A FTK solution method for high-precision CORS ionospheric correction
By real-time solution and model construction of CORS ionospheric delay in the cloud, combined with edge computing, the accuracy and security problems of ionospheric delay information acquisition in the existing technology are solved, and high-precision GNSS positioning is achieved.
Patent Information
- Application Number
- CN202210424768.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-04-21
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2042-04-21
AI Technical Summary
In the existing civil positioning solution methods, the acquisition of ionospheric delay information has problems such as long fixed ambiguity, high risk of position exposure and low positioning accuracy.
Using the FTK solution method of high-precision CORS ionospheric correction, by obtaining the dual-frequency GNSS observation data of the CORS site, real-time solution and ionospheric delay model are carried out in the cloud, generating the grid ionospheric delay correction number, and using edge calculation The algorithm realizes real-time accurate coordinate calculation of monitoring sites.
It reduces the fixed time of ambiguity throughout the week, reduces the risk of position exposure, improves positioning accuracy to centimeters or even millimeters, reduces hardware costs, and improves calculation speed.
Smart Images

Figure CN114859390B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of navigation and positioning, and in particular relates to a FTK solution method for high-precision CORS ionospheric correction. Background Art
[0002] Atmospheric molecules are ionized under the radiation of solar ultraviolet rays, X-rays, etc., generating free electrons and positive and negative ions, forming the ionosphere region, which will produce reflection, refraction, scattering and absorption effects on electromagnetic wave signals passing through it, and have an important impact on remote communications, satellite navigation, aerospace, earthquake monitoring, weather forecasting, etc. The ionosphere effect is one of the most serious error sources of satellite navigation systems, so high-precision ionospheric delay information plays an important role in improving the delay correction accuracy of single-frequency users and the convergence speed of precise positioning of dual-frequency / multi-frequency users.
[0003] In the existing civil positioning solution method, the method for obtaining ionospheric delay information is to reversely solve and obtain it through the different effects of the ionosphere on two electromagnetic waves of different frequencies. This solution method has the following problems:
[0004] 1) The integer ambiguity fixation time is relatively long. One reason is that the solution needs to be performed locally. Due to portability and other requirements, the local computing power of the local hardware is relatively limited, which makes the integer ambiguity fixation time relatively long.
[0005] 2) The risk of location exposure is high. The existing solution method requires actively broadcasting the device location to the platform, which has a high risk of location exposure.
[0006] 3) The positioning accuracy is average. The positioning accuracy of this solution method can only reach 10m level. Summary of the invention
[0007] The technical problem to be solved by the present invention is: in view of the problems existing in the prior art, the present invention provides a FTK solution method for high-precision CORS ionospheric correction.
[0008] The technical solution of the present invention is: a FTK solution method for high-precision CORS ionospheric correction, the method comprising the following steps:
[0009] S01. Obtain real-time GNSS observation data from the dual-frequency GNSS receiver at the CORS site;
[0010] S02, real-time resolution of the integer ambiguity of the CORS site in the cloud;
[0011] S03. Calculate the ionospheric delay of each CORS station in the cloud;
[0012] S04. Build a regional real-time ionospheric delay model in the cloud;
[0013] S05. The cloud uses the regional real-time ionospheric delay model to generate grid ionospheric delay corrections;
[0014] S06. The monitoring station calculates the approximate coordinates of the station according to the real-time GNSS observation data of the monitoring station, and then calculates the real-time ionospheric delay correction at the station through bilinear interpolation according to the received real-time ionospheric delay correction. Other error corrections (tropospheric delay, orbital clock error, etc.) also use similar algorithms;
[0015] S07. The monitoring station uses edge computing algorithms to solve the real-time and precise coordinates of the monitoring station.
[0016] Specifically, the method for real-time resolution of the integer ambiguity of the CORS site in step S02 is:
[0017] In the first step, the Melbourne-Wübbena combination is used to resolve the wide-lane ambiguity, and multiple epochs are used for smoothing, and then the accurate wide-lane double-difference integer ambiguity is directly rounded.
[0018] In the second step, the carrier phase and pseudorange observations are combined in an ionospheric-free manner, and the real solution of the ionospheric-free delay combination ambiguity is estimated using Kalman filtering.
[0019] In the third step, the fixed solution of wide-lane ambiguity and the floating-point solution and covariance of ionospheric-free delay combined ambiguity are used to restore the floating-point solution and covariance of L1 and L2 carrier ambiguities. Then, the LAMBDA algorithm is used to fix the ambiguity, and finally the double-difference integer ambiguities of L1 and L2 carriers are calculated.
[0020] Furthermore, in the first step, the Melbourne-Wübbena combination is used to achieve the resolution of wide lane ambiguity, and the calculation formula is as follows;
[0021]
[0022]
[0023] Where, L M-W is the Melbourne-Wübbena combined observation value, N WL is the wide lane integer ambiguity, f1 and f2 are the frequencies of L1 and L2 carriers, c is the speed of light in a vacuum environment, P1 and P2 are pseudorange observations, and L1 and L2 are carrier phase observations;
[0024] It uses multiple epochs for smoothing and then directly rounds off to get the accurate wide-lane double-difference integer ambiguity.
[0025] Furthermore, the specific implementation method of restoring the floating point solution and covariance of the L1 carrier ambiguity in the third step is as follows;
[0026] The ionospheric-free delay combined ambiguity is a linear combination of L1 and L2 ambiguities, and its calculation formula is as follows:
[0027]
[0028] Where f1 and f2 are the frequencies of L1 and L2 carriers, N1 is the integer ambiguity of L1 carrier, N2 is the integer ambiguity of L2 carrier, and N IF is the ionospheric delay-free combined ambiguity;
[0029] For GPS and BDS observation signals, the combined observation equation without ionospheric delay is as follows:
[0030]
[0031]
[0032] For the GLONASS navigation signal, the ionospheric delay-free combined observation equation is as follows:
[0033]
[0034]
[0035] Where IF represents the ionosphere, represents the double-difference pseudo-range observation value between satellite i, j and receiver r, m, represents the double difference carrier phase observation value between satellite i, j and receiver r, m. The superscript i, j represents the satellite number, and the subscript r, m represents the receiver number. represents the double difference observation value of the geometric distance between the satellite and the receiver, is the coefficient of the tropospheric mapping function in the propagation path, T r and T m are the zenith tropospheric delays of receiver r and receiver m, respectively, and λ i and λ j are the signal wavelengths of satellite i and satellite j respectively, and are the hardware delay terms in pseudorange and carrier observations, respectively, and is the observation noise of the corresponding pseudorange and carrier observations; is the integer ambiguity of the double-difference ionospheric combination, is the single-interval single-difference ionospheric delay-free combined ambiguity of reference satellite j, and They are and The coefficient of
[0036] The single-difference L1 and L2 ambiguities between stations can be obtained by the following formula:
[0037]
[0038]
[0039] In the formula, and is the inter-station single-difference ambiguity of the L1 and L2 carriers, and is the inter-station single-difference carrier observation value of L1 and L2 carriers, and is the inter-station single-difference pseudorange observation value of L1 and L2 carriers, is the inter-station single-difference ionospheric delay of L1, and is the signal wavelength of satellite i, and the subscript is the receiver number;
[0040] Using the fixed wide lane ambiguity and the float solution and covariance of the ionospheric delay-free combined ambiguity, the double difference ambiguity float solution and its covariance of the L1 carrier can be recovered:
[0041]
[0042] Among them, N WL is the wide lane integer ambiguity, and P is the pseudorange.
[0043] Furthermore, after the ambiguity is fixed using the LAMBDA algorithm in the third step, the ambiguity quality is controlled based on hypothesis testing, and the minimum Ω is tested given the significance level. min (N) and the next smallest Ω sec (N') is significantly different. If it is significantly the same, it is necessary to continue filtering and search and fix the ambiguity in the next epoch until the minimum Ω is found. min (N) and the next smallest Ω sec (N') is significantly different; if it is significantly different, the ambiguity N is judged to be fixed correctly;
[0044]
[0045] here, is the estimated value of the integer ambiguity, is the weight matrix, Ω min (N) is the variance of the ambiguity N, Ω sec (N') is the variance of the ambiguity N';
[0046] Using the smallest Ω min(N) and the next smallest Ω sec (N') The ratio of the two is used to determine whether the ambiguity is fixed correctly. The specific implementation method is as follows;
[0047]
[0048] Determine whether the statistic ratio falls into the rejection region. If it falls into the rejection region, it is considered that Ω sec (N') is significantly greater than Ω at the confidence level α min (N), otherwise it is considered that Ω sec (N') is not significantly greater than Ω at the confidence level α min (N);
[0049] Among them, the rejection region is W={ratio>F α (n,n)},F α (n,n) is the quantile of the upper confidence level of the F distribution, and n is the number of epochs involved in the ambiguity filtering. If the statistic ratio falls into the rejection region, it is considered that Ω sec (N') is significantly greater than Ω at the confidence level α min (N), so it is correct to judge that the set of ambiguities N is fixed;
[0050] When the L1 and wide lane double difference ambiguities are successfully fixed, the L2 double difference ambiguity can also be calculated according to the linear relationship between L1, L2 and wide lane ambiguities:
[0051]
[0052] is the double difference ambiguity of the L1 carrier, is the wide lane double difference ambiguity, is the double difference ambiguity of the L2 carrier.
[0053] Furthermore, the method for constructing the regional real-time ionospheric delay model in step S04 is:
[0054] The precise coordinate data of each CORS station and the fixed integer ambiguity are used to calculate the atmospheric delay at each CORS station, and the ionospheric delay is separated therefrom. According to the coordinate data and ionospheric delay of each CORS station, the ionospheric delay in the CORS network service area to which the CORS station belongs is firstly subjected to a spatial structural analysis, and then the spatial variability analysis is performed on the remaining residuals, and then the Kriging algorithm is used to establish a real-time ionosphere model suitable for the CORS network service area.
[0055] Furthermore, the method for generating the grid ionospheric delay correction number in step S05 is:
[0056] In the cloud, the CORS network service area to which the CORS station belongs is divided into grids, and regional plane gridding is performed; the real-time ionospheric delay correction number of each grid point is calculated using the regional ionosphere real-time model; the real-time ionospheric delay correction number of each grid point is broadcast in real time using the CORS network; the GNSS receiver at the monitoring station calculates the real-time ionospheric correction number of the monitoring station in real time according to the approximate coordinates of the monitoring station and the real-time ionospheric delay correction number of the received grid point using the bilinear interpolation method, and the real-time ionospheric correction number of the monitoring station is the user GNSS correction number.
[0057] Compared with the prior art, the present invention has the following beneficial effects:
[0058] (1) During positioning, the present invention calculates the ionospheric delay and establishes the regional ionospheric real-time model in the cloud, which has sufficient computing power, thereby reducing the integer ambiguity fixation time;
[0059] (2) The present invention adopts passive positioning during positioning, and the positioning process does not need to broadcast the location of the device to the outside, thereby reducing the risk of location exposure;
[0060] (3) When positioning, the present invention finely models the regional ionosphere real-time model and directly broadcasts the ionospheric delay correction number of each grid point, and then interpolates the user GNSS correction number using the approximate coordinates of the monitoring station and the ionospheric delay correction number at the grid point, thereby improving the positioning accuracy to the centimeter level, or even the millimeter level;
[0061] (4) Monitoring stations only need to use single-frequency receivers to achieve high-precision positioning, reducing hardware costs;
[0062] (5) The monitoring station does not need to independently calculate the ionospheric delay. Instead, it can directly use the ionospheric delay calculation results of the grid points in the cloud to directly calculate the ionospheric delay correction number of the monitoring station through interpolation, thereby improving the calculation speed and reducing the fixed time. BRIEF DESCRIPTION OF THE DRAWINGS
[0063] In order to more clearly illustrate the technical solutions in the embodiments or exemplary embodiments of the present application, the drawings required for use in the embodiments or exemplary descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application and should not be regarded as limiting the scope. For ordinary technicians in this field, other drawings can be obtained according to the drawings without paying creative work.
[0064] Figure 1 It is a schematic diagram of the method steps of the present invention. DETAILED DESCRIPTION
[0065] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Generally, the components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.
[0066] Implementation Example 1:
[0067] In order to simultaneously achieve the reduction of integer ambiguity fixation time, reduced exposure risk, improved positioning accuracy and reduced monitoring station hardware costs, reference Figure 1 In this embodiment, a high-precision CORS (Continuously Operating Reference Stations) ionospheric correction FTK (Flachen Korrektur Parameter, regional transformation number technology) solution method is provided, and the method includes the following steps:
[0068] S01. Acquire real-time GNSS observation data of dual-frequency GNSS (Global Navigation Satellite System) receivers at CORS sites;
[0069] S02, real-time resolution of the integer ambiguity of the CORS site in the cloud;
[0070] S03. Calculate the ionospheric delay of each CORS station in the cloud;
[0071] S04. Build a regional real-time ionospheric delay model in the cloud;
[0072] S05, the monitoring station generates grid ionospheric delay correction numbers using the regional real-time ionospheric delay model;
[0073] S06. The monitoring station calculates the approximate coordinates of the monitoring station according to the real-time GNSS observation data of the monitoring station, and then calculates the real-time ionospheric delay correction number at the station through bilinear interpolation according to the received real-time ionospheric delay correction number. Similar algorithms are also used for other error correction numbers.
[0074] S07. The monitoring station uses edge computing algorithms to solve the real-time and precise coordinates of the monitoring station.
[0075] In order to solve the integer ambiguity of the CORS site in real time, in this embodiment, specifically, the method for solving the integer ambiguity of the CORS site in real time in step S02 is:
[0076] In the first step, the Melbourne-Wübbena combination is used to resolve the wide-lane ambiguity, and multiple epochs are used for smoothing, and then the accurate wide-lane double-difference integer ambiguity is directly rounded.
[0077] In the second step, the carrier phase and pseudorange observations are combined in an ionospheric-free manner, and the real solution of the ionospheric-free delay combination ambiguity is estimated using Kalman filtering.
[0078] In the third step, the fixed solution of wide-lane ambiguity and the floating-point solution and covariance of ionospheric-free delay combined ambiguity are used to restore the floating-point solution and covariance of L1 and L2 carrier ambiguities. Then, the LAMBDA algorithm is used to fix the ambiguity, and finally the double-difference integer ambiguities of L1 and L2 carriers are calculated.
[0079] In the first step, the wide lane integer ambiguity is resolved by the Melbourne-Wübbena combination, and the calculation formula is:
[0080]
[0081]
[0082] Where, L M-W is the Melbourne-Wübbena combined observation value, N WL is the wide lane integer ambiguity, f1 and f2 are the frequencies of L1 and L2 carriers, c is the speed of light in a vacuum environment, P1 and P2 are pseudorange observations, and L1 and L2 are carrier phase observations.
[0083] The Melbourne-Wübbena combination in the formula can be regarded as a wide-lane observation value and Weighted average of components composition.
[0084]
[0085]
[0086] In the formula, is the double difference observation;
[0087] The ionospheric delays in the above formula are:
[0088]
[0089]
[0090] In the formula, is the ionospheric delay of the wide-lane phase observation, is the ionospheric delay of the L1 phase observation, is the ionospheric delay of the weighted average pseudorange observation.
[0091] Therefore, the use of the Melbourne-Wübbena combination to solve wide-lane ambiguities can effectively eliminate ionospheric delays. At the same time, since satellite coordinates and satellite clock errors are not used when solving wide-lane ambiguities using this combination, the use of the Melbourne-Wübbena combination to solve wide-lane ambiguities can also eliminate satellite clock errors and satellite orbit errors. Finally, this combination is not affected by the station coordinates. The effects of tropospheric delay and receiver clock errors on carrier phase observations and pseudorange observations are consistent, and are also eliminated in the Melbourne-Wübbena combination. However, since the Melbourne-Wübbena combination consists of pseudorange and carrier phase observations, its observation noise is relatively large, so it is necessary to use multiple epoch results for smoothing to obtain reliable wide-lane ambiguities. Finally, the direct rounding method can be used to obtain accurate wide-lane double-difference integer ambiguities.
[0092] In the third step, since the distance between the reference stations in the CORS network is generally tens of kilometers, the ionospheric delay in the baseline can no longer be eliminated by inter-station differentials. Moreover, the magnitude of the double-difference ionospheric delay affects the correct fixation of the ambiguity.
[0093] Here, the method of ionospheric-free delay combination is used to eliminate the ionospheric delay between reference stations, and its observation equation is as follows.
[0094]
[0095]
[0096] The superscript SYS represents different satellite navigation systems (GPS / GLONASS / BDS), IF represents the ionosphere, represents the double-difference pseudo-range observation value between satellite i, j and receiver r, m, represents the double difference carrier phase observation value between satellite i, j and receiver r, m. The superscript i, j represents the satellite number, and the subscript r, m represents the receiver number. represents the double difference observation value of the geometric distance between the satellite and the receiver, is the coefficient of the tropospheric mapping function in the propagation path, T r and T m are the zenith tropospheric delays of receiver r and receiver m, respectively, and λ i and λ j are the signal wavelengths of satellite i and satellite j respectively, represents double difference, N is the integer ambiguity, and are the hardware delay terms in pseudorange and carrier observations, respectively, and is the observation noise of the corresponding pseudorange and carrier observations.
[0097] The ionospheric-free delay combined ambiguity is a linear combination of L1 and L2 ambiguities, and its calculation formula is as follows:
[0098]
[0099] N1 is the integer ambiguity of L1 carrier, N2 is the integer ambiguity of L2 carrier, N IF is the ionospheric delay-free combined ambiguity. So for GPS and BDS observation signals, the ionospheric delay-free combined observation equation is as follows:
[0100]
[0101]
[0102] For the GLONASS navigation signal, the ionospheric delay-free combined observation equation is as follows:
[0103]
[0104]
[0105] In the formula, is the single-interval single-difference ionospheric delay-free combined ambiguity of reference satellite j, is the integer ambiguity of the double difference ionospheric combination. The value range is between 5.252-5.276 cm, so The coefficient of The maximum value is 0.024 cm.
[0106] The single-difference L1 and L2 ambiguities between stations can be obtained by the following formula:
[0107]
[0108]
[0109] In the formula, and is the inter-station single-difference ambiguity of the L1 and L2 carriers, and is the inter-station single-difference carrier observation value of L1 and L2 carriers, and is the inter-station single-difference pseudorange observation value of L1 and L2 carriers, is the inter-station single-difference ionospheric delay of L1, and is the signal wavelength of satellite i, and the subscript is the receiver number.
[0110] Since the ionospheric-free delay combination ambiguity does not have integer characteristics, the existing ambiguity search method cannot achieve the fixation of the ionospheric-free delay combination ambiguity. The floating point solution of the ionospheric-free delay combination ambiguity can be continuously solved by Kalman filtering. Since the Melbourne-Wübbena combination can quickly obtain the fixed solution of the wide lane ambiguity, we can use the fixed wide lane ambiguity and the floating point solution and covariance of the ionospheric-free delay combination ambiguity to recover the double difference ambiguity floating point solution and its covariance of the L1 carrier:
[0111]
[0112] Where P is the pseudorange. Since the double-difference ambiguity of L1 has integer characteristics, the classic LAMBDA method can be used to down-correlate the ambiguity, thereby greatly reducing the amount of calculation. The general process of the LAMBDA algorithm is as follows: a) Use integer Gauss transform on the ambiguity variance matrix to construct an integer reversible matrix Z, use the Z matrix to perform Z transform on the ambiguity, and perform down-correlation Z transform on the covariance matrix corresponding to the ambiguity; b) Search for the Z-transformed ambiguity based on the down-correlation ambiguity covariance matrix; c) Use Z -1 Restore the corresponding blur.
[0113] In the third step, the confidence of the ambiguity searched by the LAMBDA algorithm alone is not necessarily very high, that is, the fixed ambiguity cannot guarantee correctness, and further ambiguity quality control is required. Based on the hypothesis test to control the ambiguity quality, given the significance level, the smallest Ω is tested. min (N) and the next smallest Ω sec (N') is significantly different. If it is significantly the same, it is necessary to continue filtering and search and fix the ambiguity in the next epoch until the minimum Ω is found. min (N) and the next smallest Ω sec (N') is significantly different; if it is significantly different, the ambiguity N is judged to be fixed correctly.
[0114]
[0115] is the estimated value of the integer ambiguity, is the weight matrix. Generally speaking, we use the smallest Ω min (N) and the next smallest Ω sec (N') The ratio of the two is used to determine whether the ambiguity is fixed correctly.
[0116]
[0117] This method assumes that the two sets of ambiguity estimates are independent of each other and follow a normal distribution. N and N' are the expected values of the two sets of ambiguity. is the estimated value of the ambiguity, Ω min (N) and Ω sec (N') is its variance value. The rejection region is W = {ratio>F α (n,n)},F α (n,n) is the quantile of the F distribution with an upper confidence level of α, and n is the number of epochs involved in ambiguity filtering. If the statistic ratio falls into the rejection region, it is considered that Ω sec (N') is significantly greater than Ω at the confidence level α min (N), so it is correct to judge that the group of ambiguities N is fixed.
[0118] When the L1 and wide lane double difference ambiguities are successfully fixed, the L2 double difference ambiguity can also be easily calculated according to the linear relationship between L1, L2 and wide lane ambiguities:
[0119]
[0120] is the double difference ambiguity of the L1 carrier, is the wide lane double difference ambiguity, is the double difference ambiguity of the L2 carrier.
[0121] Specifically, the method for constructing the regional real-time ionospheric delay model in step S04 is: using the precise coordinate data of each CORS station and the fixed integer ambiguity, the atmospheric delay at each CORS station is calculated, and the ionospheric delay amount is separated therefrom; according to the coordinate data and ionospheric delay amount of each CORS station, the ionospheric delay amount in the CORS network service area to which the CORS station belongs is subjected to spatial structural analysis and variability analysis, and the ionospheric real-time model suitable for the CORS network service area is established by using the Kriging algorithm.
[0122] Specifically, the method for generating user GNSS correction numbers in step S05 is: dividing the CORS network service area to which the CORS site belongs into grids in the cloud, and performing regional plane grid division; using the regional ionosphere real-time model to calculate the real-time ionospheric delay correction number of each grid point; using the CORS network to broadcast the real-time ionospheric delay correction number of each grid point in real time; the GNSS receiver at the monitoring site calculates the real-time ionospheric correction number of the monitoring site in real time according to the approximate coordinates of the monitoring site and the received real-time ionospheric delay correction number of the grid point by a bilinear interpolation method, and the real-time ionospheric correction number of the monitoring site is the user GNSS correction number.
[0123] The above-mentioned embodiments only express the preferred implementation modes of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present invention. It should be pointed out that, for those skilled in the art, several modifications, improvements and substitutions can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention.
Claims
1. A high-precision CORS ionospheric correction FTK solution method, characterized in that: The method comprises the following steps: S01. Obtain real-time GNSS observation data from the dual-frequency GNSS receiver at the CORS site; S02, real-time resolution of the integer ambiguity of the CORS site in the cloud; The specific implementation of step S02 is as follows; In the first step, the Melbourne-Wübbena combination is used to resolve the wide-lane ambiguity, and multiple epochs are used for smoothing, and then the accurate wide-lane double-difference integer ambiguity is directly rounded. In the first step, the Melbourne-Wübbena combination is used to resolve the wide lane ambiguity. The calculation formula is as follows: Where, L M-W is the Melbourne-Wübbena combined observation value, N WL is the wide lane integer ambiguity, f1 and f2 are the frequencies of L1 and L2 carriers, c is the speed of light in a vacuum environment, P1 and P2 are pseudorange observations, and L1 and L2 are carrier phase observations; And use multiple epochs for smoothing, and then directly round off to get accurate wide-lane double-difference integer ambiguity; In the second step, the carrier phase and pseudorange observations are combined in an ionospheric-free manner, and the real solution of the ionospheric-free delay combination ambiguity is estimated using Kalman filtering. The third step is to use the fixed solution of the wide-lane double-difference integer ambiguity and the floating-point solution and covariance of the ionospheric delay-free combined ambiguity to restore the floating-point solution and covariance of the L1 carrier ambiguity, and then use the LAMBDA algorithm to fix the ambiguity, and finally calculate the double-difference integer ambiguity of the L1 and L2 carriers; The specific implementation method of restoring the floating point solution and covariance of the L1 carrier ambiguity in the third step is as follows; The ionospheric-free delay combined ambiguity is a linear combination of L1 and L2 ambiguities, and its calculation formula is as follows: Where f1 and f2 are the frequencies of L1 and L2 carriers, N1 is the integer ambiguity of L1 carrier, N2 is the integer ambiguity of L2 carrier, and N IF is the ionospheric delay-free combined ambiguity; For GPS and BDS observation signals, the combined observation equation without ionospheric delay is as follows: For the GLONASS navigation signal, the ionospheric delay-free combined observation equation is as follows: Where IF represents the ionosphere, represents the double-difference pseudo-range observation value between satellite i, j and receiver r, m, represents the double difference carrier phase observation value between satellite i, j and receiver r, m. The superscript i, j represents the satellite number, and the subscript r, m represents the receiver number. represents the double difference observation value of the geometric distance between the satellite and the receiver, is the coefficient of the tropospheric mapping function in the propagation path, T r and T m are the zenith tropospheric delays of receiver r and receiver m, respectively, and λ i and λ j are the signal wavelengths of satellite i and satellite j respectively, and are the hardware delay terms in pseudorange and carrier observations, respectively, and is the observation noise of the corresponding pseudorange and carrier observations; is the integer ambiguity of the double-difference ionospheric combination, is the single-interval single-difference ionospheric delay-free combined ambiguity of reference satellite j, and They are and The coefficient of The single-difference L1 and L2 ambiguities between stations are obtained by the following equations: In the formula, and is the inter-station single-difference ambiguity of the L1 and L2 carriers, and is the inter-station single-difference carrier observation value of L1 and L2 carriers, and is the inter-station single-difference pseudorange observation value of L1 and L2 carriers, is the inter-station single-difference ionospheric delay of L1, and is the signal wavelength of satellite i, and the subscript is the receiver number; Using the fixed wide lane ambiguity and the floating point solution and covariance of the ionospheric delay-free combined ambiguity, the double difference ambiguity floating point solution and its covariance of the L1 carrier are recovered: Among them, N WL is the wide lane integer ambiguity, P is the pseudorange; S03. Calculate the ionospheric delay of the CORS site in the cloud; S04. Build a regional real-time ionospheric delay model in the cloud; S05. The cloud uses the regional real-time ionospheric delay model to generate grid ionospheric delay correction numbers; S06. The monitoring station calculates the approximate coordinates of the station according to the real-time GNSS observation data of the station, and calculates the real-time ionospheric delay correction number at the station by bilinear interpolation for the received real-time ionospheric delay correction number. The same algorithm is also used for other error correction numbers to calculate the error correction number; S07. The monitoring station uses edge computing algorithms to solve the real-time and precise coordinates of the monitoring station.
2. The FTK solution method for high-precision CORS ionospheric correction according to claim 1, characterized in that: In the third step, after the ambiguity is fixed using the LAMBDA algorithm, the ambiguity quality is controlled based on hypothesis testing. Given a significance level, the smallest Ω is tested. min (N) and the next smallest Ω sec (N') is significantly different. If it is significantly the same, it is necessary to continue filtering and search and fix the ambiguity in the next epoch until the minimum Ω is found. min (N) and the next smallest Ω sec (N') is significantly different; If they are significantly different, the ambiguity N is judged to be fixed correctly; here, is the estimated value of the integer ambiguity, is the weight matrix, Ω min (N) is the variance of the ambiguity N, Ω sec (N') is the variance of the ambiguity N'; Using the smallest Ω min (N) and the next smallest Ω sec (N') The ratio of the two is used to determine whether the ambiguity is fixed correctly. The specific implementation method is as follows; Determine whether the statistic ratio falls into the rejection region. If it falls into the rejection region, it is considered that Ω sec (N') is significantly greater than Ω at the confidence level α min (N), otherwise it is considered that Ω sec (N') is not significantly greater than Ω at the confidence level α min (N); Among them, the rejection region is W={ratio>F α (n,n)},F α (n,n) is the quantile of the upper confidence level of the F distribution, and n is the number of epochs involved in the ambiguity filtering. If the statistic ratio falls into the rejection region, it is considered that Ω sec (N') is significantly greater than Ω at the confidence level α min (N), so it is correct to judge that the group of ambiguities N is fixed; When both L1 and widelane double-difference ambiguities are successfully fixed, the L2 double-difference ambiguity is calculated according to the linear relationship between L1, L2 and widelane ambiguities: is the double difference ambiguity of the L1 carrier, is the wide lane double difference ambiguity, is the double difference ambiguity of the L2 carrier.
3. The FTK solution method for high-precision CORS ionospheric correction according to claim 1, characterized in that: The method for constructing the regional real-time ionospheric delay model in step S04 is: Using the precise coordinate data of each CORS station and the fixed integer ambiguity, the atmospheric delay at each CORS station is calculated, and the ionospheric delay is separated from it; According to the coordinate data and ionospheric delay of each CORS station, the ionospheric delay in the CORS network service area to which the CORS station belongs is firstly subjected to spatial structural analysis, and then the spatial variability analysis is performed on the remaining residuals. Finally, the Kriging algorithm is used to establish a real-time ionospheric model suitable for the CORS network service area.
4. The FTK solution method for high-precision CORS ionospheric correction according to claim 1, characterized in that: The method for generating the grid ionospheric delay correction number in step S05 is: In the cloud, the CORS network service area to which the CORS station belongs is divided into grids, and regional plane gridding is performed; the real-time ionospheric delay correction number of each grid point is calculated using the regional ionosphere real-time model; the real-time ionospheric delay correction number of each grid point is broadcast in real time using the CORS network; the GNSS receiver at the monitoring station calculates the real-time ionospheric correction number of the monitoring station in real time according to the approximate coordinates of the monitoring station and the real-time ionospheric delay correction number of the received grid point using the bilinear interpolation method, and the real-time ionospheric correction number of the monitoring station is the user GNSS correction number.
5. The FTK solution method for high-precision CORS ionospheric correction according to claim 1, characterized in that: Other error corrections include tropospheric delay and orbital clock errors.
