Method and System for On-Orbit Real-Time Precise Orbit and Clock Error Determination Based on PPP-B2b Enhancement for LEO

Abstract

The present invention discloses a method and system for determining real-time precise orbits and clock offsets of LEO in orbit based on PPP-B2b enhancement, belonging to the technical field of high-precision autonomous orbit determination and clock offset estimation for low-earth orbit satellites. The present invention corrects the orbits and clock offsets of BDS-3 / GPS satellites by using the enhancement information broadcast by the B2b signal of Beidou-3. Under the support of regional BDS-3 / GPS enhancement information, when there are new BDS-3 / GPS tracking satellites, before measurement update, the epoch-to-epoch clock offset variation obtained by using the carrier phase time difference method is used as two orbit determination filtering clock offset parameters for constraint, and the receiver clock offset and undifferenced floating ambiguity parameters are separated in orbit in real time; finally, the precise orbit of LEO and the undifferenced floating ambiguity parameters output by orbit determination filtering are substituted into the phase observation equation to calculate the precise clock offset of the BDS-3 / GPS receiver. The present invention is applicable to the situation where there is only regional GNSS enhancement information support and is less affected by orbit maneuvers.

Description

Technical Field

[0001] The present invention belongs to the technical field of high-precision autonomous orbit determination and clock error estimation for low-Earth orbit satellites, and particularly relates to a method and system for determining real-time precise orbits and clock errors in orbit for LEO based on PPP-B2b augmentation. Background Art

[0002] In recent years, with the increasing number of scientific missions for low-Earth orbit satellites, many scientific missions have put forward higher and higher requirements for the orbits and clock errors of low-Earth orbit satellite platforms in terms of real-time performance, accuracy, and reliability. Spaceborne GNSS has become the main technical means for high-precision orbit determination of low-Earth orbit satellites due to its advantages such as continuous observation, high accuracy, low cost and power consumption, small volume and light weight. Through post-processing, the current precise orbit determination of low-Earth orbit satellites based on GNSS can reach the centimeter level, but it highly depends on post-precise ephemeris and precise clock error products, so the timeliness is not high.

[0003] Since the completion and opening of the Beidou-3 (BDS-3) system in July 2020, in addition to providing high-quality and reliable positioning, navigation, and timing services for global users, it also provides free static centimeter-level and dynamic decimeter-level precise point positioning (PPP) services for users. Among them, the PPP-B2b signal, as a data dissemination channel, broadcasts enhanced information such as precise orbits and clock error corrections of BDS-3 and other GNSS satellites through the Beidou-3 GEO satellite to provide services for users.

[0004] However, BDS-3 PPP-B2b is currently mainly used for low-dynamic navigation and positioning application scenarios in the Asia-Pacific region. For high-dynamic low-Earth orbit satellite scenarios, there is only regional GNSS enhancement information, which poses certain challenges to the autonomous orbit determination data processing method. In addition, due to scientific mission requirements, low-Earth orbit satellites inevitably have orbit maneuvers, and traditional simplified dynamics autonomous orbit determination methods cannot guarantee the accuracy and continuity of real-time orbits and clock errors during orbit maneuvers.

[0005] Currently, the existing technologies mainly focus on improving the orbit accuracy of low-Earth orbit satellites, etc., and the research on real-time clock error estimation for low-Earth orbit satellites is relatively scarce. Therefore, there is an urgent need for a data processing method for on-orbit real-time autonomous orbit determination and clock error estimation of spaceborne GNSS for low-Earth orbit satellites that is applicable to only regional enhancement information and is not affected by orbit maneuvers. Summary of the Invention

[0006] Aiming at the problems existing in the prior art, the present invention provides a method and system for determining real-time precise orbits and clock errors in orbit for LEO based on PPP-B2b augmentation. The present invention can be used in the situation supported by only regional GNSS enhancement information and is less affected by orbit maneuvers.

[0007] To solve the above technical problems, the present invention adopts the following technical solutions:

[0008] Method for determining real-time precise orbit and clock offset of LEO in orbit based on PPP-B2b enhancement, comprising the following steps:

[0009] (1) Receive GNSS navigation signals and PPP-B2b enhancement signals through a GNSS receiving antenna;

[0010] (2) Filter, amplify, down-convert and sample the GNSS navigation signals and PPP-B2b enhancement signals;

[0011] (3) Realize acquisition and tracking of digital intermediate frequency signals through baseband signal processing, output GNSS raw observation data, including pseudorange, carrier phase, Doppler data, output GNSS raw navigation message and PPP-B2b message;

[0012] (4) Parse GNSS broadcast ephemeris and information of orbit and clock offset corrections PPP-B2b SSR from GNSS raw navigation message and PPP-B2b message respectively, match the PPP-B2b SSR with the GNSS broadcast ephemeris, and then correct the orbit and clock offset calculated by the GNSS broadcast ephemeris to obtain real-time precise ephemeris;

[0013] (5) Complete filtering estimation of the precise orbit in orbit in real time to obtain the three-dimensional position and velocity of the LEO satellite in the Earth-centered Earth-fixed coordinate system, and the undifferenced floating-point ambiguity parameters of each satellite;

[0014] (6) Complete estimation of the precise clock offset in orbit in real time;

[0015] (7) Complete ultra-short-term prediction of the precise orbit and clock offset in orbit in real time.

[0016] Further, the specific method of step (4) is:

[0017] Parse the PPP-B2b message to obtain satellite mask, satellite orbit correction, and satellite clock offset correction;

[0018] Parse the GNSS raw navigation message to obtain the GNSS broadcast ephemeris;

[0019] Use the SSR version number to synchronously match the satellite mask, satellite orbit correction, and satellite clock offset correction, use the mask version number to synchronously match the satellite mask and satellite clock offset correction, use the mask information to parse the satellite number of the satellite clock offset correction, use the correction version number to synchronously match the satellite orbit correction and satellite clock offset correction, and use the IODN in the satellite orbit correction to match the GNSS broadcast ephemeris;

[0020] After the matching is completed, real-time precise orbits and clock offsets are calculated using satellite orbit correction numbers and satellite clock offset correction numbers to obtain real-time precise ephemerides.

[0021] Further, the specific manner of step (5) is as follows:

[0022] (501) Run two orbit determination filters based on kinematic orbit determination and simplified dynamic orbit determination simultaneously; among them, the filtering state variables of the two orbit determination filters based on kinematic orbit determination and simplified dynamic orbit determination are:

[0023]

[0024] In the formula, and are the filtering state variables based on kinematic orbit determination and simplified dynamic orbit determination respectively, and are the three-dimensional position and velocity of the low-Earth orbit satellite at time and are the BDS-3 / GPS receiver clock offset parameters respectively, is the receiver clock rate parameter, is the atmospheric drag coefficient, is the solar radiation pressure coefficient, are the empirical accelerations in the radial, tangential, and normal directions of the orbit, The th elements in the two arrays are the ambiguity parameters to be estimated for the m BDS-3 satellites and n GPS satellites observed by the low-Earth orbit satellite at time The th elements in the two arrays are the ephemeris errors of the m BDS-3 satellites and n GPS satellites observed by the low-Earth orbit satellite along their respective signal propagation paths at time

[0025] According to whether there is PPP-B2b enhancement information support, the ephemeris error along the BDS-3 / GPS signal propagation path is modeled using a piecewise random walk process;

[0026] (502) At time update, based on kinematic orbit determination, the carrier phase time difference method is used to make a one-step prediction of the satellite position, and based on simplified dynamic orbit determination, the low-Earth orbit satellite orbit dynamics method is used to make a one-step prediction of the satellite position;

[0027] (503) When there are new BDS-3 / GPS tracking satellites, before measurement update, the epoch-to-epoch clock error variation obtained by solving the carrier phase time difference method is used as a virtual observation to constrain the two orbit determination filtering clock error parameters, and the on-orbit real-time orbit determination filtering model with additional TDCP clock error constraint is as follows:

[0028] (7)

[0029] Wherein, and are the Kalman filter state variables and observation variables respectively, and are the state transition matrix and the observation matrix respectively, and are the system noise and the observation noise respectively, is from the time to the time and are respectively at the time and the BDS-3 / GPS receiver clock error parameters in the filtering state variables at the time

[0030] The extended Kalman filter is used to perform measurement update on Equation (7) to obtain the precise orbit of the low-earth orbit satellite and the undifferenced floating-point ambiguity, so as to decouple and separate the receiver clock error and the undifferenced floating-point ambiguity parameters on orbit in real time;

[0031] (504) After obtaining the results of kinematic orbit determination and simplified dynamic orbit determination, when it is detected and judged that the low-earth orbit satellite has an orbit maneuver, the result of kinematic orbit determination is output, otherwise the result of simplified dynamic orbit determination is output.

[0032] Further, the specific manner of step (6) is:

[0033] (601) Take the three-dimensional position, velocity of the low-earth orbit satellite in the earth-centered earth-fixed coordinate system and the undifferenced floating-point ambiguity parameters of each satellite that have been decoupled from the receiver clock error obtained in step (5) as known values, substitute them into the carrier phase observation equation, and solve the precise clock error of the low-earth orbit satellite receiver;

[0034] Among them, the current epoch time low-earth orbit satellite BDS-3 / GPS receiver clock error calculated from the ionosphere-free combined phase observations of the th BDS-3 satellite and the th GPS satellite is expressed as:

[0035] (8)

[0036] In the formula, and are respectively the ionosphere-free combined phase observation values of the th BDS-3 and the th GPS satellite at the current epoch, and and are the BDS-3 / GPS clock offset parameters of the low-orbit satellite calculated therefrom; and are respectively the clock offsets of the th BDS-3 and the th GPS satellite at the signal transmission time; and are respectively the undifferenced floating ambiguities of the th BDS-3 and the th GPS satellite; and are respectively the phase winding error corrections of the th BDS-3 and the th GPS satellite; and are respectively the relativistic error corrections of the th BDS-3 and the th GPS satellite; and are respectively the geometric distances between the phase centers of the transmitting and receiving antennas of the th BDS-3 and the th GPS satellite, and there are:

[0037] , ,

[0038] , , are respectively the three-dimensional positions of the low-orbit satellite, the th BDS-3 and the th GPS satellite in the Earth-centered Earth-fixed coordinate system at the current epoch;

[0039] (602) Calculate the mean and variance of the precise receiver clock offset sequence, construct tolerance conditions with the mean and variance. When the precise receiver clock offset of a certain channel exceeds the tolerance, it is considered that there is a gross error in the carrier phase observation value of this channel and it does not participate in the final calculation of the precise receiver clock offset;

[0040] If the low-orbit satellite has received a total of BDS-3 satellites and GPS satellites at the current time, then calculate Precise clock offset sequences of LEO satellites of BDS-3 at the current moment and Precise clock offset sequences of LEO satellites of GPS at the current moment , where \(i = 1,\cdots,m\), \(j = 1,\cdots,n\); then, the mean value and standard deviation of the precise clock offset sequences of LEO satellites are calculated by the following formula:

[0041] (9)

[0042] where and are the mean values of the precise clock offset sequences of LEO satellites of BDS-3 / GPS respectively, and are the standard deviations of the precise clock offset sequences of LEO satellites of BDS-3 / GPS respectively;

[0043] (603) After excluding the channels with gross errors exceeding the tolerance limit, recalculate the mean value of the precise clock offset sequence of the receiver composed of all channels with qualified tolerance, and take this mean value as the final precise clock offset of the BDS-3 / GPS receiver.

[0044] Furthermore, in step (7), for the precise orbit, dynamic orbit integration is used for prediction; for the precise clock offset, a two-dimensional model of clock offset - clock rate is used for prediction, and the time interval of ultra-short-term prediction is set to 1 min.

[0045] The LEO on-orbit real-time precise orbit and clock offset determination system based on PPP-B2b enhancement is used to implement the above-mentioned LEO on-orbit real-time precise orbit and clock offset determination method based on PPP-B2b enhancement, and includes:

[0046] A GNSS receiving antenna (1), which is used to receive GNSS navigation signals and PPP-B2b enhancement signals;

[0047] A radio frequency front-end module (2), which is used to filter, amplify, down-convert and sample GNSS navigation signals and PPP-B2b enhancement signals;

[0048] A baseband signal processing module (3), which is used to realize the acquisition and tracking of digital intermediate frequency signals, output GNSS raw observation data, including pseudorange, carrier phase, Doppler data, output GNSS raw navigation messages and PPP-B2b messages;

[0049] The real-time precise ephemeris calculation module (4) respectively parses the GNSS broadcast ephemeris and the information of orbit and clock bias correction PPP-B2b SSR from the GNSS raw navigation message and the PPP-B2b message, matches the PPP-B2b SSR with the GNSS broadcast ephemeris, and then corrects the orbit and clock bias calculated from the GNSS broadcast ephemeris to obtain the real-time precise ephemeris;

[0050] The real-time precise orbit estimation module (5) is used to perform filtering estimation of the precise orbit in orbit in real time, obtain the three-dimensional position and velocity of the low-earth orbit satellite in the geocentric-earth-fixed coordinate system, and the undifferenced floating-point ambiguity parameter of each satellite;

[0051] The real-time precise clock bias estimation module (6) is used to perform estimation of the precise clock bias in orbit in real time;

[0052] The precise clock bias prediction module (7) is used to perform ultra-short-term prediction of the precise orbit and clock bias in orbit in real time.

[0053] The beneficial effects of the present invention compared with the prior art are as follows:

[0054] (1) In the present invention, the ephemeris error along the BDS-3 / GPS signal propagation path is considered in the orbit determination filter. This parameter adopts a piecewise random walk model, which is suitable for the situation where the low-earth orbit satellite can only obtain regional augmentation information.

[0055] (2) By comparing the differences between the two results of kinematic orbit determination (KOD) and reduced-dynamic orbit determination (RDOD), the present invention detects and judges whether the low-earth orbit satellite has an orbit maneuver, and then the optimal output of the orbit determination result is not affected by the orbit maneuver.

[0056] (3) The present invention takes the low-earth orbit satellite orbit and carrier phase ambiguity as known values, inversely calculates the precise receiver clock bias through the carrier phase observation equation, and takes into account the influence of channel gross errors on the final receiver precise clock bias.

[0057] (4) The present invention uses only the Beidou-3 PPP-B2b augmentation information in the region, which can improve the orbit determination and clock bias estimation accuracy of low-earth orbit satellites in the Asia-Pacific region. At the same time, the method of the present invention can also be extended to the application of augmentation information broadcast by other navigation systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a schematic diagram of the principle of the LEO on-orbit real-time precise orbit and clock bias determination system based on PPP-B2b augmentation in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0059] The following further describes the present invention in detail with specific embodiments, but the implementation manners of the present invention are not limited thereto.

[0060] A method for determining real-time precise orbit and clock offset of LEO in orbit based on PPP-B2b enhancement is applied to low-orbit satellites. The enhanced information broadcast by the B2b signal of Beidou-3 is used to correct the orbits and clock offsets of BDS-3 / GPS satellites. With the support of regional BDS-3 / GPS enhanced information, when there are new BDS-3 / GPS tracking satellites, before measurement update, the clock offset change amount between epochs obtained by using the Time-Differenced Carrier Phase (TDCP) method is used as two orbit determination filtering clock offset parameters for constraint, and the receiver clock offset and undifferenced floating-point ambiguity parameters are separated in orbit in real time. Finally, the precise orbit of LEO and the undifferenced floating-point ambiguity parameters output by the orbit determination filtering are substituted into the phase observation equation to calculate the precise clock offset of the BDS-3 / GPS receiver.

[0061] Specifically, the method includes the following steps:

[0062] (1) Receive BDS-3 / GPS navigation signals and PPP-B2b enhancement signals through a GNSS receiving antenna;

[0063] (2) The RF front-end module completes the filtering, amplification, down-conversion and sampling (A / D conversion) of the BDS-3 / GPS navigation signals and PPP-B2b enhancement signals;

[0064] (3) The baseband signal processing module completes the acquisition and tracking of the digital intermediate frequency signals, outputs the original BDS-3 / GPS observation data, including pseudorange, carrier phase, Doppler, etc., and outputs the original BDS-3 / GPS navigation message and PPP-B2b message;

[0065] (4) Through the real-time precise ephemeris calculation module, parse the BDS-3 / GPS broadcast ephemeris, orbit and clock offset correction information PPP-B2b SSR from the BDS-3 / GPS original navigation message and PPP-B2b message respectively, match the PPP-B2b SSR with the GPS / BDS-3 broadcast ephemeris, and then correct the orbits and clock offsets calculated by the BDS-3 / GPS broadcast ephemeris to obtain the real-time precise ephemeris.

[0066] When parsing the PPP-B2b message, focus on parsing the information type 1 (satellite mask), information type 2 (satellite orbit correction), and information type 4 (satellite clock offset correction) in the PPP-B2b message.

[0067] When matching the PPP-B2b SSR with the BDS-3 / GPS broadcast ephemeris, the SSR issue number (IOD SSR) is used to synchronize the matching information types 1, 2, and 4, the mask issue number (IODP) is used to synchronize the matching information types 1 and 4, and the satellite numbers of information type 4 are resolved using the masked information. The correction issue number (IOD Corr) is used to synchronize the matching information types 2 and 4, and the IODN in information type 2 is used to match the BDS-3 / GPS broadcast ephemeris.

[0068] After the matching is completed, the real-time precise orbit and clock offset are calculated using information type 2 (satellite orbit correction) and information type 4 (satellite clock offset correction). The specific calculation process refers to the PPP-B2b ICD document for precise point positioning service signals and will not be elaborated here.

[0069] (5) The filtering estimation of the precise orbit is completed in orbit in real time through the real-time precise orbit estimation module. The specific method is as follows:

[0070] (501) Two orbit determination filters based on kinematic orbit determination (KOD) and reduced-dynamic orbit determination (RDOD) are run simultaneously. The orbit determination filtering state variables include the ephemeris error along the BDS-3 / GPS signal propagation path. This ephemeris error is modeled using a piecewise random walk model, and its random process noise is set according to the extreme value range of the change rate of this ephemeris error.

[0071] Among them, the filtering state variables of the two orbit determination filters based on kinematic orbit determination and reduced-dynamic orbit determination are selected as:

[0072] (1)

[0073] In the formula, and are the filtering state variables based on kinematic orbit determination (KOD) and reduced-dynamic orbit determination (RDOD) respectively, and are the three-dimensional position and velocity of the LEO satellite at time and are the BDS-3 / GPS receiver clock offset parameters respectively, is the receiver clock rate parameter; is the atmospheric drag coefficient, is the solar radiation pressure coefficient, is the empirical acceleration in the radial (R), tangential (T), and normal (N) directions of the orbit; are respectively The ambiguity parameters to be estimated for the m BDS-3 satellites and n GPS satellites observed by the LEO satellite at a certain moment are respectively the ephemeris errors of the m BDS-3 satellites and n GPS satellites observed by the LEO satellite along their respective signal propagation paths at a certain moment.

[0074] The ephemeris error along the BDS-3 / GPS signal propagation path can be calculated by the following formula:

[0075] (2)

[0076] where is the three-dimensional position of the LEO satellite in the Earth-centered Earth-fixed coordinate system; is the three-dimensional position and clock offset of the real BDS-3 / GPS satellite. Usually, the three-dimensional position and clock offset of the BDS-3 / GPS satellite calculated from the final precise ephemeris and clock products released by IGS can be used as the reference true value; is the three-dimensional position and clock offset of the BDS-3 / GPS satellite calculated from the BDS-3 / GPS broadcast ephemeris or after PPP-B2b SSR correction, and c is the speed of light;

[0077] The ephemeris error along the BDS-3 / GPS signal propagation path has been proven to be able to be modeled using a random walk process:

[0078] (3)

[0079] where is the three-dimensional position of the LEO satellite in the Earth-centered Earth-fixed coordinate system; is the three-dimensional position and clock offset of the real BDS-3 / GPS satellite. Usually, the three-dimensional position and clock offset of the BDS-3 / GPS satellite calculated from the final precise ephemeris and clock products released by IGS can be used as the reference true value; is the three-dimensional position and clock offset of the BDS-3 / GPS satellite calculated from the BDS-3 / GPS broadcast ephemeris or after PPP-B2b SSR correction;

[0080] and are respectively the moment and the ephemeris error along the BDS-3 / GPS signal propagation path at the moment, is the epoch time interval, is a white noise sequence with zero mean, represents the variance of the variable, is the process noise variance of White noise sequences with zero mean respectively The power spectral density of, that is, the random process noise.

[0081] When the satellite is continuously and normally tracked, The initial variance of Can be set according to The maximum value of, and its random process noise And The power spectral density of is related. Generally, it can also be roughly set according to The rate of change of. By setting an appropriate random model, make Can absorb part of the ephemeris error along the BDS-3 / GPS signal propagation path, and the corresponding real-time orbit determination accuracy will also be greatly improved.

[0082] Since PPP-B2b is restricted by the regional tracking network, it can only be used in the Asia-Pacific region at present. For low-Earth orbit satellites, only when its sub-satellite point is located in the Asia-Pacific region can it receive PPP-B2b enhancement information that matches the GPS / BDS-3 observation values. In other regions, due to the lack of PPP-B2b enhancement information, only BDS-3 / GPS broadcast ephemeris can be used. Therefore, further, according to different situations with or without PPP-B2b enhancement information support, the ephemeris error along the BDS-3 / GPS signal propagation path Is modeled using a piecewise random walk process, that is In the case of having regional PPP-B2b enhancement information support (without regional PPP-B2b enhancement information support, denoted as working condition 1), its random process noise is In the case of only having BDS-3 / GPS broadcast ephemeris (without regional PPP-B2b enhancement information support, denoted as working condition 0), its random process noise is . Among them, And The specific values are respectively based on The rate of change of in working condition 1 and working condition 0 is roughly set. For example, if the rate of change ranges between plus and minus 1, the corresponding random process noise is set to 1.

[0083] (502) During time update, based on kinematic orbit determination (KOD), the carrier phase time difference method is used to perform a one-step prediction of the satellite position, and based on reduced dynamics orbit determination (RDOD), the low-Earth orbit satellite orbit dynamics method is used to perform a one-step prediction of the satellite position.

[0084] The carrier phase time difference method is to perform epoch-to-epoch difference on the phase observations of the previous and subsequent epochs when there is no cycle slip in the carrier phase observations, and the epoch-to-epoch carrier phase difference observations can be obtained:

[0085] (4)

[0086] In the formula, is the carrier phase differential observation value between epochs; are respectively the carrier phase non-differential observation values at time and time is the change in geometric distance between epochs, which is related to the LEO position and the BDS-3 / GPS navigation star position between epochs; and are respectively the changes in receiver clock error and navigation star clock error between epochs; is the change in phase winding correction between epochs; is the change in relativistic effect correction between epochs; is the multipath and observation noise change; within a very short observation time interval, the errors such as the change in BDS-3 / GPS satellite position / clock error between epochs calculated from BDS-3 / GPS ephemeris (broadcast ephemeris or its PPP-B2b corrected real-time precise ephemeris) are of the same order of magnitude as the phase noise (about mm level), so they can basically be ignored.

[0087] Taking a single BDS-3 / GPS system as an example, if the number of continuously observed BDS-3 / GPS satellites reaches more than 4, let the change in LEO satellite position / clock error between epochs be , where are the components of the change in LEO satellite position between epochs in the three axes of the Earth-centered Earth-fixed coordinate system, are respectively the changes in BDS-3 / GPS receiver clock error between epochs.

[0088] Then the least squares method can be used to solve it, which can be expressed as:

[0089] (5)

[0090] In the formula, is the observation coefficient matrix of epoch difference; is the residual vector of epoch difference observation value; is the weight matrix of epoch difference observation value.

[0091] Based on kinematic orbit determination (KOD), the TDCP method is used to make a one-step prediction of the satellite position, that is, there is

[0092] (6)

[0093] In the formula, are respectively the components of the LEO satellite in the three axes of the Earth-centered Earth-fixed coordinate system at time are respectively The components of the LEO satellite on the three axes in the Earth-Centered Earth-Fixed (ECEF) coordinate system at a certain moment.

[0094] Based on the Reduced Dynamics Orbit Determination (RTOD), during the time update, the satellite position is predicted using dynamic orbit integration, which is well-known in the industry and will not be elaborated here.

[0095] (503) Before the measurement update, the change in clock bias between epochs obtained by solving using the Time-Differenced Carrier Phase (TDCP) method is used as a virtual observation to constrain the two orbit determination filtering clock bias parameters, and the receiver clock bias and undifferenced floating ambiguity parameters are separated in real-time in orbit.

[0096] Because the Time-Differenced Carrier Phase (TDCP) method eliminates the ambiguity parameter by differencing the carrier phase observations of two consecutive epochs, and the change in position and clock bias between epochs with high precision can be obtained through least squares estimation, the solved change in clock bias is independent of the ambiguity parameter and directly reflects the actual change of the receiver clock. The solution accuracy of the change in clock bias is related to factors such as the noise of the carrier phase observation, multipath error, BDS-3 / GPS real-time ephemeris error, and satellite geometric distribution, and usually an accuracy of cm or even mm level can be obtained.

[0097] To decouple the LEO phase clock bias and the ambiguity parameter, based on the real-time precise orbit determination filtering model, when the newly tracked on-board BDS-3 / GPS observations participate in the orbit determination filtering, the change in clock bias between epochs solved by TDCP is used as a constraint equation to participate in the filtering solution. Therefore, the on-orbit real-time orbit determination filtering model with additional TDCP clock bias constraint is as follows:

[0098] (7)

[0099] In the formula, and are the state variables and observation variables of the Kalman filter respectively; and are the state transition matrix and the observation matrix respectively; and are the system noise and the observation noise respectively; is the change in the BDS-3 / GPS receiver clock bias between epochs solved by the TDCP method from the moment to the moment by formula (5); and are the BDS-3 / GPS receiver clock bias parameters in the filtering state variables at the moment and the

[0100] The measurement update of Equation (7) is performed using the Extended Kalman Filter (EKF) to obtain the precise orbit of the LEO and the undifferenced float ambiguity.

[0101] (504) After obtaining the results based on Kinematic Orbit Determination (KOD) and Reduced Dynamics Orbit Determination (RDOD), when it is detected and determined that the LEO satellite has an orbit maneuver, the result of Kinematic Orbit Determination (KOD) is output; otherwise, the result based on Reduced Dynamics Orbit Determination (RDOD) is output.

[0102] (6) The estimation of the precise clock offset is completed in real-time on orbit through the real-time precise clock offset estimation module. The specific method is as follows:

[0103] (601) The three-dimensional position and velocity of the LEO satellite in the Earth-Centered Earth-Fixed (ECEF) coordinate system and the undifferenced float ambiguity parameter of each satellite output in step (5) are used as known values, and substituted into the carrier phase observation equation to solve the precise clock offset of the LEO satellite receiver.

[0104] From the th BDS-3 satellite and the th GPS satellite, the clock offset of the LEO satellite BDS-3 / GPS receiver at the current epoch calculated from the ionosphere-free combined phase observations can be expressed as:

[0105] (8)

[0106] In Equation (8), and are respectively the clock offset parameters of the LEO satellite BDS-3 / GPS calculated from the ionosphere-free combined phase observations of the th BDS-3 and the th GPS satellite at the current epoch; and are respectively the clock offsets of the th BDS-3 and the th GPS satellite at the signal transmission time; and are respectively the geometric distances between the phase centers of the transmitting and receiving antennas of the , , and th BDS-3 / th GPS satellite; are respectively the three-dimensional positions of the LEO satellite and the th BDS-3 / th GPS satellite in the Earth-Centered Earth-Fixed coordinate system at the current epoch; and are respectively the th BDS-3 / th GPS satellite; The undifferenced float ambiguities of BDS-3 / GPS satellites are obtained by filtering and estimation in step (5); and are the phase winding error corrections of the th BDS-3 / th GPS satellite respectively; and are the relativistic error corrections of the th BDS-3 / th GPS satellite respectively.

[0107] (602) Calculate the mean and variance of the receiver precise clock error sequence obtained above, construct the tolerance conditions with the mean and variance. When the receiver precise clock error of a certain channel exceeds the tolerance, it is considered that there is a gross error in the carrier phase observation value of this channel, so it does not participate in the final calculation of the receiver precise clock error.

[0108] If a total of BDS-3 satellites and GPS satellites are received by the LEO satellite at the current moment, then the BDS-3 precise clock error sequences of the LEO satellite at the current moment can be calculated by formula (8), and GPS precise clock error sequences of the LEO satellite at the current moment. Then, the mean value and standard deviation of the LEO satellite precise clock error sequence can be calculated by the following formula:

[0109] (9)

[0110] where and are the mean values of the BDS-3 / GPS precise clock error sequences of the LEO satellite respectively, and are the standard deviations of the BDS-3 / GPS precise clock error sequences of the LEO satellite respectively. Taking BDS-3 as an example, when the precise clock error of the LEO satellite calculated by a certain BDS-3 satellite does not satisfy , then this BDS-3 satellite does not participate in the final solution of the LEO satellite precise clock error.

[0111] (603) After excluding the channels with gross errors exceeding the tolerance above, recalculate the mean value of the receiver precise clock error sequence composed of all channels with qualified tolerance, and take this mean value as the final BDS-3 / GPS receiver precise clock error.

[0112] (7) The ultra-short-term prediction of precise orbit and clock bias is completed in orbit in real time through the precise orbit and clock bias prediction module. Specifically, for the precise orbit, dynamic orbit integration is used for prediction, and for the precise clock bias, a two-dimensional model of clock bias and clock rate is used for prediction. The time interval of the ultra-short-term prediction is set to 1 minute.

[0113] To implement the above method, a LEO on-orbit real-time precise orbit and clock determination system based on PPP-B2b enhancement as shown in Figure 1 can be adopted, which includes:

[0114] A GNSS receiving antenna (1) for receiving GNSS navigation signals and PPP-B2b enhancement signals;

[0115] A radio frequency front-end module (2) for filtering, amplifying, down-converting and sampling GNSS navigation signals and PPP-B2b enhancement signals;

[0116] A baseband signal processing module (3) for realizing the acquisition and tracking of digital intermediate frequency signals, outputting GNSS raw observation data, including pseudorange, carrier phase, Doppler data, and outputting GNSS raw navigation messages and PPP-B2b messages;

[0117] A real-time precise ephemeris calculation module (4) for parsing GNSS broadcast ephemeris and information on orbit and clock bias corrections PPP-B2b SSR from GNSS raw navigation messages and PPP-B2b messages respectively, matching the PPP-B2b SSR with the GNSS broadcast ephemeris, and then correcting the orbit and clock bias calculated from the GNSS broadcast ephemeris to obtain a real-time precise ephemeris;

[0118] A real-time precise orbit estimation module (5) for completing the filtering estimation of the precise orbit in orbit in real time to obtain the three-dimensional position and velocity of the LEO satellite in the Earth-centered Earth-fixed coordinate system, as well as the undifferenced floating ambiguity parameters of each satellite;

[0119] A real-time precise clock bias estimation module (6) for completing the estimation of the precise clock bias in orbit in real time;

[0120] A precise clock bias prediction module (7) for completing the ultra-short-term prediction of the precise orbit and clock bias in orbit in real time.

[0121] In summary, the present invention corrects the orbits and clock biases of BDS-3 / GPS satellites by using the enhanced information broadcast by the BDS-3 B2b signal. With the support of regional BDS-3 / GPS enhanced information, when there are new BDS-3 / GPS tracking satellites, before measurement update, the change in clock bias between epochs obtained by using the carrier-phase time-differencing method is used as two orbit determination filtering clock bias parameters for constraint, and the receiver clock bias and undifferenced floating ambiguity parameters are separated in real time in orbit. Finally, the precise orbit of the LEO and the undifferenced floating ambiguity parameters output by the orbit determination filtering are substituted into the phase observation equation to calculate the precise clock bias of the BDS-3 / GPS receiver. The present invention is applicable to the situation where there is only regional GNSS enhanced information support and is less affected by orbit maneuvers.

[0122] The above are only the embodiments of the present invention, and do not limit the protection scope of the present invention. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be included in the protection scope of the present invention.

Claims

1. A method for determining LEO on-orbit real-time precise orbit and clock errors based on PPP-B2b enhancement, characterized in that: The steps include: (1) Receive GNSS navigation signals and PPP-B2b enhancement signals through the GNSS receiving antenna; (2) Filter, amplify, down-convert and sample GNSS navigation signals and PPP-B2b enhancement signals; (3) Capture and track digital intermediate frequency signals through baseband signal processing, output GNSS raw observation data, including pseudorange, carrier phase, Doppler data, and output GNSS raw navigation messages and PPP-B2b messages; (4) Parse the GNSS broadcast ephemeris and orbit and clock correction information PPP-B2b SSR from the GNSS original navigation message and PPP-B2b message respectively, match the PPP-B2b SSR with the GNSS broadcast ephemeris, and then correct the orbit and clock error calculated by the GNSS broadcast ephemeris to obtain real-time precise ephemeris; (5) The precise orbit is estimated by filtering in real time on-orbit, and the three-dimensional position and velocity of the low-orbit satellite in the Earth-centered Earth-fixed system, as well as the undifferenced floating-point ambiguity parameters of each satellite, are obtained. The specific method is as follows: (501) Two orbit determination filters based on kinematic orbit determination and based on simplified dynamic orbit determination are run simultaneously; wherein the filter state quantities of the two orbit determination filters based on kinematic orbit determination and based on simplified dynamic orbit determination are: In the formula, and are the filter state quantities based on kinematic orbit determination and simplified dynamic orbit determination, respectively. and for The three-dimensional position and velocity of the low-orbit satellite at any moment; and They are BDS-3 / GPS receiver clock error parameters, is the receiver clock rate parameter, is the atmospheric drag coefficient, is the solar pressure coefficient, are the empirical accelerations in the radial, tangential and normal directions of the orbit, The first of the two arrays The elements are The ambiguity parameters to be estimated for m BDS-3 satellites and n GPS satellites observed by the low-orbit satellite at the moment, The first of the two arrays The elements are The ephemeris errors of m BDS-3 satellites and n GPS satellites observed by low-orbit satellites at the moment along their respective signal propagation paths; Depending on whether there is PPP-B2b augmentation information support, the ephemeris error along the BDS-3 / GPS signal propagation path is The piecewise random walk process is used for modeling; (502) When updating the time, the satellite position is predicted in one step using the carrier phase time difference method based on the kinematic orbit determination, and the satellite position is predicted in one step using the low-orbit satellite orbit dynamics method based on the simplified dynamic orbit determination; (503) When there is a new BDS-3 / GPS tracking satellite, before the measurement update, the inter-epoch clock error variation obtained by the carrier phase time difference method is used as a virtual observation value to constrain the two orbit determination filter clock error parameters. The on-orbit real-time orbit determination filter model with additional TDCP clock error constraints is obtained as follows: （7） In the formula, and are the Kalman filter state and observation, respectively. and are the state transfer matrix and the observation matrix respectively, and are system noise and observation noise respectively, for Time has come The change in BDS-3 / GPS receiver clock error between epochs calculated by the carrier phase time difference method; and They are Moment and BDS-3 / GPS receiver clock error parameters in the moment filter state; The extended Kalman filter is used to measure and update equation (7) to obtain the precise orbit and non-differenced floating-point ambiguity of the low-orbit satellite, thereby decoupling and separating the receiver clock error and non-differenced floating-point ambiguity parameters in real time on orbit; (504) After obtaining the results of orbit determination based on kinematics and orbit determination based on simplified dynamics, when it is detected and determined that the low-orbit satellite has an orbital maneuver, the result of kinematics orbit determination is output; otherwise, the result of simplified dynamics orbit determination is output; (6) Estimating precise clock errors in real time on orbit; (7) Real-time ultra-short-term prediction of precise orbit and clock error on-orbit.

2. The method for determining LEO on-orbit real-time precise orbit and clock error based on PPP-B2b enhancement according to claim 1 is characterized in that: The specific method of step (4) is: Parse the PPP-B2b message to obtain the satellite mask, satellite orbit correction number, and satellite clock correction number; Parse the GNSS original navigation message to obtain the GNSS broadcast ephemeris; Use the SSR version number to synchronize and match the satellite mask, satellite orbit correction number and satellite clock correction number, use the mask version number to synchronize and match the satellite mask and satellite clock correction number, and use the mask information to parse the satellite number of the satellite clock correction number, use the correction number version number to synchronize and match the satellite orbit correction number and satellite clock correction number, and use the IODN in the satellite orbit correction number to match the GNSS broadcast ephemeris; After the matching is completed, the satellite orbit correction number and satellite clock correction number are used to calculate the real-time precise orbit and clock error, and obtain the real-time precise ephemeris.

3. The method for determining LEO on-orbit real-time precise orbit and clock error based on PPP-B2b enhancement according to claim 1 is characterized in that: The specific method of step (6) is: (601) Substituting the three-dimensional position and velocity of the low-orbit satellite in the Earth-centered Earth-fixed system obtained in step (5) and the undifferenced floating-point ambiguity parameters of each satellite that have been decoupled from the receiver clock error as known values ​​into the carrier phase observation equation to solve the precise clock error of the low-orbit satellite receiver; Among them, the BDS-3 satellite and The clock error of the low-orbit satellite BDS-3 / GPS receiver at the current epoch calculated from the ionosphere-free combined phase observations of the GPS satellites is expressed as: （8） In the formula, and The current epoch time is BDS-3, Ionosphere-free combined phase observations of GPS satellites and Calculated clock error parameters of low-orbit satellite BDS-3 / GPS; and The signal transmission time is BDS-3, GPS satellite clock error; and Respectively BDS-3, Undifferenced floating point ambiguities of GPS satellites; and Respectively BDS-3, Phase winding error correction of GPS satellites; and Respectively BDS-3, Relativistic error correction of GPS satellites; and Respectively BDS-3, The geometric distance between the phase center of the transmitting antenna and the phase center of the receiving antenna of a GPS satellite, and: ， ， , , They are the low-orbit satellite at the current epoch, BDS-3, The three-dimensional position of the GPS satellites in the Earth-centered Earth-fixed coordinate system; (602) Calculate the mean and variance of the receiver precision clock error sequence, and construct the error limit condition with the mean and variance. When the receiver precision clock error of a channel exceeds the error limit, it is considered that the carrier phase observation value of this channel has a gross error and does not participate in the final receiver precision clock error calculation; If the current low-orbit satellite receives a total of BDS-3 and GPS satellites, then the formula (8) is used to calculate The current low-orbit satellite BDS-3 precise clock error sequence and The current low-orbit satellite GPS precise clock error sequence , i=1,...,m, j=1,...,n; then, the mean value and standard deviation of the low-orbit satellite precise clock error sequence are calculated by the following formula: （9） in, and are the average values ​​of the BDS-3 / GPS precise clock error sequence of the low-orbit satellite, and are the standard deviations of the BDS-3 / GPS precision clock error series of low-orbit satellites; (603) After excluding the channels with gross errors that exceed the limit, recalculate the mean of the receiver precision clock error sequence composed of all channels with qualified limits, and use the mean as the final BDS-3 / GPS receiver precision clock error.

4. The method for determining LEO on-orbit real-time precise orbit and clock error based on PPP-B2b enhancement according to claim 1 is characterized in that: In step (7), for the precise orbit, the dynamical orbit integral is used for prediction; for the precise clock error, the clock error-clock rate two-dimensional model is used for prediction, and the time interval for ultra-short-term prediction is set to 1 minute.

5. The LEO on-orbit real-time precise orbit and clock determination system based on PPP-B2b enhancement is characterized by: The method for determining the LEO on-orbit real-time precise orbit and clock error based on PPP-B2b enhancement as claimed in any one of claims 1 to 4 comprises: A GNSS receiving antenna (1), used for receiving GNSS navigation signals and PPP-B2b enhancement signals; A radio frequency front-end module (2) is used to filter, amplify, down-convert and sample GNSS navigation signals and PPP-B2b enhancement signals; The baseband signal processing module (3) is used to realize the capture and tracking of the digital intermediate frequency signal, output the original GNSS observation data, including pseudorange, carrier phase, Doppler data, and output the original GNSS navigation message and PPP-B2b message; The real-time precise ephemeris calculation module (4) parses the GNSS broadcast ephemeris and orbit and clock correction information PPP-B2b SSR from the GNSS original navigation message and the PPP-B2b message, matches the PPP-B2b SSR with the GNSS broadcast ephemeris, and then corrects the orbit and clock error calculated by the GNSS broadcast ephemeris to obtain the real-time precise ephemeris; A real-time precise orbit estimation module (5) is used to complete the filtering estimation of the precise orbit in real time on orbit, and obtain the three-dimensional position and velocity of the low-orbit satellite in the Earth-centered Earth-fixed system, as well as the undifferenced floating-point ambiguity parameters of each satellite; A real-time precise clock error estimation module (6), used to estimate the precise clock error in real time on orbit; The precise clock error prediction module (7) is used to complete the ultra-short-term prediction of precise orbit and clock error in real time on orbit.

Images