一种改进的平方根滤波在卫星钟差实时解算中的应用

摘要

Currently,the Kalman filter is widely used to calculate satellite clock offset in real-time based on the un-differenced method.Square root filter can enhance the stability of numerical calculation,and avoiding filtering divergence,but the inversion computation time of matrix will take longer.The sequential algorithm can improve computational efficiency for avoiding direct matrix inversion.Taking the advantages of these two algorithms,a new square root filter method is proposed based on the sequential algorithm;it is then applied to satellite clock offset calculating in real time.The result shows that the accuracy of satellite clock offset is better than 0.2 ns,and the calculation time is reduced by 40 ％.%目前普遍采用卡尔曼滤波方法来实现非差法卫星钟差的实时解算.平方根滤波可以增强数值计算的稳定性,避免滤波发散,但矩阵的求逆会耗费较长的计算时间;序贯算法可以避免对矩阵直接求逆,能提升计算效率.综合两种算法的优点,提出一种基于序贯算法的平方根滤波器,并应用于卫星钟差的实时解算,得到的卫星钟差精度优于0.2 ns,且计算时长缩短40％.