Session C2: Advances in High Accuracy Positioning

Enhanced Precise Point Positioning Using all Operational Satellite-Based SSR Data
Cheng-Wei Wang, National Cheng Kung University; Nobuaki Kubo, Tokyo University of Marine Science and Technology; Shau-Shiun Jan, National Cheng Kung University
Date/Time: Wednesday, Sep. 18, 2:58 p.m.

Real-time estimation of precise ephemeris and satellite clock bias poses a formidable challenge for precise point positioning (PPP) [1], potentially leading to degraded positioning accuracy and longer convergence time. As a promising technology for decades, PPP preserved both satisfactory performance and flexible operation; however, weekly-processing product was essential in the past. Inaccurate satellite positions and clock biases dominate satellite-specified errors, which might require several hours for the navigation filter to converge. These factors eventually result in mean position errors in the order of meters [2, 3], which is of utmost importance. Without access to the Internet and precise product, achieving lane-level accuracy with kinematic PPP is extremely challenging, especially in dense urban environments. To mitigate the long-latency problem and reach real-time satellite precise orbit determination, the next-generation augmentation services from Galileo, Quasi-Zenith Satellite System (QZSS) and BeiDou have become available. The provided free-of-charge, modernized correction messages transmitted from satellite, enable users to conduct state-space representation (SSR) augmented PPP [4, 5]. Thus, the satellite-side errors could be significantly reduced [3, 6, 7], thereby facilitating implementation of real-time PPP. Existing studies [3, 8, 9] have demonstrated the competitive results compared to traditional post-processed PPP. Consequently, the rapid growth of autonomous driving applications has positioned this emerging technology as a focal point of research in recent years.
Not every visible satellite can be augmented, which depends on availability of valid corrections: A single augmentation message from single constellation might limit use of SSR PPP, as it cannot adapt to varying environments or different regions. QZSS provides Multi-GNSS Advanced Orbit and Clock Augmentation (MADOCA)-PPP [6] using the L6E signal (1278.75 MHz) while BeiDou-3 provides B2b-PPP [10] using the B2b signal (1207.14 MHz). Both are available in Asia-Pacific region, however, restricting users to constellation-dependent coverage area. Furthermore, the first B2b-PPP solution can only be implemented after completing issue-of-data (IOD) matching strategies [10]. This indicates that users may need to wait for up to one broadcast period, which is 48 sec, to receive complete correction messages. On the other hand, Galileo supports the first global augmentation service, High Accuracy Service (HAS)-PPP [11], using E6-B signal (1278.75 MHz). The message dissemination principle involves distributing the encoded HAS pages to multiple satellites at different receiving times. Simultaneous and continuous reception of HAS messages is required for conducting HAS-PPP. The more Galileo satellites are visible, the shorter the message decoding time. If users cannot complete reception of HAS message within 150 sec, HAS-PPP might fail to be applied when the signal is severely blocked [11]. To mutually compensate deficiencies in each augmentation message, this paper aims to leverage all operational satellite-based SSR product. Our proposed positioning engine can augment and integrate five SSR-corrected constellations, including GPS, QZSS, GLONASS, Galileo and BeiDou. The general PPP solutions will not be limited by regions and decoding elapsed time. However, the proposed method faces the following challenge: only survey-grade receivers that can simultaneously receive and output raw navigation bits from L6E, B2b and E6-B signal. Therefore, we employ two low-cost software defined receiver (SDR), PocketSDR [12], to receive L6E/E6-B and B2b signal at the same time. Building upon our previous work [13], we have integrated a customized SSR message decoding module and successfully obtained the correction information.
This work aims to address the following question: Which SSR product should be selected to augment the common constellation? Specifically, MADOCA service provides correction messages for GPS, GLONASS, Galileo and QZSS. HAS provides correction messages for GPS and Galileo, while the B2b service provides for GPS and BeiDou. To maximize the availability, HAS is chosen as targeted SSR to augment Galileo. On the other hand, GPS is the commonly augmented constellation for which all three SSR products provide the correction message. Therefore, as the first part of this paper, the targeted SSR will be selected by conducting error analysis. Existing research [3, 14, 15] evaluated the performance (satellite orbit, clock bias, ranging error, etc.) of each single service; however, the actual performance of satellite-based MADOCA message has not yet been explored since its trial service began in 2022. Hence, we aim to evaluate all the SSR-supported constellation in this paper, by taking the satellite orbit, clock bias and signal-in-space-ranging-error (SISRE) as indicators. International GNSS Station (IGS) final product is chosen as reference while at least five consecutive days SSR message will be collected and analyzed. Second, the optimization-based approach, factor graph optimization (FGO) [16] is chosen to estimate the system states. FGO theoretically has the same performance as extended Kalman filter (EKF) [17]. However, the assumption is not consistent due to measurement marginalization and batch processing in FGO. This approach has recently captured more attention as it outperforms the EKF and provides a sensor integration framework [18, 19]. In this paper, L1 and L5 frequencies of pseudorange and carrier phase are applied with the ionospheric-free combination. To address the continuity issue arising due to the absence of L5 signals from several GPS satellites, the single frequency measurements, L1 pseudorange and carrier phase are aided with ionospheric error estimation. Furthermore, the L1 Doppler measurements are included to estimate receiver velocity. We have found that the Doppler-aiding model greatly reduces the convergence time in kinematic datasets, especially after the signal interrupted conditions [2]. The positioning accuracy can therefore improve in the order of meters.
The proposed method is validated using several kinematic datasets in different environments, including suburban and urban canyons. In addition to the two SDRs for SSR message collection, a low-cost receiver, Septentrio mosaic-X5 receiver is employed to collect multi-GNSS raw observations at 1 Hz. The ground truth is produced by the RTK/INS system composed of Northrop Grumman LN200C and NovAtel PwrPak7. In conclusion, the evaluation in this paper includes:
(1) Conducting the error analysis on all SSR-supported constellations, including QZSS MADOCA, Galileo HAS and BeiDou-3 B2b services.
(2) Comparing the service availability among QZSS MADOCA-PPP, Galileo HAS-PPP, and BeiDou-3 B2b-PPP in Taiwan.
(3) The positioning performance of the proposed all operational satellite-based SSR-augmented PPP under varying environments. The result is compared to conventional single point positioning (SPP), conventional PPP, EKF-based SSR-augmented PPP, and FGO-based SSR-augmented PPP.
The results demonstrate the improvement in navigation availability through enhanced amount of valid SSR corrections, and continuity achieved by single frequency-aiding observation model. Our positioning engine is regardless of regions and decoding elapsed time with improving positioning accuracy.
References:
[1] Zumberge, J. F., Heflin, M. B., Jefferson, D. C., Watkins, M. M., & Webb, F. H. (1997). Precise point positioning for the efficient and robust analysis of GPS data from large networks. Journal of Geophysical Research: Solid Earth, 102(B3), 5005-5017. doi:https://doi.org/10.1029/96JB03860
[2] Cheng-Wei, W. (2023). Analysis of using MADOCA product on tightly coupled GNSS/INS. (M.S.). National Cheng Kung University, Tainan, Taiwan.
[3] Zhou, H., Wang, L., Fu, W., Han, Y., Li, T., Li, W., & Chen, R. (2022). Real-time single-frequency precise point positioning using BDS-3 PPP-B2b corrections. Measurement, 205, 112178. doi:https://doi.org/10.1016/j.measurement.2022.112178
[4] Hirokawa, R., Fernández-Hernández, I., & Reynolds, S. (2021). PPP/PPP-RTK open formats: Overview, comparison, and proposal for an interoperable message. NAVIGATION, 68(4), 759-778. doi:https://doi.org/10.1002/navi.452
[5] Hirokawa, R., Hauschild, A., & Everett, T. (2023). Python Toolkit for Open PPP/PPP-RTK Services. Paper presented at the Proceedings of the 36th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2023).
[6] Kawate, K., Igarashi, Y., Yamada, H., Akiyama, K., Okeya, M., Takiguchi, H., Murata, M., Sasaki, T., Matsushita, S., Miyoshi, S., Miyoshi, M., & Kogure, S. (2023). MADOCA: Japanese precise orbit and clock determination tool for GNSS. Advances in Space Research, 71(10), 3927-3950. doi:https://doi.org/10.1016/j.asr.2023.01.060
[7] Parra, C., Schütz, A., Pany, T., Hugentobler, U., & Baumann, S. (2023). The Galileo High Accuracy Service: Assessment of the Quality of Corrections and Preliminary PPP Performance. Paper presented at the Proceedings of the 36th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2023).
[8] Miguez, J., Oliveira-Salgueiro, F., Psychas, D., Galletti, E., Zoccarato, P., Blonski, D., & Garcia-Molina, J. A. (2023). Precise Point Positioning Performance Assessment Setup for Galileo High Accuracy Service.
[9] Wang, G., Kuang, C., Cai, C., Yi, Z., & Dai, W. (2018). Real-time PPP based on the Quasi-Zenith Satellite System MADOCA-LEX signal. IET Radar, Sonar & Navigation, 12(5), 494-498. doi:https://doi.org/10.1049/iet-rsn.2017.0387
[10] China Satellite Navigation Office. BeiDou Navigation Satellite System Signal in Space Interface Control Document: Precise Point Positioning Service Signal PPP-B2b (Version 1.0). http://en.beidou.gov.cn/SYSTEMS/ICD/202008/P020200803538771492778.pdf
[11] European Union Space Program Agency (EUSPA) (2022) Galileo high accuracy service signal-in-space interface control document (HAS SIS ICD). https://www.gsc-europa.eu/sites/default/files/sites/all/files/Galileo_HAS_SIS_ICD_v1.0.pdf
[12] Takasu, T. (2022). Development of QZSS L6 Receiver without Pilot Signal by using SDR. Paper presented at the IPNTJ Annual Conference.
[13] Wang, C.-W., & Jan, S.-S. (2023). Development of Kalman Filter-Based Software Receiver for QZSS L6 and Galileo E6-B Signals. Paper presented at the Proceedings of the 36th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2023).
[14] Martini, I., Susi, M., Cucchi, L., & Fernandez-Hernandez, I. (2023). Galileo high accuracy service performance and anomaly mitigation capabilities. GPS Solutions, 28(1), 25. doi:10.1007/s10291-023-01555-w
[15] Zhang, S., Du, S., Li, W., & Wang, G. (2019). Evaluation of the GPS Precise Orbit and Clock Corrections from MADOCA Real-Time Products. Sensors, 19(11). doi:10.3390/s19112580
[16] Kschischang, F. R., Frey, B. J., & Loeliger, H. A. (2001). Factor graphs and the sum-product algorithm. IEEE Transactions on Information Theory, 47(2), 498-519. doi:10.1109/18.910572
[17] Zhu, F., Xu, Z., Zhang, X., Zhang, Y., Chen, W., & Zhang, X. (2024). On State Estimation in Multi-Sensor Fusion Navigation: Optimization and Filtering. ArXiv, abs/2401.05836.
[18] Suzuki, T. (2023). Precise Position Estimation Using Smartphone Raw GNSS Data Based on Two-Step Optimization. Sensors, 23(3). doi:10.3390/s23031205
[19] Wen, W., Pfeifer, T., Bai, X., & Hsu, L.-T. (2021). Factor graph optimization for GNSS/INS integration: A comparison with the extended Kalman filter. NAVIGATION, 68(2), 315-331. doi:https://doi.org/10.1002/navi.421