Carrier-Phase RAIM Algorithm Based on a Vector Autoregressive Model

摘要

A new CRAIM algorithm is designed for multiple faults detection and identification in RTK positioning using vector auto regression (VAR) model, and the corresponding protection level is given. Firstly, the RTK positioning process is introduced, and then we fit the double differenced carrier phase observations with VAR model. Based on the model, we predict the observations and obtain their prediction residuals which are used to detect and identify the potential faults in the observations. If a prediction residual of a certain observation is deviated from the normal range, we demonstrate that the corresponding observation is abnormal. Secondly, according to the accurate calculation of missed detection, a quantitative analysis method is designed for the reliability of the faults detection and identification algorithm. Thirdly, a new allocation method for integrity risk of CRAIM is designed. Then starting from the integrity risk, an algorithm for computing protection levels is given. Examples illustrate that the new algorithm can handle many types of faults, such as the faults caused by serious multipath, incorrect ambiguity resolution and the errors of atmospheric propagation. Besides, the probability of missed detection is investigated under the condition of different faults sizes and observational times. The results demonstrate that probability of missed detection satisfies the requirements of users. Finally, the availabilities for the new algorithm in different flight phase are analyzed by the computation of protection levels.