A troposphere delay accuracy transfer method and system thereof

Abstract

The application relates to a tropospheric delay precision transmission method and system, which comprises the following steps: obtaining the elevation variation component observation value of each reference station based on the horizontal spatial component value and the measured tropospheric delay value of each reference station in a GNSS positioning system; establishing a parameterized function model based on the elevation variation component observation value of each reference station and the station elevation, solving the parameter estimation value of the parameterized function model by adopting a weighted least square method, and calculating the variance-covariance matrix of the parameter estimation value; calculating the elevation variation component standard deviation of the position of a user terminal in the GNSS positioning system by the error propagation law based on the variance-covariance matrix; taking the elevation variation component standard deviation as precision information, and taking the tropospheric delay prior value obtained based on the parameter estimation value as data input together to participate in the positioning solution of the user terminal. The application avoids the problem of parameter correlation enhancement caused by the lack of precision information, and can still realize high-precision ambiguity fixing and positioning solution under complex terrain conditions.

Description

Technical Field

[0001] This application relates to the field of high-precision positioning technology of Global Navigation Satellite System (GNSS), specifically to a method and system for transferring tropospheric delay accuracy. Background Technology

[0002] In high-precision GNSS positioning, atmospheric delay error is the main source of error affecting positioning accuracy. Tropospheric delay, as a non-dispersive error, cannot be completely eliminated by linear combination of dual-frequency observations. Traditional tropospheric delay processing methods are mainly divided into two categories: one is to use empirical models (such as the Saastamoinen model and the Hopfield model) for single-station correction; the other is to generate regional corrections based on a reference station network using spatial interpolation methods.

[0003] However, these traditional methods have the following technical drawbacks: empirical models fail to fully consider the significant impact of elevation factors on tropospheric delay, resulting in a significant decrease in correction accuracy in areas with drastic elevation changes; network interpolation methods typically treat tropospheric delay as a planar distribution, ignoring the inherent physical characteristics of tropospheric delay as it changes with elevation. This approach leads to interpolation results that cannot accurately reflect the vertical variation of tropospheric delay in areas with varying elevations, further reducing correction accuracy.

[0004] Existing technologies lack reliable accuracy information for providing tropospheric correction values ​​to users, making it impossible for users to impose reasonable prior constraints on these parameters during parameter calculation. Due to this lack of accuracy, the strong correlation between ambiguity parameters and tropospheric parameters cannot be effectively eliminated, leading to a reduced success rate in ambiguity fixation, a problem particularly pronounced in complex terrain conditions. In complex terrain, tropospheric delays vary drastically, and the lack of accuracy information makes it difficult for users to distinguish between the true signal and delay errors, resulting in a significant increase in the failure rate of ambiguity fixation. Summary of the Invention

[0005] This application provides a method and system for transmitting tropospheric delay accuracy, which can solve the problem that traditional methods in the prior art cannot effectively transmit the accuracy information of tropospheric delay to the user end, resulting in the user end being unable to use this accuracy information for accurate parameter calculation.

[0006] In a first aspect, embodiments of this application provide a method for transferring tropospheric delay accuracy, comprising: Based on the horizontal spatial component values ​​and measured tropospheric delay values ​​of each reference station in the GNSS positioning system, the observed values ​​of the elevation change component of each reference station are obtained. Based on the elevation change component observations and station elevations of each reference station, a parameterized function model is established. The weighted least squares method is used to solve the parameter estimates of the parameterized function model, and the variance-covariance matrix of the parameter estimates is calculated. Based on the variance-covariance matrix, the standard deviation of the elevation variation component of the user's location in the GNSS positioning system is calculated by the error propagation law. The standard deviation of the elevation variation component is used as the accuracy information, and together with the tropospheric delay prior value obtained based on parameter estimation, it is used as data input to participate in the user-end positioning solution.

[0007] In conjunction with the first aspect, in one implementation, based on the horizontal spatial component values ​​and measured tropospheric delay values ​​of each reference station in the GNSS positioning system, the observed elevation change component values ​​of each reference station are obtained, specifically including: Based on an empirical model, the horizontal spatial component values ​​of each reference station in the GNSS positioning system are calculated. The elevation change component of each reference station is obtained by calculating the difference between the measured tropospheric delay value and the horizontal spatial component value.

[0008] In conjunction with the first aspect, in one implementation, before calculating the horizontal spatial component values ​​of each reference station in the GNSS positioning system based on an empirical model, the method further includes: Based on the location information of each reference station, one model is selected from the candidate models as the empirical model.

[0009] In conjunction with the first aspect, in one implementation, the parameterized function model is: ; Where ΔZTD is the observed value of the elevation change component of the reference station, H is the station elevation of the reference station, and a and b are estimated parameters.

[0010] In conjunction with the first aspect, in one implementation, the weighted least squares method is used to solve for the parameter estimates of the parameterized function model, and the variance-covariance matrix of the parameter estimates is calculated, specifically including: A design matrix is ​​constructed based on the elevation component observations and station elevations of each reference station; A weight matrix is ​​constructed based on the absolute difference between the elevation of each reference station and the average elevation. By combining the design matrix, weight matrix, and elevation component observations, the parameter estimates of the parameterized function model are obtained using the weighted least squares method. The residual vector is calculated using the elevation component observations and the estimated parameters obtained from the solution. Calculate the unit weight variance using the residual vector and the weight matrix; Based on the unit weight variance and the design matrix, calculate the variance-covariance matrix of the parameter estimates.

[0011] In conjunction with the first aspect, in one implementation, based on the variance-covariance matrix, the standard deviation of the elevation variation component of the user's location in the GNSS positioning system is calculated using the error propagation law, specifically including: Based on the number of reference stations in the GNSS positioning system and the number of parameters to be estimated in the parametric function model, calculate the extended unit weight variance scaling factor. The extended unit weight variance scaling factor and variance-covariance matrix are introduced into the accuracy transfer model based on the error propagation law, and the standard deviation of the elevation variation component of the user's location in the GNSS positioning system is calculated.

[0012] In conjunction with the first aspect, in one implementation, the formula for calculating the extended unit weight variance scaling factor K is: ; Where n is the number of reference stations in the GNSS positioning system, and t is the number of parameters to be estimated in the parametric function model.

[0013] In conjunction with the first aspect, in one implementation, the standard deviation of the elevation variation component is used as accuracy information, and together with the tropospheric delay prior value obtained based on parameter estimation, it is used as data input to participate in the user-end positioning calculation, specifically including: Use the standard deviation of elevation variation components as accuracy information; Based on the parameter estimates of the parameterized function model and the station elevation of the user's location, the prior value of tropospheric delay is obtained. The accuracy information and the prior value of tropospheric delay are used as virtual observation values ​​to participate in the user-end positioning solution.

[0014] In conjunction with the first aspect, in one implementation, the accuracy information and the prior value of tropospheric delay are used as virtual observations to participate in the user-end positioning calculation, specifically including: Constraint weights are determined based on accuracy information; Based on the tropospheric delay prior value and constraint weights, an observation equation containing virtual observations of the tropospheric delay is constructed. The constructed observation equations are integrated into the user-end positioning solution model and participate in the user-end positioning solution.

[0015] Secondly, embodiments of this application provide a tropospheric delay accuracy transfer system, comprising: a first module for acquiring the elevation change component observation values ​​of each reference station based on the horizontal spatial component values ​​and measured tropospheric delay values ​​of each reference station in the GNSS positioning system; a second module for establishing a parameterized function model based on the elevation change component observation values ​​and station elevations of each reference station, solving the parameter estimates of the parameterized function model using the weighted least squares method, and calculating the variance-covariance matrix of the parameter estimates; a third module for calculating the standard deviation of the elevation change component of the user's location in the GNSS positioning system based on the variance-covariance matrix and using the error propagation law; and a fourth module for using the standard deviation of the elevation change component as accuracy information, and using it together with the prior tropospheric delay values ​​obtained based on the parameter estimates as data inputs to participate in the user's positioning calculation.

[0016] By rationally decomposing tropospheric delay and precisely modeling elevation components, the accuracy information of the reference station network can be accurately transmitted to the user end, thus forming strong constraints in the user-end positioning calculation. Since there is a clear physical relationship between elevation components and station elevations, the elevation components at different elevation locations can be accurately predicted through parametric function models. The standard deviation of the elevation components calculated by the error propagation law reflects the accuracy information of the predicted value, enabling the user end to perform more accurate positioning calculations based on the accuracy assessment of the reference station network. This precise transmission of accuracy information effectively reduces the correlation between ambiguity parameters and tropospheric parameters, because the accuracy information provides reliable prior constraints for parameter calculation, avoiding the problem of enhanced parameter correlation caused by the lack of accuracy information in traditional methods. Thus, high-precision ambiguity fixation and positioning calculation can still be achieved under complex terrain conditions.

[0017] The beneficial effects of the technical solutions provided in this application include: This application provides a method and system for transmitting tropospheric delay accuracy. By rationally decomposing the tropospheric delay and precisely modeling the elevation components, the accuracy information of the reference station network can be accurately transmitted to the user end, thereby forming strong constraints in the user end's positioning calculation. Since there is a clear physical relationship between the elevation components and the station elevations, the elevation components at different elevation locations can be accurately predicted through a parametric function model. The standard deviation of the elevation components calculated by the error propagation law reflects the accuracy information of the predicted value, enabling the user end to perform more accurate positioning calculations based on the accuracy assessment of the reference station network. This accurate transmission of accuracy information effectively reduces the correlation between ambiguity parameters and tropospheric parameters, because the accuracy information provides reliable prior constraints for parameter calculations, avoiding the problem of enhanced parameter correlation caused by the lack of accuracy information in traditional methods. Thus, high-precision ambiguity fixing and positioning calculations can still be achieved under complex terrain conditions. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the tropospheric delay accuracy transfer method of this application. Detailed Implementation

[0019] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0020] This application provides a method and system for transmitting tropospheric delay accuracy, which can solve the problem that traditional methods in the prior art cannot effectively transmit the accuracy information of tropospheric delay to the user end, resulting in the user end being unable to use this accuracy information for accurate parameter calculation.

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0022] In a first aspect, embodiments of this application provide a method for transferring tropospheric delay accuracy, comprising: 101: Based on the horizontal spatial component values ​​and measured tropospheric delay values ​​of each reference station in the GNSS positioning system, obtain the observed values ​​of the elevation change component of each reference station; 102: Based on the elevation change component observations and station elevations of each reference station, a parameterized function model is established. The weighted least squares method is used to solve the parameter estimates of the parameterized function model, and the variance-covariance matrix of the parameter estimates is calculated. 103: Based on the variance-covariance matrix, the standard deviation of the elevation variation component of the user's location in the GNSS positioning system is calculated using the error propagation law; 104: The standard deviation of the elevation change component is used as the accuracy information, and together with the tropospheric delay prior value obtained based on parameter estimation, it is used as data input to participate in the user-end positioning solution.

[0023] In this GNSS-enhanced positioning system, the elevation variation component observations are first obtained based on the measured tropospheric delay and horizontal spatial component values ​​of each reference station. The horizontal spatial component values ​​are calculated using empirical models (such as the Saastamoinen model, Hopfield model, or UNB3m model). These models primarily reflect the large-scale planar delay caused by regional atmospheric conditions and are independent of station elevation. The elevation variation component observations are obtained by calculating the difference between the measured tropospheric delay and the horizontal spatial component values. This separation method effectively distinguishes the planar component and the elevation-related component in the tropospheric delay, making it particularly suitable for areas with large elevation variations and avoiding correction errors caused by neglecting the influence of elevation in traditional methods.

[0024] In the parameterized function model establishment stage, the observed elevation change components of each reference station are compared with the corresponding station elevations in the linear function model. ;in, The elevation change component observations of the reference station. The station elevation for reference. , For parameters to be estimated, when using the weighted least squares method for adjustment calculations, the construction of the weight matrix P takes into account the differences in station elevations, specifically expressed as follows: , in The network average elevation or the elevation of the main reference station. It is the cofactor matrix of the baseline. Let be the cofactor of the i-th reference station; and Hi be the station elevation of the i-th reference station. The optimal estimates of parameters a and b are obtained by solving using the least squares method, and the variance-covariance matrix of the parameter estimates is calculated, which contains the accuracy information of the model parameters. The design of this weight matrix considers the impact of station elevation differences on the model parameter estimation. The greater the difference between the reference station elevation and the network average elevation, the larger the corresponding cofactor QLi value and the smaller the weight 1 / QLi. This design ensures that reference stations with large elevation differences receive appropriate weights in the modeling, avoiding excessive influence of stations with similar elevations on the model parameter estimation.

[0025] The diagonal elements of the weight matrix P are represented as follows: ; This weighting mechanism assigns higher weights to reference stations whose elevations are far from the network average in the modeling process, as these stations contribute more to the modeling of elevation-related components and more effectively constrain the estimation of the slope parameter b. The weight matrix is ​​a diagonal matrix, indicating that the observations of each reference station are independent and do not consider spatial correlations. This aligns with the physical characteristic that elevation variation components are primarily influenced by the vertical atmospheric structure.

[0026] In the accuracy transfer step, based on the variance-covariance matrix obtained in step 102, an extended unit weight variance scaling factor is introduced. A precision transfer model is constructed, where n is the number of reference stations and t is the number of parameters to be estimated. The factor K is the core of precision transfer, used to adjust the precision transfer weights from the reference station network to the user end. Based on the error propagation law, the standard deviation (STD) of the elevation variation component at a specific location at the user end is calculated. The specific process includes constructing the user end design vector. The error propagation matrix A is calculated, along with the extended cofactor matrix integrating the reference station and the user terminal. When the reference station network is sparse, the K factor automatically amplifies the accuracy error to avoid transmission distortion; when the reference stations are dense, the K factor appropriately reduces the error range to ensure the reliability of accuracy transmission, enabling the user terminal to obtain tropospheric delay elevation component accuracy information that matches its location.

[0027] When calculating the user's location, the calculation will be based on the parameter estimates a, b, and the user's own elevation. The calculated tropospheric delay prior value (e.g., ΔZTDu=a+b·) The virtual observations, along with their corresponding standard deviation (STD), are input into the positioning system as virtual observations. During the positioning process, these virtual observations form strong constraints, effectively reducing the correlation between the tropospheric delay parameter, elevation coordinate parameter, and phase ambiguity parameter. This constraint makes it easier to fix the ambiguity parameter to integer values, especially under complex terrain conditions, significantly improving the success rate of ambiguity fixation and positioning accuracy, while also accelerating the convergence speed. This enables the GNSS augmented positioning system to provide more stable and reliable high-precision positioning services in practical applications.

[0028] This method is applicable to both network RTK and PPP-RTK augmented positioning modes. Therefore, this application, through reasonable decomposition of tropospheric delay and precise modeling of elevation components, ensures that the accuracy information of the reference station network is accurately transmitted to the user end, thus forming strong constraints in the user-end positioning calculation. Since there is a clear physical relationship between elevation components and station elevations, the elevation components at different elevation locations can be accurately predicted through a parametric function model. The standard deviation of the elevation components calculated using the error propagation law reflects the accuracy information of this prediction, enabling the user end to perform more accurate positioning calculations based on the accuracy assessment of the reference station network. This precise transmission of accuracy information effectively reduces the correlation between ambiguity parameters and tropospheric parameters, as the accuracy information provides reliable prior constraints for parameter calculation, avoiding the enhanced parameter correlation problem caused by the lack of accuracy information in traditional methods. Thus, high-precision ambiguity fixation and positioning calculations can still be achieved under complex terrain conditions.

[0029] Based on the above embodiments, in this embodiment, the elevation change component observation values ​​of each reference station are obtained based on the horizontal spatial component values ​​and the measured tropospheric delay values ​​of each reference station in the GNSS positioning system, specifically including steps 1011 to 1012: Step 1011: Based on the empirical model, calculate the horizontal spatial component values ​​of each reference station in the GNSS positioning system.

[0030] Specifically, in a GNSS-enhanced positioning system, the horizontal spatial component values ​​of each reference station are first calculated based on empirical models. The horizontal spatial component values ​​refer to the large-scale planar components of tropospheric delay caused by regional atmospheric conditions, independent of station elevation, reflecting a relatively uniform atmospheric state within the region. Mature empirical tropospheric models, such as the Saastamoinen model, the Hopfield model, or the UNB3m model, are used in the calculations. These models are constructed based on atmospheric physical properties and historical meteorological data, effectively reflecting the regional average atmospheric conditions. In practical applications, when real-time meteorological data is lacking, simplified calculations can be performed using standard atmospheric parameters. For example, the UNB3m model provides standard atmospheric parameters based on annual day and local time.

[0031] Step 1012: Obtain the elevation change component observation value of each reference station by calculating the difference between the measured tropospheric delay value and the horizontal spatial component value.

[0032] Specifically, the elevation variation component observation values ​​of each reference station are obtained by calculating the difference between the measured tropospheric delay value and the horizontal spatial component value. The measured tropospheric delay value is obtained from GNSS observation data after precise data processing, and is usually calculated using a fixed double-difference ambiguity method. The specific expression for the difference calculation is ΔZTD = ZTDmeas - ZTDplane, where ZTDmeas represents the measured tropospheric delay value.

[0033] The physical significance of the elevation variation component observation lies in its quantification of the influence of station elevation on tropospheric delay. During GNSS signal propagation, signals from low-altitude stations need to pass through a thicker layer of atmosphere, thus experiencing greater tropospheric delay. By decomposing the overall tropospheric delay into a horizontal spatial component and an elevation variation component, the spatial distribution characteristics of tropospheric delay can be described more accurately. The horizontal spatial component mainly reflects regional atmospheric conditions and is suitable for processing using planar interpolation methods; while the elevation variation component is closely related to the station elevation and requires the establishment of a dedicated elevation correlation model. This separation method is particularly suitable for areas with large elevation variations, such as mountainous or hilly areas, where traditional methods often suffer from a significant decrease in correction accuracy due to neglecting the influence of elevation.

[0034] To ensure calculation accuracy, ZTDmeas and ZTDplane should use the same reference system and time base during the difference calculation process to avoid introducing additional noise due to inconsistencies in coordinate systems or time synchronization errors.

[0035] Based on the above embodiments, in this embodiment, before calculating the horizontal spatial component values ​​of each reference station in the GNSS positioning system based on an empirical model, the method further includes: Based on the location information of each reference station, one model is selected from the candidate models as the empirical model.

[0036] Specifically, in GNSS augmentation positioning systems, selecting an appropriate empirical model based on the location information of each reference station is a crucial prerequisite for ensuring the accuracy of horizontal spatial component calculations. The location information of a reference station mainly includes its latitude, longitude, altitude, and the climate characteristics of its region. These factors directly affect the physical properties of the troposphere. For example, reference stations located in mid-to-high latitude regions (latitude greater than 30°) typically use the Saastamoinen model, which is based on atmospheric refraction theory and calculates the horizontal spatial components using station latitude, elevation, and meteorological parameters. Reference stations located near the equator or in the tropics are better suited to the UNB3m model, which considers the influence of annual day length and local time on atmospheric parameters and better reflects the significantly seasonal atmospheric characteristics of the tropics.

[0037] For reference station networks with significant altitude differences, model selection must consider elevation adaptability. When the reference station altitudes are generally below 1000 meters, the Hopfield model is preferred due to its accurate description of atmospheric refraction in low-altitude areas. However, when the network includes high-altitude stations (e.g., above 2000 meters), a simplified version of the Saastamoinen model is more suitable because it adapts better to atmospheric pressure variations in high-altitude regions. In practical applications, the system automatically analyzes the geographical distribution characteristics of the reference station network. When the network spans multiple climate zones or experiences drastic elevation changes, a zonal modeling strategy can be adopted, selecting the most suitable empirical model for each sub-region with different geographical characteristics, rather than forcing the use of a single model.

[0038] The choice of empirical model directly affects the calculation accuracy of the horizontal spatial component, and consequently the extraction quality of the elevation change component observations. The Saastamoinen model exhibits high accuracy when supported by real-time meteorological data, but in the absence of such data, the UNB3m model, with its built-in standard atmospheric parameter library (based on annual day and local time), provides more stable calculation results. While the Hopfield model has higher computational complexity, its detailed description of atmospheric vertical stratification makes it perform better in regions with complex meteorological conditions. During model selection, the system evaluates the historical performance data of each model under the current geographical environment, prioritizing models with higher verification accuracy under similar conditions.

[0039] In the specific implementation of model selection, the system first calculates the geographic center point and coverage area of ​​the reference station network, determining whether it crosses important geographic boundaries (such as the equator, the tropics, etc.). For networks with small coverage areas and gentle elevation changes, a single model is typically used; while for networks with large coverage areas or dramatic elevation changes, a zoning strategy is implemented, dividing the network into several sub-regions with similar geographic characteristics, and each sub-region independently selects the most suitable empirical model. This dynamic selection mechanism ensures the regional adaptability of the horizontal spatial component calculation, laying the foundation for the accurate extraction of subsequent elevation change components and avoiding systematic biases caused by inappropriate model selection. Especially under complex terrain conditions, reasonable model selection can significantly improve the overall accuracy of tropospheric delay correction.

[0040] In step 102, a parametric function model is established based on the observed elevation change components of each reference station and the station elevation. This parametric function model adopts the linear form ΔZTD=a+b·H. The selection of this linear model is based on actual observations of the physical characteristics of the troposphere: within a certain elevation range, the tropospheric delay changes approximately linearly with elevation because atmospheric density decreases approximately exponentially with increasing altitude, while in local regions it can be approximated as a linear change.

[0041] The method employs weighted least squares to solve for the parameter estimates of the parametric function model and calculates the variance-covariance matrix of the parameter estimates. Specifically, this involves: first, constructing a design matrix based on the elevation component observations and station elevations of each reference station; then, constructing a weight matrix based on the absolute difference between the elevations of each reference station and the average elevation; next, combining the design matrix, weight matrix, and elevation component observations, using weighted least squares to solve for the parameter estimates of the parametric function model; then, using the elevation component observations and the solved parameter estimates, calculating the residual vector; and finally, calculating the unit weight variance using the residual vector and the weight matrix; and finally, calculating the variance-covariance matrix of the parameter estimates based on the unit weight variance and the design matrix.

[0042] Specifically, in solving the parametric function model, a design matrix is ​​first constructed based on the observed elevation components and station elevations of each reference station. The design matrix Br adopts the form [1H], where the first column is a vector of all 1s corresponding to the constant term 'a' in the model, and the second column contains the slope parameter 'b' corresponding to the actual elevation values ​​of each reference station. For example, when the elevations of reference stations A, B, C, and D are 50 meters, 100 meters, 150 meters, and 200 meters, respectively, the design matrix is ​​a 4-row, 2-column matrix: the first row is

[150] , the second row is

[1100] , the third row is

[1150] , and the fourth row is

[1200] . This design reflects the structural characteristics of the linear model ΔZTD=a+b·H, ensuring a clear mathematical relationship between the observed values ​​and the parameters to be estimated. The number of rows in the design matrix equals the number of reference stations, and the number of columns equals the number of parameters to be estimated. Its rank must equal the number of columns to ensure that the parameters are estimable. Therefore, it is required that the elevations of the reference stations are not all the same; at least two reference stations with different elevations are needed to solve the linear model.

[0043] The weight matrix is ​​constructed based on the absolute difference between the elevation of each reference station and the network average elevation, specifically expressed as follows: , Where Href is the network average elevation, calculated as Href = (H1 + H2 + ... + Hn) / n. For example, when the elevations of the four reference stations are 50 meters, 100 meters, 150 meters, and 200 meters, the network average elevation is 125 meters. The absolute differences between each reference station and the average elevation are 75 meters, 25 meters, 25 meters, and 75 meters, respectively, with corresponding weights of 1 / 75, 1 / 25, 1 / 25, and 1 / 75. This weighting mechanism ensures that reference stations with elevations far from the network average receive higher weights in the modeling, as these stations contribute more to the modeling of elevation-related components and more effectively constrain the estimation of the slope parameter b. The weight matrix is ​​a diagonal matrix, indicating that the observations of each reference station are independent and spatial correlation is not considered. This aligns with the physical characteristic that elevation variation components are primarily influenced by the vertical atmospheric structure.

[0044] Parameter estimation is achieved by solving for the parameter vector using weighted least squares. The calculation formula is: Where Lr is the vector of elevation component observations. This formula minimizes the weighted sum of squared residuals V. The P·V parameter estimates are statistically optimal. Taking four reference stations as an example, when the observed elevation components are 1.2 mm, 0.9 mm, 0.6 mm, and 0.3 mm, the calculated parameter estimates might be a = 1.5 mm and b = -0.006 mm / m, indicating that for every 100 m increase in elevation, the tropospheric delay decreases by approximately 0.6 mm. Parameter a reflects the tropospheric delay shift at the reference elevation and is significantly affected by meteorological conditions; parameter b is mainly related to the vertical structure of the atmosphere and is relatively stable on short timescales. During the solution process, (Br... The PBr matrix must be invertible, which requires that the elevation distribution of the reference stations not be too concentrated, otherwise the matrix will be ill-conditioned and the parameter estimates will be unstable.

[0045] The residual vector passes through The calculated residual represents the difference between the model's predicted values ​​and the actual observed values. For example, when the parameter estimates are a = 1.5 mm and b = -0.006 mm / m, for a reference station at an elevation of 50 meters, the model's predicted value is 1.5 - 0.006 × 50 = 1.2 mm. If the actual observed value is also 1.2 mm, then the residual is 0. Analyzing the residual vector is an important means of evaluating the model's fit quality. Ideally, the residuals should exhibit a random distribution without significant systematic bias. When the absolute values ​​of the residuals are generally large or show regular changes, it indicates that the linear model may be insufficient to describe the relationship between elevation and tropospheric delay, and a more complex model form may need to be considered, such as a piecewise linear model or the introduction of quadratic terms.

[0046] Unit weighted variance through = The value is calculated as / (nt), where n is the number of reference stations and t is the number of parameters to be estimated (t=2 in this embodiment). The unit weight variance reflects the internal consistency accuracy of the observations. When its value is close to 1, it indicates that the weight allocation of the observations is reasonable and the model matches the observation data well. When the value is significantly greater than 1, it indicates that the accuracy of the observations is overestimated or that there is a systematic error in the model. When the value is significantly less than 1, it indicates that the accuracy of the observations is underestimated.

[0047] The variance-covariance matrix of the parameter estimates is obtained through D= ·(Br ·P·Br) - ¹The calculated matrix contains information on the accuracy of the parameter estimates. The diagonal elements represent the variance of each parameter, and the standard deviation is obtained by taking the square root, reflecting the reliability of the parameter estimates; the off-diagonal elements represent the covariance between parameters, reflecting the correlation between parameters.

[0048] Based on the above embodiments, in this embodiment, the standard deviation of the elevation variation component of the user's location in the GNSS positioning system is calculated using the variance-covariance matrix and the error propagation law, specifically including steps 1031 to 1032: Step 1031: Calculate the extended unit weight variance scaling factor based on the number of reference stations in the GNSS positioning system and the number of parameters to be estimated in the parameterized function model.

[0049] The formula for calculating the extended unit weight variance scaling factor K is: ; Where n is the number of reference stations in the GNSS positioning system, and t is the number of parameters to be estimated in the parameterized function model (2 in this embodiment, i.e., parameters a and b in the linear model ΔZTD=a+b·H).

[0050] The mathematical derivation of this factor is as follows: First, we define the concept of unit weight variance. Unit weight variance reflects the internal consistency accuracy of observations, and its calculation formula is: ; Where V is the residual vector, P is the weight matrix, n is the number of reference stations, and t is the number of parameters to be estimated.

[0051] When considering the inclusion of the user terminal in the network, it is necessary to calculate the unit weight variance that includes n+1 points (n reference stations and 1 user terminal). and The relationship between them can be derived using the following formula:

[0052] Since the user terminal is located within the reference station network, the characteristics of its interpolation residuals are similar to those of the residuals of surrounding reference stations. Based on the error distribution pattern, the following assumptions can be made:

[0053] Therefore, and The following expressions are between them: ; Therefore, the extended unit weight variance scaling factor K is defined as: ; The physical meaning of the K-factor lies in its reflection of the proportion of change in unit weight variance when expanding from n reference stations to n+1 points including the user terminal. Its value is always less than 1 and approaches 1 as the number of reference stations increases. When the number of reference stations is small, the K value is small, indicating that the reference station network has a weaker constraint on the accuracy of the user terminal. When the number of reference stations is large, the K value is close to 1, indicating that the reference station network can more accurately transmit accuracy information. The design of the K-factor considers the impact of the geometric distribution characteristics of the reference station network on accuracy transmission. When the reference station network is sparse or unevenly distributed, the K-factor automatically adjusts to reflect this uncertainty, avoiding the transmission of overly optimistic accuracy information to the user terminal.

[0054] Step 1032: Introduce the extended unit weight variance scaling factor and variance-covariance matrix into the accuracy transfer model constructed based on the error propagation law, and calculate the standard deviation of the elevation change component of the user's location in the GNSS positioning system.

[0055] Specifically, the construction of the accuracy transfer model organically combines the K-factor with the variance-covariance matrix. The specific process is as follows: First, define the interpolation error. : ; in, Design vectors for the user end. For parameter estimation vectors, This represents the true value of the elevation component on the user's end.

[0056] Will Substituting into the above equation, we get: ; in, , For identity matrix, subscript Indicates a reference station. This refers to the user side.

[0057] According to the law of error propagation, the variance of the interpolation error is: ; in, Let be the variance-covariance matrix of the observations.

[0058] Will Substituting into the above equation, we get: ; Furthermore, Substituting, we finally get: = ·Q(n+1)×(n+1) Calculate the variance of the interpolation error at the user end.

[0059] This formula is based on a strict error propagation law and takes into full account the impact of reference station observation accuracy, network geometry distribution, and user terminal location on accuracy transfer.

[0060] The standard deviation of the elevation change component at the user end is obtained through The calculated value quantifies the uncertainty of the tropospheric delay elevation component at the user end. Unlike traditional methods that only provide tropospheric delay correction values ​​without precision information, this application transmits the precision information of the reference station network to the user end through a rigorous statistical model, enabling the user end to obtain a personalized precision assessment based on its own location characteristics. In practical applications, when the user is located inside the reference station network and the elevation is close to the network average, the STD value is small; when the user is located at the network edge or the elevation differs significantly from the network average, the STD value increases accordingly, accurately reflecting the impact of location on correction precision. This precision transmission mechanism based on the error propagation law effectively solves the fundamental problem of lacking precision information in existing technologies, providing reliable prior precision information for user end positioning calculations.

[0061] Based on the above embodiments, in this embodiment, the standard deviation of the elevation variation component is used as accuracy information, and together with the tropospheric delay prior value obtained based on parameter estimation, it is used as data input to participate in the user-end positioning calculation, specifically including steps 1041 to 1043: Step 1041: Use the standard deviation of the elevation variation component as accuracy information.

[0062] In the user-end positioning calculation process, the standard deviation of the elevation variation component is directly used as accuracy information to construct reliable prior constraints. This standard deviation is a measure of the uncertainty of the user-end elevation variation component calculated in step 103 using the error propagation law. This accuracy information directly reflects the coverage quality of the reference station network to the user-end location and the stability of atmospheric conditions, and has significant value in positioning calculation. Unlike traditional methods that only provide tropospheric delay correction values ​​without accuracy assessment, this method transmits the accuracy information of the reference station network to the user end through a rigorous statistical model, enabling the user end to obtain personalized accuracy assessments based on its own location characteristics. When the user is located inside the reference station network and the elevation is close to the network average, the standard deviation is small; when the user is located at the network edge or the elevation differs significantly from the network average, the standard deviation increases accordingly, accurately reflecting the impact of location on correction accuracy.

[0063] Step 1042: Based on the parameter estimates of the parameterized function model and the station elevation of the user's location, obtain the tropospheric delay prior value.

[0064] The prior value of tropospheric delay is obtained based on parameter estimates from the parametric function model and the elevation calculation of the user terminal site. Specifically, the parameters a and b obtained in step 102 and the approximate elevation Hu obtained by the user terminal through preliminary positioning are substituted into the linear model. The approximate elevation of the user terminal can be obtained through various methods, including coarse positioning based on GNSS pseudorange observations, digital elevation model queries, or input of external elevation information. In the actual system, this prior value calculation process is performed in real time and dynamically updated as the user's location changes, ensuring that the tropospheric delay correction always matches the current geographical location. It is worth noting that the physical meaning of the prior value is the elevation-related tropospheric delay component of the user terminal location, rather than the complete tropospheric delay. The complete tropospheric delay also needs to include the horizontal spatial component (calculated through an empirical model).

[0065] Step 1043: Use the accuracy information and the tropospheric delay prior value as virtual observation values ​​to participate in the user-end positioning solution. Specifically, this includes: determining the constraint weights based on the accuracy information; constructing an observation equation that includes the tropospheric delay virtual observation value based on the tropospheric delay prior value and the constraint weights; and integrating the constructed observation equation into the user-end positioning solution model to participate in the user-end positioning solution.

[0066] Specifically, the process of using accuracy information and prior tropospheric delay values ​​as virtual observations in the positioning solution first determines constraint weights based on the accuracy information. The formula for calculating the constraint weights is w = 1 / σ², where σ is the standard deviation of the elevation variation component. This weight reflects the reliability of the prior information; the smaller the standard deviation, the larger the weight, indicating that the prior value is more reliable and should be given higher confidence in the solution. Subsequently, based on the prior tropospheric delay values ​​and constraint weights, an observation equation incorporating virtual tropospheric delay observations is constructed. In the double-difference observation model, the added virtual observation equation is Truij = ΔZTDu, where Truij represents the double-difference tropospheric delay, and ΔZTDu is the calculated prior value. This equation is incorporated into the observation equation system through the weight w, forming a strong constraint on the tropospheric delay parameter.

[0067] Finally, the constructed virtual observation equations are integrated into the user-end positioning solution model, participating in the solution together with the original GNSS observation equations. Within the least squares or Kalman filter framework, the virtual observation equations, together with other observation equations, constitute an extended set of observation equations, whose matrix form is [observation equation;Truij] = [design matrix; 0]·[parameters] + [observation noise; virtual observation noise]. This integration effectively reduces the correlation between tropospheric delay parameters, elevation coordinate parameters, and phase ambiguity parameters, making it easier to fix the ambiguity parameters as integer values.

[0068] The complete observation equations are as follows: Dual-frequency carrier phase observation equation: ; ; Dual-frequency pseudorange observation equation: ; ; Tropospheric Delayed Virtual Observation Equation: ; in: , The wavelengths of the first and second frequencies are respectively. , Satellites , With receiver , The double-difference carrier phase observations between , Satellites , With receiver , The double-difference pseudorange observations between For satellite , With receiver , The double difference geometric distance between them The double difference for tropospheric delay. , For double-difference integer ambiguity, The double difference for ionospheric delay. , For the first and second frequencies, The tropospheric delay prior value is obtained based on parameter estimation and user-end elevation calculation.

[0069] In the actual solution process, the tropospheric delay virtual observation equation Based on its standard deviation Assign corresponding weights This creates a strong constraint on the tropospheric delay parameter. This constraint effectively reduces the correlation between the tropospheric delay parameter and the elevation coordinate parameter and phase ambiguity parameter, significantly improving the success rate of ambiguity fixation and positioning accuracy.

[0070] Especially in complex terrain conditions, when there is a significant elevation difference between the reference station and the user terminal, this constraint can significantly improve the success rate of ambiguity fixation and positioning accuracy.

[0071] To clearly illustrate the present invention, a hypothetical CORS network environment will be used as the first embodiment for detailed description below. Those skilled in the art will understand that this embodiment is for illustrative purposes only and should not be construed as limiting the invention. This example is based on a known CORS network environment, which includes a primary reference station M and three secondary reference stations A, B, and C. The rover U is located within this network.

[0072] This step first acquires GNSS observation data from the primary reference station M and auxiliary reference stations A, B, and C in the CORS network. By fixing the double-difference ambiguity, the measured tropospheric delay values ​​for each station are obtained through precise data processing. .

[0073] The Saastamoinen model was selected to calculate the large-scale planar components of each reference station. Including: the main reference station Auxiliary reference station: .

[0074] The elevation component observations of each reference station are extracted by using the difference between the measured tropospheric delay and the plane component: Elevation component observations of the main reference station M: ; Elevation component observations of auxiliary reference station A: ; Elevation component observations of auxiliary reference station B: ; Elevation component observations of auxiliary reference station C: ; The elevation component observation vector is obtained: .

[0075] Based on elevation data from various reference stations: , , , Establish a linear function model: ; in, This represents the tropospheric delay component in the elevation direction. Indicates elevation. , This indicates that the parameters are estimated, and the parameter estimates can be solved using the least squares method.

[0076] Based on the structural characteristics of the linear model and the elevation data of each reference station, a design matrix B is constructed to establish the linear relationship between the observed values ​​and the parameters to be estimated, in the following form: ; To balance the weight contribution of different elevation reference stations in the modeling, the network average elevation is calculated based on the elevation data of each reference station. : ; Construct the weight matrix P: ; Solving the parameter vector using the least squares method The specific calculation formula is as follows: ; The residual vector is calculated as follows: ; Calculate the unit weight variance based on the residual vector and the weight matrix: ; Based on the approximate elevation H obtained by the rover U through preliminary positioning (the accuracy meets the model input requirements), and referring to the construction logic of the reference station design matrix, a user-end design vector B is constructed to establish the relationship between the user-end elevation and its components. The specific form is as follows: ; Substituting the obtained model parameter vector into the linear model, the elevation correction ΔZTD for the tropospheric delay at the user end is calculated. This correction is the core parameter for tropospheric delay compensation at the user end, and the specific calculation formula is as follows: ; Based on the error propagation law, the error propagation matrix A is calculated. This matrix is ​​used to characterize the propagation law of the correction amount of the reference station's observation error to the user end. The specific calculation formula is as follows: ; Based on statistical theory, an extended unit weight variance scaling factor K is introduced, and the specific calculation formula is as follows: ; To integrate the accuracy information from the reference stations and the user terminal, an extended cofactor matrix Q is constructed, which includes four reference stations and one user terminal. This matrix is ​​a diagonal matrix, and its specific form is as follows: ; Calculate the variance of the interpolation error at the user end. This comprehensively reflects the error level of the correction amount on the user side. The specific calculation formula is as follows: ; The standard deviation (STD) of the user-end elevation component is obtained by taking the square root of the variance of the interpolation error. The specific calculation formula is as follows: ; Using the calculated ΔZTD (tropospheric delay correction) and STD (standard deviation), a tropospheric weighted model is constructed based on the weighting principle of the observation equation to achieve accurate compensation for tropospheric delay.

[0077] Next, the ionospheric delay correction for rover U is calculated using the optimal interpolation method. .

[0078] Then As virtual observations, their standard deviations are used as prior accuracy information and assigned corresponding weights.

[0079] Then, an observation equation containing virtual observations with tropospheric delay is constructed, and the location solution is obtained by weighted least squares method.

[0080] In some optional embodiments, the observation equations specifically include: ; ; ; ; ; in, It is a double-difference geometric distance. This is the correction amount for tropospheric delay interpolation. and This represents the double-difference integer ambiguity.

[0081] By solving the observation equation system containing virtual observations using the weighted least squares method, the strong constraints formed by these virtual observations can effectively reduce the correlation between the tropospheric delay parameter, elevation coordinate parameter, and phase ambiguity parameter, thereby significantly improving the success rate and convergence speed of phase ambiguity fixation. Ultimately, high-precision real-time positioning of the rover station can still be achieved even in complex terrain or harsh observation environments.

[0082] The above description is only a preferred embodiment of this application, but the scope of protection of this invention is not limited thereto. Any person skilled in the art can make equivalent substitutions or changes based on the technical solution and inventive concept of this invention within the scope of the technology disclosed in this invention. Whether based on different numbers of reference station networks, different empirical models or different adjustment methods, they should all be covered within the scope of protection of this invention.

[0083] This application overcomes the shortcomings of traditional methods that ignore the influence of elevation and lack accuracy transfer. Through reasonable decomposition, precise modeling and accuracy transfer of tropospheric delay, it effectively reduces the strong correlation between parameters, significantly improves the success rate of ambiguity fixation and positioning accuracy, and is particularly suitable for high-precision positioning applications of network RTK and PPP-RTK under complex terrain conditions.

[0084] The functions of each module in the above-mentioned tropospheric delay accuracy transfer system correspond to the steps in the above-mentioned tropospheric delay accuracy transfer method embodiment, and their functions and implementation processes will not be described in detail here.

[0085] Thirdly, embodiments of this application provide a tropospheric delay accuracy transmission device, which can be a personal computer (PC), laptop computer, server, or other device with data processing capabilities.

[0086] In this embodiment of the application, the tropospheric delay accuracy transfer device may include a processor, a memory, a communication interface, and a communication bus.

[0087] The communication bus can be of any type and is used to interconnect the processor, memory, and communication interface.

[0088] The communication interface includes input / output (I / O) interfaces, physical interfaces, and logical interfaces used for interconnecting devices within the tropospheric delay accuracy transfer device, as well as interfaces used for interconnecting the tropospheric delay accuracy transfer device with other devices (such as other computing devices or user equipment). Physical interfaces can be Ethernet interfaces, fiber optic interfaces, ATM interfaces, etc.; user equipment can be displays, keyboards, etc.

[0089] Memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.

[0090] The processor can be a general-purpose processor, which can call the tropospheric delay precision transfer program stored in memory and execute the tropospheric delay precision transfer method provided in the embodiments of this application. For example, the general-purpose processor can be a central processing unit (CPU). The method executed when the tropospheric delay precision transfer program is called can be referred to in the various embodiments of the tropospheric delay precision transfer method of this application, and will not be repeated here.

[0091] Fourthly, embodiments of this application also provide a computer-readable storage medium.

[0092] The present application has a computer-readable storage medium storing a tropospheric delay precision transfer program, wherein when the tropospheric delay precision transfer program is executed by a processor, it implements the steps of the tropospheric delay precision transfer method as described above.

[0093] The method implemented when the tropospheric delay accuracy transfer procedure is executed can be referred to in various embodiments of the tropospheric delay accuracy transfer method of this application, and will not be repeated here.

[0094] It should be noted that the sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0095] The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not indicate a sequence, nor do they limit "first," "second," and "third" to different types.

[0096] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.

[0097] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.

[0098] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.

[0099] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.

[0100] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for transferring tropospheric delay accuracy, characterized in that, It includes: Based on the horizontal spatial component values ​​and measured tropospheric delay values ​​of each reference station in the GNSS positioning system, the observed values ​​of the elevation change component of each reference station are obtained. Based on the elevation change component observations and station elevations of each reference station, a parameterized function model is established. The weighted least squares method is used to solve the parameter estimates of the parameterized function model, and the variance-covariance matrix of the parameter estimates is calculated. Based on the variance-covariance matrix, the standard deviation of the elevation variation component of the user's location in the GNSS positioning system is calculated by the error propagation law. The standard deviation of the elevation variation component is used as the accuracy information, and together with the tropospheric delay prior value obtained based on parameter estimation, it is used as data input to participate in the user-end positioning solution.

2. The tropospheric delay accuracy transfer method as described in claim 1, characterized in that, Based on the horizontal spatial component values ​​and measured tropospheric delay values ​​of each reference station in the GNSS positioning system, the observed values ​​of the elevation variation components of each reference station are obtained, specifically including: Based on an empirical model, the horizontal spatial component values ​​of each reference station in the GNSS positioning system are calculated. The elevation change component of each reference station is obtained by calculating the difference between the measured tropospheric delay value and the horizontal spatial component value.

3. The tropospheric delay accuracy transfer method as described in claim 2, characterized in that, Before calculating the horizontal spatial component values ​​of each reference station in the GNSS positioning system based on an empirical model, the method further includes: Based on the location information of each reference station, one model is selected from the candidate models as the empirical model.

4. The tropospheric delay accuracy transfer method as described in claim 1, characterized in that, The parameterized function model is as follows: ； in, The elevation change component observations of the reference station. The station elevation for reference. , For parameters with estimation.

5. The tropospheric delay accuracy transfer method as described in claim 1, characterized in that, The weighted least squares method is used to solve for the parameter estimates of the parameterized function model, and the variance-covariance matrix of the parameter estimates is calculated, specifically including: A design matrix is ​​constructed based on the elevation component observations and station elevations of each reference station; A weight matrix is ​​constructed based on the absolute difference between the elevation of each reference station and the average elevation. By combining the design matrix, weight matrix, and elevation component observations, the parameter estimates of the parameterized function model are obtained using the weighted least squares method. The residual vector is calculated using the elevation component observations and the estimated parameters obtained from the solution. Calculate the unit weight variance using the residual vector and the weight matrix; Based on the unit weight variance and the design matrix, calculate the variance-covariance matrix of the parameter estimates.

6. The tropospheric delay accuracy transfer method as described in claim 1, characterized in that, Based on the variance-covariance matrix, the standard deviation of the elevation variation component of the user's location in the GNSS positioning system is calculated using the error propagation law, specifically including: Based on the number of reference stations in the GNSS positioning system and the number of parameters to be estimated in the parametric function model, calculate the extended unit weight variance scaling factor. The extended unit weight variance scaling factor and variance-covariance matrix are introduced into the accuracy transfer model based on the error propagation law, and the standard deviation of the elevation variation component of the user's location in the GNSS positioning system is calculated.

7. The tropospheric delay accuracy transfer method as described in claim 6, characterized in that, The formula for calculating the extended unit weight variance scaling factor K is as follows: ； Where n is the number of reference stations in the GNSS positioning system, and t is the number of parameters to be estimated in the parametric function model.

8. The tropospheric delay accuracy transfer method as described in claim 1, characterized in that, The standard deviation of the elevation variation component is used as accuracy information, and together with the tropospheric delay prior value obtained based on parameter estimation, it is used as data input to participate in the user-end positioning solution, specifically including: Use the standard deviation of elevation variation components as accuracy information; Based on the parameter estimates of the parameterized function model and the station elevation of the user's location, the prior value of tropospheric delay is obtained. The accuracy information and the prior value of tropospheric delay are used as virtual observation values ​​to participate in the user-end positioning solution.

9. The tropospheric delay accuracy transfer method as described in claim 8, characterized in that, Accuracy information and prior tropospheric delay values ​​are used as virtual observations in the user-end positioning calculation, specifically including: Constraint weights are determined based on accuracy information; Based on the tropospheric delay prior value and constraint weights, an observation equation containing virtual observations of the tropospheric delay is constructed. The constructed observation equations are integrated into the user-end positioning solution model and participate in the user-end positioning solution.

10. A tropospheric delay accuracy transfer system, characterized in that, It includes: The first module is used to obtain the elevation change component observation value of each reference station based on the horizontal spatial component value and the measured value of tropospheric delay of each reference station in the GNSS positioning system. The second module is used to establish a parameterized function model based on the elevation change component observations and station elevations of each reference station, and to solve the parameter estimates of the parameterized function model using the weighted least squares method, and to calculate the variance-covariance matrix of the parameter estimates. The third module is used to calculate the standard deviation of the elevation variation component of the user's location in the GNSS positioning system based on the variance-covariance matrix and the error propagation law. The fourth module is used to take the standard deviation of the elevation change component as accuracy information, and together with the tropospheric delay prior value obtained based on parameter estimation, it is used as data input to participate in the user-end positioning solution.

Images