A New Method of Signal De-Noising by Multi-Resolution Analysis Based on Discrete Element
Fuyong Lin · Journal of Huaqiao University · 2013
By constructing a discrete base,multi-resolution analysis methods are applied to denoise the noise signal.The proposed method of discrete base can be explicitly represented,and has symmetric characteristics.The calculation is greatly reduced through the cycle matrix inverse matrix method.Comparing different denoising methods,the signal denoising effect is assessed from two aspects of signal-to-noise ratio(SNR) and mean square error(MSE).Experimental results indicate that this method shows good characteristics in signal denoising aspect relative to wavelet analysis method.Denoising effect is obvious,and can achieve good signal-to-noise ratio and mean square error when discrete base coefficient is near 0.75.