基于二代小波的光纤陀螺实时降噪方法研究

在以光纤陀螺为主要惯性敏感元件的捷联惯导系统中,陀螺输出信号中的非确定性随机漂移的实时滤除,对提高实际系统的初始对准精度及导航精度均具有重要的意义。考虑到传统小波阈值法的去噪性能及实时性问题,提出了一种基于第二代小波变换,并结合硬阈值、强制降噪和带滑动数据窗的光纤陀螺信号实时降噪改进方案。利用MATLAB进行了正弦信号和光纤陀螺输出信号的模拟实时降噪实验,并与一代小波实时降噪方案进行了比较,验证了改进方案在理论计算速度大幅提升的前提下,降噪性能得到提高,进而减小了系统输出的姿态误差。

Research on real-time de-noising of FOG based on second generation wavelet transform

In the application of SINS using fiber optic gyroscope(FOG) as the main inertial measurement sensor,the real-time de-noising of the FOG non-deterministic random drift significantly improves the SINS initial alignment and navigation accuracy.Considering the restriction of traditional wavelet de-noising performance and the real-time problem,an improved method for the real-time de-noising of the FOG output signal using the second generation wavelet transform combined with hard threshold,mandatory noise reduction and sliding data window is proposed.Sinusoidal signal and FOG output signal experiments were carried out using MATLAB.Compared with traditional wavelet real-time de-noising performance,the experiment results show that the proposed method can both enhance the de-noising performance and reduce the SINS output attitude error under the premise of great theoretical computing speed increasing.