连续样条小波变换解析高噪声重叠分析信号

采用连续样条小波变换方法对化学中高噪声重叠信号中峰位置和峰数目的提取进行了分析，讨论了信噪比、重叠程度、信号相邻峰高比对重叠信号解析的影响．结果表明：对于信噪比低至1的高噪声信号，采用连续样条小波变换处理能够得到重叠信号的峰位置和峰数目；连续样条小波变换对于高噪声重叠信号具有较好的解析能力，为低含量多组分物质的复合信号的定性定量分析提供了基础；该方法与离散小波变换法相比，数据点无须为2^n(n为整数)，并且不需要对信号进行多次分解，因此简单快速。

Resolution of High-Noisy Overlapped Analytical Signals by Means of Continuous Spline Wavelet Transform

The continuous spline wavelet transform is employed to the resolution of high-noisy overlapped signals in analytical chemistry to obtain the location and the number of the overlapped peaks. The effects of the signal-to-noisy ratio (SNR), the overlapping degree and the peak height ratio between two adjacent peaks on the resolution are then discussed. The results show that, by using the continuous spline wavelet transform, the location and number of the overlapped peaks can be obtained when the SNR of the signal is down to 1. Moreover, the high-noisy overlapped signals can be resolved well, thus providing references to the qualitative and quantitative analyses of the complex signals of low-content multi-component substances. It is also indicated that, comparing with the discrete wavelet transform, the continuous spline wavelet transform is of the advantages of simplicity and quickness because a data number 2 n (n is an integer) and a multi-resolution are not needed.