The study of the improved wavelet thresholding with translation invariant de-noising on capillary electrophoresis signal

摘要

According to the characters of CE signal, improved wavelet thresholding with translation invariant was described to denoise for microchip capillary electrophoresis in this paper. The improved thresholding function has muti-derivative compared with the soft thresholding. The denoising method can not only remain the geometrical characteristics and keep the amplitudes of the original electrophoresis signal waveforms efficiently, but also suppress impulsive noise to some extent. In order to suppress Pseudo-Gibbs phenomena combining the translation-invariant de-noising was proposed. The experimental results indicated that the proposed methods in the paper have good signal-to-noise ratio through testing the actual electrophoresis signal. The baseline of the denoised signal is smooth and the peak shape has not distortion.