PPP-B2b-based real-time precise positioning methods, systems, and media for marine applications

By combining hybrid differential single-station Kalman filtering and signal-to-noise ratio stochastic model with PPP-B2b correction and broadcast ephemeris, an observation model is constructed, which solves the problem of low positioning accuracy of PPP-B2b at sea and realizes real-time high-precision positioning at sea, which is suitable for maritime emergency search and rescue and energy exploration.

CN117826200BActive Publication Date: 2026-11-17QINGDAO INST OF MARINE GEOLOGY +1
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202410032551.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-01-10
Publication Date
2026-11-17
Estimated Expiration
2044-01-10

AI Technical Summary

Technical Problem

PPP-B2b technology has low positioning accuracy at sea and is difficult to apply to real-time precision positioning. In particular, it suffers from severe multipath errors due to large ships and the mirror effect of water surface reflected signals, which affects positioning accuracy.

Method used

A hybrid differential single-station Kalman filter method is adopted, combined with PPP-B2b corrections and broadcast ephemeris, to construct a signal-to-noise ratio stochastic model for data preprocessing and error correction. A combined observation model of non-differential deionization and epoch-differential deionization is constructed for time and measurement updates to improve positioning accuracy.

Benefits of technology

It achieves real-time high-precision positioning at sea, provides stable and reliable positioning services, solves the problem of low positioning accuracy in PPP-B2b at sea, is suitable for environments without network communication, and has economic advantages.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117826200B_ABST
    Figure CN117826200B_ABST
Patent Text Reader

Abstract

This invention discloses a method, system, and medium for real-time precise positioning at sea based on PPP-B2b, belonging to the field of information technology services. First, using PPP-B2b corrections broadcast by BeiDou-3 GEO satellites and combined with broadcast ephemeris data, precise clock errors and orbits based on PPP-B2b corrections are recovered. Then, GNSS observation data is acquired, and for satellites with and without PPP-B2b corrections, non-differential anionospheric combined observation models and epoch-differential anionospheric combined observation models are constructed respectively. The PPP-B2b precise clock errors and orbits are applied to the non-differential anionospheric combined observation model, and a signal-to-noise ratio stochastic model is constructed to calculate satellite weights. Finally, time updates are performed based on the epoch-differential anionospheric combined observation model, and measurement updates are performed based on the non-differential anionospheric combined observation model. Hybrid differential single-station Kalman filtering is used for parameter estimation to complete real-time positioning at sea.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of information technology service technology, specifically relating to a method, system and medium for real-time precision positioning at sea based on PPP-B2b. Background Technology

[0002] With the rapid development of GNSS (Global Navigation Satellite System), high-precision GNSS positioning technology has been widely applied in fields such as maritime search and rescue and marine exploration. The high-precision positioning information it acquires is a fundamental prerequisite for all marine development activities. Currently, there are two main types of GNSS high-precision positioning technology: Real-Time Kinematic (RTK) and Precise Point Positioning (PPP). The PPP method can acquire high-precision positioning information using only a single receiver, making it easier to implement at sea compared to RTK technology. Currently, real-time PPP is commonly implemented using the IGS (International GNSS Service) RTS service, but the broadcasting of RTS signals relies on a network, making it difficult to apply at sea.

[0003] Currently, some international commercial companies have developed satellite broadcasting services that can provide users with real-time, high-precision positioning results even when network communication is difficult at sea. Current commercial high-precision positioning services mainly include the OmniStar system, NavCom's Starfire system, Trimble's RTX system, and Fugro's StarFix system. However, these commercial services often require expensive licensing fees, which is not conducive to low-cost PPP high-precision positioning applications for the general public. In 2020, BDS-3 released its PPP Enhanced Service, namely the PPP-B2b Special Service. This service broadcasts correction information via geostationary orbit satellites, providing free and publicly available high-precision positioning services to users in surrounding areas. Furthermore, the PPP-B2b service is not limited by network communication and has the potential to achieve low-cost, real-time, high-precision positioning in maritime areas.

[0004] For example, the invention patent with publication number [CN114280644A] discloses a precise point positioning system and method based on PPP-B2b service. First, it receives PPP-B2b signals broadcast by the BeiDou-3 system, along with observation data and broadcast ephemeris. The received PPP-B2b signals are decoded and preprocessed to obtain correction values. Then, the observation data is corrected based on the correction values, and the known terms in the observation data are solved using the ionosphere-free model. Finally, the corresponding precise point positioning result is output according to user requirements. The invention patent with publication number [CN115616615A] discloses a low-cost single-frequency GNSS receiver precise positioning method enhanced by PPP-B2b. It receives navigation signals from GPS / BDS-3 satellites through ground equipment, corrects the broadcast ephemeris using real-time enhanced correction values ​​from PPP-B2b broadcast by the BeiDou-3 GEO satellite, and completes real-time single-frequency precise point positioning in the ground equipment using a real-time single-frequency precise point positioning algorithm model employing PPP-B2b correction values.

[0005] However, in the marine environment, large ships and the water surface will generate large reflection signals. The specular reflection formed by these smooth surfaces will produce a larger multipath error than the diffuse reflection formed by rough surfaces, thus reducing the positioning accuracy of PPP-B2b technology at sea. Therefore, PPP-B2b technology has not yet been applied to real-time precision positioning at sea. Summary of the Invention

[0006] This invention addresses the problem of low positioning accuracy and difficulty in applying PPP-B2b services at sea. It proposes a real-time precise positioning method, system, and medium based on PPP-B2b at sea. After calculating satellite weights using a random signal-to-noise ratio model, a hybrid differential single-station Kalman filter is used for PPP-B2b positioning, thereby improving the positioning accuracy of PPP-B2b at sea. This provides a stable, reliable, and inexpensive real-time positioning method for users in maritime environments without network communication.

[0007] This invention is achieved using the following technical solution: a real-time precise positioning method for the sea based on PPP-B2b, comprising the following steps:

[0008] Step A involves performing precise ephemeris calculations based on the BeiDou-3 PPP-B2b service, in conjunction with the broadcast ephemeris. This mainly includes the calculation of precise orbits and clock errors, including:

[0009] 1) Express the precise orbital corrections in the satellite orbital coordinate system as precise orbital corrections in the geocentric-geocentric coordinate system:

[0010]

[0011] Among them, [δO r δOa δO c ] T These are the orbital corrections in the radial, tangential, and normal star-fixed (ACR) coordinate systems; [δO x δO y δO z ] T This refers to the orbital corrections for the satellite within the Earth Center Earth Fixed (ECEF) reference frame.

[0012]

[0013] In the formula, r and To calculate satellite position and satellite velocity using broadcast ephemeris.

[0014] 2) Calculate the precision orbit based on PPP-B2b service:

[0015]

[0016] in, This represents the precise coordinates of the satellite within the Earth-Center Earth Fixed (ECEF) reference frame, recovered from the PPP-B2b corrections. This represents the satellite's coordinates in a geocentric reference frame, calculated from the broadcast ephemeris.

[0017] 3) Calculate the precise clock bias based on PPP-B2b service by combining the satellite clock bias provided by the broadcast ephemeris.

[0018]

[0019] in, This represents the precise clock error recovered by the PPP-B2b correction. C0 represents the clock bias calculated from the broadcast ephemeris, C0 represents the clock bias correction parameter obtained from the PPP-B2b message, and c represents the speed of light.

[0020] Step B: Acquire GNSS observation data, perform data preprocessing and error correction. For satellites with PPP-B2b corrections and those without, construct non-differential ionospheric combined observation models and epoch-differential ionospheric combined observation models respectively. Apply PPP-B2b precise clock errors and orbits to the non-differential ionospheric combined observation model, and then construct a signal-to-noise ratio stochastic model to calculate satellite weights. Specifically, this includes:

[0021] 1) Data preprocessing mainly includes cycle slip detection and repair and gross error removal. Cycle slips are detected using the ionospheric residual method, and gross errors are detected using the MW combined residual method. Error correction mainly includes tropospheric delay, relativistic effect correction, phase entanglement, Earth rotation correction, tidal effect correction, etc., all of which are corrected using existing models.

[0022] 2) By applying PPP-B2b correction products, the precise position and clock bias of the satellite are calculated. Based on the original pseudorange observations in the GNSS observation data, the non-differential de-ionospheric combined pseudorange observations are calculated, and the original phase observations are used to calculate the non-differential de-ionospheric combined phase observations. This further yields the linearized non-differential de-ionospheric combined observation model:

[0023]

[0024] In the formula, the superscripts G and C in each symbol represent the GPS and BDS-3 systems, respectively; the superscript s represents the satellite end; the subscript r represents the receiver end; and the subscript IF represents the deionization combination. and These represent the non-differential anionospheric pseudorange and phase observation values ​​of GPS satellites minus the station-satellite distance and various error corrections; Represents the direction cosine vector from the user end to the satellite end, the station-satellite distance, and the sum of these values. Obtained through receiver approximate coordinates and precise orbit recovered by PPP-B2b; δr e This represents the increment of the receiver's position relative to its approximate coordinates in the geocentric-fixed coordinate system; Indicates receiver clock bias; m w This represents the tropospheric wetted delay projection function, using the global projection function; δT w λ is the tropospheric wet delay in the zenith direction; IF Indicates the combined wavelength for deionization; To eliminate ionospheric combined ambiguity; δISB G-C This refers to the inter-system deviation parameter; and The noise values ​​for the non-differential ionospheric composite pseudorange and phase observations are respectively represented and estimated as zero-mean white noise.

[0025] Furthermore, the unspecified satellite clock bias is obtained from the precise clock bias recovered by PPP-B2b. Errors such as relativistic effects, Sagnac effects, Shapiro delay, phase entanglement, and tidal effects have been corrected using existing models. The unknown parameter in the above equation includes the receiver coordinate increment δr. e GPS system receiver clock bias Zenit tropospheric wet delay δT w Inter-system deviation parameter δISB G-C and ambiguity and As estimated as parameters to be estimated in the non-differential ionospheric combined observation model:

[0026]

[0027] Where x represents the parameter vector to be estimated in the non-differential anionospheric combined observation model, if n GPS satellites and m BDS-3 satellites are observed simultaneously in the current epoch, then the complete non-differential anionospheric combined observation model is expressed as:

[0028]

[0029] Where, p s and l s These represent the n GPS satellites and m BDS-3 satellites, respectively. and The vector composed of H represents the coefficient matrix of the non-differential anionization combined observation model; and These represent the n GPS satellites and m BDS-3 satellites, respectively. and The vector formed is represented as:

[0030]

[0031]

[0032]

[0033] For satellites without PPP-B2b corrections, epoch-difference deionized combined phase observations are calculated using the original phase observations. An epoch-difference deionized combined observation model is then constructed using broadcast ephemeris data. Specifically:

[0034]

[0035] in, This represents the epoch-difference deionization combined phase observation value minus various error corrections; Δρ k,k-1 This represents the difference in station-satellite distance between consecutive epochs. In the epoch-difference deionization combined observation model, the satellite positions required to calculate the station-satellite distance are obtained from the broadcast ephemeris. This represents the change in receiver clock bias between consecutive epochs; This indicates that the noise in the epoch difference of the combined ionospheric phase observations is eliminated. In the absence of cycle slips, the ambiguity change value of the combined ionospheric phase observations is 0. The inter-system deviation parameters have a strong time correlation with the zenith tropospheric wet delay, which has been eliminated by the epoch difference.

[0036] The parameters to be estimated for the epoch-differential de-ionospheric combined observation model include the receiver's adjacent epoch displacement and the receiver clock bias variation:

[0037]

[0038] Among them, dz k,k-1 dr represents the vector of parameters to be estimated in the epoch-difference deionization combined observation model. k,k-1 Let represent the change in receiver position before and after an epoch. Assuming j satellites without PPP-B2b corrections, the complete epoch-difference deionization combined observation model is expressed as:

[0039]

[0040] Where, ΔL s Represents j satellites The vector formed; This represents the coefficient matrix of the epoch-difference deionization combined observation model; Represents j satellites The vector formed by these vectors.

[0041] 3) Due to factors such as the location of each satellite, the signal quality varies. A reasonable random model is needed to weight each satellite. The accuracy of the model directly affects the PPP-B2b positioning accuracy; therefore, an accurate random model is essential for PPP-B2b data processing. Significant multipath effects at sea will severely impact PPP-B2b positioning accuracy. The signal-to-noise ratio (SNR) of satellite signals is correlated with multipath propagation. Therefore, this scheme constructs a random SNR model to reduce the impact of multipath effects on PPP-B2b positioning accuracy. The constructed random SNR model is as follows:

[0042]

[0043] In the formula, and C represents the noise variance of the original pseudorange observations and the original phase observations, respectively; p and C l For model coefficients, C / N0 s Let be the signal-to-noise ratio of satellite s;

[0044] Furthermore, the noise variance of the non-differential deionized ionospheric combined pseudorange and phase observations applied to measurement updates is:

[0045]

[0046] in,

[0047]

[0048] In the formula, the subscripts i and j in each symbol represent the two frequencies involved in the deionization combination; and These represent the noise variance of the satellite s-undifferentiated deionized ionospheric combined pseudorange and the phase observation, respectively. f i and f j These represent the magnitudes of frequency i and frequency j, respectively. URA represents the noise variance of PPP-B2b precision products; URA represents the user ranging accuracy factor broadcast by PPP-B2b products.

[0049] The noise variance of the epoch difference deionization combined phase observations applied to time updates is:

[0050]

[0051] in, This represents the noise variance of the satellite epoch-difference deionization phase observation.

[0052] Step C: Finally, based on the signal-to-noise ratio random model, time updates are performed using the epoch-difference deionization combined observation model, and measurement updates are performed using the non-difference deionization combined observation model. Hybrid differential single-station Kalman filtering is used for parameter estimation to complete real-time positioning at sea.

[0053] 1) For satellites without BeiDou PPP-B2b corrections, the receiver displacement relative to the previous epoch is calculated based on the epoch differential deionization combined observation model, and time updates are performed. For satellites with BeiDou PPP-B2b corrections, the measurement update is performed using the non-differential deionization combined observation model. Finally, the parameter estimation is completed using hybrid differential single-station Kalman filtering.

[0054] Kalman filtering consists of two steps: time update and measurement update.

[0055] The formula for time updating based on the epoch-difference deionization combined observation model and the signal-to-noise ratio stochastic model is as follows:

[0056]

[0057]

[0058] in, These represent the estimated state parameters and covariance matrix at time k-1, respectively. and dx represents the predicted state parameters and their corresponding covariance matrix at epoch k, respectively. k,k-1 , This represents the state changes and their covariance matrix. The state changes consist of the receiver's adjacent epoch displacement, the receiver clock bias change, the inter-system bias parameter change, the tropospheric zenith wet delay change, and the ambiguity change.

[0059] The receiver's adjacent epoch displacement and receiver clock bias variation values ​​are parameters dz estimated using the least squares method from the epoch-difference de-ionospheric combined observation model. k,k-1 The tropospheric zenith wet delay variation value is obtained from the tropospheric dynamic noise. When no cycle slip occurs, the inter-system deviation parameter and ambiguity are estimated as constants.

[0060] Measurement updates are based on a non-differential combined ionospheric observation model and a signal-to-noise ratio stochastic model.

[0061] y k =H k ·x k +∈ k ,∈ k ~N(0,R k )

[0062]

[0063]

[0064]

[0065] Among them, y k H is the observation vector. k Here is the coefficient matrix of the non-differential deionization combined model, x k Represents the state vector, ∈ k R is the non-differential deionization combined observation noise vector. k For the observed noise vector ∈ k The covariance matrix, G k It is the gain matrix. and This refers to the updated parameter estimates and their covariance matrix.

[0066] 2) After adjusting the parameters to be estimated using a hybrid differential single-station Kalman filter, the receiver coordinate increment, GPS system receiver clock error, zenith tropospheric wet delay, inter-system bias parameters, and ambiguity parameters are obtained. Then, the adjustment results are subjected to quality control. The maximum positioning residual is searched for and quality analysis is performed. If the maximum residual is greater than a certain threshold, it is considered to have gross error. The corresponding satellite is removed and the filtering is repeated until all residuals are less than the threshold. Finally, the receiver coordinate increment obtained after quality control is added to the approximate coordinates to obtain the receiver coordinates and output them, realizing real-time precise single-point positioning service for maritime users.

[0067] The present invention also proposes a positioning system based on a PPP-B2b real-time precision positioning method for the sea, the positioning system including a PPP calculation module and a PPP-B2b decoding module;

[0068] The PPP-B2b decoding module is used to combine broadcast ephemeris data to achieve PPP-B2b decoding and IOD matching, and to obtain real-time precise orbit and clock error correction values ​​for PPP-B2b.

[0069] The PPP solution module is used to preprocess and correct errors in GNSS observation data, and to construct non-differential ionospheric combined observation models and epoch-differential ionospheric combined observation models and their corresponding stochastic models. It applies real-time precise orbit and clock error corrections, performs measurement updates based on the non-differential ionospheric combined observation model, and performs time updates based on the epoch-differential ionospheric combined observation model. It uses hybrid differential single-station Kalman filtering to complete parameter estimation, thereby achieving real-time positioning solution at sea.

[0070] The present invention also proposes a computer-readable medium storing a computer program thereon, characterized in that, when the computer program is executed by a processor, it implements the steps of a PPP-B2b-based real-time precision positioning method for the sea.

[0071] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0072] This scheme obtains the PPP-B2b correction data broadcast by BeiDou-3 GEO satellites and combines it with broadcast ephemeris to recover the precise clock error and orbit based on the PPP-B2b correction data. After acquiring GNSS observation data and completing data preprocessing and error correction, it applies the precise clock error and orbit based on the PPP-B2b correction data to construct non-differential anionospheric combined observation models and epoch-differential anionospheric combined observation models for satellites with and without PPP-B2b correction data, respectively, and constructs a signal-to-noise ratio stochastic model to calculate satellite weights. Time updates are performed based on the epoch-differential anionospheric combined observation model and the signal-to-noise ratio stochastic model, and measurement updates are performed based on the non-differential anionospheric combined observation model and the signal-to-noise ratio stochastic model. Hybrid differential single-station Kalman filtering is used for parameter estimation to achieve real-time high-precision positioning at sea.

[0073] By utilizing orbit and clock corrections broadcast via PPP-B2b service, combined with broadcast ephemeris data to recover precise ephemeris and precise clock errors, and employing a signal-to-noise ratio stochastic model and hybrid differential single-station Kalman filtering for real-time high-precision positioning at sea, the problem of low positioning accuracy and difficulty in real-time application of PPP-B2b at sea is solved. It can provide accurate and reliable real-time positioning services at sea, and in the future, it can be applied to fields such as maritime emergency search and rescue centers and offshore energy exploration, which will generate significant economic benefits. Attached Figure Description

[0074] Figure 1 This is a flowchart illustrating the real-time precision positioning method at sea according to an embodiment of the present invention.

[0075] Figure 2 This is a comparison chart of the positioning error of PPP-B2b and the positioning error using commercial satellite station differential technology in a marine positioning experiment according to an embodiment of the present invention;

[0076] Figure 3 This is a comparison chart of the convergence time of PPP-B2b positioning and the convergence time of commercial satellite station differential positioning in a marine positioning experiment according to an embodiment of the present invention. Detailed Implementation

[0077] To better understand the above-described objects, features, and advantages of the present invention, the present invention will be further described below in conjunction with the accompanying drawings and embodiments. Many specific details are set forth in the following description to provide a thorough understanding of the present invention; however, the present invention may be practiced in other ways than those described herein, and therefore, the present invention is not limited to the specific embodiments disclosed below.

[0078] Example 1: This example proposes a real-time precise positioning method for marine applications based on PPP-B2b, including the following steps:

[0079] Step A: Obtain the PPP-B2b correction data broadcast by the BeiDou-3 GEO satellite, and combine it with the broadcast ephemeris data to restore the precise clock error and orbit based on the PPP-B2b correction data;

[0080] To ensure the correlation between the information content broadcast by different information types, as well as the correlation between correction data and broadcast ephemeris, the information is identified by four version numbers: Spatial State Expression Data Period Number (IOD SSR), Pseudo-Random Noise Code Data Period Number (IODP), Navigation Data Period Number (IODN), and Correction Data Identification and Data Period Number (IOD Corr), to facilitate matching and use.

[0081] (1) The orbit correction data in the PPP-B2b message has an epoch time interval of 48s, and the broadcast orbit correction information includes the orbit correction vector. The satellite position correction vector can be calculated using the orbit correction values. After the IODN of the orbit correction information is successfully matched with the IODC in the broadcast ephemeris navigation message, the precise orbit correction data in the satellite orbit coordinate system can be expressed as the precise orbit correction data in the geocentric-ground-fixed coordinate system:

[0082]

[0083] Among them, [δO r δO a δO v ]T These are the orbital corrections in the radial, tangential, and normal star-fixed (ACR) coordinate systems; [δO x δO y δO z ] T This refers to the orbital corrections for the satellite within the Earth-Centered Earth Fixed (ECEF) reference frame. r e a e c The unit vectors corresponding to the radial, tangential, and normal directions, respectively, can be expressed as:

[0084]

[0085] In the formula, r and To calculate satellite position and satellite velocity using broadcast ephemeris;

[0086] Calculations are performed on precision orbits based on PPP-B2b services:

[0087]

[0088] in This represents the precise coordinates of the satellite within the Earth-Center Earth Fixed (ECEF) reference frame, recovered from the PPP-B2b corrections. This represents the satellite's coordinates in a geocentric reference frame, calculated from the broadcast ephemeris.

[0089] (2) The epoch time interval of the clock correction number in the PPP-B2b clock correction information is 6 seconds; after the clock correction information is successfully matched with the broadcast ephemeris, the clock correction parameter C0 contained in the message can be used to correct the clock error parameter calculated from the broadcast ephemeris. Calculation of precise clock errors based on PPP-B2b service:

[0090]

[0091] in, This represents the precise clock error recovered by the PPP-B2b correction. This represents the satellite clock error calculated from the broadcast ephemeris. C0 represents the clock error correction parameter obtained from the PPP-B2b message, and c represents the speed of light. When the orbit correction number in the PPP-B2b message and the IOD Corr in the clock error correction information are the same, and their IOD Corr can be matched with the IODC in the navigation message, the orbit and clock error correction numbers can be used for error correction.

[0092] Step B: Acquire GNSS observation data, perform data preprocessing and error correction, construct non-differential ionospheric combined observation model and epoch-differential ionospheric combined observation model for satellites with PPP-B2b correction and satellites without PPP-B2b correction, respectively, and apply the PPP-B2b precise clock error and orbit calculated in Step A to the non-differential ionospheric combined observation model, and then construct a signal-to-noise ratio stochastic model to calculate satellite weights;

[0093] (1) Data preprocessing mainly includes cycle slip detection and repair and gross error removal. Cycle slips are detected by ionospheric residual method and gross errors are detected by MW combined residual method. Error correction mainly includes tropospheric delay, relativistic effect correction, phase entanglement, Earth rotation correction, tidal effect correction, etc., all of which are corrected using existing models and will not be elaborated on here.

[0094] (2) In the non-differential de-ionospheric combined observation model, the precise position and clock error of the satellite are calculated by applying the PPP-B2b correction product. The non-differential de-ionospheric combined pseudorange observation value is calculated based on the original pseudorange observation value in the GNSS observation data, and the non-differential de-ionospheric combined phase observation value is calculated based on the original phase observation value. The linearized non-differential de-ionospheric combined observation model is then obtained.

[0095]

[0096] In the formula, the superscripts G and C in each symbol represent the GPS and BDS-3 systems, respectively; the superscript s represents the satellite end; the subscript r represents the receiver end; and the subscript IF represents the deionization combination. and These represent the non-differential anionospheric pseudorange and phase observation values ​​of GPS satellites minus the station-satellite distance and various error corrections; Represents the direction cosine vector from the user end to the satellite end, the station-satellite distance, and the sum of these values. Obtained through receiver approximate coordinates and precise orbit recovered by PPP-B2b; δr e This represents the increment of the receiver's position relative to its approximate coordinates in the geocentric-fixed coordinate system; Indicates receiver clock bias; m w This represents the tropospheric wetted delay projection function, using the global projection function; δT w λ is the tropospheric wet delay in the zenith direction; IF Indicates the combined wavelength for deionization; To eliminate ionospheric combined ambiguity; δISB G-C This refers to the inter-system deviation parameter; and The noise values ​​for the non-differential ionospheric composite pseudorange and phase observations are respectively represented and estimated as zero-mean white noise.

[0097] Furthermore, the unspecified satellite clock bias is obtained from the precise clock bias recovered by PPP-B2b. Errors such as relativistic effects, Sagnac effects, Shapiro delay, phase entanglement, and tidal effects have been corrected using existing models. The unknown parameters in the above equation include the receiver coordinate increment δr. e GPS system receiver clock bias Zenit tropospheric wet delay δT w Inter-system deviation parameter δISB G-C and ambiguity and As estimated as parameters to be estimated in the non-differential ionospheric combined observation model:

[0098]

[0099] Where x represents the parameter vector to be estimated in the non-differential anionospheric combined observation model. If n GPS satellites and m BDS-3 satellites are observed simultaneously in the current epoch, then the complete non-differential anionospheric combined observation model is expressed as:

[0100]

[0101] Where, p s and l s These represent the n GPS satellites and m BDS-3 satellites, respectively. and The vector composed of H represents the coefficient matrix of the non-differential anionization combined observation model; and These represent the n GPS satellites and m BDS-3 satellites, respectively. and The vector formed is represented as:

[0102]

[0103]

[0104]

[0105] Currently, the PPP-B2b service only broadcasts satellite corrections for GPS and BDS-3 systems. Satellites from systems such as Galileo and GLONASS lack PPP-B2b corrections, making it impossible to obtain precise clock biases and precise ephemeris data. For these satellites without PPP-B2b corrections, epoch-difference deionized combined phase observations are calculated using the original phase observations. An epoch-difference deionized combined observation model is then constructed using the broadcast ephemeris. Specifically:

[0106]

[0107] in, This represents the epoch-difference deionization combined phase observation value minus various error corrections; Δρ k,k-1 This represents the difference in station-satellite distance between consecutive epochs. In the epoch-difference deionization combined observation model, the satellite positions required to calculate the station-satellite distance are obtained from the broadcast ephemeris. This represents the change in receiver clock bias between consecutive epochs; This indicates that the noise in the epoch difference of the combined ionospheric phase observations is eliminated. In the absence of cycle slips, the ambiguity change value of the combined ionospheric phase observations is 0. The inter-system deviation parameters have a strong time correlation with the zenith tropospheric wet delay, which has been eliminated by the epoch difference.

[0108] The parameters to be estimated for the epoch-differential de-ionospheric combined observation model include the receiver's adjacent epoch displacement and the receiver clock bias variation:

[0109]

[0110] Among them, dz k,k-1 dr represents the vector of parameters to be estimated in the epoch-difference deionization combined observation model. k,k-1 This represents the change in receiver position before and after an epoch. Suppose there are j satellites without PPP-B2b corrections, then the complete epoch-difference deionization combined observation model can be expressed as:

[0111]

[0112] Where, ΔL s Represents j satellites The vector formed; This represents the coefficient matrix of the epoch-difference deionization combined observation model; Represents j satellites The vector formed by these vectors.

[0113] (3) Due to factors such as the location of each satellite, the signal quality varies. A reasonable random model is needed to weight each satellite. The accuracy of the model directly affects the PPP-B2b positioning accuracy; therefore, an accurate random model is essential for PPP-B2b data processing. Significant multipath effects at sea will severely impact PPP-B2b positioning accuracy. The signal-to-noise ratio (SNR) of satellite signals is correlated with multipath effects. Therefore, this scheme constructs a random SNR model to reduce the impact of multipath effects on PPP-B2b positioning accuracy. The constructed random SNR model is as follows:

[0114]

[0115] In the formula, and C represents the noise variance of the original pseudorange observations and the original phase observations, respectively; p and C l For model coefficients, C / N0 s Let be the signal-to-noise ratio of satellite s;

[0116] Furthermore, the noise variance of the non-differential deionized ionospheric combined pseudorange and phase observations applied to measurement updates is:

[0117]

[0118] in,

[0119]

[0120] In the formula, the subscripts i and j in each symbol represent the two frequencies involved in the deionization combination; and These represent the noise variance of the satellite s-undifferentiated deionized ionospheric combined pseudorange and the phase observation, respectively. f i and f j These represent the magnitudes of frequency i and frequency j, respectively. URA represents the noise variance of PPP-B2b precision products; URA represents the user ranging accuracy factor broadcast by PPP-B2b products.

[0121] The noise variance of the epoch difference deionization combined phase observations applied to time updates is:

[0122]

[0123] in, This represents the noise variance of the satellite epoch-difference deionization phase observation.

[0124] Step C involves updating measurements based on the non-differential ionospheric combined observation model and the signal-to-noise ratio stochastic model, and updating time based on the epoch-differential ionospheric combined observation model and the signal-to-noise ratio stochastic model. Parameter estimation is performed using a hybrid differential single-station Kalman filter to complete real-time positioning at sea.

[0125] (1) For satellites without BeiDou PPP-B2b corrections, the displacement relative to the previous epoch is calculated based on the epoch differential deionization combined observation model, and time is updated accordingly; for satellites with BeiDou PPP-B2b corrections, the measurement is updated using the non-differential deionization combined observation model, and finally the parameter estimation is completed using the hybrid differential single-station Kalman filter.

[0126] Kalman filtering consists of two steps: time update and measurement update.

[0127] The formula for time updating based on the epoch-difference deionization combined observation model and the signal-to-noise ratio stochastic model is as follows:

[0128]

[0129]

[0130] in, These represent the estimated state parameters and covariance matrix at time k-1, respectively. and dx represents the predicted state parameters and their corresponding covariance matrix at epoch k, respectively. k,k-1 , This represents the state changes and their covariance matrix. The state changes consist of the receiver's adjacent epoch displacement, the receiver clock error change, the tropospheric zenith wet delay change, the inter-system bias parameter change, and the ambiguity change.

[0131] The receiver's adjacent epoch displacement and receiver clock bias variation values ​​are parameters dz estimated using the least squares method from the epoch-difference de-ionospheric combined observation model. k,k-1 The tropospheric zenith wet delay variation value is obtained from the tropospheric dynamic noise. When no cycle slip occurs, the inter-system deviation parameter and ambiguity are estimated as constants.

[0132] Measurement updates are based on a non-differential ionospheric combined observation model and a signal-to-noise ratio stochastic model.

[0133] y k =H k ·x k +∈ k ,∈ k ~N(0,R k )

[0134]

[0135]

[0136]

[0137] Among them, y k H is the observation vector. k Here is the coefficient matrix of the non-differential deionization combined model, x k Represents the state vector, ∈ k R is the non-differential deionization combined observation noise vector. k For the observed noise vector ∈ k The covariance matrix, G k It is the gain matrix. and This refers to the updated parameter estimates and their covariance matrix.

[0138] (2) After adjusting the parameters to be estimated using hybrid differential single-station Kalman filtering, the receiver coordinate increment, GPS system receiver clock error, zenith tropospheric wet delay, inter-system bias parameters, and ambiguity parameters are obtained. Then, the adjustment results are quality controlled, and the maximum positioning residual is found for quality analysis. If the maximum residual is greater than a certain threshold, it is considered that there is a gross error. The corresponding satellite is removed and the filtering is repeated until all residuals are less than the threshold. Finally, the receiver coordinate increment obtained after quality control is added to the approximate coordinates to obtain the receiver coordinates and output them, so as to realize real-time precise single-point positioning service for maritime users.

[0139] The precision positioning method disclosed in this embodiment utilizes the orbit, clock error, and code deviation correction numbers broadcast by the PPP-B2b service to obtain high-precision positioning information at sea in real time. This solves the problem of low positioning accuracy of PPP-B2b at sea, which makes it difficult to apply in real time. It provides a stable, reliable, and inexpensive real-time high-precision positioning method for users in maritime environments without network communication, and provides technical support for fields such as maritime rescue and marine energy exploration.

[0140] Example 2: Based on the positioning method described in Example 1, this example proposes a positioning system based on the PPP-B2b real-time precision positioning method for the sea. The positioning system includes a PPP calculation module and a PPP-B2b decoding module.

[0141] Combination Figure 1 As shown, the PPP-B2b decoding module combines broadcast ephemeris data to achieve PPP-B2b decoding and IOD matching, and obtains real-time PPP-B2b orbit and clock error corrections. The PPP calculation module is used to preprocess and correct errors in GNSS observation data. Then, for satellites with and without PPP-B2b corrections, it constructs an undifferentiated anionospheric combined observation model and an epoch-differential anionospheric combined observation model, respectively, and constructs a signal-to-noise ratio stochastic model to calculate satellite weights. Time updates are performed based on the epoch-differential anionospheric combined observation model and the signal-to-noise ratio stochastic model, and measurement updates are performed based on the undifferentiated anionospheric combined observation model and the signal-to-noise ratio stochastic model. Hybrid differential single-station Kalman filtering is used for parameter estimation to achieve real-time precise positioning at sea.

[0142] Example 3: Based on the positioning method proposed in Example 1, this example provides a computer device and a computer-readable storage medium storing a computer program. The computer device includes a memory and a processor. The memory stores the computer program. When the computer program is executed by the processor, the processor performs the steps of the PPP-B2b-based real-time precise positioning method for the sea as described in Example 1.

[0143] The following examples illustrate the application scenarios of the PPP-B2b-based real-time precise positioning method at sea:

[0144] To evaluate the performance of the PPP-B2b-based real-time precise positioning method at sea, the positioning results of commercial real-time satellite differential positioning technology were compared.

[0145] The marine GNSS positioning experiment was conducted on March 6, 2023 (65 days of the year) from 00:40:00 to 05:40:00 (GPS time) in Jiaozhou Bay, Qingdao, Shandong Province. A FRLL-PlUS receiver was used to acquire PPP-B2b messages, a NovAtel PwrPark7D receiver to acquire rover GNSS observations, and a NavCom SF3050 receiver to acquire satellite differential positioning results. Simultaneously, a base station was set up on the shore, and a Sinan M300 receiver was used to acquire base station GNSS observations. RTK calculations using InertialExplorer (IE 8.90) software were used as reference positions. All acquired GNSS observations included raw pseudorange and phase observations from both GPS, GLONASS, Galileo, BDS2, and BDS3 systems, with a sampling rate of 1 second. The positioning accuracy and convergence speed of Starfire and PPP-B2b differential positioning methods were evaluated. The criteria for convergence were set as follows: positioning errors in the east and north directions less than 30 cm, and vertical positioning errors less than 60 cm, maintained for at least one minute. Since both positioning methods have a certain convergence time, the results were selected one hour after the start of the experiment for positioning accuracy evaluation, with an evaluation period of four hours.

[0146] Table 1. Statistics on Positioning Accuracy of Marine Positioning Experiments

[0147]

[0148] Table 1 presents the Bias and RMS statistics of the positioning error sequences for the satellite-based differential and PPP-B2b positioning methods in four dimensions: north-south, east-west, vertical, and 3D. Additionally, Figure 2 and Figure 3 The positioning error sequences of satellite-based differential positioning and PPP-B2b, as well as the convergence times of the two methods, are presented respectively. It can be seen that by employing a signal-to-noise ratio stochastic model and hybrid differential single-station Kalman filtering, PPP-B2b achieves overall positioning accuracy and convergence time at sea comparable to commercial satellite-based differential positioning technology. It provides a stable, reliable, and cost-effective real-time location calculation method for users in marine environments without network communication, and can be widely applied in fields such as marine energy exploration.

[0149] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments for application in other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for real-time precise positioning at sea based on PPP-B2b, characterized in that, Includes the following steps: Step A: Obtain the PPP-B2b correction data broadcast by the BeiDou-3 system, and combine it with the broadcast ephemeris data to restore the precise clock error and orbit based on the PPP-B2b correction data; Step B: Acquire GNSS observation data and perform data preprocessing and error correction. For satellites with PPP-B2b corrections and satellites without PPP-B2b corrections, construct non-differential deionization combined observation models and epoch-differential deionization combined observation models respectively. Apply the precise clock error and orbit based on PPP-B2b corrections to the non-differential deionization combined observation model, and then construct a signal-to-noise ratio stochastic model to calculate satellite weights. Step C: Measurement updates are performed based on the non-differential ionospheric combined observation model and the signal-to-noise ratio stochastic model. Time updates are performed based on the epoch-differential ionospheric combined observation model and the signal-to-noise ratio stochastic model. Parameter estimation is performed using a hybrid differential single-station Kalman filter to complete real-time positioning at sea.

2. The PPP-B2b based real-time precise positioning at sea method according to claim 1, characterized in that, In step B, the construction process of the non-differential ionospheric combined observation model is as follows: For satellites with PPP-B2b corrections, the precise position and clock bias of the satellite are calculated by applying the PPP-B2b correction product. Based on the raw pseudorange observations in the GNSS observation data, the non-differential de-ionospheric combined pseudorange observations are calculated, and the non-differential de-ionospheric combined phase observations are calculated from the raw phase observations. This further yields the linearized non-differential de-ionospheric combined observation model. where the superscripts G and C in each symbol represent GPS and BDS-3 system respectively, the superscript s represents satellite end, the subscript r represents receiver end, and the subscript IF represents ionosphere-free combination; represents the ionosphere-free combination pseudorange observation of GPS satellite minus station-satellite distance and each error correction, represents the ionosphere-free combination phase observation of GPS satellite minus station-satellite distance and each error correction; represents the direction cosine vector from user end to satellite end; δr e represents the increment of receiver position relative to approximate coordinates in the geocentric and geodetic coordinate system; represents the receiver clock error; m w denotes the tropospheric wet delay projection function; δT w is the tropospheric wet delay in zenith direction; λ IF denotes the ionosphere-free combination wavelength; is the ionosphere-free combination ambiguity; δISB G-C is the inter-system bias parameter; and denote the ionosphere-free combination pseudorange and phase observation noise, respectively; The unknown parameters in the above equations include the receiver coordinate increment dr e , the GPS system receiver clock error , the zenith tropospheric wet delay δT w , the inter-system bias parameter δISB G-C , and the ambiguity and are estimated as the estimated parameters in the non-difference ionosphere combined observation model: Where x represents the parameter vector to be estimated in the non-differential anionospheric combined observation model, if n GPS satellites and m BDS-3 satellites are observed simultaneously in the current epoch, then the complete non-differential anionospheric combined observation model is expressed as: Where, p s and l s These represent the n GPS satellites and m BDS-3 satellites, respectively. and The vector formed; and These represent the n GPS satellites and m BDS-3 satellites, respectively. and The vector is composed of H, which represents the coefficient matrix of the non-differential ionospheric combined observation model.

3. The PPP-B2b based real-time precise positioning at sea method according to claim 1, characterized in that, In step B, the construction process of the epoch-difference deionization combined observation model is as follows: For satellites without PPP-B2b corrections, epoch-difference deionized combined phase observations are calculated using the original phase observations, and an epoch-difference deionized combined observation model is constructed using broadcast ephemeris. wherein, denotes the ionosphere-free combination phase observations with epoch difference minus errors; k,k-1 denotes the difference in station-star distance between the epochs; denotes the change in receiver clock bias between the epochs; denotes the noise in the ionosphere-free combination phase observations with epoch difference. The parameters to be estimated for the epoch-differential de-ionospheric combined observation model include the receiver's adjacent epoch displacement and the receiver clock bias variation: where dz k,k-1 is the estimated parameter vector of the ionosphere-free combination observation model, dr k,k-1 is the receiver position change between epochs, and j is the number of satellites without PPP-B2b corrections. The complete ionosphere-free combination observation model is represented as where ΔL s represents the vector composed of the ionospheric delays of the j satellites; represents the coefficient matrix of the ionosphere-free combination observation model; represents the vector composed of the ionospheric delays of the j satellites; represents the vector composed of the ionospheric delays of the j satellites;​ 4. The PPP-B2b based real-time precise positioning at sea method according to claim 1, characterized in that, In step B, a signal-to-noise ratio (SNR) stochastic model is established to calculate satellite weights. The constructed SNR stochastic model is as follows: wherein and respectively denote the noise variance of the raw pseudorange and raw phase observations; C p and C l are model coefficients, C / N0 s is the signal-to-noise ratio of satellite s. Furthermore, the noise variance of the non-differential deionized ionospheric combined pseudorange and phase observations applied to measurement updates is: In the formula, the subscripts i and j in each symbol represent the two frequencies involved in the deionization combination; and These represent the noise variance of the satellite s-undifferentiated deionized ionospheric combined pseudorange and the phase observation, respectively. f i and f j These represent the magnitudes of frequency i and frequency j, respectively. This represents the noise variance of PPP-B2b precision products. URA represents the user ranging accuracy factor broadcast by the PPP-B2b product. The noise variance of the epoch difference deionization combined phase observations applied to time updates is: in, This represents the noise variance of the satellite epoch-difference deionization phase observation.

5. The PPP-B2b based real-time precise positioning at sea method according to claim 1, characterized in that, Step C is implemented in the following manner: (1) Time updates are performed based on the epoch-difference deionization combined observation model and the signal-to-noise ratio stochastic model: wherein, respectively represent the state parameter estimation value and the covariance matrix at the epoch k-1, and respectively represent the state parameter prediction value and the corresponding covariance matrix at the epoch k, dx k,k-1 , represent the state change and the covariance matrix thereof; (2) Measurement updates are performed based on the non-differential ionospheric combined observation model and the signal-to-noise ratio stochastic model: y k = H k · x k + ∈ k , ∈ k ~ N(0, R k ) Among them, y k H is the observation vector. k The coefficient matrix of the non-differential deionization combined model is x. k Represents the state vector, ∈ k R is the non-differential deionization combined observation noise vector. k For the observed noise vector ∈ k The covariance matrix, G k It is the gain matrix. and This refers to the updated parameter estimates and their covariance matrix.

6. The real-time precise positioning method for marine applications based on PPP-B2b according to claim 1, characterized in that, Step A is specifically implemented in the following manner: (1) Express the precise orbital correction in the satellite orbital coordinate system as the precise orbital correction in the geocentric and Earth-fixed coordinate system: where [δO r δO a δO c ] T are the orbit correction numbers in the radial, tangential and normal star-fixed coordinate system;[δO x δO y δO z ] T are the orbit correction numbers in the satellite's Earth-Centered Earth-Fixed reference frame;[e r e a e c ] are the unit vectors in the radial, tangential and normal directions, respectively; (2) Precision orbit calculation based on PPP-B2b service: in This represents the precise coordinates of the satellite within a geocentric reference frame, recovered from the PPP-B2b corrections. This represents the satellite's coordinates in a geocentric, geofixed reference frame, calculated from the broadcast ephemeris. (3) Calculate the precision clock bias based on PPP-B2b by combining the satellite clock bias provided by the broadcast ephemeris: in, This represents the precise clock error recovered by the PPP-B2b correction. C0 represents the clock bias calculated from the broadcast ephemeris, C0 represents the clock bias correction parameter obtained from the PPP-B2b message, and c represents the speed of light.

7. The real-time precise positioning method for marine applications based on PPP-B2b according to claim 1, characterized in that, In step C, after adjusting the parameters to be estimated using a hybrid differential single-station Kalman filter, the receiver coordinate increment, GPS system receiver clock error, zenith tropospheric wet delay, inter-system bias parameters, and ambiguity parameters are obtained. Then, the adjustment results are subjected to quality control, and the maximum positioning residual is searched for and quality analysis is performed until all residuals are less than the threshold. Finally, the receiver coordinate increment obtained after quality control is added to the approximate coordinates to realize real-time precise single-point positioning service for maritime users.

8. A positioning system based on the PPP-B2b-based real-time precise positioning method for marine applications according to any one of claims 1-7, characterized in that, The positioning system includes a PPP resolution module and a PPP-B2b decoding module; The PPP-B2b decoding module is used to combine broadcast ephemeris data to achieve PPP-B2b decoding and IOD matching, and to obtain real-time PPP-B2b orbit and clock correction values; The PPP solution module is used to preprocess and correct errors in GNSS observation data. It constructs an unequal-differenced ionospheric combined observation model for satellites with PPP-B2b corrections, combines it with the constructed signal-to-noise ratio stochastic model, and applies real-time orbit and clock error corrections for measurement updates. For satellites without PPP-B2b corrections, it constructs an epoch-differenced ionospheric combined observation model, combines it with the signal-to-noise ratio stochastic model for time updates, and uses hybrid differential Kalman filter parameter estimation to achieve real-time positioning at sea.

9. A computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the PPP-B2b-based real-time precision positioning method for maritime applications as described in any one of claims 1 to 7.

Citation Information

Patent Citations

  • PPP-B2b enhanced low-cost single-frequency GNSS receiver precision positioning method

    CN115616615A

  • Precision point positioning system and method based on PPP-B2b service

    CN114280644A

  • Real-time co-seismic displacement speed calculation method and system based on Beidou PPP-B2b service

    CN116973974A