Waveform Estimation of Fractal Signals Using Optimum Soft Threshold Technique

A new method based on the optimum soft threshold technique is proposed in this paper for wavelet estimation of fractal signals which are polluted by external white noise. The optimum soft threshold technique is employed in each scale of wavelet decomposition to estimate the multi-scale input fractal signals. The estimates of the input fractal signal in each scale are then synthesized to restore the original signal. Since it doesn¿t need to know the parameters of fractal signal and the statistical characteristic of external white noise in advance, this method is practical and maneuverable in many situations, besides that, the simulation results also show that this method is robust to the error of SNR estimated.