论文部分内容阅读
根据毛细管电泳信号中噪声与真实信号的模极大值特性之间存在较大差异的特点，提出了一种利用二进样条小波变换模极大值算法去噪的方法．研究了信号的半峰宽、不同信噪比以及阈值的选取对去噪结果的影响，并对模拟含噪信号和含噪毛细管电泳信号去噪前后的结果进行了比较．实验结果表明，该法能有效地消除谱带较宽的毛细管电泳等信号中存在的噪声．
According to the large difference between the mode-maximum property of the noise and the real signal in the capillary electrophoresis signal, a method of using the binary spline wavelet transform modulus maxima algorithm to denoise is proposed. The influence of the half width of the signal, the different signal to noise ratio and the threshold selection on the denoising results was studied. The results before and after denoising of the simulated noisy signal and the noisy capillary electrophoresis signal were compared. Experimental results show that the proposed method can effectively eliminate the presence of noise in the wideband capillary electrophoresis signal.