Ionospheric delay correction method and device based on regional ground-based GNSS and occultation data

Abstract

The application discloses a method and device for ionospheric delay correction based on regional ground-based GNSS and occultation data, and the method comprises the following steps: acquiring double-frequency observation data of ground-based GNSS in a target region and calculating VTEC; constructing a regional ionospheric TEC model by using a spherical cap harmonic function, calculating VTEC on each grid point in the target region according to a preset resolution, and screening effective grid points according to the number of piercing points of the grid points to form a regional grid constraint data set; taking the regional grid constraint data as virtual observation values, fusing the virtual observation values with global occultation vertical total electron content observation values, constructing a joint observation equation, and solving real-time broadcast ionospheric model coefficients; and calculating ionospheric delay correction values according to the real-time broadcast ionospheric model coefficients. According to the application, the regional VTEC grid constraint is constructed and is cooperatively inverted with global occultation data, so that the numerical ill-conditioned problem caused by direct fusion of multi-source heterogeneous data is effectively relieved, and on the premise of not depending on an empirical background field, the global precision, stability and dynamic adaptability of the model are significantly improved.

Description

Technical Field

[0001] This invention belongs to the field of ionospheric time delay correction technology, and particularly relates to an ionospheric time delay correction method and apparatus based on regional ground-based GNSS and occultation data. Background Technology

[0002] Ionospheric delay is one of the main error sources affecting the positioning accuracy of single-frequency users in Global Navigation Satellite Systems (GNSS). To support real-time correction of this error for global users, major GNSS systems broadcast parameterized global ionospheric models in their navigation messages. For example, the US GPS uses the 8-parameter Klobuchar model, the EU Galileo uses the 3-parameter NeQuick-G model, and China's BeiDou-3 system (BDS-3) broadcasts the 9-parameter BeiDou Global Ionospheric Model (BDGIM). These models describe the spatiotemporal variations of global vertical total electron content (VTEC) through low-dimensional parameters, offering advantages such as high computational efficiency and strong compatibility, and playing a crucial role in improving single-frequency positioning performance.

[0003] However, the observational foundations upon which model parameter estimations rely differ significantly between different systems. Broadcast models of GPS and Galileo typically calculate coefficients based on globally distributed ground-based observation networks provided by the International GNSS Service (IGS), offering strong global sampling support. In contrast, BDGIM's operational use primarily relies on a high-density ground-based GNSS monitoring network within China, with sparse overseas stations and severely insufficient global observational coverage. To compensate for this limitation, current BDGIM incorporates a long-term statistical climatological background field (such as a Fourier expansion model) as a priori constraint during parameter estimation to maintain the structural integrity of the global ionosphere. Nevertheless, such background fields only reflect the average climatological state of the ionosphere and lack dynamic response capabilities to short-term disturbances (such as solar flares and geomagnetic storms). During periods of high solar activity, the actual ionospheric state deviates significantly from the climatological background, leading to a sharp increase in model correction residuals and insufficient overall robustness.

[0004] In recent years, the development of low Earth orbit (LEO) satellite occultation technology (such as COSMIC-2, FY-3, and Tianmu-1) has provided new independent data sources for global ionospheric modeling. This technology can obtain ionospheric electron density profiles with near-global distribution and high vertical resolution, and further invert VTEC information to effectively cover traditional ground-based observation blind spots such as the ocean and polar regions.

[0005] Existing research has attempted to incorporate occultation observation data into global ionospheric modeling to improve model accuracy and physical consistency in data-sparse regions. However, most existing fusion methods are based on globally distributed ground-based GNSS observation networks, achieving model optimization through joint inversion with occultation data. This strategy is well-suited for systems with globally uniformly deployed ground-based monitoring stations (such as GPS), but it doesn't fully consider the practical constraint that ground-based GNSS observations are only densely deployed in limited areas (e.g., the BeiDou system mainly relies on stations in China and surrounding areas). In such regional observation scenarios, directly incorporating high-density regional ground-based GNSS VTEC data (approximately every 2 hours) would be problematic. ) and sparse global VTEC occultation observations (on the order of approximately When these two types of data are used together for least squares inversion, the severe imbalance in spatial coverage and order of magnitude between them will result in a highly ill-conditioned normal equation matrix (condition number can reach 100%). The above results in unstable or even divergent calculations, making it difficult to meet the requirements of broadcast ionospheric models for reliability, robustness, and real-time updates. On the other hand, while relying solely on occultation data to construct a global model can achieve ground-independent global coverage, it cannot effectively utilize high-precision, high-spatiotemporal-resolution ground-based GNSS observations within the region, thus limiting further improvements in the accuracy of regional ionospheric delay correction.

[0006] Therefore, there is an urgent need for a new modeling method that can effectively integrate global occultation data to construct a high-precision, highly stable broadcast ionospheric model that is independent of or weakly dependent on the climate background field, under the condition that only regional ground-based GNSS observations are supported, thereby improving the system's service consistency and disturbance adaptability on a global scale. Summary of the Invention

[0007] The purpose of this invention is to overcome the above-mentioned defects and problems in the prior art and provide an ionospheric time delay correction method and device based on regional ground-based GNSS and occultation data. This method effectively alleviates the numerical ill-conditioning problem caused by the direct fusion of multi-source heterogeneous data by constructing regional VTEC grid constraints and co-inverting with global occultation data. Without relying on empirical background fields, it significantly improves the global accuracy, stability and dynamic adaptability of the model.

[0008] To achieve the above objectives, the technical solution of the present invention is:

[0009] In a first aspect, the present invention provides an ionospheric delay correction method based on regional ground-based GNSS and occultation data, comprising:

[0010] Acquire ground-based GNSS dual-frequency observation data within the target area, extract the total electron content along the oblique path, and convert it into the total vertical electron content using a projection function;

[0011] A regional ionospheric TEC model was constructed using a spherical cap harmonic function. Based on the observed vertical total electron content of all effective puncture points in the region, the coefficients to be estimated in the regional ionospheric TEC model were determined by least squares fitting.

[0012] Within the target area, the total vertical electron content at each grid point is calculated at a preset resolution, and effective grid points are selected based on the number of puncture points at each grid point to form a regional grid constraint dataset.

[0013] Electron density profiles from global occultation events retrieved from multi-source low Earth orbit satellites were obtained, and after quality control, they were integrated to obtain the observed values ​​of global occultation vertical total electron content.

[0014] Regional grid-constrained data are used as virtual observations and fused with global occultation vertical total electron content observations to construct a joint observation equation. The weighted least squares method is then used to solve the real-time broadcasting coefficients of the broadcast ionosphere model.

[0015] The user receiver receives the real-time broadcast coefficients and broadcast ephemeris, and combines them with preset fixed parameters to calculate the vertical total electron content at the puncture point. Then, it converts the total electron content of the oblique path through the ionospheric projection function, thereby determining the ionospheric delay correction number on the GNSS signal propagation path.

[0016] Preferably, the regional ionosphere TEC model is as follows:

[0017] ;

[0018] In the formula, The vertical total electron content; and These are the latitude and longitude of the puncture point in the geographic coordinate system, respectively; and These are the latitude and longitude of the region's center, respectively. for Spend The regularized Legendre function of order 1; It is the highest order; and The coefficients of the TEC model for the ionosphere in the region to be estimated are denoted as .

[0019] Preferably, the step of calculating the total vertical electron content at each grid point within the target area at a preset resolution, and selecting effective grid points based on the number of puncture points at each grid point to form a regional grid-constrained dataset includes:

[0020] Within the target area, the total vertical electron content at each grid point is calculated at a resolution of 2.5° latitude × 5° longitude. Valid grid points are selected based on the criterion that the number of puncture points at each grid point is greater than or equal to 10, thus forming a regional grid-constrained dataset.

[0021] Preferably, the virtual observations are fused with global occultation vertical total electron content observations at a weight ratio of 2:1 to 4:1.

[0022] Preferably, the broadcast ionosphere model is as follows:

[0023] ;

[0024] ;

[0025] ;

[0026] ;

[0027] ;

[0028] ;

[0029] ;

[0030] In the formula, The vertical total electron content; This is the delayed forecast value for the ionosphere; The parameters for the broadcast ionosphere model to be determined are: The function value is calculated based on the user's puncture point location and observation time; For regularization functions; For the standard Legendre function; and These represent the latitude and longitude of the ionospheric intersection point in the Sun-fixed geomagnetic coordinate system; These are the model prediction coefficients stored in the user receiver; This is a function value calculated based on preset curing parameters.

[0031] Preferably, a design matrix is ​​constructed, and the weighted least squares method is used to solve for the real-time broadcast coefficients of the broadcast ionosphere model:

[0032] ;

[0033] In the formula, The real-time broadcast coefficients to be solved; The design matrix consists of a row corresponding to a vertical total electron content observation point, a column corresponding to a model basis function, and matrix elements representing the function values ​​of the basis functions corresponding to the observation points. ; This is the matrix transpose symbol; For observation weight array; These are observations, including virtual observations and global occultation vertical total electron content observations.

[0034] Preferably, the quality control includes:

[0035] Calculate the relative rate of change of electron density between adjacent height points for each effective electron density profile;

[0036] The height axis is divided into several intervals. For each effective electron density profile, a robust percentile index of the relative electron density change rate is statistically analyzed in each height interval to generate noise upper limit threshold, sawtooth amplitude threshold and extreme mutation threshold.

[0037] Based on the noise upper limit threshold, extreme mutation threshold, and sawtooth amplitude threshold, a three-level progressive judgment process is established, including high noise background criteria, extreme single-point jump criteria, and valid sawtooth structure criteria. If any criterion is triggered, the electron density profile is determined to be invalid and discarded.

[0038] Secondly, the present invention provides an ionospheric delay correction device based on regional ground-based GNSS and occultation data. The device is used to implement the aforementioned ionospheric delay correction method based on regional ground-based GNSS and occultation data. The device includes:

[0039] The regional grid-constrained dataset acquisition module is used to acquire ground-based GNSS dual-frequency observation data within the target area, extract the total electron content along the oblique path, and convert it into the total vertical electron content using a projection function. A regional ionospheric TEC model is constructed using a spherical cap harmonic function. Based on the observed total vertical electron content at all effective puncture points within the area, the coefficients to be estimated in the regional ionospheric TEC model are determined through least-squares fitting. Within the target area, the total vertical electron content at each grid point is calculated at a preset resolution, and effective grid points are selected based on the number of puncture points at each grid point, forming the regional grid-constrained dataset.

[0040] The RO-VTEC observation acquisition module is used to acquire electron density profiles from global occultation events retrieved from multi-source low Earth orbit satellites. After quality control, the data are integrated to obtain the global occultation vertical total electron content observations.

[0041] The real-time broadcast coefficient acquisition module is used to integrate regional grid-constrained data as virtual observations with global occultation vertical total electron content observations to construct a joint observation equation, and to solve the real-time broadcast coefficients of the broadcast ionosphere model using the weighted least squares method.

[0042] The ionospheric delay correction acquisition module is used by the user receiver to receive real-time broadcast coefficients and broadcast ephemeris, combine them with preset solidification parameters, calculate the vertical total electron content at the puncture point, and then convert it into the oblique path total electron content through the ionospheric projection function, thereby determining the ionospheric delay correction on the GNSS signal propagation path.

[0043] Thirdly, the present invention provides an ionospheric delay correction device based on regional ground-based GNSS and occultation data, including a memory and a processor;

[0044] The memory is used to store computer program code and transmit the computer program code to the processor;

[0045] The processor is configured to execute, according to instructions in the computer program code, an ionospheric delay correction method based on regional ground-based GNSS and occultation data as described above.

[0046] Fourthly, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements an ionospheric delay correction method based on regional ground-based GNSS and occultation data as described above.

[0047] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0048] This invention discloses an ionospheric time delay correction method and apparatus based on regional ground-based GNSS and occultation data. By compressing high-dimensional ground-based observations into low-dimensional grid virtual observations, the condition number of the normal equations is reduced, ensuring the stability and reliability of the inversion process. The model is entirely driven by measured data, accurately responding to short-term ionospheric disturbances and maintaining high accuracy even during periods of high solar activity. The regional grid inherits the high-resolution advantage of ground-based GNSS, while occultation data provides global constraints; the two complement each other, significantly improving the correction performance globally, especially in data-sparse regions. Therefore, the proposed method, by constructing regional VTEC grid constraints and co-inverting with global occultation data, effectively alleviates the numerical ill-conditioning problem caused by the direct fusion of multi-source heterogeneous data, significantly improving the model's global accuracy, stability, and dynamic adaptability without relying on empirical background fields. Attached Figure Description

[0049] Figure 1 This is a flowchart of the ionospheric delay correction method based on regional ground-based GNSS and occultation data proposed in this invention.

[0050] Figure 2 This is a comparison diagram of the normal equation condition numbers of the ionospheric delay correction method and the direct fusion method based on regional ground-based GNSS and occultation data proposed in this embodiment of the invention.

[0051] Figure 3 This is a flowchart illustrating the quality control of electron density profile data as proposed in an embodiment of the present invention.

[0052] Figure 4 This is a schematic diagram illustrating the analysis of Case 1, which shows the qualified electron density profile proposed in the embodiments of the present invention.

[0053] Figure 5 This is a schematic diagram illustrating the analysis of Case 2, which shows the qualified electron density profile proposed in the embodiments of the present invention.

[0054] Figure 6 This is a schematic diagram illustrating the analysis of Case 1, a non-compliant electron density profile proposed in an embodiment of the present invention.

[0055] Figure 7 This is a schematic diagram illustrating the analysis of Case 2, which shows an electron density profile of substandard quality, as presented in an embodiment of the present invention.

[0056] Figure 8 This is a schematic diagram illustrating the analysis of Case 3, a non-compliant electron density profile proposed in an embodiment of the present invention.

[0057] Figure 9 This is a structural block diagram of the ionospheric delay correction device based on regional ground-based GNSS and occultation data proposed in this invention.

[0058] Figure 10 This is a structural block diagram of the ionospheric delay correction device based on regional ground-based GNSS and occultation data proposed in this invention. Detailed Implementation

[0059] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0060] See Figure 1 This invention provides an ionospheric delay correction method based on regional ground-based GNSS and occultation data, comprising:

[0061] S1. Obtain ground-based GNSS dual-frequency observation data within the target area, extract the total electron content along the oblique path, and convert it into the total vertical electron content using a projection function;

[0062] A regional ionospheric TEC model was constructed using a spherical cap harmonic function. Based on the observed vertical total electron content of all effective puncture points in the region, the coefficients to be estimated in the regional ionospheric TEC model were determined by least squares fitting.

[0063] Within the target area, the total vertical electron content at each grid point is calculated at a preset resolution, and effective grid points are selected based on the number of puncture points at each grid point to form a regional grid constraint dataset.

[0064] S2. Obtain the electron density profile from the global occultation event inversion from multi-source low Earth orbit satellites, and integrate it after quality control to obtain the observed value of the global occultation vertical total electron content.

[0065] S3. The regional grid-constrained data is used as virtual observations and fused with global occultation vertical total electron content observations to construct a joint observation equation. The weighted least squares method is then used to solve the real-time broadcast coefficients of the broadcast ionosphere model.

[0066] S4. The user receiver receives the real-time broadcast coefficients and broadcast ephemeris, and calculates the vertical total electron content at the puncture point by combining the preset solidified parameters. Then, it converts the total electron content of the oblique path through the ionospheric projection function, thereby determining the ionospheric delay correction number on the GNSS signal propagation path and realizing high-precision ionospheric delay correction.

[0067] The purpose of this invention is to overcome the shortcomings of existing broadcast ionospheric models, such as insufficient global coverage, strong dependence on the climatic background field, and poor robustness during ionospheric disturbances, when relying solely on regional ground-based GNSS observations. This invention provides a broadcast ionospheric delay correction method based on the fusion of regional ground-based GNSS and global occultation data. This method effectively alleviates the numerical ill-conditioning caused by the direct fusion of multi-source heterogeneous data by constructing regional VTEC grid constraints and co-inverting them with global occultation data. Without relying on empirical background fields, it significantly improves the model's global accuracy, stability, and dynamic adaptability.

[0068] Furthermore, ground-based GNSS dual-frequency observation data within the target area are acquired. The total electron content along the oblique path is extracted based on the phase smoothing pseudorange method or the non-combined precise point positioning (PPP) method, and then converted into the total vertical electron content using a projection function.

[0069] ;

[0070] In the formula, This is a projection function used to convert the total electron content of the vertical and oblique ionosphere; This represents the total electron content along the oblique path. The vertical total electron content; The zenith angle of the satellite at the receiver; The average radius of the Earth; This represents the height of the ionosphere.

[0071] Specifically, dual-frequency observation data is acquired within a two-hour time window prior to the target time, collected by densely deployed ground-based GNSS monitoring stations (such as the 31 BeiDou monitoring stations in China). Carrier phase smoothing is applied to the raw observations to suppress multipath noise, and the differential code bias (DCB) product estimated from the previous day is used to correct receiver and satellite hardware delays. The slant path total electron content (STEC) is converted to vertical total electron content (VTEC) based on a geometric mapping function, with the puncture point height set at 350 km. After processing, approximately 170,000 valid VTEC observations are obtained every two hours.

[0072] Furthermore, a regional ionospheric (TEC) model is constructed using spherical cap harmonic functions:

[0073] ;

[0074] In the formula, The vertical total electron content; and These are the latitude and longitude of the puncture point in the geographic coordinate system, respectively; and These are the latitude and longitude of the region's center, respectively. for Spend The regularized Legendre function of order 1; It is the highest order; and The coefficients of the TEC model for the ionosphere in the region to be estimated are denoted as .

[0075] Furthermore, within the target region (e.g., China: 70°E-140°E, 15°N-55°N), the total vertical electron content at each grid point is calculated at a resolution of 2.5° latitude × 5° longitude. Then, effective grid points are selected based on the criterion that each grid point has at least 10 puncture point observation data around it, forming a highly reliable regional grid-constrained dataset.

[0076] Within the target area, a total of 320 grid points were generated. After removing grid points with fewer than 10 puncture points, 185 effective grid points with sufficient observational support were retained, and their VTEC values ​​were used as virtual observations in subsequent fusion and inversion.

[0077] If a grid point has too few surrounding puncture points, it means that the accuracy of the model fitting for that grid point is low. If it is used in subsequent modeling processes, it will reduce the accuracy of the entire global model. This invention removes grid points with fewer than 10 surrounding puncture points. After removing these invalid grid points, the TEC information of the other grid points is used as virtual observations and input into the subsequent broadcast ionosphere model. Therefore, the TEC information of these grid points is called regional grid constraint data.

[0078] Furthermore, electron density profiles retrieved from occultation events using multiple low Earth orbit (LEO) satellites (including Tianmu-1, COSMIC-2, and FY-3) were collected. Quality control was performed on each profile (including a three-level progressive screening based on high noise, extreme jumps, and jagged structure criteria) to remove outliers. The qualified electron density profiles were integrated from 90 km to 800 km to obtain the occultation path TEC. Using the NeQuick-2 model, the standard GNSS path TEC (ground to 20200 km) and the occultation path TEC (ground to LEO orbit altitude) were calculated at the same spatiotemporal point. The ratio of these two was used as a scaling factor to normalize the occultation TEC, making it comparable to ground-based VTEC. Approximately 1000 valid normalized occultation VTEC observations were obtained every 2 hours, with a uniform global distribution.

[0079] Normalized ionospheric TEC data for:

[0080] ;

[0081] In the formula, TEC data of the ionosphere obtained by surface integration of electron density profile from occultation; TEC data for the ground-to-GNSS satellite orbital altitude range calculated based on the NeQuick background model; This is TEC data for the orbital altitude range from ground to low Earth orbit (LEO) satellites, calculated based on the NeQuick background model.

[0082] By employing TEC normalization processing technology, the technical challenge of inconsistent spatial altitude coverage between occultation observation TEC and ground-based GNSS observation TEC was solved.

[0083] Furthermore, the generated 185 regional grid VTECs are used as virtual observations and fused with occultation VTEC observations according to weights. The virtual observations are fused with global occultation vertical total electron content (RO-VTEC) observations at a weight ratio of 2:1 to 4:1. In practical applications, because the TEC observations obtained from ground-based GNSS data are more accurate than those obtained from occultation data, the weight ratio of virtual observations to global occultation vertical total electron content observations is set at 3:1.

[0084] Furthermore, a broadcast ionospheric model is constructed using the BeiDou-3 BDGIM architecture, specifically as follows:

[0085] ;

[0086] ;

[0087] ;

[0088] ;

[0089] ;

[0090] ;

[0091] ;

[0092] In the formula, The vertical total electron content; This is the delayed forecast value for the ionosphere; The parameters for the broadcast ionosphere model to be determined are: The function value is calculated based on the user's puncture point location and observation time; For regularization functions; For the standard Legendre function; and These represent the latitude and longitude of the ionospheric intersection point in the Sun-fixed geomagnetic coordinate system; These are the model prediction coefficients stored in the user receiver. Calculated based on the non-spreading coefficient; The function value is calculated based on preset solidified parameters (the solidified parameters include 17 items such as solar activity reference, seasonal modulation terms and geomagnetic offset constant, which are pre-stored in the user receiver).

[0093] Furthermore, a design matrix is ​​constructed, and the weighted least squares method is used to solve for the real-time broadcast coefficients of the broadcast ionospheric model:

[0094] ;

[0095] In the formula, Nine BDGIM real-time broadcast coefficients were obtained to be determined as the real-time broadcast coefficients to be solved. The design matrix is ​​a nonlinear observational matrix used to solve the real-time broadcast coefficients of the broadcast ionosphere model after fusing regional grid-constrained data (virtual observations) with global occultation vertical total electron content observations. Each row of the design matrix corresponds to a vertical total electron content observation point (from the regional grid or occultation), and each column corresponds to a model basis function. The matrix elements are the function values ​​of the basis functions corresponding to the observation points. ; This is the matrix transpose symbol; For observation weight array; These are observations, including virtual observations and global occultation vertical total electron content observations.

[0096] Furthermore, the estimated nine coefficients are encoded in BDGIM message format and broadcast every two hours via BeiDou navigation satellites. The specific process for calculating the ionospheric delay correction at the user end is as follows:

[0097] S401, the GNSS receiver receives broadcast ephemeris from navigation satellites, which includes satellite orbit parameters, clock error information, and nine ionospheric delay correction parameters generated by this invention;

[0098] S402. Based on the satellite orbit parameters and the approximate position of the receiver, calculate the latitude, longitude, and corresponding satellite elevation angle of the signal penetration point in the solar-fixed geomagnetic coordinate system.

[0099] S403. Combining the 9 broadcast parameters with the 17 BDGIM solidified parameters pre-stored in the receiver, the vertical total electron content (VTEC) at the puncture point is calculated by substituting them into the BDGIM model.

[0100] S404. Convert VTEC to tilted total electron content (STEC) using the ionospheric projection function, and then determine the ionospheric delay correction at the L1 frequency.

[0101] See Figure 2 Compared with the direct fusion method of multi-source heterogeneous data, the present invention compresses ground-based observations into low-dimensional grid virtual observations (185 nodes), reducing the condition number of the normal equations to about 21, thus ensuring stable convergence of the inversion process.

[0102] Table 1 shows a comparison of the ionospheric delay correction accuracy of different broadcast ionospheric delay correction techniques. Klobuchar represents the 8-parameter model used by GPS; NeQuick-G represents the 3-parameter model used by Galileo; BDGIM represents the current 9-parameter model of BeiDou-3; and Fusion-BDGIM represents the two-step fusion method proposed in this invention. In tests conducted during the high solar activity period in August 2024 (F10.7≈180):

[0103] RMSE in China: The present invention (Fusion-BDGIM) achieved 9.45 TECu, a 36.4% improvement over BDGIM (14.86 TECu).

[0104] Global average RMSE: This invention (Fusion-BDGIM) has a RMSE of 9.92 TECu, a 41.4% improvement over BDGIM (16.92 TECu);

[0105] The condition number for the normal equations is 21 in this invention (Fusion-BDGIM), and the direct fusion method is as follows: .

[0106] The results show that, while maintaining the same number of broadcast parameters (9) as the current BDGIM, the present invention significantly improves the global correction accuracy, while BDGIM suffers from obvious smoothing distortion due to its reliance on the climate background field.

[0107] Table 1. Comparison of Ionospheric Delay Correction Accuracy of Different Broadcast Ionospheric Delay Correction Techniques

[0108]

[0109] Currently, the BeiDou-3 BDGIM model primarily relies on ground-based GNSS data from China for operational use, with sparse overseas stations necessitating the introduction of a climatological background field to maintain the global field structure. However, this background field lacks responsiveness to short-term ionospheric disturbances, leading to a significant increase in correction residuals during periods of high solar activity. This invention addresses the technical challenge of existing broadcast ionospheric models, which struggle to balance global accuracy and dynamic adaptability with only regional ground-based observation capabilities, by proposing a modeling method that fuses regional ground-based GNSS and global occultation data. This method fully leverages the high precision of regional ground-based GNSS and the global coverage of occultation data. Through a two-step fusion strategy (regional modeling, gridding, and joint inversion), it fundamentally alleviates the ill-conditioned problems in the normal equations caused by direct fusion of multi-source heterogeneous data, significantly improving the model's accuracy, stability, and robustness globally without relying on an empirical background field.

[0110] The advantages of this invention include: ① Completely alleviating numerical ill-conditioning problems. By compressing high-dimensional ground-based observations into low-dimensional grid virtual observations (e.g., 185 nodes), the condition number of the normal equations is reduced from > ① The accuracy is reduced to approximately 20-30 to ensure a stable and reliable inversion process. ② It eliminates dependence on the climatic background field. The model is entirely driven by measured data, accurately responding to short-term ionospheric disturbances and maintaining high accuracy even during periods of high solar activity. ③ It balances regional accuracy with global coverage. The regional grid inherits the high-resolution advantage of ground-based GNSS, while occultation data provides global constraints. The two complement each other, significantly improving correction performance globally, especially in data-sparse areas. ④ It is compatible with existing broadcast architectures: the output consists of standard low-dimensional parameters (such as the 9 parameters of BDGIM), allowing seamless integration with GNSS systems such as BeiDou, and possessing direct engineering application value. ⑤ It enhances system autonomy and robustness. Even in the absence of overseas ground stations, it can still maintain global service capabilities through occultation data, improving the independent operation capability of the navigation system.

[0111] Furthermore, under certain conditions, the vertical electron density profile obtained from occultation inversion may exhibit non-physical sawtooth fluctuations. These anomalies often stem from numerical instability, noise amplification, or multipath effects in the inversion algorithm, and do not represent the true ionospheric structure. Failure to identify and remove these anomalies will directly impact the accuracy and stability of subsequent modeling. Current occultation data processing workflows generally lack specific checks for such sawtooth profiles, leading to the misuse of anomalous data and weakening the reliability of the model.

[0112] See Figure 3 Quality control is performed to obtain qualified electron density profile data, including:

[0113] S201. Calculate the relative rate of change of electron density between adjacent height points of each effective electron density profile.

[0114] S202. Divide the height axis into several intervals. For each effective electron density profile, calculate the robust percentile index of the relative electron density change rate in each height interval to generate the noise upper limit threshold, sawtooth amplitude threshold and extreme mutation threshold.

[0115] S203. Based on the noise upper limit threshold, extreme mutation threshold, and sawtooth amplitude threshold, a three-level progressive judgment process is formulated, including high noise background criteria, extreme single-point jump criteria, and valid sawtooth structure criteria. If any criterion is triggered, the electron density profile is determined to be invalid and discarded.

[0116] To overcome the problem that a globally fixed threshold cannot adapt to the non-uniformity of ionospheric height, this invention constructs a height-stratified dynamic threshold system based on historical high-quality electron profile data (such as long-term observations from COSMIC, Tianmu-1, Fengyun-3, the Yunyao constellation, and ionospheric altimeters). This invention is applicable to the automatic quality screening of multi-source observation data such as radio occultation (RO), ionospheric altimeter, and GNSS inversion, and is particularly useful for identifying non-physical structural anomalies such as high-frequency sawtooth oscillations and abrupt changes.

[0117] The core idea of ​​this invention is that local changes in the real ionospheric profile exhibit high dependence and statistical regularity, while non-physical anomalies manifest as random or periodic, drastic fluctuations unrelated to altitude. Based on this, this invention proposes a detection strategy combining segmented dynamic thresholds and multiple criteria fusion. Specifically, it includes: constructing a highly adaptive dynamic threshold system: utilizing a large amount of historical "normal" profile data, calculating robust percentile indices of relative electron density change rates for each altitude range to generate three types of thresholds—noise upper limit threshold, sawtooth amplitude threshold, and extreme abrupt change threshold; designing a three-level progressive discrimination process: sequentially screening for high-noise background, extreme single-point jumps, and valid sawtooth structures; triggering any condition results in an invalid profile. Compared to traditional methods, this invention has the advantages of high adaptability, low false alarm rate, and high detection rate, making it particularly suitable for quality control of multi-source ionospheric data in complex space environments.

[0118] Furthermore, the original electron density profile data is first processed using conventional quality control procedures to remove obviously invalid data. Then, through manual visual screening, abnormal profiles with non-physical oscillation characteristics such as "sawtooth" are further removed, ultimately forming a pure "normal profile" training set, i.e., an effective electron density profile dataset.

[0119] Furthermore, the method for calculating the relative rate of change of electron density is as follows:

[0120] For each valid electron density profile, calculate the absolute value of electron density between adjacent height points:

[0121] ;

[0122] In the formula, This represents the absolute difference in electron density. For height point electron density; For height point electron density;

[0123] Normalizing based on the smaller electron density value between two adjacent points, we obtain the relative rate of change of electron density:

[0124] ;

[0125] In the formula, It represents the rate of change of relative electron density; This indicates the smaller electron density value between two adjacent points.

[0126] Furthermore, robust percentile indices of relative electron density change rate are statistically analyzed within each height interval to generate noise upper limit threshold, sawtooth amplitude threshold, and extreme abrupt change threshold, including:

[0127] The rate of change of relative electron density was statistically analyzed within each altitude interval (e.g., altitude intervals were divided into intervals of 50 km along the altitude axis). The 75th percentile (P75), 95th percentile (P95), and maximum value (Max) are used to obtain the first, second, and third statistics, respectively. These statistics are then combined across all electron density profiles within the same height range. The 75th percentile of the first statistic is used to obtain the sawtooth amplitude threshold (P75_of_P75), the 75th percentile of the second statistic is used to obtain the noise upper limit threshold (P75_of_P95), and the 75th percentile of the third statistic is multiplied by 1.2 to obtain the extreme mutation threshold (1.2×P75_of_Max). This results in a dynamic threshold table that varies with height, used for anomaly detection in subsequent profiles.

[0128] This invention, through hierarchical statistics and a robust percentile strategy, accurately identifies non-physical jagged anomalies while preserving normal physical fluctuations, significantly improving the adaptability and reliability of quality control.

[0129] Based on the aforementioned dynamic thresholds, this invention designs a three-level progressive criterion, whereby the profile is deemed invalid if any criterion is triggered. First, the upper limit threshold for noise is used to assess whether the profile is in an overall high-noise state. Then, the extreme mutation threshold is used to check whether there are drastic single-point jumps in the profile. Finally, the sawtooth amplitude threshold is used to identify more deceptive "regular sawtooth" anomalies (manifested as alternating rising / falling oscillations).

[0130] Furthermore, the high-noise background criterion (overall fluctuation intensity screening) is as follows:

[0131] The proportion of the relative electron density change rate exceeding the upper limit threshold of the noise level. :

[0132] ;

[0133] In the formula, It represents the rate of change of relative electron density; For height point The upper limit threshold for noise; This represents the total number of data points in the electron density profile data.

[0134] If the proportion exceeds a preset threshold (e.g., 20%), it indicates that the electron density profile is contaminated by high-frequency noise and lacks physical reliability, thus the electron density profile is deemed invalid.

[0135] Furthermore, the extreme single-point transition criterion (non-physical transition screening) is as follows:

[0136] Extreme mutation thresholds are used to identify points with abnormally high rates of change in relative electron density. If two or more height points exist within the effective height range that satisfy the following conditions:

[0137] ;

[0138] In the formula, It represents the rate of change of relative electron density; For height point The extreme mutation threshold;

[0139] This indicates that the electron density profile contains a non-physical abrupt change, and the electron density profile is deemed invalid.

[0140] Furthermore, the effective sawtooth structure criterion (periodic oscillation identification) adopts a joint discrimination strategy of structure + amplitude, specifically as follows:

[0141] (1) Detect sign reversal. Calculate the sign change of the electron density difference sequence to locate potential sawtooth vertices (i.e., locate potential local maxima / minimum points);

[0142] (2) Verify that the amplitude meets the standard. For each sign flip point, if at least one of the two adjacent relative electron density change rates is not lower than the sawtooth amplitude threshold zigzagThrVec, then it is a valid sawtooth vertex;

[0143] (3) Continuity assessment. The maximum continuous length of the effective sawtooth vertices is statistically analyzed. If the maximum continuous length is greater than or equal to the set threshold (the set threshold is 5), it indicates that there is a systematic, non-physical periodic oscillation in the electron density profile, and the electron density profile is judged to have sawtooth anomalies.

[0144] Furthermore, the essence of sawtooth anomalies is that electron density alternately rises and falls with height, forming a continuous sequence of local extrema. This invention accurately locates potential sawtooth vertices through the following steps; the method for determining the potential sawtooth vertices is as follows:

[0145] (1) Calculate the first-order difference (reflecting the local trend of change):

[0146] Given electron density profile and corresponding height The first-order difference (i.e., change) between adjacent height points is calculated as follows:

[0147] ;

[0148] In the formula, This represents the change in electron density between two adjacent altitude points; For height point electron density; For height point electron density;

[0149] (2) Extract the difference symbol sequence :

[0150] ;

[0151] (3) Check if the symbol has changed:

[0152] Calculate the difference of symbol sequences :

[0153] ;

[0154] like This indicates that at the altitude point At this point, the trend of electron density change reverses:

[0155] If the difference sign sequence changes from +1 to -1, it indicates the height point. This is a local maximum (peak);

[0156] If the difference sign sequence changes from -1 to +1, it indicates the height point. This is a local minimum point (valley);

[0157] The locations of local maxima and minima are potential sawtooth vertices. By combining amplitude and continuity constraints, the true ionospheric structure and pseudo-oscillations can be effectively distinguished.

[0158] This invention, through the fusion of segmented dynamic thresholds and multi-level criteria, can efficiently and accurately identify non-physical sawtooth oscillations and extreme jumps, significantly improving the quality control capability of data before ionospheric modeling; it has the following advantages: (1) Strong adaptability: the threshold is dynamically adjusted with height, taking into account both the stability of the E layer and the high gradient characteristics of the F layer; (2) Low false alarm rate: through structural features (sign flip) and amplitude dual verification, it avoids misjudging the real gradient as an anomaly; (3) High detection rate: it covers a variety of anomaly types such as high noise, extreme jumps, and periodic oscillations; (4) Strong engineering applicability: the algorithm is simple and easy to integrate into the existing ionospheric data processing flow.

[0159] The conventional electron density profile quality control method uses the following four quality control indicators to screen the raw electron density profile data.

[0160] (1) Average relative deviation:

[0161] ;

[0162] In the formula, The average relative deviation; This represents the total number of electron density samples in a single profile; For the first Electron density inversion value at each sampling point; This is the background value obtained by filtering with a 7-point moving average.

[0163] If a certain cross-section If the value is greater than 0.25, it is judged as an abnormal profile and is removed.

[0164] (2) Noise factor:

[0165] ;

[0166] In the formula, Noise factor; This refers to the number of sampling points within an altitude range above 300km.

[0167] like If the profile is deemed to have excessive fluctuations in the height region, it is marked as a problem profile.

[0168] (3) Top electron density gradient:

[0169] ;

[0170] In the formula, The top electron density gradient; and The values ​​are electron density at altitudes of 490 km and 420 km, respectively, with an altitude interval of 70 km.

[0171] like If the density at the top decreases too slowly, it may reflect an inversion bias, and the profile is marked as a problem profile.

[0172] (4) and Reasonable range:

[0173] Based on existing research experience It is usually located between 180 and 450 km, while The reasonable range is to If any parameter exceeds the above range, the entire profile will be considered abnormal and discarded.

[0174] Based on the above four criteria, any electron density profile that meets any of the rejection conditions is marked as a problem profile.

[0175] Figure 4 and Figure 5The application effect of the abnormal profile quality discrimination method proposed in this invention is demonstrated by taking the profiles ionPrf_TM01.2024.215.10.48.E24_0001.0001.nc (quality qualified electron density profile case 1) and ionPrf_TM01.2024.215.00.01.G03_0001.0001.nc (quality qualified electron density profile case 2) as examples. Figure 4 (a) and Figure 5 (a) is a schematic diagram of electron density distribution with height, from Figure 4 (a) and Figure 5 (a) It can be seen that both electron density profiles exhibit smooth and continuous variation characteristics, without obvious sawtooth oscillations; Figure 4 (b) and Figure 5 (b) is a schematic diagram showing the distribution of electron density change and dynamic threshold with height. Figure 4 (b) and Figure 5 (b) It can be seen that the corresponding relative electron density change rate does not exceed the three dynamic thresholds (sawtooth amplitude threshold, noise upper limit threshold, and extreme mutation threshold) set by the present invention throughout the entire height range. The results show that the present invention can accurately identify and retain qualified profiles that conform to physical properties.

[0176] Figure 6 Taking the typical sawtooth-shaped abnormal electron density profile ionPrf_TM01.2024.215.00.57.R24_0001.0001.nc (Case 1 of substandard electron density profile) as an example, Figure 7 Taking the typical sawtooth-shaped abnormal electron density profile ionPrf_TM02.2024.215.00.57.R20_0001.0001.nc (Case 2 of substandard electron density profile) as an example, Figure 8 Taking the typical sawtooth-shaped abnormal electron density profile ionPrf_TM01.2024.215.00.35.G05_0001.0001.nc (Case 3 of unqualified electron density profile) as an example, the application effect of the abnormal profile quality discrimination method proposed in this invention is demonstrated. Figure 6 (a) Figure 7 (a) and Figure 8 (a) is a schematic diagram of electron density distribution with height. Figure 6 (b) Figure 7 (b) and Figure 8 (b) is a schematic diagram showing the distribution of electron density change and dynamic threshold with height. Figure 6 (a) Figure 7 (a) and Figure 8 (a) It can be seen that all three profiles exhibit obvious non-physical sawtooth-like undulation characteristics, among which Figure 6 (a) and Figure 7 The oscillations in the cross-section shown in (a) are particularly violent; by Figure 6 (b) Figure 7 (b) and Figure 8 (b) It is evident that the corresponding relative electron density change rate significantly exceeds the three dynamic thresholds set by this invention in multiple height ranges—including the sawtooth amplitude threshold (P75_of_P75), the noise upper limit threshold (P75_of_P95), and the extreme mutation threshold (1.2×P75_of_Max). The results show that this invention can effectively identify such high-frequency oscillation anomalies and mark them as unqualified profiles.

[0177] Table 2 lists the traditional quality control indicators corresponding to the three typical sawtooth-shaped anomalous electron density profiles mentioned above. Among them, the R24 profile, with a maximum height of only 489 km, does not fully cover the 420-490 km interval, therefore the top gradient cannot be calculated; the remaining indicators are as follows: the mean relative deviation (MRD) are 0.07, 0.08, and 0.03, respectively, all far below the rejection threshold of 0.25; the upper-level noise factor Δ is respectively... , and Both are less than 0.01; the top gradients of R20 and G05 are respectively and Both meet the "rapid descent" requirement; hmF2 (294.95-350.00 km) and NmF2 ( All three sections are within the physically reasonable range. Therefore, under existing industry standards, these three sections, which clearly exhibit jagged anomalies, did not trigger any traditional quality control criteria and were incorrectly judged as "qualified" sections.

[0178] Table 2. Statistics of Quality Judgment Indicators for Traditional Methods

[0179]

[0180] This comparison clearly demonstrates that traditional methods are ineffective in identifying jagged structural anomalies, while the present invention significantly improves the detection capability of such hidden anomalies by fusing segmented dynamic thresholds with structural features.

[0181] See Figure 9 The present invention also provides an ionospheric delay correction device based on regional ground-based GNSS and occultation data. The device is used to implement the aforementioned ionospheric delay correction method based on regional ground-based GNSS and occultation data. The device includes:

[0182] The regional grid-constrained dataset acquisition module is used to acquire ground-based GNSS dual-frequency observation data within the target area, extract the total electron content along the oblique path, and convert it into the total vertical electron content using a projection function. A regional ionospheric TEC model is constructed using a spherical cap harmonic function. Based on the observed total vertical electron content at all effective puncture points within the area, the coefficients to be estimated in the regional ionospheric TEC model are determined through least-squares fitting. Within the target area, the total vertical electron content at each grid point is calculated at a preset resolution, and effective grid points are selected based on the number of puncture points at each grid point, forming the regional grid-constrained dataset.

[0183] The RO-VTEC observation acquisition module is used to acquire electron density profiles from global occultation events retrieved from multi-source low Earth orbit satellites. After quality control, the data are integrated to obtain the global occultation vertical total electron content observations.

[0184] The real-time broadcast coefficient acquisition module is used to integrate regional grid-constrained data as virtual observations with global occultation vertical total electron content observations to construct a joint observation equation, and to solve the real-time broadcast coefficients of the broadcast ionosphere model using the weighted least squares method.

[0185] The ionospheric delay correction acquisition module is used by the user receiver to receive real-time broadcast coefficients and broadcast ephemeris, combine them with preset solidification parameters, calculate the vertical total electron content at the puncture point, and then convert it into the oblique path total electron content through the ionospheric projection function, thereby determining the ionospheric delay correction on the GNSS signal propagation path.

[0186] See Figure 10 The present invention also provides an ionospheric delay correction device based on regional ground-based GNSS and occultation data, including a memory and a processor;

[0187] The memory is used to store computer program code and transmit the computer program code to the processor;

[0188] The processor is configured to execute, according to instructions in the computer program code, an ionospheric delay correction method based on regional ground-based GNSS and occultation data as described above.

[0189] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements an ionospheric delay correction method based on regional ground-based GNSS and occultation data as described above.

[0190] Generally, the computer instructions for implementing the method of the present invention can be carried on any combination of one or more computer-readable storage media. Non-transitory computer-readable storage media can include any computer-readable medium except for the signal itself, which is temporarily propagating.

[0191] Computer-readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EKROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0192] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. These programming languages ​​include object-oriented programming languages—such as Java, Smalltalk, and C++—as well as conventional procedural programming languages—such as the "C" language or similar programming languages. In particular, Python, suitable for neural network computation, and platform frameworks such as TensorFlow and PyTorch can be used. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer or to an external computer (e.g., via the Internet using an Internet service provider) through any type of network, including a local area network (LAN) or a wide area network (WAN).

[0193] The aforementioned equipment and non-transitory computer-readable storage media can be found in the detailed description of an ionospheric delay correction method based on regional ground-based GNSS and occultation data and its beneficial effects, which will not be repeated here.

[0194] Although embodiments of the present invention have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for ionospheric time delay correction based on regional ground-based GNSS and occultation data, characterized in that, include: Acquire ground-based GNSS dual-frequency observation data within the target area, extract the total electron content along the oblique path, and convert it into the total vertical electron content using a projection function; A regional ionospheric TEC model was constructed using a spherical cap harmonic function. Based on the observed vertical total electron content of all effective puncture points in the region, the coefficients to be estimated in the regional ionospheric TEC model were determined by least squares fitting. Within the target area, the total vertical electron content at each grid point is calculated at a preset resolution, and effective grid points are selected based on the number of puncture points at each grid point to form a regional grid constraint dataset. Electron density profiles from global occultation events retrieved from multi-source low Earth orbit satellites were obtained, and after quality control, they were integrated to obtain the observed values ​​of global occultation vertical total electron content. Regional grid-constrained data are used as virtual observations and fused with global occultation vertical total electron content observations to construct a joint observation equation. The weighted least squares method is then used to solve the real-time broadcasting coefficients of the broadcast ionosphere model. The user receiver receives the real-time broadcast coefficients and broadcast ephemeris, and combines them with preset fixed parameters to calculate the vertical total electron content at the puncture point. Then, it converts the total electron content of the oblique path through the ionospheric projection function, thereby determining the ionospheric delay correction number on the GNSS signal propagation path.

2. The ionospheric time delay correction method based on regional ground-based GNSS and occultation data according to claim 1, characterized in that, The regional ionosphere TEC model is as follows: ； In the formula, The vertical total electron content; and These are the latitude and longitude of the puncture point in the geographic coordinate system, respectively; and These are the latitude and longitude of the region's center, respectively. for Spend The regularized Legendre function of order 1; It is the highest order; and The coefficients of the TEC model for the ionosphere in the region to be estimated are denoted as .

3. The ionospheric time delay correction method based on regional ground-based GNSS and occultation data according to claim 1, characterized in that, Within the target area, the total vertical electron content at each grid point is calculated at a preset resolution, and effective grid points are selected based on the number of puncture points at each grid point to form a regional grid-constrained dataset, including: Within the target area, the total vertical electron content at each grid point is calculated at a resolution of 2.5° latitude × 5° longitude. Valid grid points are selected based on the criterion that the number of puncture points at each grid point is greater than or equal to 10, thus forming a regional grid-constrained dataset.

4. The ionospheric time delay correction method based on regional ground-based GNSS and occultation data according to claim 1, characterized in that, The virtual observations are fused with global occultation vertical total electron content observations at a weight ratio of 2:1 to 4:

1.

5. The ionospheric time delay correction method based on regional ground-based GNSS and occultation data according to claim 1, characterized in that, The broadcast ionosphere model is as follows: ； ； ； ； ； ； ； In the formula, The vertical total electron content; This is the delayed forecast value for the ionosphere; The parameters for the broadcast ionosphere model to be determined are: The function value is calculated based on the user's puncture point location and observation time; For regularization functions; For the standard Legendre function; and These represent the latitude and longitude of the ionospheric intersection point in the Sun-fixed geomagnetic coordinate system; These are the model prediction coefficients stored in the user receiver; This is a function value calculated based on preset curing parameters.

6. The ionospheric time delay correction method based on regional ground-based GNSS and occultation data according to claim 5, characterized in that, Construct the design matrix and use the weighted least squares method to solve for the real-time broadcast coefficients of the broadcast ionospheric model: ； In the formula, The real-time broadcast coefficients to be solved; The design matrix consists of a row corresponding to a vertical total electron content observation point, a column corresponding to a model basis function, and matrix elements representing the function values ​​of the basis functions corresponding to the observation points. ; This is the matrix transpose symbol; For observation weight array; These are observations, including virtual observations and global occultation vertical total electron content observations.

7. The ionospheric time delay correction method based on regional ground-based GNSS and occultation data according to claim 1, characterized in that, The quality control includes: Calculate the relative rate of change of electron density between adjacent height points for each effective electron density profile; The height axis is divided into several intervals. For each effective electron density profile, a robust percentile index of the relative electron density change rate is statistically analyzed in each height interval to generate noise upper limit threshold, sawtooth amplitude threshold and extreme mutation threshold. Based on the noise upper limit threshold, extreme mutation threshold, and sawtooth amplitude threshold, a three-level progressive judgment process is established, including high noise background criteria, extreme single-point jump criteria, and valid sawtooth structure criteria. If any criterion is triggered, the electron density profile is determined to be invalid and discarded.

8. An ionospheric time delay correction device based on regional ground-based GNSS and occultation data, characterized in that, The apparatus is used to implement the method according to any one of claims 1 to 7, the apparatus comprising: The regional grid-constrained dataset acquisition module is used to acquire ground-based GNSS dual-frequency observation data within the target area, extract the total electron content along the oblique path, and convert it into the total vertical electron content using a projection function. A regional ionospheric TEC model is constructed using a spherical cap harmonic function. Based on the observed total vertical electron content at all effective puncture points within the area, the coefficients to be estimated in the regional ionospheric TEC model are determined through least-squares fitting. Within the target area, the total vertical electron content at each grid point is calculated at a preset resolution, and effective grid points are selected based on the number of puncture points at each grid point, forming the regional grid-constrained dataset. The RO-VTEC observation acquisition module is used to acquire electron density profiles from global occultation events retrieved from multi-source low Earth orbit satellites. After quality control, the data are integrated to obtain the global occultation vertical total electron content observations. The real-time broadcast coefficient acquisition module is used to integrate regional grid-constrained data as virtual observations with global occultation vertical total electron content observations to construct a joint observation equation, and to solve the real-time broadcast coefficients of the broadcast ionosphere model using the weighted least squares method. The ionospheric delay correction acquisition module is used by the user receiver to receive real-time broadcast coefficients and broadcast ephemeris, combine them with preset solidification parameters, calculate the vertical total electron content at the puncture point, and then convert it into the oblique path total electron content through the ionospheric projection function, thereby determining the ionospheric delay correction on the GNSS signal propagation path.

9. An ionospheric time delay correction device based on regional ground-based GNSS and occultation data, characterized in that, Including memory and processor; The memory is used to store computer program code and transmit the computer program code to the processor; The processor is configured to execute the method as described in any one of claims 1 to 7 according to instructions in the computer program code.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images