A global ionospheric model precision factor generation method and device

Abstract

The application discloses a global ionospheric model precision factor generation method and device, and belongs to the technical field of satellite navigation and ionospheric modeling. The method comprises the following steps: inversing ionospheric vertical total electron content based on carrier phase smoothing pseudo-range; extracting ionospheric unmodeled error by using a global ionospheric grid map; analyzing the skewness and kurtosis distribution characteristics of the error; combining experience error optimization and measured error driving to reconstruct the ionospheric model precision factor, and improving the reliability and stability of the precision factor by introducing monitoring point classification, scale adjustment factor, inflation factor time smoothing and other mechanisms; further proposing an ionospheric error precision consistency map and an RMS envelope probability index to verify the consistency of the precision factor and the measured error. The application can generate a precision factor that truly reflects the statistical characteristics of ionospheric error, provides effective support for the construction of an ionospheric random model in high-precision GNSS positioning, and significantly improves the positioning precision and convergence performance.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of satellite navigation and ionospheric modeling, and particularly relates to a global ionospheric model precision factor generation method and device. BACKGROUND

[0002] Global Navigation Satellite System (GNSS) plays a key role in high-precision positioning, especially in real-time precise point positioning (PPP). Although systematic errors such as satellite clock bias and orbit error can be effectively weakened by differential technology or model correction, the residual part of unmodeled errors such as ionospheric delay is still an important factor restricting the positioning accuracy and convergence speed due to its complex spatiotemporal variation characteristics. These unmodeled errors will cause the observation residuals to show temporal correlation, reducing the reliability of parameter estimation.

[0003] In GNSS positioning, ionospheric delay is one of the main error sources. Using ionosphere-free combinations or parameterized modeling can eliminate most of the first-order delay, but high-order effects and ionospheric disturbances are difficult to accurately describe. Empirical models such as the International Reference Ionosphere (IRI), Klobuchar model, and Global Ionospheric Maps (GIMs) can provide some degree of correction, but there are still limitations in characterizing the complex spatiotemporal characteristics of ionospheric errors. To improve positioning performance, a reasonable ionospheric random model needs to be established, and the key is to obtain precision information that can truly reflect the uncertainty of ionospheric model correction values.

[0004] Currently, there are two main types of ionospheric error reference information: one is error envelope for integrity monitoring, such as the grid vertical ionospheric error (GIVE) broadcast by the Satellite Based Augmentation System (SBAS). GIVE constructs error boundaries through conservative algorithms and is mainly used for anomaly data filtering, so it is not suitable for random model construction in precise positioning. The second is the precision information attached to the ionospheric model itself, such as the Root Mean Square (RMS) maps of GIMs published by the International GNSS Service (IGS) analysis centers. However, the RMS maps of different agencies differ greatly in physical meaning and generation algorithm, and their consistency and reliability are questionable. More importantly, this type of precision information is based on the internal consistency precision estimated by the error propagation law, which is too "optimistic" and fails to objectively reflect the actual error level at the grid point, resulting in limited effectiveness in high-precision GNSS positioning applications. The inflation factor introduced in existing methods relies heavily on empirical settings and lacks adaptability to the non-Gaussian statistical characteristics of ionospheric unmodeled errors.

[0005] Therefore, it is urgent to develop a precision factor generation method that can accurately characterize the actual error of global ionospheric models to support reliable ionospheric random model construction, thereby effectively improving the precision and reliability of GNSS precise positioning. SUMMARY

[0006] To solve the above technical problems, the present application provides a global ionospheric model precision factor generation method and device, which can extract the statistical characteristics of unmodeled errors from ionospheric residuals and generate more realistic and reliable precision factor information, thereby providing a unified precision reference for global ionospheric models, supporting reasonable ionospheric random model construction, and significantly improving the precision and convergence performance of GNSS positioning.

[0007] To achieve the above-mentioned purpose, the technical scheme adopted by the present application is as follows:

[0008] A global ionospheric model precision factor generation method, the method comprising:

[0009] Step S1, based on carrier phase smoothing pseudorange inversion to obtain the vertical total electron content VTEC measured value at the ionospheric piercing point;

[0010] Step S2, using global ionospheric grid map GIM, the VTEC model value at the piercing point is obtained by interpolation, and the difference between the VTEC measured value and the VTEC model value is calculated as ionospheric unmodeled error;

[0011] Step S3, analyze the distribution characteristics of the ionospheric unmodeled error;

[0012] Step S4, based on the distribution characteristics, the internal consistency precision of GIM model is optimized, and the actual observation residual at the piercing point is used for driving update, and the ionospheric model precision factor is calculated;

[0013] Step S5, the availability of the precision factor is judged by ionospheric error precision consistency, if available, go to the next step;

[0014] Step S6, output the final global ionospheric model precision factor.

[0015] On the other hand, the present application provides a global ionospheric model precision factor generation device, comprising:

[0016] Inversion module, for based on carrier phase smoothing pseudorange inversion to obtain the vertical total electron content VTEC measured value at the ionospheric piercing point;

[0017] Interpolation module, for using global ionospheric grid map GIM, the VTEC model value at the piercing point is obtained by interpolation, and the difference between the VTEC measured value and the VTEC model value is calculated as ionospheric unmodeled error;

[0018] an analysis module, configured to analyze distribution characteristics of the ionospheric unmodeled error, optimize the internal consistency accuracy of the GIM model based on the distribution characteristics, and calculate an ionospheric model accuracy factor by driving and updating using actual observation residuals at the piercing points;

[0019] an output module, configured to determine the availability of the accuracy factor by using ionospheric error accuracy consistency, and output a final global ionospheric model accuracy factor if the accuracy factor is available.

[0020] In a third aspect, the present application provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the aforementioned global ionospheric model accuracy factor generation method.

[0021] In a fourth aspect, the present application provides a computer-readable storage medium having stored executable instructions, which, when executed by a processor, enable the processor to implement the aforementioned global ionospheric model accuracy factor generation method.

[0022] The present application has the following beneficial effects:

[0023] The accuracy factor is calculated with high authenticity: the present application breaks through the limitations of the prior art relying on empirical parameters and Gaussian assumptions, and generates the accuracy factor by analyzing the non-Gaussian statistical characteristics (such as skewness and kurtosis) of the ionospheric unmodeled error, adaptively optimizing the model post-error, and driving and updating using actual observation residuals, so that the accuracy factor can more truly reflect the actual error level of the ionospheric model.

[0024] The reliability and practicality of the accuracy factor are improved: by introducing the classification mechanism of monitoring points and unmonitoring points, the scale adjustment factor, and the time domain smoothing operation of the inflation factor, the over-optimistic estimation of the accuracy information is effectively avoided, the continuity and stability of the accuracy information in space and time are ensured, and the practical needs of constructing the ionospheric random model in high-precision GNSS positioning are better met.

[0025] Effective accuracy factor verification means are provided: the ionospheric error accuracy consistency diagram (IEAD) and the RMS envelope probability (RMSBP) index are innovatively proposed, which can quantitatively evaluate the consistency of the accuracy factor and the measured error distribution, and provide intuitive and rigorous criteria for the effectiveness and reliability of the accuracy factor. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 a flowchart of the global ionospheric model accuracy factor generation method of the present application;

[0027] Figure 2A skewness and kurtosis sequence diagram of ionospheric error in the embodiment is shown;

[0028] Figure 3 An ionospheric precision factor distribution map obtained based on the application is shown;

[0029] Figure 4 An ionospheric error precision consistency diagram is shown. DETAILED DESCRIPTION

[0030] The application is further described below in combination with the drawings and embodiments.

[0031] The application provides a global ionospheric model precision factor generation method and device, which is based on carrier phase smoothed pseudorange observation and satellite hardware bias correction product, extracts ionospheric vertical total electron content (VTEC) observation in combination with projection function; calculates VTEC model value by using global ionospheric grid map (GIM) product, obtains ionospheric unmodeled error; analyzes ionospheric unmodeled error distribution characteristics and noise characteristics, calculates ionospheric unmodeled error skewness, kurtosis and the like; reconstructs ionospheric model precision factor by using empirical error optimization and actual error driving; proposes an ionospheric error precision consistency diagram judgment method to verify the effectiveness of the precision factor. The method has the advantages of high fitting degree and accurate ionospheric model precision representation, and is suitable for navigation enhancement application scenarios.

[0032] As shown in Figure 1 , a global ionospheric model precision factor generation method of the application specifically includes the following steps:

[0033] Step S1, ionospheric vertical total electron content (VTEC) inversion based on carrier phase smoothed pseudorange; wherein multi-frequency observation data are obtained by using a ground-based GNSS receiver, and carrier phase is used to smooth pseudorange observation to weaken the influence of multipath effect and observation noise; ionospheric slant total electron content (STEC) is extracted at ionospheric piercing point positions by using satellite hardware delay correction product; ionospheric projection function is introduced to convert the extracted STEC into VTEC at the piercing point.

[0034] Step S2, ionospheric unmodeled error extraction based on global ionospheric grid map (GIM); wherein according to the grid point total electron content information provided in the GIM product, in combination with the ionospheric piercing point position information, the bilinear interpolation method is used to obtain the VTEC model estimation value at the ionospheric piercing point, the model value is compared with the measured VTEC obtained based on the carrier phase smoothed pseudorange, and the difference between the two is the ionospheric unmodeled error.

[0035] Step S3, ionospheric unmodeled error distribution characteristic analysis; calculate ionospheric unmodeled error skewness, kurtosis and the like. Skewness

[0036] Skewness Skewness is defined by the third moment, and the calculation method is as follows:

[0037] (1)

[0038] Wherein, N is the sample size, S is the sample standard deviation, X is the sample observation value, E is the sample expectation. When Skewness is greater than 0, it is considered that the distribution is positively skewed, otherwise, it is negatively skewed.

[0039] Kurtosis, also known as kurtosis coefficient, is used to represent the characteristic number of the peak value of the probability density distribution curve at the mean value. Kurtosis reflects the sharpness of the peak, and the kurtosis value Kurtosis is defined by the fourth moment, and the calculation method is as follows:

[0040] (2)

[0041] Figure 2 The skewness and kurtosis sequences of ionospheric errors in 2017 are given. The upper graph (skewness sequence) shows the time sequence change of the asymmetry of ionospheric unmodeled error distribution. The farther the skewness value deviates from 0, the greater the degree of error distribution deviating from the normal distribution. The figure directly reveals that the ionospheric error generally exists non-Gaussian characteristics, and this characteristic changes dynamically with time, which proves that the direct use of model accuracy information based on Gaussian assumption is insufficient. The lower graph (kurtosis sequence) shows the time sequence change of the sharpness of the peak of ionospheric unmodeled error distribution. The kurtosis value greater than 3 (relative to the normal distribution) indicates that the error distribution has a sharper peak and a thicker tail. This further verifies that there are more abnormal large values (heavy tail phenomenon) in the ionospheric error than expected by the normal distribution.

[0042] Step S4, based on the experience error optimization and the actual error driven ionospheric model precision factor calculation; wherein the skewness and kurtosis information of the ionospheric unmodeled error obtained by the experience error analysis is used to update the ionospheric model variance, and the actual error obtained by the ionospheric observation near the grid point is used to drive the update of the ionospheric model precision factor information.

[0043] The ionospheric model can give the model variance information at all grid points, which can be represented as the combination of the model post-error and the grid point coefficient matrix, that is:

[0044] (3)

[0045] Wherein, is the variance given by the ionospheric model; is the ionospheric model post-error; The cofactor matrix of the ionospheric model; The model coefficient matrix determined by the space-time information of the grid points. However, The degree of fitting of the model to the observations, which is applicable to observations with ideal normal distribution of residuals. However, the ionospheric residuals have obvious skewness and kurtosis characteristics, and therefore need to be further optimized as follows:

[0046] (4)

[0047] wherein, The optimized post-verification mean square error of the ionospheric model, which is equivalent to the standard deviation of the Gaussian normal distribution calculated by the post-data distribution, which can envelope the ionospheric residuals; The improvement factor determined by the skewness and kurtosis of the residual distribution, satisfying:

[0048] (5)

[0049] The grid points of the ionospheric model are divided into two categories: monitored points and unmonitored points, which are determined by the actual observation distribution near the grid points, i.e. within a search domain with the grid point A as the center and a radius of If the number of ionospheric observations is greater than the minimum threshold , the point is defined as a monitored point, otherwise as an unmonitored point. The precision factor (P) of the ionospheric model is as follows:

[0050] (6)

[0051] wherein, The standard deviation between the ionospheric observation information and the ionospheric model value within the search domain of the grid point; The inflation factor; Reflects the state of the grid point. For a monitored point, is 1, and for an unmonitored point, is 2, i.e. is a scale adjustment factor introduced to avoid overly optimistic precision information of unmonitored points; is a weight adjustment factor, satisfying the following formula:

[0052] (7)

[0053] wherein, The experience is set to 5-10, and N is the number of IPPs (IP and port) within the search domain.

[0054] The inflation factor satisfies:

[0055] (8)​​

[0056] This represents the number of observations within the search domain for each grid point. Since the accuracy of the ionospheric model should maintain continuity and stability in the time domain, the expansion factor is... Perform a smoothing operation as follows:

[0057] (9)

[0058] , These are the smoothed dilation factors for the current time step and the previous time step, respectively. The initial expansion factor was obtained from extensive post-hoc analysis. , , These are the weighting coefficients.

[0059] Figure 3 The distribution of ionospheric accuracy factors obtained based on this method is presented. The darker the color (or the denser the contour lines), the lower the uncertainty of the ionospheric model in that region (the smaller the accuracy factor value); the lighter the color, the lower the model confidence.

[0060] Step S5: Determine the availability of ionospheric accuracy factor based on the ionospheric error accuracy consistency map.

[0061] To verify the consistency between ionospheric error and the accuracy factor of the ionospheric model, the following should be satisfied:

[0062] (10)

[0063] in, For probability, For ionospheric error, The standard deviation is The probability density function of the zero-mean normal distribution. Ionospheric observation information extracted from GNSS stations The model values ​​obtained by interpolation of the ionosphere model. It is a multiple. When When the values ​​are 1, 2, and 3, The probabilities should be no less than 68.27%, 95.45%, and 99.73%, respectively. For example... Figure 4As shown, based on the "Stanford-like figure" form, an ionospheric accuracy information analysis method based on Ionospheric Errors-Accuracy Diagram (IEAD) is proposed. For convenience of statistics, the distribution probability of ionospheric residual error not exceeding 1R, 2R and 3R is respectively called 1RMSBP, 2RMSBP and 3RMSBP (RMS bounding percentage, RMSBP). The actual percentage of residual error falling within the range of ±1 times, ±2 times and ±3 times accuracy factor (i.e. 1RMSBP, 2RMSBP and 3RMSBP) is calculated. If these actual percentages are close to the expected values (about 68.27%, 95.45% and 99.73%) of the theoretical normal distribution, respectively, it indicates that the accuracy factor can truly reflect the statistical characteristics of the error, and is determined as "available". Generally, the smaller the deviation between the actual percentage and the theoretical value, the better the consistency, and the higher the usability of the accuracy factor. This step realizes visual analysis through "Ionospheric Errors-Accuracy Diagram (IEAD)".

[0064] Step S6, ionospheric model accuracy factor output.

[0065] The verified global ionospheric model accuracy factor calculated in the foregoing is packaged as a final product and output. The output data is grid data completely aligned in space and time with the global ionospheric grid model, and each grid point contains its longitude, latitude, epoch time and corresponding accuracy factor value. To ensure compatibility, the data can be packaged in standard formats such as IONEX or self-defined formats. The update frequency of the accuracy factor product is synchronized with the GIM product it depends on (such as every 1 hour or 2 hours), and can support multiple service modes such as real-time, near real-time and post-processing. The product, as auxiliary accuracy information of GIM, can be directly used in high-precision GNSS positioning data processing, providing key input for constructing a reasonable ionospheric random model, thereby effectively improving positioning accuracy and convergence speed.

[0066] On the other hand, the application provides a global ionospheric model accuracy factor generation device, which comprises various modules capable of realizing each step of the foregoing method, specifically comprising:

[0067] The inversion module is configured to obtain the VTEC measured value at the ionospheric piercing point based on carrier phase smoothing pseudorange inversion;

[0068] The interpolation module is configured to obtain the VTEC model value at the piercing point by interpolation using the global ionospheric grid map GIM, and calculate the difference between the VTEC measured value and the VTEC model value as the ionospheric unmodeled error;

[0069] An analysis module is configured to analyze distribution characteristics of the ionospheric unmodeled error, optimize the internal consistency accuracy of the GIM model based on the distribution characteristics, and calculate an ionospheric model accuracy factor by driving and updating using actual observation residuals at the piercing points.

[0070] An output module is configured to determine the usability of the accuracy factor by using ionospheric error accuracy consistency, and output a final global ionospheric model accuracy factor if the accuracy factor is usable.

[0071] In a third aspect, the present application provides an electronic device, comprising: one or more processors; a memory configured to store one or more programs; wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the global ionospheric model accuracy factor generation method described above.

[0072] In a fourth aspect, the present application provides a computer-readable storage medium having stored thereon executable instructions that, when executed by a processor, enable the processor to implement the global ionospheric model accuracy factor generation method described above.

[0073] The above-described specific embodiments further illustrate the purposes, technical solutions and beneficial effects of the present application. It should be understood that the above-described specific embodiments are merely examples of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for generating accuracy factors for a global ionospheric model, characterized in that, The method includes: Step S1: Obtain the measured value of the vertical total electron content (VTEC) at the ionospheric puncture point based on carrier phase smoothing pseudorange inversion; Step S2: Using the Global Ionospheric Grid Map (GIM), the VTEC model value at the puncture point is obtained by interpolation, and the difference between the measured VTEC value and the VTEC model value is calculated as the ionospheric unmodeling error. Step S3: Analyze the distribution characteristics of the unmodeled error of the ionosphere; Step S4: Optimize the internal consistency accuracy of the GIM model based on the distribution characteristics, and use the actual observation residuals at the puncture points for driving updates to calculate the ionospheric model accuracy factor; wherein, an improvement factor is calculated based on the skewness and kurtosis, and the posterior mean square error of the GIM model is optimized using the improvement factor; including: dividing the GIM grid points into monitored points and unmonitored points: within an area centered on the grid point and with a set radius as the search range, if the number of ionospheric puncture point observations is greater than a set threshold, then the grid point is defined as a monitored point; otherwise, it is an unmonitored point; Based on the optimized post-hoc mean square error and the root mean square error of the observation residuals within the grid point search domain, the basic accuracy estimate is calculated. The basic accuracy estimate is then multiplied by a scale adjustment factor set according to the grid point status, a weight adjustment factor related to the number of puncture point observations within the search domain, and a time-smoothed dilation factor to obtain the ionospheric model accuracy factor. The scaling factor takes a first value for monitored points and a second value greater than the first value for unmonitored points; the weighting factor is negatively correlated with the number of puncture point observations within the search domain; the inflation factor is calculated smoothly based on the number of observations at the current and historical times to maintain temporal stability. Step S5: Determine the availability of the accuracy factor using the consistency of ionospheric error accuracy. If available, proceed to the next step. Step S6: Output the final global ionospheric model accuracy factor.

2. The method for generating accuracy factors for a global ionosphere model according to claim 1, characterized in that, Step S1 includes: acquiring multi-frequency observation data using a ground-based GNSS receiver, processing the observation values ​​using carrier phase smoothing pseudorange technology; extracting the total ionospheric electron content (STEC) at the ionospheric puncture point using a satellite hardware delay correction product; and converting the STEC into the measured VTEC value at the puncture point using an ionospheric projection function.

3. The method for generating the accuracy factor of a global ionosphere model according to claim 1, characterized in that, In step S3, analyzing the distribution characteristics includes at least calculating the skewness and kurtosis of the unmodeled error of the ionosphere; wherein the skewness is used to measure the asymmetry of the error distribution, and the kurtosis is used to measure the sharpness of the peak of the error distribution curve.

4. The method for generating accuracy factors for a global ionosphere model according to claim 1, characterized in that, Step S5 specifically involves calculating the actual percentages of ionospheric residuals falling within the range of 1, 2, and 3 times the accuracy factor. If these actual percentages are close to the theoretical expected value of the standard normal distribution, then the accuracy factor is determined to be usable.

5. A global ionospheric model accuracy factor generation apparatus, used to perform the method according to any one of claims 1-4, characterized in that, include: The inversion module is used to obtain the measured value of the vertical total electron content (VTEC) at the ionospheric puncture point based on carrier phase smoothing pseudorange inversion; The interpolation module is used to obtain the VTEC model value at the puncture point by interpolation using the global ionospheric grid map GIM, and calculate the difference between the measured VTEC value and the VTEC model value as the ionospheric unmodeling error. The analysis module is used to analyze the distribution characteristics of the unmodeled error of the ionosphere; based on the distribution characteristics, the internal consistency accuracy of the GIM model is optimized, and the actual observation residuals at the puncture point are used for driving the update to calculate the accuracy factor of the ionosphere model. The output module is used to determine the availability of the accuracy factor based on the consistency of ionospheric error accuracy. If available, it outputs the final global ionospheric model accuracy factor.

6. An electronic device, characterized in that, include: One or more processors; Memory, used to store one or more programs; When one or more programs are executed by the one or more processors, the one or more processors implement the global ionospheric model accuracy factor generation method according to any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, It stores executable instructions that, when executed by a processor, enable the processor to implement the global ionospheric model accuracy factor generation method as described in any one of claims 1-4.

Images