Denoising Research of Doppler Blood Flow Signals with Wavelet Transform
Yikai Shi, Liang Deng, Qin Yao, Bo Tang · 2010
The signals collected by Doppler blood flow instrument have the noise inevitably due to measurement system, signals interference and turbulent flow of blood flow. The chaotic signals are non-stationary signals, and contain some of the peaks and mutations. The denoising method using the traditional Fourier transform analysis is powerless. The wavelet analysis can simultaneously carry on the analysis in the time frequency range to the signals, so that it can more effectively distinguish the mutant part of the signals and noise to achieve signals denoising. This article used threshold denoising principle based on wavelet transform. Denoising signals are reconstructed in order to achieve the denoising purpose. The denoising system applied the db9 wavelet of Matlab7.0 Toolbox and soft denoising method for collection of the primary waveform to wavelet analysis. Results illustrate that this method can extract characteristics of signals effectively and have good value in practice.