提出一种连续小波变换(CWT)用于频率急剧变化的瞬态位移干涉仪信号处理的方法。连续小波变换利用对不同频率具有不同分辨本领的特性,对信号进行时频分析,计算出信号频率,进而恢复出信号的速度。当在可变时间窗口宽度内小波的中心频率和信号局部频率相等或接近时,小波变换系数幅值最大,据此能判断出信号的频率。用连续小波变换方法对计算机模拟的位移干涉信号进行处理,恢复的速度相对误差在3%,从爆轰实验的光纤位移干涉仪信号中也能够准确地恢复出物体的速度历史。

A transient displacement interferometer signal processing method based on the continuous wavelet transform (CWT) is proposed. The interference signal is analyzed by CWT, and the frequency and the velocity of the signal can be deduced. One main characteristic of CWT is its adaptive nature for signals with large variations in frequency. In the flexible time window, if the varying central frequency is equal or close to the instantaneous frequency of local signals, the modulus of CWT is maximal, so the local frequency is determined. The processing results of simulated interference signals show that the relative error is less than 3%. Similarly, velocity history can be precisely recovered from signals of fiber displacement interferometer in detonation experiment.