一种用于磁流体陀螺微弱信号检测的小波降噪方法研究

Research of the Wavelet Denoising Applied to the Weak Signal Detection of the Magnetofluid Gyroscope

新型磁流体陀螺可以同时满足卫星微角颤振在轨测量的亚微弧度测量精度及千赫兹带宽的技术要求,优于当前卫星系统中所用陀螺,是用于卫星微角颤振测量的理想传感器。但是其输出信号极其微弱并且淹没在大量噪声中导致无法有效地实现检测角速度的功能。提出了一种基于自相关的小波阈值去噪算法提取微弱信号。该方法通过计算小波分解各层细节分量的自相关系数,自动确定最优小波分解层数,并通过该自相关系数选择最优去噪阈值。对传统阈值去噪算法与所提出的改进小波算法进行了对比实验,实验结果表明该方法能将输出信号信噪比提高7 dB^10 dB,适用于所设计的磁流体陀螺微弱信号的检测提取。

New magnetofluid gyroscope which is superior to the gyroscope used in the current satellite system can meet the technical requirements of the submicron measurement accuracy and the kHz bandwidth in the In￣orbit measurement of the satellite micro angle vibration,It’s an ideal sensor widely used in measurement of satellite micro angle vibration. But its output signal which is drowned into a lot of noise is so weak that the angular velocity can’t be detected effectively.A wavelet denoising method is proposed using self￣correlation to extract the weak signal. This method can choose the optimal wavelet decomposition level and denoising threshold based on the self￣correlation co￣efficient of the detail components. A comparative experiment was made between the traditional wavelet denoising method and the method proposed in this paper. The results show that the SNR of the output signal could be im￣proved by 7 dB^10 dB and it is suitable for the weak signal detection of the magnetofluid gyroscope.