基于小波分析法的光纤陀螺系统标定方法

Calibration method of FOG system based on wavelet theory

光纤陀螺的随机噪声很大程度地影响系统的标定精度,通常的解决方法是增加平均次数、延长采样时间等,不能从根本上解决问题。利用小波变换阈值滤波对光纤陀螺系统标定的测试数据进行预先处理,采用Allan方差法分析滤波前后的测试数据,发现滤波后陀螺各项随机误差系数均得到明显抑制。用消噪后的陀螺信号进行标定处理,结果表明仅用原数据量的十分之一即可获得相同的标定精度,减少了测试时间,提高了标定效率。

FOG＇s random noise greatly influences the calibration precision of system. The usual solving method of prolonging sampling time, increasing average fi＇equency can＇t ultimately solve the problem. By pre- processing the calibrated testing data of FOG system using wavelet transform threshold-value filtring, all the random error coefficient will be restrained obviously, which is found by analysing the fore-and-aft testing data using Allan variance method. The gyro signal after de-noising was then used to calibration processing, and the result indicates that it could obtain the same calibration accuracy by using only tenth of the original data. In addtion, it also reduce the testing time and improve the calibration efficiency.