Wavelet Denoising Alogorithm And Implementation Technology Of Wavelet IP Core

Wavelet analysis is an advanced technology in signal acquisition and processingarea, whose superiority over traditional Fourier transform in non-stationary signalprocessing is particularly noteworthy. As an important application of wavelet, waveletdenoising is becoming a forefront topic of signal processing area and of great value bothin theoretical research and engineering application.This paper gives a brief introduction to the basic theory of wavelet, and thenmainly focus on the two keystones that affect the performance of wavelet denoising: theestimation of decompose level and mother wavelet select. Generally speaking, thethreshold value is determined by experiences or single elected function, which lacks ofcomprehensive consideration and can’t get a satisfied result. Therefore, this paper focuson the threshold estimation of denoising algorithm and estimates the optimal thresholdthrough the processing of searching the best threshold and the comparation betweendifferent threshold algorithms. Following that, this paper gives a detailed analysis andconclusion. In practical engineering applications, wavelet denoising need to considercomprehensively according to the signal’s features and the denoising requirements.Because of wavelet transform’s complexity, software can’t be satisfactory whenfacing the requirement of real time signal processing. To solve this problem, this paperintroduces the design method of soft IP core and implements the Mallat algorithm andreconstruction algorithm on FPGA. Through its application in practical signal denoising,it is verified to be effective.