A method and system for correcting ionospheric slant delay

Abstract

The application provides a method and system for ionospheric slant delay correction, wherein the method comprises: obtaining initial dual-frequency pseudo-range observation data, initial first carrier phase observation value, initial second carrier phase observation value, signal-to-noise ratio and satellite elevation angle; determining the dual-frequency pseudo-range observation data score level based on the above data, if the first dual-frequency pseudo-range observation score level is obtained, obtaining the historical ionospheric state and the historical state estimation error covariance matrix, and obtaining the real-time ionospheric state vector, the Kalman gain and the multi-source observation vector based on the above data; correcting the real-time ionospheric state vector based on the above data to obtain the standard ionospheric state vector; obtaining the corrected ionosphere and satellite elevation angle projection based on the standard ionospheric state vector, and calculating the ionospheric slant delay. The method for ionospheric slant delay correction provided by the application avoids the insufficient ionospheric delay correction precision caused by the limited GNSS terminal computing power and bandwidth, and realizes the high-precision positioning of GNSS.

Description

Technical Field

[0001] This invention relates to the field of global navigation satellite system positioning technology, and in particular to a method and system for correcting ionospheric slant delay. Background Technology

[0002] In Global Navigation Satellite System (GNSS) positioning technology, ionospheric delay is one of the main sources of error affecting positioning accuracy, especially in low-cost terminal applications where its impact is more significant. Traditional ionospheric delay correction methods (such as the Klobuchar model) can only correct about 60%-70% of the delay error, and the error can surge to over 100% during geomagnetic storms or space weather events, making it difficult to meet the requirements of high-precision positioning. Meanwhile, low-cost terminals themselves suffer from uneven antenna gain and susceptibility to environmental interference, resulting in significant noise in the observation data, which further exacerbates the difficulty of ionospheric inversion.

[0003] Under the current technological background, although methods such as carrier phase smoothing pseudorange have improved GNSS positioning accuracy to some extent, the following limitations still exist: First, data utilization is low. Existing technologies mostly use static thresholds to remove low-quality observations, which means that only a portion of the data is usable in complex environments such as urban canyons. Furthermore, the lack of integration of multi-dimensional parameters such as signal-to-noise ratio (SNR), elevation angle, and multipath error to dynamically evaluate the observation data further reduces the effective utilization rate of the data. Second, dynamic adaptability is poor. Smoothing algorithms with fixed window lengths cannot adaptively adjust when satellite motion speed changes, and are prone to amplifying errors due to cycle slip accumulation. Finally, there is insufficient collaboration between terminal computing power and cloud. High-precision skew delay (STEC) modeling requires real-time updates of grid parameters, but low-cost terminals are limited by computing power and bandwidth, making it difficult to efficiently process massive amounts of data. The lack of a traditional cloud-based collaborative architecture further restricts the balance between real-time positioning and accuracy, failing to meet the actual needs of low-cost terminals for high-precision positioning. Summary of the Invention

[0004] The present invention aims to provide a method and system for correcting ionospheric slant delay, so as to solve the above-mentioned technical problems, avoid insufficient ionospheric delay correction accuracy due to the limited computing power and bandwidth of GNSS terminals, and realize high-precision positioning of GNSS.

[0005] To address the aforementioned technical problems, this invention provides a method for correcting ionospheric slant delay, comprising: Acquire the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, the initial second carrier phase observation value, the signal-to-noise ratio corresponding to the initial dual-frequency pseudorange observation data, and the satellite elevation angle corresponding to the initial dual-frequency pseudorange observation data; Based on the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, the initial second carrier phase observation value, the signal-to-noise ratio, and the satellite elevation angle, the dual-frequency pseudorange observation data rating level corresponding to the initial dual-frequency pseudorange observation data is determined; If the rating level of the dual-frequency pseudorange observation data is the first dual-frequency pseudorange observation rating level, then the historical ionospheric state and the historical state estimation error covariance matrix are obtained, and based on the historical ionospheric state, the historical state estimation error covariance matrix, the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value and the initial second carrier phase observation value, the real-time ionospheric state vector, Kalman gain and multi-source observation vector are obtained. The real-time ionospheric state vector is corrected based on Kalman gain and multi-source observation vectors to obtain the standard ionospheric state vector; Based on the standard ionospheric state vector and the preset grid, the corrected ionosphere and satellite elevation angle projection are obtained, and the ionospheric slant delay is calculated based on the satellite elevation angle projection and the corrected ionosphere.

[0006] In the above scheme, by combining initial dual-frequency pseudorange observation data, initial first carrier phase observation values, initial second carrier phase observation values, signal-to-noise ratio, and satellite elevation angle to determine the scoring level of dual-frequency pseudorange observation data, the quality of the initial dual-frequency pseudorange observation data can be accurately judged, providing a scientific basis for subsequent graded difference processing. Next, when the dual-frequency pseudorange observation data score level is the first dual-frequency pseudorange observation level, by combining historical ionospheric states, historical state estimation error covariance matrix, initial first carrier phase observation values, and initial second carrier phase observation values, the real-time ionospheric state vector, Kalman gain, and multi-source observation vector are obtained. This allows for direct local computing power to complete the preliminary modeling and parameter preparation of the ionospheric state, eliminating the need to upload massive amounts of raw data to the cloud and avoiding insufficient ionospheric delay correction accuracy due to limitations in GNSS terminal computing power and bandwidth. Then, by correcting the real-time ionospheric state vector using Kalman gain and multi-source observation vectors, the influence of observation noise and model errors can be reduced, outputting a high-precision standard ionospheric state vector and preventing insufficient ionospheric delay correction accuracy. Finally, the ionosphere was corrected using the standard ionospheric state vector and a preset grid, and the satellite elevation angle projection was obtained. The ionospheric slant delay was then calculated by correcting the ionosphere and the satellite elevation angle projection, thus achieving high-precision GNSS positioning.

[0007] Furthermore, it also includes: If the rating level of the dual-frequency pseudorange observation data is the second dual-frequency pseudorange observation rating level, then the satellite motion velocity is obtained, and the smoothing window length is calculated based on the satellite motion velocity and the preset environmental interference coefficient. Based on the smoothing window length, the initial dual-frequency pseudorange observation data is smoothed to obtain smoothed dual-frequency pseudorange observation data. Based on the smoothed dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value, the ionospheric slant delay is calculated.

[0008] In the above scheme, by acquiring the satellite's motion velocity and calculating the smoothing window length in conjunction with a preset environmental interference coefficient, the smoothing window length can be adaptively adjusted to adapt to changes in satellite motion and the actual environment, thus solving the problem of poor dynamic adaptability of traditional fixed window length algorithms. Next, smoothing the initial dual-frequency pseudorange observation data using the smoothing window length effectively suppresses multipath noise and random errors in the initial dual-frequency pseudorange observation data, improving the reliability of the initial dual-frequency pseudorange observation data at the second-level rating, and laying a data foundation for subsequent ionospheric slant delay calculations. Then, by calculating the ionospheric slant delay using the smoothed dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value, an optimized ionospheric slant delay is obtained, avoiding errors caused by direct calculation of low-quality data and achieving high-precision positioning.

[0009] Furthermore, it also includes: If the dual-frequency pseudorange observation data score is the third dual-frequency pseudorange observation score, then the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value are subjected to geometric distance error elimination processing to obtain standard dual-frequency pseudorange observation data, standard first carrier phase observation value, and standard second carrier phase observation value. The ionospheric slant delay is calculated based on standard dual-frequency pseudorange observation data, standard first carrier phase observations, and standard second carrier phase observations.

[0010] In the above scheme, by performing geometric distance error elimination processing on the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value, geometric distance interference terms unrelated to the ionosphere can be removed, resulting in clean ionospheric delay-related observation data, laying the foundation for subsequent high-precision calculations. Then, the ionospheric slant delay is calculated using standard dual-frequency pseudorange observation data, standard first carrier phase observation value, and standard second carrier phase observation value, directly obtaining the corresponding ionospheric slant delay and improving correction efficiency.

[0011] Further, determining the dual-frequency pseudorange observation data rating level corresponding to the initial dual-frequency pseudorange observation data based on the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, the initial second carrier phase observation value, the signal-to-noise ratio, and the satellite elevation angle includes: Calculate the multipath error index based on the initial dual-frequency pseudorange observation data; Construct a dynamic scoring model; Based on signal-to-noise ratio, satellite elevation angle, multipath error index and dynamic scoring model, the score of dual-frequency pseudorange observation data is calculated. Based on the preset first signal-to-noise ratio threshold, the preset second signal-to-noise ratio threshold, the preset first satellite elevation angle threshold, the preset second satellite elevation angle threshold, the preset first multipath error index threshold, the preset second multipath error index threshold, the signal-to-noise ratio, the satellite elevation angle, and the multipath error index, the scoring level of the initial dual-frequency pseudorange observation data is determined.

[0012] In the above scheme, the multipath error index is calculated using initial dual-frequency pseudorange observation data, quantifying the degree of interference from multipath effects on the initial dual-frequency pseudorange observation data and supplementing key dimension indicators for data quality assessment. Next, by constructing a dynamic scoring model, quantitative correlation rules are established between signal-to-noise ratio (SNR), satellite elevation angle, multipath error index, and data quality, providing a standardized basis for subsequent scoring and grading based on dual-frequency pseudorange observation data. Then, by calculating the dual-frequency pseudorange observation data score using SNR, satellite elevation angle, multipath error index, and the dynamic scoring model, the quality of the observation data can be quantitatively assessed, avoiding the limitations of traditional static threshold judgments. Finally, by presetting a first SNR threshold, a second SNR threshold, a first satellite elevation angle threshold, a second satellite elevation angle threshold, a first multipath error index threshold, and a second multipath error index threshold, combined with SNR, satellite elevation angle, and multipath error index, the scoring level of the initial dual-frequency pseudorange observation data is determined, enabling the initial dual-frequency pseudorange observation data to be classified, providing clear classification support for subsequent differentiated processing strategies.

[0013] Furthermore, the calculation of the dual-frequency pseudorange observation data score based on signal-to-noise ratio, satellite elevation angle, multipath error index, and dynamic scoring model includes: Obtain the receiver's real-time environment; Based on the real-time environment of the receiver, obtain the dominant weight coefficient value; The score of dual-frequency pseudorange observation data is calculated based on the dominant weight coefficient, signal-to-noise ratio, satellite elevation angle, multipath error index, and dynamic scoring model.

[0014] In the above scheme, by acquiring the receiver's real-time environment and identifying the current scene and weather, a scenario-based basis can be provided for dynamically adjusting the dominant weight coefficient value. Next, by acquiring the dominant weight coefficient value through the receiver's real-time environment, the dynamic scoring model can be adaptively optimized. Then, by calculating the dual-frequency pseudorange observation data score using the dominant weight coefficient value, signal-to-noise ratio, satellite elevation angle, multipath error index, and dynamic scoring model, the resulting dual-frequency pseudorange observation data score better reflects real-time environmental characteristics, improving the accuracy and adaptability of the classification judgment.

[0015] Further, if the rating level of the dual-frequency pseudorange observation data is the first dual-frequency pseudorange observation rating level, then the historical ionospheric state and the historical state estimation error covariance matrix are obtained. Based on the historical ionospheric state, the historical state estimation error covariance matrix, the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value, the real-time ionospheric state vector, Kalman gain, and multi-source observation vector are obtained, including: If the rating level of the dual-frequency pseudorange observation data is the first dual-frequency pseudorange observation rating level, then the initial ionospheric slant delay is calculated based on the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value. Obtain the historical ionospheric state and the historical state estimation error covariance matrix, and obtain the real-time ionospheric state vector based on the historical ionospheric state, the preset state transition matrix and the preset noise; Calculate the prior state error covariance matrix based on the historical state estimation error covariance matrix; The initial puncture point position of the initial dual-frequency pseudorange observation data is calculated based on the initial dual-frequency pseudorange observation data, and an observation matrix is ​​constructed based on the initial puncture point position and the preset grid points. Calculate the Kalman gain based on the prior state error covariance matrix and the observation matrix; The initial ionospheric slant delay, pre-acquired ground station data, pre-acquired occultation data, and pre-acquired geomagnetic index are fused to obtain a multi-source observation vector.

[0016] In the above scheme, if the dual-frequency pseudorange observation data score level is the first dual-frequency pseudorange observation score level, then the initial ionospheric slant delay is calculated using the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value. This can serve as the core input for subsequent multi-source data fusion in the cloud, improving the completeness and accuracy of ionospheric delay correction in complex scenarios, while also adapting to the application requirements of low-cost terminals with limited computing power. Next, by obtaining the historical ionospheric state and the historical state estimation error covariance matrix, and combining it with a preset state transition matrix and preset noise to obtain the real-time ionospheric state vector, the current estimated ionospheric state can be optimized, improving the timeliness and accuracy of the real-time ionospheric state vector. Then, by calculating the prior state error covariance matrix using the historical state estimation error covariance matrix, the uncertainty of the prior ionospheric state vector can be quantified, providing data support for the weight allocation of the Kalman filter. Subsequently, the initial puncture point location is calculated using initial dual-frequency pseudorange observation data, and an observation matrix is ​​constructed by combining it with preset grid points. This establishes a mapping relationship between the grid VTEC and the zenith path VTEC, achieving dimensional matching between the state vector and the observation data. Next, the Kalman gain is calculated using the prior state error covariance matrix and the observation matrix, balancing the weight ratios of prior state and observation data to ensure the reliability of subsequent filtering and correction results. Finally, the initial ionospheric slant delay, pre-acquired ground station data, pre-acquired occultation data, and pre-acquired geomagnetic indices are fused to obtain a multi-source observation vector. This integrates multi-dimensional, high-precision data, improving the robustness of ionospheric state estimation and avoiding limitations caused by a single data source.

[0017] Further, the step of obtaining the corrected ionosphere and satellite elevation angle projection based on the standard ionospheric state vector and a preset grid, and calculating the ionospheric slant delay based on the satellite elevation angle projection and the corrected ionosphere, includes: The standard ionospheric state vector is filled into a preset grid to obtain VTEC grid data; The location of the puncture point is determined based on VTEC grid data and pre-acquired puncture points; Based on the location of the puncture point, the spherical distance between the puncture point and the electron density of the grid ionosphere are obtained; Based on the spherical distance of the puncture point and the electron density of the grid ionosphere, a corrected ionosphere is obtained; Obtain the satellite elevation angle of the puncture point and calculate the satellite elevation angle projection; Ionospheric slant delay is calculated based on satellite elevation angle projection and corrected ionospheric parameters.

[0018] In the above scheme, by filling the standard ionospheric state vector into a pre-defined grid, spatialized VTEC grid data can be generated, realizing a regionalized representation of the ionospheric state and providing basic data support for local interpolation calculations of GNSS terminals. Next, by determining the puncture point location using the VTEC grid data and pre-acquired puncture points, the key geometric points where satellite signals penetrate the ionosphere can be identified, establishing a correspondence between the grid data and the actual observation scenario of the terminal. Then, by obtaining the spherical distance of the puncture point and the ionospheric electron density of the grid, the core parameters required for interpolation calculations are extracted, laying a data foundation for accurately solving the local ionospheric delay. Subsequently, by obtaining the spherical distance of the puncture point and the ionospheric electron density of the grid, the corrected ionosphere can be obtained, reducing the errors caused by grid discretization. Finally, by obtaining the satellite elevation angle of the puncture point and calculating the satellite elevation angle projection, a conversion relationship between the vertical total electron content and the oblique path electron content is established, which can adapt to the actual propagation path of the satellite signal. Finally, by calculating the ionospheric slant delay through satellite elevation angle projection and ionospheric correction, a high-precision ionospheric delay correction value for the GNSS terminal in the actual observation scenario is output, avoiding the problem of insufficient correction accuracy caused by the terminal's limited computing power, and realizing high-precision GNSS positioning.

[0019] This invention provides a correction system for ionospheric slant delay, comprising a dual-frequency observation data acquisition module, a quality classification determination module, a parameter calculation module, an ionospheric state correction module, and an ionospheric slant delay solution module, specifically: The dual-frequency observation data acquisition module is used to acquire initial dual-frequency pseudorange observation data, initial first carrier phase observation value, initial second carrier phase observation value, signal-to-noise ratio corresponding to the initial dual-frequency pseudorange observation data, and satellite elevation angle corresponding to the initial dual-frequency pseudorange observation data; The quality grading determination module is used to determine the dual-frequency pseudorange observation data rating level corresponding to the initial dual-frequency pseudorange observation data based on the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, the initial second carrier phase observation value, the signal-to-noise ratio, and the satellite elevation angle. The parameter calculation module is used to obtain the historical ionospheric state and historical state estimation error covariance matrix if the rating level of the dual-frequency pseudorange observation data is the first dual-frequency pseudorange observation rating level, and to obtain the real-time ionospheric state vector, Kalman gain and multi-source observation vector based on the historical ionospheric state, historical state estimation error covariance matrix, initial dual-frequency pseudorange observation data, initial first carrier phase observation value and initial second carrier phase observation value; The ionospheric state correction module is used to correct the real-time ionospheric state vector based on Kalman gain and multi-source observation vector to obtain a standard ionospheric state vector. The ionospheric slant delay calculation module is used to obtain the corrected ionosphere and satellite elevation angle projection based on the standard ionospheric state vector and a preset grid, and to calculate the ionospheric slant delay based on the satellite elevation angle projection and the corrected ionosphere.

[0020] This invention provides an ionospheric slant delay correction system. In practical applications, it only requires a quality grading module. By combining initial dual-frequency pseudorange observation data, initial first carrier phase observation values, initial second carrier phase observation values, signal-to-noise ratio, and satellite elevation angle, the system determines the rating level of the dual-frequency pseudorange observation data. This allows for accurate quality assessment of the initial dual-frequency pseudorange observation data, providing a scientific basis for subsequent grading difference processing. Next, a parameter calculation module is used. When the dual-frequency pseudorange observation data rating level is the first dual-frequency pseudorange observation level, it combines historical ionospheric state, historical state estimation error covariance matrix, initial first carrier phase observation values, and initial second carrier phase observation values ​​to obtain the real-time ionospheric state vector, Kalman gain, and multi-source observation vector. This system can directly rely on local computing power to complete the preliminary modeling and parameter preparation of the ionospheric state, eliminating the need to upload massive amounts of raw data to the cloud and avoiding insufficient ionospheric delay correction accuracy due to limitations in GNSS terminal computing power and bandwidth. Then, an ionospheric state correction module is employed to correct the real-time ionospheric state vector using Kalman gain and multi-source observation vectors. This reduces the impact of observation noise and model errors, outputting a high-precision standard ionospheric state vector and preventing insufficient ionospheric delay correction accuracy. Finally, an ionospheric slant delay calculation module is used to correct the ionosphere using the standard ionospheric state vector and a pre-defined grid, and to obtain the satellite elevation angle projection. By correcting the ionosphere and the satellite elevation angle projection, the ionospheric slant delay is calculated, achieving high-precision GNSS positioning.

[0021] Furthermore, the parameter calculation module is also used for: If the rating level of the dual-frequency pseudorange observation data is the second dual-frequency pseudorange observation rating level, then the satellite motion velocity is obtained, and the smoothing window length is calculated based on the satellite motion velocity and the preset environmental interference coefficient. Based on the smoothing window length, the initial dual-frequency pseudorange observation data is smoothed to obtain smoothed dual-frequency pseudorange observation data. Based on the smoothed dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value, the ionospheric slant delay is calculated.

[0022] In the above scheme, by acquiring the satellite's motion velocity and calculating the smoothing window length in conjunction with a preset environmental interference coefficient, the smoothing window length can be adaptively adjusted to adapt to changes in satellite motion and the actual environment, thus solving the problem of poor dynamic adaptability of traditional fixed window length algorithms. Next, smoothing the initial dual-frequency pseudorange observation data using the smoothing window length effectively suppresses multipath noise and random errors in the initial dual-frequency pseudorange observation data, improving the reliability of the initial dual-frequency pseudorange observation data at the second-level rating, and laying a data foundation for subsequent ionospheric slant delay calculations. Then, by calculating the ionospheric slant delay using the smoothed dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value, an optimized ionospheric slant delay is obtained, avoiding errors caused by direct calculation of low-quality data and achieving high-precision positioning.

[0023] Furthermore, the parameter calculation module is also used for: If the dual-frequency pseudorange observation data score is the third dual-frequency pseudorange observation score, then the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value are subjected to geometric distance error elimination processing to obtain standard dual-frequency pseudorange observation data, standard first carrier phase observation value, and standard second carrier phase observation value. The ionospheric slant delay is calculated based on standard dual-frequency pseudorange observation data, standard first carrier phase observations, and standard second carrier phase observations.

[0024] In the above scheme, by performing geometric distance error elimination processing on the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value, geometric distance interference terms unrelated to the ionosphere can be removed, resulting in clean ionospheric delay-related observation data, laying the foundation for subsequent high-precision calculations. Then, the ionospheric slant delay is calculated using standard dual-frequency pseudorange observation data, standard first carrier phase observation value, and standard second carrier phase observation value, directly obtaining the corresponding ionospheric slant delay and improving correction efficiency. Attached Figure Description

[0025] Figure 1 A flowchart illustrating a method for correcting ionospheric slack delay according to an embodiment of the present invention; Figure 2 This is an architectural diagram of an ionospheric slant delay correction system provided in an embodiment of the present invention. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] This embodiment provides a method for correcting ionospheric slant delay; please refer to the flowchart for the method. Figure 1 ,include: Step S1: Obtain the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, the initial second carrier phase observation value, the signal-to-noise ratio corresponding to the initial dual-frequency pseudorange observation data, and the satellite elevation angle corresponding to the initial dual-frequency pseudorange observation data; Step S2: Based on the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, the initial second carrier phase observation value, the signal-to-noise ratio, and the satellite elevation angle, determine the dual-frequency pseudorange observation data rating level corresponding to the initial dual-frequency pseudorange observation data; Step S3: If the rating level of the dual-frequency pseudorange observation data is the first dual-frequency pseudorange observation rating level, then obtain the historical ionospheric state and the historical state estimation error covariance matrix, and based on the historical ionospheric state, the historical state estimation error covariance matrix, the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value, obtain the real-time ionospheric state vector, Kalman gain, and multi-source observation vector; Step S4: Correct the real-time ionospheric state vector based on Kalman gain and multi-source observation vector to obtain the standard ionospheric state vector; Step S5: Based on the standard ionospheric state vector and the preset grid, obtain the corrected ionosphere and satellite elevation angle projection, and calculate the ionospheric slant delay based on the satellite elevation angle projection and the corrected ionosphere.

[0028] In this embodiment, initial dual-frequency pseudorange observation data (P1 and P2) and initial first carrier phase observation values ​​are acquired. Initial second carrier phase observations The signal-to-noise ratio (SNR) of the initial dual-frequency pseudorange observation data is obtained through a GNSS receiver. SNR represents the ratio of received signal strength to noise intensity, measured in dB-Hz. In practice, the low-cost GNSS chip in the receiver directly estimates and outputs the SNR value for each tracked satellite through its internal signal processing link. The terminal can directly obtain this SNR value by parsing the receiver's standard data output protocol (such as NMEA-0183 or a manufacturer-defined protocol). By receiving broadcast ephemeris data from the satellites, the instantaneous three-dimensional coordinates (usually in geocentric-fixed coordinate system, ECEF) of each satellite in space are calculated based on the ephemeris parameters to obtain the satellite position. Then, through point positioning or using the result of the previous successful positioning, the approximate three-dimensional coordinates of the GNSS receiver itself, i.e., the receiver position, are obtained. Then, based on the receiver position and the satellite position, a lightweight geometry library on the embedded system is used to calculate the vector of the satellite relative to the receiver. The vector is then transformed from the ECEF coordinate system to a station-centered coordinate system (ENU) with the receiver as the origin. The angle between this vector and the local horizontal plane (northeast plane) is the satellite elevation angle corresponding to the initial dual-frequency pseudorange observation data. The calculation formula is: el = arcsin(zenith direction component of unit vector).

[0029] By combining initial dual-frequency pseudorange observation data, initial first carrier phase observations, initial second carrier phase observations, signal-to-noise ratio, and satellite elevation angle to determine the scoring level of dual-frequency pseudorange observation data, the quality of the initial dual-frequency pseudorange observation data can be accurately judged, providing a scientific basis for subsequent graded difference processing. Next, when the dual-frequency pseudorange observation data score is the first dual-frequency pseudorange observation grade, a pre-set terminal-cloud collaborative system is used to combine historical ionospheric states, historical state estimation error covariance matrix, initial first carrier phase observations, and initial second carrier phase observations to obtain real-time ionospheric state vectors, Kalman gain, and multi-source observation vectors. This allows for direct local computing power to complete the preliminary modeling and parameter preparation of the ionospheric state without the need to upload massive amounts of raw data to the cloud, avoiding insufficient ionospheric delay correction accuracy due to limited computing power and bandwidth of GNSS terminals. This is suitable for high-precision positioning needs in low-computing-power scenarios such as shared bicycles and drones. The terminal unit in the terminal-cloud collaborative system is integrated into a low-cost GNSS receiver, such as the U-blox M10 chip, which includes a dual-band antenna, a dynamic hierarchical processor, and a lightweight Conv-LSTM interpolation model. The cloud server in the terminal-cloud collaborative system deploys a GPU cluster, running a multi-station joint Kalman filter, a hierarchical Bayesian network (HBN) multi-source data fusion unit, and a regional ionospheric grid solution unit. The communication unit in the terminal-cloud collaborative system adopts the NB-IoT protocol, compressing data through differential coding to ensure that the data packet size is ≤256 bytes and the transmission delay is ≤500 ms.

[0030] Then, by correcting the real-time ionospheric state vector using Kalman gain and multi-source observation vectors, the influence of observation noise and model errors can be reduced, resulting in a high-precision standard ionospheric state vector and preventing insufficient accuracy in ionospheric delay correction. Finally, the ionosphere is corrected using the standard ionospheric state vector and a pre-defined grid, and the satellite elevation angle projection is obtained. By correcting the ionosphere and the satellite elevation angle projection, the ionospheric slant delay is calculated, achieving high-precision GNSS positioning.

[0031] Furthermore, it also includes: If the rating level of the dual-frequency pseudorange observation data is the second dual-frequency pseudorange observation rating level, then the satellite motion velocity is obtained, and the smoothing window length is calculated based on the satellite motion velocity and the preset environmental interference coefficient. Based on the smoothing window length, the initial dual-frequency pseudorange observation data is smoothed to obtain smoothed dual-frequency pseudorange observation data. Based on the smoothed dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value, the ionospheric slant delay is calculated.

[0032] In this embodiment, the smoothing window length is calculated by acquiring the satellite's motion speed and combining it with a preset environmental interference coefficient, specifically as follows: (vsat represents the satellite's velocity, which can be calculated in real-time from ephemeris data; k is a preset environmental interference coefficient, which can be 0.8 (urban canyon) or 1.2 (open environment). This allows for adaptive adjustment of the smoothing window length to adapt to changes in satellite motion and the actual environment, solving the problem of poor dynamic adaptability in traditional fixed window length algorithms. Next, by employing a carrier phase smoothing pseudorange algorithm with a smoothing window length to smooth the initial dual-frequency pseudorange observation data, multipath noise and random errors in the initial dual-frequency pseudorange observation data can be effectively suppressed, improving the reliability of the initial dual-frequency pseudorange observation data at the second-level dual-frequency pseudorange observation classification, laying a data foundation for subsequent ionospheric slant delay calculations. Simultaneously, during the smoothing process, the initial first carrier phase observation value, the initial second carrier phase observation value, and the multipath error index (MPI) are monitored in real-time. If the carrier phase change rate exceeds 0.05 cycles / s, or the MPI abrupt change is greater than 0.2m, it indicates a cycle slip or anomaly has been detected. At this time, the smoothing window length is immediately reset to the initial value L0= 10, to ensure the reliability of the smoothing results. Then, the dual-frequency pseudorange observation data ( , The ionospheric slant delay is calculated using the initial first carrier phase observations and the initial second carrier phase observations, specifically as follows: ,in, , This yields an optimized ionospheric slant delay, which avoids errors caused by direct calculation of low-quality data and achieves high-precision positioning.

[0033] Furthermore, it also includes: If the dual-frequency pseudorange observation data score is the third dual-frequency pseudorange observation score, then the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value are subjected to geometric distance error elimination processing to obtain standard dual-frequency pseudorange observation data, standard first carrier phase observation value, and standard second carrier phase observation value. The ionospheric slant delay is calculated based on standard dual-frequency pseudorange observation data, standard first carrier phase observations, and standard second carrier phase observations.

[0034] In this embodiment, by performing geometric distance error elimination processing on the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value, geometric distance interference terms unrelated to the ionosphere can be removed, resulting in pure ionospheric delay-related observation data, laying the foundation for subsequent high-precision calculations. Next, the ionospheric slant delay is calculated using standard dual-frequency pseudorange observation data, standard first carrier phase observation value, and standard second carrier phase observation value, directly yielding the corresponding ionospheric slant delay, specifically: ,in, , , This improves calibration efficiency.

[0035] Further, determining the dual-frequency pseudorange observation data rating level corresponding to the initial dual-frequency pseudorange observation data based on the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, the initial second carrier phase observation value, the signal-to-noise ratio, and the satellite elevation angle includes: Calculate the multipath error index based on the initial dual-frequency pseudorange observation data; Construct a dynamic scoring model; Based on signal-to-noise ratio, satellite elevation angle, multipath error index and dynamic scoring model, the score of dual-frequency pseudorange observation data is calculated. Based on the preset first signal-to-noise ratio threshold, the preset second signal-to-noise ratio threshold, the preset first satellite elevation angle threshold, the preset second satellite elevation angle threshold, the preset first multipath error index threshold, the preset second multipath error index threshold, the signal-to-noise ratio, the satellite elevation angle, and the multipath error index, the scoring level of the initial dual-frequency pseudorange observation data is determined.

[0036] In this embodiment, the multipath error index (MPI) is calculated using initial dual-frequency pseudorange observation data, specifically as follows: , This study quantifies the degree of interference from multipath effects on the initial dual-frequency pseudorange observation data, supplementing key dimensions of data quality assessment. Next, by constructing a dynamic scoring model, quantitative correlation rules were established between signal-to-noise ratio (SNR), satellite elevation angle (el), multipath error index (MPI), and data quality. Specifically: ,in, As the signal-to-noise ratio weight, it is set to 0.5. The satellite elevation angle weight is set to 0.3. The multipath error index weight is set to 0.2 to provide a standardized basis for subsequent scoring and grading based on dual-frequency pseudorange observation data. Then, the dual-frequency pseudorange observation data score is calculated using signal-to-noise ratio, satellite elevation angle, multipath error index, and dynamic scoring model. This enables a quantitative assessment of the observation data quality, avoiding the limitations of traditional static threshold judgment. Finally, by combining the preset first signal-to-noise ratio threshold, the preset second signal-to-noise ratio threshold, the preset first satellite elevation angle threshold, the preset second satellite elevation angle threshold, the preset first multipath error index threshold, and the preset second multipath error index threshold with the signal-to-noise ratio, satellite elevation angle, and multipath error index, the scoring level of the initial dual-frequency pseudorange observation data is determined. This allows for the classification of the initial dual-frequency pseudorange observation data, providing clear classification support for subsequent differentiated processing strategies. Specifically, in this embodiment, the preset first signal-to-noise ratio threshold is 40 dB, the preset second signal-to-noise ratio threshold is 30 dB, the preset first satellite elevation angle threshold is 30°, the preset second satellite elevation angle threshold is 15°, the preset first multipath error index threshold is 0.3 m, and the preset second multipath error index threshold is 0.6 m.When SNR > 40 dB, el > 30°, and MPI < 0.3 m, the initial dual-frequency pseudorange observation data is rated as the third dual-frequency pseudorange observation sub-level, ensuring that this type of initial dual-frequency pseudorange observation data can be directly used for ionospheric slant delay calculation, improving correction efficiency; when 30 dB ≤ SNR ≤ 40 dB, 15° ≤ el ≤ 30°, and 0.3 m ≤ MPI ≤ 0.6 m, the initial dual-frequency pseudorange observation data is rated as the second dual-frequency pseudorange observation sub-level; when SNR > 40 dB, 15° ≤ el ≤ 30°, and MPI < 0.3 m, the initial dual-frequency pseudorange observation data is rated as the second dual-frequency pseudorange observation sub-level; when SNR > 40 dB, 15° ≤ el ≤ 30°, and 0.3 m, the initial dual-frequency pseudorange observation data is rated as the third dual-frequency pseudorange observation sub-level; when SNR > 40 dB, 15° ≤ el ≤ 30°, and MPI < ... When m ≤ MPI ≤ 0.6 m, the initial dual-frequency pseudorange observation data is rated as the second dual-frequency pseudorange observation grade; when 30 dB ≤ SNR ≤ 40 dB, 15° ≤ el ≤ 30°, and MPI < 0.3 m, the initial dual-frequency pseudorange observation data is rated as the second dual-frequency pseudorange observation grade; when 30 dB ≤ SNR ≤ 40 dB, el > 30°, and MPI < 0.3 m, the initial dual-frequency pseudorange observation data is rated as the second dual-frequency pseudorange observation grade; when 30 dB ≤ SNR ≤ 40 dB, el > 30°, and 0.3 m ≤ MPI ≤ 0.6 m, the initial dual-frequency pseudorange observation data is rated as the second dual-frequency pseudorange observation grade; when SNR > 40 dB, el > 30°, and 0.3 m ≤ MPI ≤ 0.6 m, the initial dual-frequency pseudorange observation data is rated as the second dual-frequency pseudorange observation grade; when SNR > 40 dB, el > 30°, and 0.3 m ≤ MPI ≤ 0.6 m, the initial dual-frequency pseudorange observation data is rated as the second dual-frequency pseudorange observation grade; when SNR > 40 dB, el > 30°, and 0.3 m ≤ MPI ≤ 0.6 m, the initial dual-frequency pseudorange observation data is rated as the second dual-frequency pseudorange observation grade; when SNR > 40 dB, el > 30°, and 0.3 m ≤ MPI ≤ 0.6 m, the initial dual-frequency pseudorange observation data is rated as the second dual-frequency pseudorange observation grade. When the initial dual-frequency pseudorange observation data is at m, the dual-frequency pseudorange observation data score level corresponding to the initial dual-frequency pseudorange observation data is the second dual-frequency pseudorange observation sub-level; an adaptive smoothing optimization processing strategy can be matched to the initial dual-frequency pseudorange observation data of the second dual-frequency pseudorange observation sub-level to ensure correction accuracy. When SNR < 30 dB, el < 15°, or MPI > 0.6 m, the dual-frequency pseudorange observation data score level corresponding to the initial dual-frequency pseudorange observation data is the first dual-frequency pseudorange observation sub-level. The initial dual-frequency pseudorange observation data of the first dual-frequency pseudorange observation sub-level can be uploaded to the cloud for multi-source fusion processing to avoid correction failure caused by insufficient terminal computing power.

[0037] Furthermore, the calculation of the dual-frequency pseudorange observation data score based on signal-to-noise ratio, satellite elevation angle, multipath error index, and dynamic scoring model includes: Obtain the receiver's real-time environment; Based on the real-time environment of the receiver, obtain the dominant weight coefficient value; The score of dual-frequency pseudorange observation data is calculated based on the dominant weight coefficient, signal-to-noise ratio, satellite elevation angle, multipath error index, and dynamic scoring model.

[0038] In this embodiment, by acquiring the receiver's real-time environment and identifying the current scene and weather, such as whether it is an urban canyon environment or a geomagnetic storm event, a scenario-based basis can be provided for dynamically adjusting the dominant weight coefficient value. Next, by acquiring the dominant weight coefficient value through the receiver's real-time environment, the dynamic scoring model can be adaptively optimized. Specifically, since the residual error in the urban canyon environment is dominated by the multipath effect, the weight of the multipath error index is adjusted accordingly. The signal strength is increased by 20%. However, during geomagnetic storm events (when the pre-acquired geomagnetic index Kp ≥ 5), signal attenuation worsens, therefore the weight of the signal-to-noise ratio is adjusted. A 15% improvement is made to ensure that the classification strategy adapts to changes in interference. Then, the dual-frequency pseudorange observation data score is calculated by using the dominant weight coefficient value, signal-to-noise ratio, satellite elevation angle, multipath error index, and dynamic scoring model. The resulting dual-frequency pseudorange observation data score can better reflect the real-time environmental characteristics, improving the accuracy and adaptability of the classification judgment.

[0039] Further, if the rating level of the dual-frequency pseudorange observation data is the first dual-frequency pseudorange observation rating level, then the historical ionospheric state and the historical state estimation error covariance matrix are obtained. Based on the historical ionospheric state, the historical state estimation error covariance matrix, the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value, the real-time ionospheric state vector, Kalman gain, and multi-source observation vector are obtained, including: If the rating level of the dual-frequency pseudorange observation data is the first dual-frequency pseudorange observation rating level, then the initial ionospheric slant delay is calculated based on the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value. Obtain the historical ionospheric state and the historical state estimation error covariance matrix, and obtain the real-time ionospheric state vector based on the historical ionospheric state, the preset state transition matrix and the preset noise; Calculate the prior state error covariance matrix based on the historical state estimation error covariance matrix; The initial puncture point position of the initial dual-frequency pseudorange observation data is calculated based on the initial dual-frequency pseudorange observation data, and an observation matrix is ​​constructed based on the initial puncture point position and the preset grid points. Calculate the Kalman gain based on the prior state error covariance matrix and the observation matrix; The initial ionospheric slant delay, pre-acquired ground station data, pre-acquired occultation data, and pre-acquired geomagnetic index are fused to obtain a multi-source observation vector.

[0040] In this embodiment, if the dual-frequency pseudorange observation data score level is the first dual-frequency pseudorange observation score level, the initial dual-frequency pseudorange observation data belonging to the first dual-frequency pseudorange observation score level, along with the corresponding initial first carrier phase observation value and initial second carrier phase observation value, are first compressed into differential coding format via the NB-IoT communication module to ensure that the data packet size is ≤256 bytes and the transmission delay is ≤500 ms before being uploaded to the cloud server. Subsequently, the initial ionospheric slant delay is calculated using the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value. This can serve as the core input for subsequent multi-source data fusion in the cloud, improving the completeness and accuracy of ionospheric delay correction in complex scenarios, while also adapting to the application requirements of low-cost terminals with limited computing power. Next, the historical ionospheric state is obtained through the state estimation results of the previous filtering cycle (time k-1). and historical state estimation error covariance matrix Combined with the preset state transition matrix F k and preset noise w k The real-time ionospheric state vector is obtained as follows: ,in This represents the real-time ionospheric state vector (such as the TEC value at a grid point), specifically the vertical total electron content (VTEC) value of each grid point within the region. This is a preset state transition matrix. The above process can optimize the current estimated ionospheric state and improve the timeliness and accuracy of the real-time ionospheric state vector. The historical state estimation error covariance matrix quantifies the state estimate value at the previous time step. The uncertainty or confidence level is represented by the diagonal elements, which represent the estimated variance of the VTEC value for each grid point, while the off-diagonal elements reflect the correlation between the estimated errors of the VTEC values ​​for different grid points. The preset state transition matrix F k Typically, based on the time-dependent presupposition of the ionosphere, random walk or first-order Gaussian-Markov process models are used to reflect the time-varying characteristics of VTEC. Then, the prior state error covariance matrix is ​​calculated using the historical state estimation error covariance matrix, specifically: , The empirical state error covariance matrix at time k reflects the results based solely on the model. After prediction, the predicted state value at the current time. Uncertainty; To predetermine the noise covariance, the uncertainty of the prior ionospheric state vector can be quantified, providing data support for the weight allocation of the Kalman filter. This uncertainty stems from two parts: the uncertainty estimated at the previous time step. Through the model The propagation, and the uncertainty of the model itself, namely the pre-defined noise covariance. Subsequently, the initial puncture point location is calculated using the initial dual-frequency pseudorange observation data, and an observation matrix is ​​constructed by combining it with preset grid points. Specifically, based on the relationship between the initial puncture point location and the preset grid points, a projection matrix is ​​constructed, and the grid VTEC is converted into zenith path VTEC, establishing a mapping relationship between the grid VTEC and the zenith path VTEC, thus achieving dimensional matching between the state vector and the observation data. Next, the Kalman gain is calculated using the prior state error covariance matrix and the observation matrix. ,in, For Kalman gain, The observation matrix is ​​constructed based on the geometric relationship between the puncture point and the VTEC grid for each observation. This allows for the integration of multi-dimensional, high-precision data, improving the robustness of ionospheric state estimation and avoiding limitations caused by a single data source. To pre-determine the observation noise covariance, a diagonal matrix can be pre-defined based on the historical accuracy statistics of each data source. The observation noise variance of the initial dual-frequency pseudorange observation data for the first dual-frequency pseudorange observation level is set to (0.5TECU)², which balances the weighting of prior states and observation data, ensuring the reliability of subsequent filtering and correction results. Finally, the initial ionospheric slant delay, pre-acquired ground station data, pre-acquired occultation data, and pre-acquired geomagnetic index are fused. After homogenizing the data from different sources, the resulting multi-source observation vector is obtained, ensuring that all components establish effective observation equations with the ionospheric state (grid VTEC). Specifically: observation vector It is composed of the initial ionospheric slant delay, pre-acquired ground station data, pre-acquired occultation data, and pre-acquired geomagnetic index. , in, For the initial ionospheric slack delay, i=1…M, To obtain base station data in advance, j=1…N, To obtain occultation data, l=1…L To pre-acquire the geomagnetic index, the pre-acquired ground station data is set to (0.1 TECU)², the pre-acquired occultation data is set to (0.3 TECU)², and the pre-acquired geomagnetic index is set to (0.5)². The pre-acquired geomagnetic index, as a "virtual observation" describing the overall level of ionospheric disturbance, can be correlated with the VTEC level or disturbance intensity of the entire region by establishing an empirical observation equation. For example, the observation equation can be expressed as: Where h is a vector representing the ionospheric state. (Grid VTEC) is mapped to a linear or nonlinear function related to the Kp index domain, where v is the preset observation noise, so that the pre-acquired geomagnetic index Kp can be used as a constraint to influence the filtering solution, especially improving the robustness of the model during geomagnetic storms.

[0041] Further, the step of obtaining the corrected ionosphere and satellite elevation angle projection based on the standard ionospheric state vector and a preset grid, and calculating the ionospheric slant delay based on the satellite elevation angle projection and the corrected ionosphere, includes: The standard ionospheric state vector is filled into a preset grid to obtain VTEC grid data; The location of the puncture point is determined based on VTEC grid data and pre-acquired puncture points; Based on the location of the puncture point, the spherical distance between the puncture point and the electron density of the grid ionosphere are obtained; Based on the spherical distance of the puncture point and the electron density of the grid ionosphere, a corrected ionosphere is obtained; Obtain the satellite elevation angle of the puncture point and calculate the satellite elevation angle projection; Ionospheric slant delay is calculated based on satellite elevation angle projection and corrected ionospheric parameters.

[0042] In this embodiment, the standard ionospheric state vector is filled into several preset grids within a target area divided according to a preset grid. The resulting standard ionospheric state vector is then encoded using the internationally recognized RTCM (Radio Technical Commission for Maritime Services) standard protocol, compressed into differential coding format via the NB-IoT communication module, and sent to the terminal. The terminal receives the differentially coded compressed ionospheric grid parameters from the cloud and generates spatialized VTEC grid data through inverse distance model interpolation, achieving a regionalized representation of the ionospheric state and providing basic data support for local interpolation calculations of the GNSS terminal. The preset grid is a 0.5°×0.5° latitude and longitude grid, with a grid data update rate of 5 minutes / period to ensure real-time requirements. Next, the puncture point location is determined using the VTEC grid data and pre-acquired puncture points. Specifically, the approximate position of the receiver is first obtained, and the terminal obtains its own latitude, longitude, and elevation through single-point positioning or by using the previously successful positioning result. ,in, The longitude of the receiver. The geodetic latitude of the receiver. The receiver's geodetic height is determined and converted to spatial rectangular coordinates. Next, the satellite position is acquired: based on the received broadcast ephemeris, the satellite's instantaneous coordinates in space are calculated. Then, the puncture point coordinates are calculated: using a standard single-layer ionospheric model (usually assumed to be a thin spherical shell at a height of 350 km), the puncture point is the intersection of the line connecting the receiver and the satellite with this single-layer model, thus obtaining the pre-acquired puncture point. Based on the VTEC grid data, the puncture point location is determined, enabling the identification of key geometric points where the satellite signal penetrates the ionosphere, establishing a correspondence between the grid data and the actual observation scene at the terminal. Then, the spherical distance of the puncture point and the grid ionospheric electron density are obtained through the puncture point location, extracting the core parameters required for interpolation calculations, laying the data foundation for accurately solving the local ionospheric delay. The spherical distance of the puncture point can be selected as the spherical distance from the puncture point to the nearest four grid points. Subsequently, the corrected ionosphere is obtained through the spherical distance of the puncture point and the grid ionospheric electron density, specifically: ,in To correct the ionosphere, Let the inverse distance weight be the i-th grid point. Let d be the ionospheric electron density of the i-th lattice point. i The spherical distance to the puncture point helps mitigate errors introduced by grid discretization. Then, the satellite elevation angle of the puncture point is obtained, and the satellite elevation angle projection is calculated, specifically as follows: Where el is the satellite elevation angle of the puncture point, z' is the zenith angle at the puncture point, R is the average radius of the Earth (6371km), and H is the height of a single layer of the ionosphere (350km). The correction factor, typically 0.9782, is used to compensate for the effects of Earth's oblateness. A conversion relationship between the vertical total electron content and the oblique path electron content is established to adapt to the actual propagation path of satellite signals. Finally, the ionospheric oblique delay is calculated using satellite elevation angle projection and a corrected ionosphere, outputting a high-precision ionospheric delay correction value for the GNSS terminal in actual observation scenarios. Specifically: This avoids the problem of insufficient correction accuracy caused by the limited computing power of the terminal, and realizes high-precision GNSS positioning.

[0043] This embodiment provides a correction system for ionospheric slant delay. Please refer to [link to relevant documentation]. Figure 2 It includes a dual-frequency observation data acquisition module, a quality grading determination module, a parameter calculation module, an ionospheric state correction module, and an ionospheric oblique delay solution module, specifically: The dual-frequency observation data acquisition module is used to acquire initial dual-frequency pseudorange observation data, initial first carrier phase observation value, initial second carrier phase observation value, signal-to-noise ratio corresponding to the initial dual-frequency pseudorange observation data, and satellite elevation angle corresponding to the initial dual-frequency pseudorange observation data. The quality grading determination module is used to determine the dual-frequency pseudorange observation data rating level corresponding to the initial dual-frequency pseudorange observation data based on the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, the initial second carrier phase observation value, the signal-to-noise ratio, and the satellite elevation angle. The parameter calculation module is used to obtain the historical ionospheric state and historical state estimation error covariance matrix if the rating level of the dual-frequency pseudorange observation data is the first dual-frequency pseudorange observation rating level, and to obtain the real-time ionospheric state vector, Kalman gain and multi-source observation vector based on the historical ionospheric state, historical state estimation error covariance matrix, initial dual-frequency pseudorange observation data, initial first carrier phase observation value and initial second carrier phase observation value; The ionospheric state correction module is used to correct the real-time ionospheric state vector based on Kalman gain and multi-source observation vector to obtain a standard ionospheric state vector. The ionospheric slant delay calculation module is used to obtain the corrected ionosphere and satellite elevation angle projection based on the standard ionospheric state vector and a preset grid, and to calculate the ionospheric slant delay based on the satellite elevation angle projection and the corrected ionosphere.

[0044] This embodiment provides an ionospheric slant delay correction system. In practical applications, it only requires a dual-frequency observation data acquisition module to acquire initial dual-frequency pseudorange observation data (P1 and P2) and initial first carrier phase observation values. Initial second carrier phase observations The signal-to-noise ratio (SNR) of the initial dual-frequency pseudorange observation data is obtained through a GNSS receiver. SNR represents the ratio of received signal strength to noise intensity, measured in dB-Hz. In practice, the low-cost GNSS chip in the receiver directly estimates and outputs the SNR value for each tracked satellite through its internal signal processing link. The terminal can directly obtain this SNR value by parsing the receiver's standard data output protocol (such as NMEA-0183 or a manufacturer-defined protocol). By receiving broadcast ephemeris data from the satellites, the instantaneous three-dimensional coordinates (usually in geocentric-fixed coordinate system, ECEF) of each satellite in space are calculated based on the ephemeris parameters to obtain the satellite position. Then, through point positioning or using the result of the previous successful positioning, the approximate three-dimensional coordinates of the GNSS receiver itself, i.e., the receiver position, are obtained. Then, based on the receiver position and the satellite position, a lightweight geometry library on the embedded system is used to calculate the vector of the satellite relative to the receiver. The vector is then transformed from the ECEF coordinate system to a station-centered coordinate system (ENU) with the receiver as the origin. The angle between this vector and the local horizontal plane (northeast plane) is the satellite elevation angle corresponding to the initial dual-frequency pseudorange observation data, calculated using the formula: el = arcsin(zenith component of the unit vector). Then, a quality grading module is used to determine the rating level of the dual-frequency pseudorange observation data by combining the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, the initial second carrier phase observation value, the signal-to-noise ratio, and the satellite elevation angle. This allows for accurate quality assessment of the initial dual-frequency pseudorange observation data, providing a scientific basis for subsequent grading difference processing. Next, the parameter calculation module is used. When the dual-frequency pseudorange observation data score level is the first dual-frequency pseudorange observation score level, the real-time ionospheric state vector, Kalman gain and multi-source observation vector are obtained by combining the historical ionospheric state, historical state estimation error covariance matrix, initial first carrier phase observation value and initial second carrier phase observation value through the preset terminal-cloud collaborative system. It can directly rely on local computing power to complete the preliminary modeling and parameter preparation of the ionospheric state without having to upload massive amounts of raw data to the cloud. This avoids insufficient ionospheric delay correction accuracy due to the limited computing power and bandwidth of GNSS terminals. It is suitable for high-precision positioning needs in low-computing-power scenarios such as shared bicycles and drones. The terminal unit in the terminal-cloud collaborative system is integrated into a low-cost GNSS receiver, such as the U-blox M10 chip, which includes a dual-band antenna, a dynamic hierarchical processor, and a lightweight Conv-LSTM interpolation model. The cloud server in the terminal-cloud collaborative system deploys a GPU cluster, running a multi-station joint Kalman filter, a hierarchical Bayesian network (HBN) multi-source data fusion unit, and a regional ionospheric grid solution unit. The communication unit in the terminal-cloud collaborative system adopts the NB-IoT protocol, compressing data through differential coding to ensure that the data packet size is ≤256 bytes and the transmission delay is ≤500 ms.Then, an ionospheric state correction module is employed to correct the real-time ionospheric state vector using Kalman gain and multi-source observation vectors. This reduces the impact of observation noise and model errors, outputting a high-precision standard ionospheric state vector and preventing insufficient ionospheric delay correction accuracy. Finally, an ionospheric slant delay calculation module is used to correct the ionosphere using the standard ionospheric state vector and a pre-defined grid, and to obtain the satellite elevation angle projection. By correcting the ionosphere and the satellite elevation angle projection, the ionospheric slant delay is calculated, achieving high-precision GNSS positioning.

[0045] Furthermore, the parameter calculation module is also used for: If the rating level of the dual-frequency pseudorange observation data is the second dual-frequency pseudorange observation rating level, then the satellite motion velocity is obtained, and the smoothing window length is calculated based on the satellite motion velocity and the preset environmental interference coefficient. Based on the smoothing window length, the initial dual-frequency pseudorange observation data is smoothed to obtain smoothed dual-frequency pseudorange observation data. Based on the smoothed dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value, the ionospheric slant delay is calculated.

[0046] In this embodiment, the smoothing window length is calculated by acquiring the satellite's motion velocity and combining it with a preset environmental interference coefficient. Specifically: (vsat represents the satellite's velocity, which can be calculated in real-time from ephemeris data; k is a preset environmental interference coefficient, which can be 0.8 (urban canyon) or 1.2 (open environment). This allows for adaptive adjustment of the smoothing window length to adapt to changes in satellite motion and the actual environment, solving the problem of poor dynamic adaptability in traditional fixed window length algorithms. Next, by employing a carrier phase smoothing pseudorange algorithm with a smoothing window length to smooth the initial dual-frequency pseudorange observation data, multipath noise and random errors in the initial dual-frequency pseudorange observation data can be effectively suppressed, improving the reliability of the initial dual-frequency pseudorange observation data at the second-level rating, laying a data foundation for subsequent ionospheric slant delay calculations. Simultaneously, during the smoothing process, the initial first carrier phase observation value, the initial second carrier phase observation value, and the multipath error index (MPI) are monitored in real-time. If the carrier phase change rate exceeds 0.05 cycles / s, or the MPI abrupt change is greater than 0.2...) m represents the detection of cycle slips or anomalies. At this point, the smoothing window length is immediately reset to the initial value L0=10 to ensure the reliability of the smoothing results. Then, the dual-frequency pseudorange observation data is smoothed ( , The ionospheric slant delay is calculated using the initial first carrier phase observations and the initial second carrier phase observations, specifically as follows: , This yields an optimized ionospheric slant delay, which avoids errors caused by direct calculation of low-quality data and achieves high-precision positioning.

[0047] Furthermore, the parameter calculation module is also used for: If the dual-frequency pseudorange observation data score is the third dual-frequency pseudorange observation score, then the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value are subjected to geometric distance error elimination processing to obtain standard dual-frequency pseudorange observation data, standard first carrier phase observation value, and standard second carrier phase observation value. The ionospheric slant delay is calculated based on standard dual-frequency pseudorange observation data, standard first carrier phase observations, and standard second carrier phase observations.

[0048] In this embodiment, by performing geometric distance error elimination processing on the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value, geometric distance interference terms unrelated to the ionosphere can be removed, resulting in pure ionospheric delay-related observation data, laying the foundation for subsequent high-precision calculations. Next, the ionospheric slant delay is calculated using standard dual-frequency pseudorange observation data, standard first carrier phase observation value, and standard second carrier phase observation value, directly yielding the corresponding ionospheric slant delay, specifically: ,in, , , This improves calibration efficiency.

[0049] Furthermore, the quality grading determination module is used to determine the dual-frequency pseudorange observation data rating level corresponding to the initial dual-frequency pseudorange observation data based on the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, the initial second carrier phase observation value, the signal-to-noise ratio, and the satellite elevation angle, including: Calculate the multipath error index based on the initial dual-frequency pseudorange observation data; Construct a dynamic scoring model; Based on signal-to-noise ratio, satellite elevation angle, multipath error index and dynamic scoring model, the score of dual-frequency pseudorange observation data is calculated. Based on the preset first signal-to-noise ratio threshold, the preset second signal-to-noise ratio threshold, the preset first satellite elevation angle threshold, the preset second satellite elevation angle threshold, the preset first multipath error index threshold, the preset second multipath error index threshold, the signal-to-noise ratio, the satellite elevation angle, and the multipath error index, the scoring level of the initial dual-frequency pseudorange observation data is determined.

[0050] In this embodiment, the multipath error index (MPI) is calculated using initial dual-frequency pseudorange observation data, specifically as follows: , This study quantifies the degree of interference from multipath effects on the initial dual-frequency pseudorange observation data, supplementing key dimensions of data quality assessment. Next, by constructing a dynamic scoring model, quantitative correlation rules were established between signal-to-noise ratio (SNR), satellite elevation angle (el), multipath error index (MPI), and data quality. Specifically: ,in, As the signal-to-noise ratio weight, it is set to 0.5. The satellite elevation angle weight is set to 0.3. The multipath error index weight is set to 0.2 to provide a standardized basis for subsequent scoring and grading based on dual-frequency pseudorange observation data. Then, the dual-frequency pseudorange observation data score is calculated using signal-to-noise ratio, satellite elevation angle, multipath error index, and dynamic scoring model. This enables a quantitative assessment of the observation data quality, avoiding the limitations of traditional static threshold judgment.

[0051] Finally, by combining the preset first signal-to-noise ratio threshold, the preset second signal-to-noise ratio threshold, the preset first satellite elevation angle threshold, the preset second satellite elevation angle threshold, the preset first multipath error index threshold, and the preset second multipath error index threshold with the signal-to-noise ratio, satellite elevation angle, and multipath error index, the scoring level of the initial dual-frequency pseudorange observation data is determined. This allows for the classification of the initial dual-frequency pseudorange observation data, providing clear classification support for subsequent differentiated processing strategies. Specifically, in this embodiment, the preset first signal-to-noise ratio threshold is 40 dB, the preset second signal-to-noise ratio threshold is 30 dB, the preset first satellite elevation angle threshold is 30°, the preset second satellite elevation angle threshold is 15°, the preset first multipath error index threshold is 0.3 m, and the preset second multipath error index threshold is 0.6 m.

[0052] When SNR>40 dB, el>30° and MPI<0.3 m, the initial dual-frequency pseudorange observation data corresponds to the third dual-frequency pseudorange observation data rating level, which can ensure that this type of initial dual-frequency pseudorange observation data can be directly used for ionospheric slant delay calculation and improve correction efficiency. When 30 dB≤SNR≤40 dB, 15°≤el≤30°, and 0.3 m≤MPI≤0.6 m, the rating level of the initial dual-frequency pseudorange observation data is the second dual-frequency pseudorange observation rating level. When SNR>40 dB, 15°≤el≤30° and MPI<0.3 m, the rating level of the initial dual-frequency pseudorange observation data is the second dual-frequency pseudorange observation rating level. When SNR>40 dB, 15°≤el≤30° and 0.3 m≤MPI≤0.6 m, the rating level of the initial dual-frequency pseudorange observation data is the second dual-frequency pseudorange observation rating level. When 30 dB≤SNR≤40 dB, 15°≤el≤30° and MPI<0.3 m, the rating level of the initial dual-frequency pseudorange observation data is the second dual-frequency pseudorange observation rating level. When 30 dB≤SNR≤40 dB, el>30° and MPI<0.3 m, the rating level of the initial dual-frequency pseudorange observation data is the second dual-frequency pseudorange observation rating level. When 30 dB≤SNR≤40 dB, el>30° and 0.3 m≤MPI≤0.6 m, the rating level of the initial dual-frequency pseudorange observation data is the second dual-frequency pseudorange observation rating level. When SNR>40 dB, el>30° and 0.3 m≤MPI≤0.6 m, the initial dual-frequency pseudorange observation data corresponds to the second dual-frequency pseudorange observation sub-level. An adaptive smoothing optimization processing strategy can be matched to the initial dual-frequency pseudorange observation data of the second dual-frequency pseudorange observation sub-level to ensure correction accuracy.

[0053] When SNR < 30 dB, el < 15°, or MPI > 0.6 m, the initial dual-frequency pseudorange observation data corresponds to the first dual-frequency pseudorange observation sub-level. The initial dual-frequency pseudorange observation data at the first dual-frequency pseudorange observation sub-level can be uploaded to the cloud for multi-source fusion processing, avoiding correction failures caused by insufficient terminal computing power.

[0054] Furthermore, the calculation of the dual-frequency pseudorange observation data score based on signal-to-noise ratio, satellite elevation angle, multipath error index, and dynamic scoring model includes: Obtain the receiver's real-time environment; Based on the real-time environment of the receiver, obtain the dominant weight coefficient value; The score of dual-frequency pseudorange observation data is calculated based on the dominant weight coefficient, signal-to-noise ratio, satellite elevation angle, multipath error index, and dynamic scoring model.

[0055] In this embodiment, by acquiring the receiver's real-time environment and identifying the current scene and weather, such as whether it is an urban canyon environment or a geomagnetic storm event, a scenario-based basis can be provided for dynamically adjusting the dominant weight coefficient value. Next, by acquiring the dominant weight coefficient value through the receiver's real-time environment, the dynamic scoring model can be adaptively optimized. Specifically, since the residual error in the urban canyon environment is dominated by the multipath effect, the weight of the multipath error index is adjusted accordingly. The signal strength is increased by 20%. However, during geomagnetic storm events (when the pre-acquired geomagnetic index Kp ≥ 5), signal attenuation worsens, therefore the weight of the signal-to-noise ratio is adjusted. A 15% improvement is made to ensure that the classification strategy adapts to changes in interference. Then, the dual-frequency pseudorange observation data score is calculated by using the dominant weight coefficient value, signal-to-noise ratio, satellite elevation angle, multipath error index, and dynamic scoring model. The resulting dual-frequency pseudorange observation data score can better reflect the real-time environmental characteristics, improving the accuracy and adaptability of the classification judgment.

[0056] Further, the parameter calculation module is used to, if the rating level of the dual-frequency pseudorange observation data is the first dual-frequency pseudorange observation rating level, obtain the historical ionospheric state and the historical state estimation error covariance matrix, and based on the historical ionospheric state, the historical state estimation error covariance matrix, the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value, obtain the real-time ionospheric state vector, Kalman gain, and multi-source observation vector, including: If the rating level of the dual-frequency pseudorange observation data is the first dual-frequency pseudorange observation rating level, then the initial ionospheric slant delay is calculated based on the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value. Obtain the historical ionospheric state and the historical state estimation error covariance matrix, and obtain the real-time ionospheric state vector based on the historical ionospheric state, the preset state transition matrix and the preset noise; Calculate the prior state error covariance matrix based on the historical state estimation error covariance matrix; The initial puncture point position of the initial dual-frequency pseudorange observation data is calculated based on the initial dual-frequency pseudorange observation data, and an observation matrix is ​​constructed based on the initial puncture point position and the preset grid points. Calculate the Kalman gain based on the prior state error covariance matrix and the observation matrix; The initial ionospheric slant delay, pre-acquired ground station data, pre-acquired occultation data, and pre-acquired geomagnetic index are fused to obtain a multi-source observation vector.

[0057] In this embodiment, if the dual-frequency pseudorange observation data score level is the first dual-frequency pseudorange observation score level, the initial dual-frequency pseudorange observation data belonging to the first dual-frequency pseudorange observation score level, along with the corresponding initial first carrier phase observation value and initial second carrier phase observation value, are first compressed into differential coding format via the NB-IoT communication module to ensure that the data packet size is ≤256 bytes and the transmission delay is ≤500 ms before being uploaded to the cloud server. Subsequently, the initial ionospheric slant delay is calculated using the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value. This can serve as the core input for subsequent multi-source data fusion in the cloud, improving the completeness and accuracy of ionospheric delay correction in complex scenarios, while also adapting to the application requirements of low-cost terminals with limited computing power. Next, the historical ionospheric state is obtained through the state estimation results of the previous filtering cycle (time k-1). and historical state estimation error covariance matrix Combined with the preset state transition matrix F k and preset noise w k The real-time ionospheric state vector is obtained as follows: ,in This represents the real-time ionospheric state vector (such as the TEC value at a grid point), specifically the vertical total electron content (VTEC) value of each grid point within the region. This is the preset state transition matrix.

[0058] The above process optimizes the current estimated ionospheric state, improving the timeliness and accuracy of the real-time ionospheric state vector. The historical state estimation error covariance matrix quantifies the state estimate from the previous time step. The uncertainty or confidence level is represented by the diagonal elements, which represent the estimated variance of the VTEC value for each grid point, while the off-diagonal elements reflect the correlation between the estimated errors of the VTEC values ​​for different grid points. The preset state transition matrix F k Typically, based on the time-dependent presupposition of the ionosphere, random walk or first-order Gaussian-Markov process models are used to reflect the time-varying characteristics of VTEC. Then, the prior state error covariance matrix is ​​calculated using the historical state estimation error covariance matrix, specifically: , The empirical state error covariance matrix at time k reflects the results based solely on the model. After making the prediction, the predicted state value at the current time. Uncertainty; To predetermine the noise covariance, the uncertainty of the prior ionospheric state vector can be quantified, providing data support for the weight allocation of the Kalman filter. This uncertainty stems from two parts: the uncertainty estimated at the previous time step. Through the model The propagation, and the uncertainty of the model itself, namely the pre-defined noise covariance. .

[0059] Subsequently, the initial puncture point location is calculated using the initial dual-frequency pseudorange observation data, and an observation matrix is ​​constructed based on the preset grid points. Specifically, a projection matrix is ​​constructed according to the relationship between the initial puncture point location and the preset grid points, and the grid VTEC is converted into zenith path VTEC, establishing a mapping relationship between the grid VTEC and the zenith path VTEC, thus achieving dimensional matching between the state vector and the observation data. Next, the Kalman gain is calculated using the prior state error covariance matrix and the observation matrix. ,in, For Kalman gain, The observation matrix is ​​constructed based on the geometric relationship between the puncture point and the VTEC grid for each observation. This allows for the integration of multi-dimensional, high-precision data, improving the robustness of ionospheric state estimation and avoiding limitations caused by a single data source. To pre-determine the observation noise covariance, a diagonal matrix can be pre-defined based on the historical accuracy statistics of each data source. The observation noise variance of the initial dual-frequency pseudorange observation data for the first dual-frequency pseudorange observation level is set to (0.5TECU)², which balances the weighting of prior states and observation data, ensuring the reliability of subsequent filtering and correction results. Finally, the initial ionospheric slant delay, pre-acquired ground station data, pre-acquired occultation data, and pre-acquired geomagnetic index are fused. After homogenizing the data from different sources, the resulting multi-source observation vector is obtained, ensuring that all components establish effective observation equations with the ionospheric state (grid VTEC). Specifically: observation vector It is composed of the initial ionospheric slant delay, pre-acquired ground station data, pre-acquired occultation data, and pre-acquired geomagnetic index. ,in, For the initial ionospheric slack delay, i=1…M, To obtain base station data in advance, j=1…N, To obtain occultation data, l=1…L To pre-acquire the geomagnetic index, the pre-acquired ground station data is set to (0.1 TECU)², the pre-acquired occultation data is set to (0.3 TECU)², and the pre-acquired geomagnetic index is set to (0.5)². The pre-acquired geomagnetic index, as a "virtual observation" describing the overall level of ionospheric disturbance, can be correlated with the VTEC level or disturbance intensity of the entire region by establishing an empirical observation equation. For example, the observation equation can be expressed as: Where h is a vector representing the ionospheric state. (Grid VTEC) is mapped to a linear or nonlinear function related to the Kp index domain, where v is the preset observation noise, so that the pre-acquired geomagnetic index Kp can be used as a constraint to influence the filtering solution, especially improving the robustness of the model during geomagnetic storms.

[0060] Further, the ionospheric slant delay calculation module is used to obtain the corrected ionosphere and satellite elevation angle projection based on the standard ionospheric state vector and a preset grid, and to calculate the ionospheric slant delay based on the satellite elevation angle projection and the corrected ionosphere, including: The standard ionospheric state vector is filled into a preset grid to obtain VTEC grid data; The location of the puncture point is determined based on VTEC grid data and pre-acquired puncture points; Based on the location of the puncture point, the spherical distance between the puncture point and the electron density of the grid ionosphere are obtained; Based on the spherical distance of the puncture point and the electron density of the grid ionosphere, a corrected ionosphere is obtained; Obtain the satellite elevation angle of the puncture point and calculate the satellite elevation angle projection; Ionospheric slant delay is calculated based on satellite elevation angle projection and corrected ionospheric parameters.

[0061] In this embodiment, a standard ionospheric state vector is filled into several preset grids within a target area divided according to a preset grid. The resulting standard ionospheric state vector is then encoded using the internationally recognized RTCM (Radio Technical Commission for Maritime Services) standard protocol, compressed into differential encoding format via the NB-IoT communication module, and sent to the terminal. The terminal receives the differentially encoded compressed ionospheric grid parameters from the cloud and generates spatialized VTEC grid data through inverse distance model interpolation. This achieves a regionalized representation of the ionospheric state, providing basic data support for local interpolation calculations by the GNSS terminal. The preset grid is a 0.5° × 0.5° latitude and longitude grid, with a grid data update rate of 5 minutes per period to ensure real-time requirements.

[0062] Next, the location of the puncture point is determined using VTEC grid data and pre-acquired puncture points. Specifically, the approximate location of the receiver is first obtained, and the terminal obtains its own latitude, longitude, and elevation through single-point positioning or by using the previously successful positioning result. ,in, The longitude of the receiver. The geodetic latitude of the receiver. The receiver's geodetic height is determined and converted to spatial rectangular coordinates. Next, the satellite position is acquired: based on the received broadcast ephemeris, the satellite's instantaneous coordinates in space are calculated. Then, the puncture point coordinates are calculated: using a standard single-layer ionospheric model (usually assumed to be a thin spherical shell at a height of 350 km), the puncture point is the intersection of the line connecting the receiver and the satellite with this single-layer model, thus obtaining the pre-acquired puncture point. Based on the VTEC grid data, the puncture point location is determined, enabling the identification of key geometric points where the satellite signal penetrates the ionosphere, and establishing a correspondence between the grid data and the actual observation scene at the terminal.

[0063] Then, by obtaining the spherical distance between the puncture points and the ionospheric electron density of the grid, the core parameters required for interpolation calculations were extracted, laying a data foundation for accurately solving the local ionospheric delay. The spherical distance between the puncture points can be selected as the spherical distance from the puncture point to the four nearest grid points. Subsequently, the corrected ionosphere is obtained using the spherical distance between the puncture points and the ionospheric electron density of the grid, specifically: ,in To correct the ionosphere, Let be the inverse distance weight of the i-th grid point. Let d be the ionospheric electron density of the i-th lattice point. i The spherical distance to the puncture point helps mitigate errors introduced by grid discretization. Then, the satellite elevation angle of the puncture point is obtained, and the satellite elevation angle projection is calculated, specifically as follows: Where el is the satellite elevation angle of the puncture point, z' is the zenith angle at the puncture point, R is the average radius of the Earth (6371km), and H is the height of a single layer of the ionosphere (350km). The correction factor, typically 0.9782, is used to compensate for the effects of Earth's oblateness. A conversion relationship between the vertical total electron content and the oblique path electron content is established to adapt to the actual propagation path of satellite signals. Finally, the ionospheric oblique delay is calculated using satellite elevation angle projection and a corrected ionosphere, outputting a high-precision ionospheric delay correction value for the GNSS terminal in actual observation scenarios. Specifically: This avoids the problem of insufficient correction accuracy caused by the limited computing power of the terminal, and realizes high-precision GNSS positioning.

[0064] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for correcting ionospheric slack delay, characterized in that, include: Acquire the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, the initial second carrier phase observation value, the signal-to-noise ratio corresponding to the initial dual-frequency pseudorange observation data, and the satellite elevation angle corresponding to the initial dual-frequency pseudorange observation data; Based on the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, the initial second carrier phase observation value, the signal-to-noise ratio, and the satellite elevation angle, the dual-frequency pseudorange observation data rating level corresponding to the initial dual-frequency pseudorange observation data is determined; If the rating level of the dual-frequency pseudorange observation data is the first dual-frequency pseudorange observation rating level, then the historical ionospheric state and the historical state estimation error covariance matrix are obtained, and based on the historical ionospheric state, the historical state estimation error covariance matrix, the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value and the initial second carrier phase observation value, the real-time ionospheric state vector, Kalman gain and multi-source observation vector are obtained. The real-time ionospheric state vector is corrected based on Kalman gain and multi-source observation vectors to obtain the standard ionospheric state vector; Based on the standard ionospheric state vector and the preset grid, the corrected ionosphere and satellite elevation angle projection are obtained, and the ionospheric slant delay is calculated based on the satellite elevation angle projection and the corrected ionosphere.

2. The method for correcting ionospheric slack delay according to claim 1, characterized in that, Also includes: If the rating level of the dual-frequency pseudorange observation data is the second dual-frequency pseudorange observation rating level, then the satellite motion velocity is obtained, and the smoothing window length is calculated based on the satellite motion velocity and the preset environmental interference coefficient. Based on the smoothing window length, the initial dual-frequency pseudorange observation data is smoothed to obtain smoothed dual-frequency pseudorange observation data. Based on the smoothed dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value, the ionospheric slant delay is calculated.

3. The method for correcting ionospheric slack delay according to claim 1, characterized in that, Also includes: If the dual-frequency pseudorange observation data score is the third dual-frequency pseudorange observation score, then the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value are subjected to geometric distance error elimination processing to obtain standard dual-frequency pseudorange observation data, standard first carrier phase observation value, and standard second carrier phase observation value. The ionospheric slant delay is calculated based on standard dual-frequency pseudorange observation data, standard first carrier phase observations, and standard second carrier phase observations.

4. The method for correcting ionospheric slack delay according to claim 1, characterized in that, The determination of the dual-frequency pseudorange observation data rating level corresponding to the initial dual-frequency pseudorange observation data based on the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, the initial second carrier phase observation value, the signal-to-noise ratio, and the satellite elevation angle includes: Calculate the multipath error index based on the initial dual-frequency pseudorange observation data; Construct a dynamic scoring model; Based on signal-to-noise ratio, satellite elevation angle, multipath error index and dynamic scoring model, the score of dual-frequency pseudorange observation data is calculated. Based on the preset first signal-to-noise ratio threshold, the preset second signal-to-noise ratio threshold, the preset first satellite elevation angle threshold, the preset second satellite elevation angle threshold, the preset first multipath error index threshold, the preset second multipath error index threshold, the signal-to-noise ratio, the satellite elevation angle, and the multipath error index, the scoring level of the initial dual-frequency pseudorange observation data is determined.

5. The method for correcting ionospheric slack delay according to claim 4, characterized in that, The scoring of dual-frequency pseudorange observation data is calculated based on signal-to-noise ratio, satellite elevation angle, multipath error index, and dynamic scoring model, including: Obtain the receiver's real-time environment; Based on the real-time environment of the receiver, obtain the dominant weight coefficient value; The score of dual-frequency pseudorange observation data is calculated based on the dominant weight coefficient, signal-to-noise ratio, satellite elevation angle, multipath error index, and dynamic scoring model.

6. The method for correcting ionospheric slack delay according to claim 1, characterized in that, If the rating level of the dual-frequency pseudorange observation data is the first dual-frequency pseudorange observation rating level, then the historical ionospheric state and the historical state estimation error covariance matrix are obtained. Based on the historical ionospheric state, the historical state estimation error covariance matrix, the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value, the real-time ionospheric state vector, Kalman gain, and multi-source observation vector are obtained, including: If the rating level of the dual-frequency pseudorange observation data is the first dual-frequency pseudorange observation rating level, then the initial ionospheric slant delay is calculated based on the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value. Obtain the historical ionospheric state and the historical state estimation error covariance matrix, and obtain the real-time ionospheric state vector based on the historical ionospheric state, the preset state transition matrix and the preset noise; Calculate the prior state error covariance matrix based on the historical state estimation error covariance matrix; The initial puncture point position of the initial dual-frequency pseudorange observation data is calculated based on the initial dual-frequency pseudorange observation data, and an observation matrix is ​​constructed based on the initial puncture point position and the preset grid points. Calculate the Kalman gain based on the prior state error covariance matrix and the observation matrix; The initial ionospheric slant delay, pre-acquired ground station data, pre-acquired occultation data, and pre-acquired geomagnetic index are fused to obtain a multi-source observation vector.

7. The method for correcting ionospheric slack delay according to claim 1, characterized in that, The step of obtaining the corrected ionosphere and satellite elevation angle projection based on the standard ionospheric state vector and a preset grid, and calculating the ionospheric slant delay based on the satellite elevation angle projection and the corrected ionosphere, includes: The standard ionospheric state vector is filled into a preset grid to obtain VTEC grid data; The location of the puncture point is determined based on VTEC grid data and pre-acquired puncture points; Based on the location of the puncture point, the spherical distance between the puncture point and the electron density of the grid ionosphere are obtained; Based on the spherical distance of the puncture point and the electron density of the grid ionosphere, a corrected ionosphere is obtained; Obtain the satellite elevation angle of the puncture point and calculate the satellite elevation angle projection; Ionospheric slant delay is calculated based on satellite elevation angle projection and corrected ionospheric parameters.

8. A correction system for ionospheric slack delay, characterized in that, It includes a dual-frequency observation data acquisition module, a quality classification judgment module, a parameter calculation module, an ionospheric state correction module, and an ionospheric oblique delay solution module, specifically: The dual-frequency observation data acquisition module is used to acquire initial dual-frequency pseudorange observation data, initial first carrier phase observation value, initial second carrier phase observation value, signal-to-noise ratio corresponding to the initial dual-frequency pseudorange observation data, and satellite elevation angle corresponding to the initial dual-frequency pseudorange observation data; The quality grading determination module is used to determine the dual-frequency pseudorange observation data rating level corresponding to the initial dual-frequency pseudorange observation data based on the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, the initial second carrier phase observation value, the signal-to-noise ratio, and the satellite elevation angle. The parameter calculation module is used to obtain the historical ionospheric state and historical state estimation error covariance matrix if the rating level of the dual-frequency pseudorange observation data is the first dual-frequency pseudorange observation rating level, and to obtain the real-time ionospheric state vector, Kalman gain and multi-source observation vector based on the historical ionospheric state, historical state estimation error covariance matrix, initial dual-frequency pseudorange observation data, initial first carrier phase observation value and initial second carrier phase observation value; The ionospheric state correction module is used to correct the real-time ionospheric state vector based on Kalman gain and multi-source observation vector to obtain a standard ionospheric state vector. The ionospheric slant delay calculation module is used to obtain the corrected ionosphere and satellite elevation angle projection based on the standard ionospheric state vector and a preset grid, and to calculate the ionospheric slant delay based on the satellite elevation angle projection and the corrected ionosphere.

9. The ionospheric slack delay correction system according to claim 8, characterized in that, The parameter calculation module is also used for: If the rating level of the dual-frequency pseudorange observation data is the second dual-frequency pseudorange observation rating level, then the satellite motion velocity is obtained, and the smoothing window length is calculated based on the satellite motion velocity and the preset environmental interference coefficient. Based on the smoothing window length, the initial dual-frequency pseudorange observation data is smoothed to obtain smoothed dual-frequency pseudorange observation data. Based on the smoothed dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value, the ionospheric slant delay is calculated.

10. The ionospheric slack delay correction system according to claim 8, characterized in that, The parameter calculation module is also used for: If the dual-frequency pseudorange observation data score is the third dual-frequency pseudorange observation score, then the initial dual-frequency pseudorange observation data, the initial first carrier phase observation value, and the initial second carrier phase observation value are subjected to geometric distance error elimination processing to obtain standard dual-frequency pseudorange observation data, standard first carrier phase observation value, and standard second carrier phase observation value. The ionospheric slant delay is calculated based on standard dual-frequency pseudorange observation data, standard first carrier phase observations, and standard second carrier phase observations.

Images