短数据量动态光散射颗粒测量信号去噪方法

在动态光散射测量中,采用自相关法对测量信号进行去噪,其去噪效果受数据量影响。根据噪声和信号的不同特点,采用小波包变换对信号进行去噪,能够提高信噪比,改善粒径反演结果。采用两种去噪方法,对粒径为100 nm颗粒的散射信号进行去噪并反演,小波包去噪法能够改善粒径误差0.88%~6.41%。在不同数据量下,由两种去噪法的反演结果对比看出,在短数据量时,小波包去噪效果更好,当数据量大于1×106时,两种去噪法效果相差不大。因此,小波包去噪法更适合于短数据量的动态光散射颗粒测量。

Denoising method adapted for dynamic light scattering particles measure signal of short data

In dynamic light scattering(DLS) measurement, auto correlation method is used for signal denoising, whose denoising effect is affected by data amount. Based on different characteristics of noises and signals, wavelet packet transform is used in the signal , which can improve signal to noise ratio and particle inversion results. The DLS signal of 100 nm particles is denoised by the two methods and the auto correlation functions of the denoised signals are respectively inverted. The inversion error can be improved by 0.88%~6.41% after denoising by wavelet packet. Denoising results of different data amounts by the two methods show that, wavelet packet transform is more effective for short data, and when the data amount is greater than 1×10⁶, the effects of the two denoising method