基于自适应时频峰值滤波的光纤陀螺去噪算法
发布时间：2019-06-05 09:21
【摘要】：为减小光纤陀螺输出信号噪声、提高惯导系统精度,提出了光纤陀螺信号自适应时频峰值滤波算法。对光纤陀螺信号进行初始变换并调制,采用伪Wigner-Ville分布对调制信号进行时频分析,给出了一种自适应的伪Wigner-Ville分布最优窗长获取准则,通过局部峰值搜索实现编码信号的瞬时频率估计进而还原出有用信号,实现了光纤陀螺噪声的去除。详细对比了小波方法与自适应时频峰值滤波算法并分析了两者的去噪效果。仿真结果和实际数据验证表明:自适应时频峰值滤波算法能有效减小光纤陀螺输出噪声,信噪比比小波滤波改善1~3 dB;特别对于高动态信号,该算法滤波后的信号能够有效地跟踪原始信号。
[Abstract]:In order to reduce the noise of fiber optic gyroscope output signal and improve the accuracy of inertial navigation system, an adaptive time-frequency peak filtering algorithm for fiber optic gyroscope signal is proposed. The fiber optic gyroscope (fog) signal is initially transformed and moduled. the time-frequency analysis of the modulation signal is carried out by using the pseudo-Wigner-Ville distribution, and an adaptive criterion for obtaining the optimal window length of the pseudo-Wigner-Ville distribution is given. The instantaneous frequency estimation of the encoded signal is realized by local peak search, and then the useful signal is reduced, and the noise removal of fiber optic gyroscope is realized. The wavelet method and adaptive time-frequency peak filtering algorithm are compared in detail, and the denoising effect of wavelet method and adaptive time-frequency peak filtering algorithm is analyzed. The simulation results and practical data verify that the adaptive time-frequency peak filtering algorithm can effectively reduce the output noise of fiber optic gyroscope, and the signal-to-noise ratio wavelet filtering can improve the signal-to-noise ratio by 1 鈮，
本文编号：2493419
[Abstract]:In order to reduce the noise of fiber optic gyroscope output signal and improve the accuracy of inertial navigation system, an adaptive time-frequency peak filtering algorithm for fiber optic gyroscope signal is proposed. The fiber optic gyroscope (fog) signal is initially transformed and moduled. the time-frequency analysis of the modulation signal is carried out by using the pseudo-Wigner-Ville distribution, and an adaptive criterion for obtaining the optimal window length of the pseudo-Wigner-Ville distribution is given. The instantaneous frequency estimation of the encoded signal is realized by local peak search, and then the useful signal is reduced, and the noise removal of fiber optic gyroscope is realized. The wavelet method and adaptive time-frequency peak filtering algorithm are compared in detail, and the denoising effect of wavelet method and adaptive time-frequency peak filtering algorithm is analyzed. The simulation results and practical data verify that the adaptive time-frequency peak filtering algorithm can effectively reduce the output noise of fiber optic gyroscope, and the signal-to-noise ratio wavelet filtering can improve the signal-to-noise ratio by 1 鈮，
本文编号：2493419
本文链接：https://www.wllwen.com/kejilunwen/wltx/2493419.html
