Adequate determination of a band of wavelet threshold for noise cancellation using particle swarm optimization

Noise reduction problem is addressed by this study. Recently, wavelet thresholding has become popular and has gotten much attention among a number of de-noisy approaches. The most of threshold determination are developed from universal method proposed by Donoho. But, some shortcomings of the determination are caused from several incorrectly estimated factors and the lack of adaptability for whole frequency. By the reason, this paper replaces a universal threshold by multi-thresholds for matching the coefficients of each wavelet segment, and then the band of threshold will be fined by particle swarm optimization (PSO). Because original signals and noise are mutually independent, an objective function of PSO is created to evaluate the second order correlation and high order correlation. In order to confirm the validity and efficiency of the proposed algorithm, several simulations which include four benchmarks with high or low noise degree are designed. Moreover, the performance of proposed algorithm will have compared with that of other existing algorithms.