Influence Functions of Colored Noises on Kinematic Positioning--Taking the AR Model of First Class As an Example

The accuracy and reliability of the kinematic positioning are affected by not only the random noises and systematic errors, but also the colored noises related to time. Any theory or technique based on the hypothesis of Gaussian white noises ignoring the colored noises cannot guarantee the actual reliability of the parameter estimates. On the basis of both the sequential adjustment and the Kalman filtering, the influence functions (IF) of the colored noises on the parameters are established and derived in theory. Both the colored measurement noises and the state model errors are included. It is shown by the calculation that the influences of the colored noises on the model parameters are significant and the expressions of the influence functions are reliable and correct. 