介绍了小波分析方法在逐束团束流位置测量系统中的应用。小波分析方法在高频处频率窗口较宽,具有较高的时间分辨力,使用小波分析可分离并提取信号的振荡成分及基线漂移成分，各成分在时间轴上的位置与原信号相同，原有的线性关系保持不变，在处理非平稳信号时不会造成信号明显劣化,如幅度失真和相位偏差。基于小波分解和重构的时间序列多分辨力滤波处理非平稳信号时不会造成信号明显劣化，保证了追踪束团振荡强度、相位、频率和振荡模式随时间变化的结果更真实可信。在合肥光源中，小波分析方法成功用于横向振荡振幅包络的提取及增长率、阻尼率的计算，也可用于提取横向振荡振幅包络及计算增长率和阻尼率，为机器研究、束流诊断和逐束团反馈系统调试提供了准确的依据。

During beam injection in Hefei Light Source(HLS), the amplitude of betatron oscillation varies rapidly, and the frequency varies greatly. It is a typical non-stationary process. Using the traditional method to process the time series signal acquired by the bunch-by-bunch measurement system may result in amplitude distortion and phase deviation. Wavelet analysis provides multi-resolution representations of the bunch signal. Subsequently, the oscillation component and baseline drift component can be separated. The locations on timeline and the linear relationships of the components remain the same as the original signal. Compared with the traditional method, the data processing method based on wavelet analysis is capable of analyzing a non-stationary signal and de-noising a signal without appreciable degradation. It will not bring about conspicuous amplitude distortion and phase deviation, and ensures that the subsequent tracking of the evolution of oscillation amplitude, phase, frequency and mode in the time domain is more reliable. Furthermore, the method has been successfully used to extract the amplitude envelope of betatron oscillation and to calculate growth rate and damping rate. It provides an accurate foundation for machine studies, beam diagnosis and the commissioning of bunch-by-bunch feedback system.