A continuous time transfer method based on beidou-3 PPP-B2b information

Abstract

A continuous time transfer method based on Beidou No.3 PPP-B2b information, the steps are as follows: S1. If the observation data is continuous, use Beidou No.3 PPP-B2b information to solve PPP and obtain time transfer result; S2. If the observation data is abnormal and causes instantaneous interruption of observation data, use the clock difference result predicted by the historical clock difference data of the receiver as the time transfer result of the current epoch. High-precision time transfer can be realized without relying on the Internet, and the situation of short-term interruption of signals caused by shielding and instantaneous interference is considered. By modeling the receiver clock, the re-convergence speed of PPP is accelerated, continuous time transfer is realized, the application range of GNSS precise timing / time transfer technology in the field of national defense and military is effectively expanded, and the availability and reliability of GNSS time transfer in the instantaneous interference environment are improved.

Description

Technical Field

[0001] This invention relates to the field of GNSS time transfer, and in particular to a high-precision continuous time transfer method based on BeiDou-3 PPP-B2b information. Background Technology

[0002] With the development of global navigation satellite systems, GNSS precise time synchronization technology has been widely used in the time and frequency field. Since the introduction of Precise Point Positioning (PPP) technology, many scholars at home and abroad have applied PPP technology to the field of time transfer. However, due to limitations in post-processing satellite precise orbit and clock error products, its application has mainly been limited to post-processing modes. In 2013, IGS real-time service broadcast real-time satellite orbit error and clock error corrections via the Internet, and the results showed that the inter-station time synchronization accuracy calculated using IGS real-time products could reach 0.3 ns. Ge et al. used BNC software to receive multi-system real-time satellite orbit and clock error products broadcast by the IGS analysis center via the Internet, which can achieve high-precision real-time time transfer.

[0003] The PPP time transfer algorithm based on IGS real-time data stream has greatly promoted the application of PPP technology in the field of real-time time transfer. However, there are many limiting environments in the defense and military fields. Many weapon systems are physically isolated from the Internet and cannot obtain real-time data stream products. This significantly restricts the real-time application of PPP time transfer technology in the field of high-precision time service. Moreover, the environment faced by time synchronization / time transfer equipment is also more complex. There are natural environments such as mountains and buildings, as well as electromagnetic radiation equipment such as radio stations and mobile base stations. There are also various intentional electromagnetic interferences. This results in numerous, densely distributed limiting environments in the time, space, and frequency domains that significantly affect the performance of time synchronization equipment. This further increases the complexity of the environment in which time synchronization terminals are used, which can easily cause satellite observation signal interruption. This leads to frequent re-convergence of PPP or failure to achieve continuous time transfer, while continuous time transfer sequence is necessary for high-precision time transfer. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing PPP time transfer technology, which relies on IGS real-time data stream products and whose time transfer results are easily discontinuous due to environmental influences. This invention proposes a continuous time transfer method based on BeiDou-3 PPP-B2b information that does not rely on the Internet and can overcome instantaneous signal blockage.

[0005] To achieve the above-mentioned objectives, the present invention adopts the following technical solution:

[0006] A continuous time transfer method based on BeiDou-3 PPP-B2b information, the steps of which are as follows:

[0007] S1: If the observation data is continuous, use the BeiDou-3 PPP-B2b information to perform PPP calculation and obtain the time transfer result;

[0008] The PPP time transfer uses a dual-frequency de-ionization combination, and the mathematical model is as follows:

[0009]

[0010]

[0011] In the formula: j represents the satellite number, Φ and P represent the combined observations of carrier phase and pseudorange ionospheric desaturation, ρ is the geometric distance from the satellite to the receiver, c is the speed of light, and δt j Satellite clock bias, where δt is the receiver clock time propagation result relative to BeiDou time, λ and N are the ionospheric decoupling carrier length and ambiguity, respectively, and T is the tropospheric delay. For carrier and pseudorange measurement noise.

[0012] S2: If the observation data is abnormal and causes a momentary interruption, the clock difference result predicted by the receiver's historical clock difference data is used as the time transfer result for the current epoch.

[0013] When the observation data is recovered, the predicted time transfer result of the current epoch is used as a constraint to constrain the clock error parameter to be estimated, assisting the PPP time transfer algorithm, accelerating the reconvergence of the data for solution, and thus realizing continuous time transfer.

[0014] Furthermore, step S1 includes the following steps:

[0015] S11. Data Acquisition

[0016] The system acquires observation data, broadcast ephemeris and PPP-B2b information, and uses the real-time satellite orbit, clock error and inter-symbol error correction information provided by the PPP-B2b signal to correct the satellite orbit, satellite clock error and satellite inter-symbol error, and obtains the final satellite orbit, clock error results and corrected observation values.

[0017] S12. Data Preprocessing

[0018] Data preprocessing involves removing outliers from the observation data and detecting cycle slips. Cycle slip detection uses TurboEdit and the ionospheric residual method to mark cycle slips.

[0019] S13. Data Processing

[0020] Data processing corrects for errors in solid tides, ocean tides, phase entanglement, tropospheric delay, Earth's rotation, relativistic effects, and receiver antenna phase center.

[0021] S14. Parameter Estimation

[0022] Parameter estimation is the process of solving for unknown parameters such as receiver position, receiver clock error, wet component of tropospheric zenith delay, and deionization combined ambiguity.

[0023] S15. Examine the residual results of the Kalman filter output. If the maximum residual is greater than the threshold, it is judged as a gross error, removed, and the Kalman filter is recalculated until the residual test is passed.

[0024] S16. Output and store the calculated time transfer results.

[0025] Further, step S2 includes the following steps:

[0026] S21. Before performing time-transfer calculations, determine whether the observed data is continuous;

[0027] S22. If the data is continuous, then the data processing method in step one shall be used;

[0028] S23. If the data is abnormally interrupted only in the current epoch, mark it, and use the stored receiver clock bias data to build a receiver clock bias model to predict the clock bias result in the current epoch.

[0029] The steps for predicting the current epoch clock difference result in step S23 are as follows:

[0030] Establish a receiver clock bias model:

[0031] x(t) = a0 + a1(t-t0) + a2(t-t0) 2 +ε x (t)

[0032] In the formula, a0, a1, and a2 represent clock error, clock speed, and clock drift, respectively, and ε x (t) represents the uncertain component that varies randomly;

[0033] The clock bias model matrix is ​​established as follows:

[0034] X = Ha + e

[0035]

[0036] Considering the strong inter-epoch correlation of clock error results, the closer the forecast is to the current observation epoch, the greater its contribution to the forecast result. To improve forecast accuracy, a forgetting factor matrix is ​​introduced, and a recursive forgetting factor least squares algorithm is used for forecasting. The steps are as follows:

[0037] Introducing W = diag(λ) n-1 ,λ n-2 ,,λ 0 ), 0<λ≤1, and calculate and T m (m≥3)

[0038]

[0039] Iterative calculation of time k (k = m+1, m+2, ..., n) T k :

[0040]

[0041] Predict the current epoch t k The time difference, or the result of time transmission:

[0042]

[0043] The forecast time-transfer results are stored, and the current forecast results can be output to ensure the continuity of the time-transfer results.

[0044] S24. If the previous observation data was interrupted, and the observation data of the current epoch is restored to normal, the time transfer result of the current epoch can be predicted according to the established receiver clock bias model in step S22. The predicted time transfer result of the current epoch is used as a constraint condition to constrain the clock bias parameter to be estimated, assist the PPP time transfer algorithm, speed up the re-convergence of data for solution, and realize continuous time transfer.

[0045] Compared with the prior art, the present invention has the following advantages:

[0046] This invention utilizes B2b information broadcast by BeiDou-3 for time transmission, overcoming the shortcomings of existing PPP time transmission technology, which relies on IGS real-time data stream products and whose time transmission results are easily discontinuous due to environmental influences. It can achieve high-precision time transmission without internet access. Furthermore, it considers situations where signal interruptions occur due to obstructions or transient interference. By modeling the receiver clock, it accelerates the reconvergence speed of PPP, achieving continuous time transmission. This effectively expands the application scope of GNSS precision time synchronization / time transmission technology in the defense and military fields, and improves the availability and reliability of GNSS time transmission under transient interference environments. Attached Figure Description

[0047] Figure 1 Real-time PPP algorithm based on BeiDou-3 PPP-B2b information;

[0048] Figure 2 A continuous-time information transmission method based on BeiDou-3 PPP-B2b. Detailed Implementation

[0049] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings.

[0050] like Figure 1 , Figure 2 As shown, the continuous-time information transmission method based on BeiDou-3 PPP-B2b includes two main steps, as follows:

[0051] Step 1: When observing continuous data, follow the... Figure 1 The process involves calculations to acquire observational data, broadcast ephemeris, and PPP-B2b information, followed by PPP solution to obtain the time transfer results. PPP time transfer employs a dual-frequency ionosphere desiccation combination, and its mathematical model is as follows:

[0052]

[0053]

[0054] To facilitate the solution, the observation equation can be linearized to obtain:

[0055] V = AX - L

[0056] Where: X = [X, Y, Z, c·δt, ztd, N] 1 N 2 ,,N m X, Y, Z are the receiver's three-dimensional coordinates, ztd is the zenith tropospheric delay, V is the observation residual vector, and A represents the design matrix.

[0057] Step Two: If abnormal observation data causes a momentary interruption in the observation data, proceed according to... Figure 2 The process involves calculations, using the clock error results predicted from the receiver's historical clock error data as the time transfer result for the current epoch;

[0058] When the observation data is recovered, the predicted time transfer result of the current epoch is used as a constraint to constrain the clock error parameter to be estimated, assisting the PPP time transfer algorithm, accelerating the reconvergence of the data for solution, and thus realizing continuous time transfer.

[0059] To clearly demonstrate the steps of this invention, the detailed steps of step S1, PPP solution based on B2b information, are given below:

[0060] (1) Data Acquisition: Acquire observation data, broadcast ephemeris, and PPP-B2b information. Using the real-time satellite orbit, clock bias, and inter-symbol bias correction information provided by the PPP-B2b signal to correct the satellite orbit, clock bias, and inter-symbol bias, obtaining the corrected satellite orbit, clock bias results, and corrected observation values ​​is crucial for achieving high-precision time transfer. The correction method is as follows:

[0061] a) Satellite orbit correction

[0062] Satellite position vector X calculated from broadcast ephemeris brdc and orbital correction vector Δ X The corrected satellite position Δ can be calculated. orbit The calculation formula is as follows:

[0063] X orbit =X brdc -Δ X

[0064] Where, Δ X The orbital correction vector Δ can be provided by B2b information. orbit The calculation is as follows:

[0065]

[0066]

[0067] e A =e C ×r R

[0068] Δ X =[e R e A e C ]·Δ orbit

[0069] In the formula r, These are the satellite position vector and velocity vector calculated from the broadcast ephemeris, respectively. R e A e C These are the radial, tangential, and normal unit vectors, respectively.

[0070] b) Satellite clock correction

[0071] Clock error correction parameters are correction parameters relative to broadcast ephemeris clock errors.

[0072]

[0073] Where t brdc Satellite clock corrections calculated for broadcast ephemeris, t s C0 represents the corrected satellite clock bias correction value, and C0 represents the clock bias correction parameter provided by the B2b information.

[0074] c) Satellite inter-symbol offset correction

[0075] B2b information provides satellite inter-symbol DCB for different signals. sig It can be used for the original observation value P sigAfter making the appropriate corrections, the corrected observation value P is calculated. corr The following is a public notice:

[0076] P corr =P sig -DCB sig

[0077] (2) Data Preprocessing. Data preprocessing mainly achieves quality control of time-transfer solution, detects gross errors in the observation data, removes gross errors and outliers, and detects cycle slips. Cycle slip detection uses TurboEdit and the ionospheric residual method to mark the locations of cycle slips in the observation data. No cycle slips are repaired. During time-transfer solution, an ambiguity parameter is added at the corresponding location for estimation.

[0078] (3) Data Processing. Data processing involves modeling and correcting various error sources related to the satellite, receiver, and signal propagation path. This primarily involves correcting errors related to phase entanglement, relativistic effects, polar tides, solid tides, ocean tides, tropospheric delay, and Earth's rotation to improve time transfer accuracy. It is worth noting that the satellite position reference point provided by the PPP-B2b correction information is already the phase center of the satellite antenna; therefore, the time transfer algorithm based on BeiDou-3 PPP-B2b information does not need to correct for satellite antenna phase deviation during data processing.

[0079] (4) The Kalman filter method is used to estimate the unknown parameters of the receiver position, the time transfer result of the receiver clock relative to the BDT, the wet component of the tropospheric zenith delay and the ionospheric de-ambiguity in real time. The weight ratio of pseudorange to carrier phase is 1 / 10000 when solving. If the receiver position is fixed, the coordinate parameters are treated as constants.

[0080] (5) The residual results of the Kalman filter output are checked. If the maximum residual is greater than the threshold, it is judged as a gross error, removed and the Kalman filter is recalculated until the residual check is passed.

[0081] (6) Output and store the time transfer results obtained from the solution.

[0082] Furthermore, step S2 above includes the following steps:

[0083] (1) Before performing time transfer calculation, determine whether the observation data is continuous;

[0084] (2) If the data is continuous, then the data processing method in step one shall be used;

[0085] (3) If the data is interrupted only in the current epoch, mark it and use the stored receiver clock bias data to build a receiver clock bias model to predict the clock bias result in the current epoch.

[0086] The clock bias model is as follows:

[0087] x(t) = a0 + a1(t-t0) + a2(t-t0) 2 +ε x (t)

[0088] In the formula, a0, a1, and a2 represent clock error, clock speed, and clock drift, respectively, and ε x (t) represents the randomly varying uncertainty component. The clock error model matrix form is as follows:

[0089] X = Ha + e

[0090]

[0091] Considering the strong inter-epoch correlation of clock error results, the closer the forecast is to the current observation epoch, the greater its contribution to the forecast result. To improve forecast accuracy, a forgetting factor matrix is ​​introduced, and a recursive forgetting factor least squares algorithm is used for forecasting. The steps are as follows:

[0092] a) Introduce W = diag(λ) n-1 ,λ n-2 ,,λ 0 ), 0<λ≤1, and calculate and T m (m≥3)

[0093]

[0094] b) Iteratively calculate time k (k = m+1, m+2, ..., n) T k :

[0095]

[0096] c) Predict the current epoch t k The time difference, or the result of time transmission:

[0097]

[0098] The forecast time-transfer results are stored, and the current forecast results can be output to ensure the continuity of the time-transfer results.

[0099] (4) If the observation data of the previous epoch was interrupted and the observation data of the current epoch is restored to normal, then based on the original observation equation, the time transfer result of the current epoch can be predicted according to the receiver clock error model established in step (2). The predicted time transfer result of the current epoch is used as a constraint condition to constrain the clock error parameter to be estimated. At this time, it is equivalent to adding a virtual observation equation on the basis of the original observation equation.

[0100] The PPP time transfer model with added clock bias constraint information is as follows:

[0101]

[0102] Solving the observation equation yields the time transfer result, which can accelerate the convergence speed of PPP time transfer solution and overcome the impact of instantaneous signal interruption on the PPP time transfer result.

Claims

1. A continuous-time transmission method based on BeiDou-3 PPP-B2b information, characterized in that, Includes the following steps: S1: If the observation data is continuous, use the BeiDou-3 PPP-B2b information to perform PPP calculation and obtain the time transfer result; PPP time transfer employs a dual-frequency de-ionization combination, and the mathematical model is as follows: In the formula: j represents the satellite number, Φ and P represent the combined observations of carrier phase and pseudorange ionospheric desaturation, ρ is the geometric distance from the satellite to the receiver, c is the speed of light, and δt j Satellite clock bias, where δt is the receiver clock time propagation result relative to BeiDou time, λ and N are the ionospheric decoupling carrier length and ambiguity, respectively, and T is the tropospheric delay. Noise is measured for carrier and pseudorange; S2: If the observation data is abnormal and causes a momentary interruption, the clock difference result predicted by the receiver's historical clock difference data is used as the time transfer result for the current epoch. When the observation data is recovered, the predicted time transfer result of the current epoch is used as a constraint to constrain the clock error parameter to be estimated, assisting the PPP time transfer algorithm, accelerating the reconvergence of the data for solution, and thus realizing continuous time transfer.

2. The continuous time transmission method based on BeiDou-3 PPP-B2b information according to claim 1, characterized in that: Step S1 includes the following steps: S11. Acquire observation data, broadcast ephemeris and PPP-B2b information, and use the real-time satellite orbit, clock error and inter-symbol error correction information provided by the PPP-B2b signal to correct the satellite orbit, satellite clock error and satellite inter-symbol error, and obtain the final satellite orbit, clock error results and corrected observation values; S12. Data preprocessing involves removing outlier data from the observation data and detecting cycle slips. Cycle slip detection uses TurboEdit and the ionospheric residual method to mark cycle slips. S13. Data processing corrects for errors in solid tides, ocean tides, phase entanglement, tropospheric delay, Earth's rotation, relativistic effects, and receiver antenna phase center. S14. Parameter estimation is to solve for the unknown parameters of receiver position, receiver clock error, tropospheric zenith delay wet component, and deionization combined ambiguity; S15. Examine the residual results of the Kalman filter output. If the maximum residual is greater than the threshold, it is judged as a gross error, removed, and the Kalman filter is recalculated until the residual test is passed. S16. Output and store the calculated time transfer results.

3. The continuous time transmission method based on BeiDou-3 PPP-B2b information according to claim 1, characterized in that: Step S2 includes the following steps: S21. Before performing time-transfer calculations, determine whether the observed data is continuous; S22. If the data is continuous, then the data processing method of step S1 shall be used; S23. If the data is abnormally interrupted only in the current epoch, mark it, and use the stored receiver clock bias data to build a receiver clock bias model to predict the clock bias result in the current epoch. S24. If the previous observation data was interrupted, and the observation data of the current epoch is restored to normal, then based on the original observation equation, the time transfer result of the current epoch can be predicted using the established receiver clock bias model in step S22. The predicted time transfer result of the current epoch can be used as a constraint condition to constrain the clock bias parameter to be estimated, assist the PPP time transfer algorithm, accelerate the reconvergence of data for solution, and realize continuous time transfer.

4. The continuous time transmission method based on BeiDou-3 PPP-B2b information according to claim 3, characterized in that: The steps for predicting the current epoch clock difference result in step S23 are as follows: a. Establish the receiver clock bias model: x(t)=a0+a1(t-t0)+a2(t-t0) 2 +ε x (t) In the formula, a0, a1, and a2 represent clock error, clock speed, and clock drift, respectively, and ε x (t) represents the uncertain component that varies randomly; b. Establish the clock error model matrix: X = Ha + e c. Introduce the forgetting factor matrix W = diag(λ) n-1 ,λ n-2 ,...,λ 0 ), 0<λ≤1, and calculate and T m (m≥3) b) Iteratively calculate time k (k = m+1, m+2, ..., n) T k : c) Predict the current epoch t k The time difference, or the result of time transmission: The recursive forgetting factor least squares algorithm is used for prediction. The predicted time-transfer results are stored and the current prediction results can be output to ensure the continuity of the time-transfer results.

Images