Currently, five new-generation BeiDou (BDS-3) experimental satellites are working in orbit and broadcast B1I, B3I, and other new signals. Precise satellite orbit determination of the BDS-3 is essential for the future global services of the BeiDou system. However, BDS-3 experimental satellites are mainly tracked by the international GNSS Monitoring and Assessment Service (iGMAS) network. Under the current constraints of the limited data sources and poor data quality of iGMAS, this study proposes an improved cycle-slip detection and repair algorithm, which is based on a polynomial prediction of ionospheric delays. The improved algorithm takes the correlation of ionospheric delays into consideration to accurately estimate and repair cycle slips in the iGMAS data. Moreover, two methods of BDS-3 experimental satellite orbit determination, namely, normal equation stacking (NES) and step-by-step (SS), are designed to strengthen orbit estimations and to make full use of the BeiDou observations in different tracking networks. In addition, a method to improve computational efficiency based on a matrix eigenvalue decomposition algorithm is derived in the NES. Then, one-year of BDS-3 experimental satellite precise orbit determinations were conducted based on iGMAS and Multi-GNSS Experiment (MGEX) networks. Furthermore, the orbit accuracies were analyzed from the discrepancy of overlapping arcs and satellite laser range (SLR) residuals. The results showed that the average three-dimensional root-mean-square error (3D RMS) of one-day overlapping arcs for BDS-3 experimental satellites (C31, C32, C33, and C34) acquired by NES and SS are 31.0, 36.0, 40.3, and 50.1 cm, and 34.6, 39.4, 43.4, and 55.5 cm, respectively; the RMS of SLR residuals are 55.1, 49.6, 61.5, and 70.9 cm and 60.5, 53.6, 65.8, and 73.9 cm, respectively. Finally, one month of observations were used in four schemes of BDS-3 experimental satellite orbit determination to further investigate the reliability and advantages of the improved methods. It was suggested that the scheme with improved cycle-slip detection and repair algorithm based on NES was optimal, which improved the accuracy of BDS-3 experimental satellite orbits by 34.07%, 41.05%, 72.29%, and 74.33%, respectively, compared with the widely-used strategy. Therefore, improved methods for the BDS-3 experimental satellites proposed in this study are very beneficial for the determination of new-generation BeiDou satellite precise orbits.

The linear combination of multi-frequency carrier-phase and pseudorange observations can form the combined observations with special properties. The type and number of combined frequencies will directly affect the characteristics of the combined observations. BDS-2 and BDS-3 broadcast three and five signals, respectively, and the study of their linear combination is of great significance for precision positioning. In this contribution, the linear combination form of multi-frequency carrier-phase observations in cycles and meters is sorted out. Seven frequency combination modes are formed, and some special combinations for positioning are searched. Then, based on the principle of minimum combined noise, a simpler search method for the optimal real coefficients of ionosphere-free (IF) combination based on the least squares (LS) principle is proposed. The general analytical expressions of optimal real coefficients for multi-frequency geometry-based and ionosphere-free (GBIF), geometry-free and ionosphere-free (GFIF), and pseudorange multipath (PMP) combinations with the first-order ionosphere delay taken into account are derived. And the expression derivation process is given when both the first-order and second-order ionospheric delays are eliminated. Based on this, the characteristics of the optimal real coefficient combination in various modes are compared and discussed. The various combinations reflect that the accuracy of the combined observations from dual-frequency (DF) to five-frequency (FF) is gradually improving. The combination coefficient becomes significantly larger after taking the second-order ionospheric delay into account. In addition, the combined accuracy of BDS-3 is better than that of BDS-2. When only the first-order ionosphere is considered, the combination attribute of (B1C, B1I, B2a) is the best among the triple-frequency (TF) combinations of BDS-3. When both the first-order and second-order ionospheric delays are considered, the (B1C, B3I, B2a) combination is recommended.

The accuracy of ultra-rapid orbits is a key parameter for the performance of GNSS (Global Navigation Satellite System) real-time or near real-time precise positioning applications. The quality of the current BeiDou demonstration system (BDS) ultra-rapid orbits is lower than that of GPS, especially for the new generational BDS-3 satellites due to the fact that the availability of the number of ground tracking stations is limited, the geographic distribution of these stations is poor, and the data processing strategies adopted are not optimal. In this study, improved data processing strategies for the generation of ultra-rapid orbits of BDS-2/BDS-3 satellites are investigated. This includes both observed and predicted parts of the orbit. First, the predicted clock offsets are taken as constraints in the estimation process to reduce the number of the unknown parameters and improve the accuracy of the parameter estimates of the orbit. To obtain more accurate predicted clock offsets for the BDS’ orbit determination, a denoising method (also called the Tikhonov regularization algorithm), inter-satellite correlation, and the partial least squares method are all incorporated into the clock offsets prediction model. Then, the Akaike information criterion (AIC) is used to determine the arc length in the estimation models by taking the optimal arc length in the estimation of the initial orbit states into consideration. Finally, a number of experiments were conducted to evaluate the performance of the ultra-rapid orbits resulting from the proposed methods. Results showed that: (1) Compared with traditional models, the accuracy improvement of the predicted clock offsets from the proposed methods were 40.5% and 26.1% for BDS-2 and BDS-3, respectively; (2) the observed part of the orbits can be improved 9.2% and 5.0% for BDS-2 and BDS-3, respectively, by using the predicted clock offsets as constraints; (3) the accuracy of the predicted part of the orbits showed a high correlation with the AIC value, and the accuracy of the predicted orbits could be improved up to 82.2%. These results suggest that the approaches proposed in this study can significantly enhance the accuracy of the ultra-rapid orbits of BDS-2/BDS-3 satellites.

Due to the traditional fixed model used in precise point positioning (PPP) solutions, multi-frequency and multi-Global Satellite Navigation System (GNSS) observations have not been fully introduced into positioning services. In consideration of the BDS-3 multi-frequency signals and the new development of other GNSS systems, a new multi-frequency and multi-GNSS PPP solution strategy should be proposed to flexibly model and use all observations. In this study, a preliminary mixed multi-frequency PPP solution strategy is analyzed and tested based on a combination of BDS-3 and GNSS observations. First, the multi-frequency observations are combined and their coefficients are rapidly estimated by least squares; then, the inter-system bias parameter and the stochastic model are introduced into the function model; and finally, the mixed PPP solution and its software are developed and verified by three groups of experiments. According to the experimental results of 96 stations and ten-day multi-GNSS experiment observations, it is indicated that the root-mean-square error of positioning and the convergence time are significantly optimized with the aid of additional frequencies, where the accuracy improvements of multi-frequency and multi-GNSS scheme in the east (E), north (N) and up (U) directions can respectively reach up to 23.2%, 13.3% and 23.8% compared with the traditional BDS-3 dual-frequency ionosphere-free (IF) PPP model; and the corresponding convergence time is reduced from 18.54 min to 13.18 min. Meanwhile, from the results of multi-frequency BDS-3 PPP experiments based on 53 stations, it is suggested that a better performance of positioning and convergence can be obtained by the mixed PPP solution, where the position RMS of the E, N and U directions are reduced by 38.2%, 23.9% and 26.3%, and the convergence time is decreased from 23.86 min to 12.43 min for the combined BDS-3 of all observations, compared with the BDS-3-only solution. Furthermore, in the vehicle experiment of multi-frequency kinematics PPP, a convergence process can be found for different scenarios of BDS-3 combination with other observations. Moreover, the residual series are different for each solution, in which reductions of 71.1%, 33.3% and 77.1% in the E, N and U directions, respectively, can be obtained compared with the traditional BDS-3 dual-frequency IF model in kinematics experiments based on multi-GNSS and multi-frequency scenarios. Therefore, it is meaningful to recommend the mixed PPP solution in the GNSS community to fully use multi-frequency and multi-GNSS observations by the adaptive combination of different observations.

GNSS ultrarapid clock biases are key inputs of rapid high-accuracy applications, especially for its prediction parts. With the fast development of the BeiDou system (BDS), the system performances are mainly represented by orbit and clock products. However, it is suggested that the BDS-predicted clock biases cannot meet the requirement of real-time or near real-time services. In this research, the BDS satellite-predicted ultrarapid clock bias products are optimized with three methods, namely, one-step strategy, intersatellite correlation, and variogram model, using the combined estimation of BDS-2 and BDS-3 satellites. Firstly, considering the traditional two-step strategy for modelling clock bias prediction, we take all terms (including trend and periodic terms) into one-step solution of model estimation based on the sparse modelling in machine learning. Secondly, because of the much more stable on-board atomic clock of BDS-3 satellites, the intersatellite correlations between BDS-2 and BDS-3 are utilized to enhance the solution of model coefficients. Thirdly, to further improve the model, the temporal correlations in model residuals are used to reconstruct the stochastic function obtained by variogram. In addition, to verify the proposed improved strategies, 12 schemes of BDS clock bias prediction experiments are designed and analyzed with different conditions. According to the results of predicted clock biases, it is indicted that (1) the stability of BDS-3 on-board clocks is more optimal compared with BDS-2, which can be used to strengthen the solution of the clock bias prediction model; (2) the one-step estimation of the clock bias model by sparse modelling can slightly increase the accuracy of prediction results; (3) both BDS-2- and BDS-3-predicted clock biases benefited each other by inserting the intersatellite correlations into the weight matrix, in which the accuracy of 18-hour period with one-step strategy can be improved by 28.6% and 27.2% for BDS-2 and BDS-3, respectively; and (4) after the introduction of the variogram model in updating the weight matrix, the clock bias prediction model is further corrected by 8.0% and 11.1% for BDS-2 and BDS-3. In summary, improved strategies for BDS ultrarapid satellites’ clock bias prediction using BDS-2 and BDS-3 integrated processing are meaningful for the current BDS ultrarapid satellites’ clock bias prediction products.