小波分析在激光多普勒信号处理中的应用

Application of wavelet analysis to laser Doppler signal processing

提出将激光多普勒测速仪(LDV)应用于车载惯性导航系统中,阐述了激光多普勒自身速度仪的基本原理和小波变换的相关理论,并运用小波变换对多普勒信号进行检测、去噪及提取多普勒频率,仿真及实验结果表明:信号进行小波分解后,对每一级小波进行阈值处理,得到了较好的去噪效果;在小波降噪常见的阈值原则中,无偏似然估计阈值和极值阈值不容易丢失信号中的有用成分,而启发式阈值原则和固定阈值原则可以更有效地去除噪声;对于提取多普勒频率而言,小波变换与快速傅里叶变换所得的结果是一致的,而小波变换不但可以求出系统自身的运动速度,而且还可以求出对应速度发生的时刻。

An idea of using laser Doppler velocimeter(LDV) to offer the velocity in vehicle inertial navigation system was proposed. The principles of measuring its own velocity of laser Doppler and relative theory of wavelet transform were expounded. Doppler signal was processed using wavelet transform, simulation and experiment showed that: after signal decomposition, processing the wavelet at each level, acquired good de-noising effect; in common threshold principles of de-noising, rigorous SURE and minimax principles did not lose the useful information easily, but heuristic SURE and fixed threshold principles acquired better de-noising effect; results of frequency extraction with wavelet transform and fast Fourier translation(FFT) were consistent, but velocity and correspondent moment could be acquired simultaneously with wavelet transform.