and then decompose the input 2D image signal by using the optimized variational mode decomposition algorithm.The decomposed image signal is composed of finite inherent mode components.Among the inherent modal components obtained,there are modes affected by noise.By visual observation,the modal component containing noise is selected.Finally,the modal component containing noise is processed by using the new threshold wavelet transform,and then the denoised mode and effective mode are reconstructed to get the denoised image signal.In this paper,two images with five different signal-to-noise ratios are used for simulation experiments.The experimental results show that compared with other algorithms,the obtained images are closer to the original image,and the PSNR is the highest.Therefore,this method can better remove the noise in the image signal.