Abstract: | In this paper, real-time validation of BeiDou measurements in un-differenced standalone mode is investigated. The Detection-Identification-Adaptation (DIA) approach was used utilizing a geometry-free observation model. Observations from each satellite are independently screened using a local modeling approach where data from each epoch are considered at a time. Main advantages of this approach include that there is no need for computation of the inter-system biases as BeiDou measurements are not integrated with other GNSS, and no satellite navigation information are needed. A reparameterized form of the unknowns is used to overcome rank deficiency of the model. Validation of the triple-frequency BeiDou data was performed in a static mode at a CORS station in Curtin University, Australia. The data span two consecutive days. The effect of multipath is investigated, as multipath can adversely affect the performance of detection and identification of outliers. The paper also presents estimation of the precision of BeiDou observations using an empirical method that utilizes the characteristics of one of the validation test statistics. Finally, the capability of the proposed method is investigated in detection and identification of artificial errors inserted in the tested BeiDou data. The errors were categorized according to their size. Results show that, with the presented stochastic information of the data, the method success rate in detection of errors in the range of 1 cycle to 9 cycles for phase data, and 1.5m to 7.5m for code observations, ranged between 87% and 99%, with an ability to identify code outliers at 92-98%. |
Published in: |
Proceedings of the ION 2013 Pacific PNT Meeting April 23 - 25, 2013 Marriott Waikiki Beach Resort & Spa Honolulu, Hawaii |
Pages: | 106 - 114 |
Cite this article: | El-Mowafy, A., "Real-Time Validation of BeiDou Observations in a Stand-alone Mode," Proceedings of the ION 2013 Pacific PNT Meeting, Honolulu, Hawaii, April 2013, pp. 106-114. |
Full Paper: |
ION Members/Non-Members: 1 Download Credit
Sign In |