Method, apparatus, and system for fixing wide-lane ambiguity in an entire network based on ionospheric relationships.

Abstract

The wide-lane ambiguity, based on ionospheric relationships, is fixed across the entire network. [Solution] A method, apparatus, and system for fixing wide lane ambiguity across an entire network, comprising: Step 1 dividing all base stations in a CORS network into a plurality of independent ambiguity calculation units; Step 2 collecting four or more groups of fixed wide lane ambiguity within the units; Step 3 constructing a regional ionospheric delay model; Step 4 determining whether there are any misfixed wide lane ambiguities based on the model, and terminating the calculation flow if there are fewer than four groups of remaining wide lane ambiguity data; Step 5 calculating candidate values ​​for wide lane ambiguity and corresponding wide lane ionospheric delays; and Step 6 confirming whether the candidate values ​​calculated in Step 5 satisfy the determination conditions.

Description

[Technical Field]

[0001] The present invention relates to the field of remote sensing disaster identification technology, and more specifically to a method, apparatus, and system for fixing wide-lane ambiguity across an entire network based on ionospheric relationships. [Background technology]

[0002] Fixing ambiguity between GNSS reference stations is fundamental to network RTK positioning. Because the distances between reference stations in a band of CORS are relatively long, the effects of ionospheric delay and tropospheric delay cannot be ignored. Generally, ambiguity estimation and fixing are performed by ionospheric-free combinations or by adding ionospheric constraints. In 1989, Blewitt and Geoffrey described in detail a method for solving the ambiguity of medium- and long-range static baselines, employing ionospheric-free combinations. The specific method is divided into the following three steps. In the first step, the ambiguity around the entire wide lane is estimated and fixed, employing Hatch-Melbourne-Wubbena combination observations. Since these combination observations do not have geometric distances, they can avoid the effects of various errors such as coordinate errors of observation stations, and are only affected by observation noise, multipath delay, and hardware delays between satellites and receivers. In the second step, the effects of ionospheric delay are eliminated using ionospheric free combinations, and the influence of tropospheric delay on the solution is reduced by estimating the wet delay of the zenith troposphere, thereby estimating the floating solution of the carrier ambiguity of the ionospheric free combination. In the third step, the floating solution and accuracy of the wide-lane ambiguity are calculated using the fixed wide-lane ambiguity and the floating solution and accuracy of the ionospheric free combination ambiguity, and the ambiguity is fixed. To accelerate the rapid fixing of ambiguity, a series of methods for rapid ambiguity fixing have been proposed since the 1990s. In 2016, Chu proposed a method for fixing ambiguity between reference stations based on three-frequency data. In this method, first, the ultra-wide lane ambiguity is fixed using pseudo-distance observations, then the wide lane ambiguity is fixed using the fixed ultra-wide lane observations, and finally the three-frequency ambiguity is solved and fixed.In 2017, Zhang proposed a method for solving the ambiguity between reference stations using an ionospheric-constrained model. This method utilizes the interrelationships of surrounding ionospheric delays to constrain the ionospheric delay parameters and solve the ambiguity between reference stations.

[0003] As described above, there is a wealth of research on resolving ambiguity between reference stations. Atmospheric interpolation models for small and medium-sized areas are essentially weighted averages of atmospheric information from each baseline or base station, and conventional models can fit well to regional atmospheric changes during calm periods. However, when atmospheric conditions become more active, atmospheric delay errors still significantly affect the user's resolution process for regional augmented positioning, and applying the above research findings to network RTK software did not yield good results. At the same time, conventional research has not fully utilized the special conditions of network RTK application scenarios. Therefore, it is necessary to fully utilize all favorable conditions, further improve the method of fixing ambiguity between reference stations, and enhance the positioning effectiveness of network RTK. [Overview of the Initiative] [Problems that the invention aims to solve]

[0004] The problem that the present invention aims to solve is to provide a method, apparatus, and system for fixing a wide lane ambiguity network based on ionospheric relationships, which can calculate the corresponding ionospheric delay using several fixed wide lane ambiguities within a CORS region, interpolate the ionospheric delay corresponding to unfixed wide lane ambiguities using the spatial relationships of the ionospheric delays, fix the remaining unfixed wide lane ambiguities using wavelengths with relatively large wide lane observations, and verify whether the fixed wide lane ambiguities have been correctly fixed. [Means for solving the problem]

[0005] To solve the above problems, the present invention employs the following technical solution.

[0006] A method for fixing wide-lane ambiguities across a network based on ionospheric relationships, comprising: Step 1: dividing all base stations in the CORS network into multiple independent ambiguity calculation units using overlapping ionospheric puncture points; Step 2: collecting four or more groups of fixed wide-lane ambiguities within each ambiguity calculation unit; Step 3: calculating the wide-lane ionospheric delay corresponding to each group of wide-lane ambiguities collected and constructing a regional ionospheric delay model; and, based on the regional ionospheric delay model, determining whether there are any misfixed wide-lane ambiguities. If misfixed, removing the misfixed wide-lane ambiguities, determining whether the remaining wide-lane ambiguity data consists of four or more groups. If there are four or more groups, repeating Step 3 to collect the remaining wide-lane ambiguities. Step 4 terminates the calculation flow if there are fewer than 4 data groups; Step 5, if no misfixation is found in the process of determining misfixation of wide lane ambiguity in Step 4, continues to determine whether there are any unfixed wide lane ambiguities in the current ambiguity calculation unit, and if there are any unfixed wide lane ambiguities, interpolates the wide lane ionospheric delay corresponding to the unfixed ambiguity using the regional ionospheric delay model in Step 3, and calculates candidate values ​​for wide lane ambiguity and corresponding wide lane ionospheric delays; Step 6 checks whether the candidate values ​​for wide lane ambiguity calculated in Step 5 satisfy the judgment conditions, and if YES, iteratively constructs the regional ionospheric delay model in addition to Step 3, and if NO, returns to Step 5 and performs the calculation to fix the wide lane ambiguity again.

[0007] The specific process for Step 1 described above is as follows:

[0008] The specific formula for determining the ionospheric puncture point is as follows: Formula (1)

number

number

number

[0009] The specific process for Step 2 described above is as follows:

[0010] First, an attempt is made to fix the ambiguity for each satellite-base station combination within the ambiguity calculation unit. If it is not possible to fix all the ambiguities, the number of ambiguities to be fixed is increased by fixing only some of them. This method of fixing only some of the ambiguities is used only to fix some of the baseline ambiguities, not all of them. Then, the residuals and variances of the observational model for the satellite-base station with fixed ambiguities are calculated, and the residuals and variances are assigned to parts of each variable, including the parameter to be estimated, according to the law of error propagation, and the residuals and variances of the ionospheric portion are extracted. Finally, the residuals of the ionospheric portion are selected and sorted, and the fixed wide-lane ambiguities with the smallest variance are selected and fixed in order and used for subsequent calculations.

[0011] The specific process for Step 3 described above is as follows:

[0012] First, process the baseline with wide-lane ambiguity fixed, and obtain the approximate ionospheric delay extracted based on the wide-lane ambiguity in the baseline, that is, the ionospheric delay of the wide lane ∇ΔI r and obtain its calculation formula as follows. Equation (2)

Number

number

number

[0014] The specific process for Step 4 described above is as follows:

[0015] After the regional ionospheric delay model was constructed, the residuals of the observational model were validated due to the existence of extraneous observations, and the validation indices were as follows:

number

[0016] If the residuals exceed the acceptable threshold, the maximum residual is removed and the model is rebuilt until residual validation passes or the number of available baselines is less than four.

[0017] In Step 5 above, after the residual check is passed, the interpolated wide-lane ionospheric values ​​for the remaining unfixed baseline are calculated using the following formula. These interpolated values ​​represent the wide-lane ionospheric results for the calculation target point, estimated using the existing wide-lane ionospheric results. ∇ΔI WL-IPν =a1·ΔX ν +a2·ΔY ν In the formula, ∇ΔI WL-IPν ΔX is the unfixed double differential ionospheric delay at baseline v. ν ΔY ν This represents the distance from each point in the reference station's central coordinate system of baseline ν to the east and north of the reference point, respectively. A step-by-step check and judgment is performed on the baseline with unfixed wide-lane ambiguity. The integer closest to the floating solution of wide-lane ambiguity is used as the center, and the system is extended by 2 or 3 cycles at each end to obtain several candidate integer solutions. These solutions are substituted into Step 3 to calculate the corresponding wide-lane ionospheric delay. If the wide-lane ionospheric delay at a given candidate point satisfies the judgment conditions in the following equation, the ambiguity is directly fixed. |∇ΔI WL-IPν -∇ΔI WLν |<0.3·μ In the formula, ∇ΔI WLν This is the double-differential ionospheric delay calculated based on candidate integer solutions.

[0018] In Step 6 above, ionospheric delay observations with wide lane ambiguity fixed for all measurement points in the current epoch are collected, model values ​​of the regional ionospheric delay model are calculated using the above observations, and the mean error and standard difference of the current model are calculated using the difference between the model calculation output and the observations. If the candidate value satisfies the criteria, first the ionospheric delay value for the region where the candidate value is located is calculated using the existing regional ionospheric delay model, and finally it is determined whether the difference between the candidate value and the regional ionospheric delay model value is less than a standard difference of 3. If YES, the regional ionospheric model is constructed iteratively; if NO, the process returns to Step 5 and the calculations to fix the wide lane ambiguity are performed again.

[0019] After Step 6 is completed, if the existing independent ambiguity calculation units experience fluctuations due to satellite motion, causing two independent ambiguity calculation units to merge or if a puncture point within a single ambiguity calculation unit splits because it no longer meets the constraints of Step 1, the regional ionospheric delay model is iteratively recalculated according to Steps 1 to 6 using the existing ionospheric delay model.

[0020] The device employs a method for fixing wide-lane ambiguities across the entire network based on the ionospheric relationships described above. The device utilizes overlapping ionospheric puncture points to divide all base stations in the CORS network into multiple independent ambiguity computing units. For each ambiguity computing unit, it provides a decision module for collecting four or more groups of fixed wide-lane ambiguities within the unit, and calculates the wide-lane ionospheric delay corresponding to each group of wide-lane ambiguities collected. It constructs a regional ionospheric delay model, determines whether or not there are misfixed wide-lane ambiguities based on the regional ionospheric delay model. If misfixation exists, it removes the fixed wide-lane ambiguities and determines whether or not there are four or more remaining data. The system determines whether there are four or more groups, and if so, repeats the above steps. If there are fewer than four groups, the calculation flow is terminated. If no misfixation is found in the process of determining misfixation of wide lane ambiguity, the system continues to determine whether there is unfixed ambiguity within the current ambiguity calculation unit. If there is, it uses a regional ionospheric delay model to interpolate the wide lane ionospheric delay corresponding to the unfixed ambiguity, and includes a fixing module for calculating candidate values ​​for wide lane ambiguity and corresponding wide lane ionospheric delays, and a modeling module that checks whether the calculated candidate values ​​satisfy the criteria, and if so, iteratively constructs the regional ionospheric model, and if so, re-executes the calculation to fix the wide lane ambiguity.

[0021] A fixing system for an entire network of wide lane ambiguity based on ionospheric relationships, comprising a processor, memory, and a fixing program for an entire network of wide lane ambiguity based on ionospheric relationships stored in the memory and executable by the processor, wherein when the above-described fixing program for an entire network of wide lane ambiguity based on ionospheric relationships is executed by the processor, the steps of the above-described method for fixing an entire network of wide lane ambiguity based on ionospheric relationships are realized. [Effects of the Invention]

[0022] The present invention provides a method, apparatus, and system for fixing wide-lane ambiguities across an entire network based on ionospheric correlation. However, the fixing speed of some wide-lane ambiguities within a region is relatively slow, and there is a possibility of misfixation of conventional wide-lane ambiguities, which adversely affects subsequent steps such as L1 ambiguity fixing and atmospheric error modeling. In response to this problem, the present invention proposes a method for fixing wide-lane ambiguities across an entire network based on ionospheric correlation. This method, based on the characteristic of strong spatial correlation of the ionosphere in overlapping puncture points, mediates the ionospheric delay extracted using wide-lane ambiguities, enabling accurate and rapid fixing of all ambiguities within a computational unit. In the process of fixing nearly one million wide-lane ambiguities, no misfixations occurred, virtually all fixings were achieved, and the unfixed rate was less than 0.1%. On average, approximately two groups of BDS observations were added to the regional augmentation correction values ​​to which this method was applied. In the three types of combined positioning modes of BDS, GPS, and GC, the average fixing rate improved by more than 10%, of which the fixing rate corresponding to positioning by a single BDS system improved by approximately 35%, and the vertical positioning accuracy improved by more than 30%. Therefore, this method can essentially avoid the adverse effects of unfixed or misfixed wide-lane ambiguities, and can lay a good foundation for subsequent GNSS regional augmentation positioning. [Brief explanation of the drawing]

[0023] The present invention will be further described below with reference to the drawings and embodiments.

[0024] [Figure 1] This is a schematic diagram of the distribution of some measurement points in a CORS network according to an embodiment of the present invention. [Figure 2] This is a comparative diagram of the ionospheric residuals of the GT14_GT18 baseline GPS PRN06 wide lane in an embodiment of the present invention. [Figure 3] This is a comparative diagram of the posterior accuracy of the ionosphere in the GPS PRN06 wide lane according to an embodiment of the present invention. [Figure 4] This is a comparative diagram of the horizontal distribution of BDS positioning for GT20 in an embodiment of the present invention, showing the case where the present invention is not adopted and the case where the present invention is adopted. [Figure 5] This is a comparison diagram of the vertical sequence of BDS positioning for GT20 in an embodiment of the present invention, comparing the case where the present invention is not adopted with the case where the present invention is adopted. [Figure 6] This is a comparative diagram of the horizontal distribution of GPS positioning in the GT20 embodiment of the present invention, showing the case where the present invention is not adopted and the case where the present invention is adopted. [Figure 7] This is a comparison diagram of the vertical GPS positioning sequence for GT20, an embodiment of the present invention, comparing the case where the present invention is not adopted with the case where the present invention is adopted. [Figure 8] This is a comparison diagram of the horizontal distribution of GC positioning for GT20 in an embodiment of the present invention, comparing the case where the present invention is not adopted with the case where the present invention is adopted. [Figure 9] This is a comparison diagram of the vertical sequence of GC positioning for GT20 in an embodiment of the present invention, comparing the case where the present invention is not adopted with the case where the present invention is adopted. [Modes for carrying out the invention]

[0025] The technical aspects of the present invention will be described in detail below with reference to the drawings and embodiments.

[0026] A method for fixing wide-lane ambiguity across an entire network based on ionospheric relationships, comprising the following steps:

[0027] Step 1: Use overlapping ionospheric puncture points to divide all base stations in the CORS network into multiple independent ambiguity computing units.

[0028] The theoretical core of this method lies in the spatial relevance of ionospheric delay. However, as distance increases, the spatial relevance of ionospheric delay gradually decreases due to the influence of ionospheric decoration effects and small to medium-sized ionospheric disturbances. Therefore, determining the boundary range is extremely important for this method.

[0029] To determine the data selection and method coverage, this specification introduces the concept of overlapping ionospheric pierce points, or IPPs. In the processing of regional ionospheric delay, it is generally assumed that the ionosphere is a thin layer of uniform density above the Earth, and that ionospheric pierce points are the intersections of GNSS signal rays with this thin layer. By analyzing the total electron content, or TEC data, of the ionosphere of globally distributed GNSS reference stations, it is determined that if two GNSS reference stations are sufficiently close to two ionospheric pierce points of the same satellite in the same epoch, the electron density gradient between the two ionospheric pierce points is very small or linearly related to the distance between the pierce points. This indicates a strong correlation between the ionospheric space between the pair of pierce points, enabling better interpolation accuracy. These two ionospheric pierce points are defined as overlapping ionospheric pierce points, and the specific determination formula for ionospheric pierce points is as follows.

number

number

number

[0030] Step 2: For each ambiguity calculation unit, collect four or more fixed wide-lane ambiguities within the unit.

[0031] This method, while designed to capture and fix wide-lane ambiguity, cannot independently fix wide-lane ambiguity itself. It assumes that some wide-lane ambiguity at certain measurement points within a unit has already been fixed. Therefore, to quickly capture and fix the remaining unfixed wide-lane ambiguity, this method requires the existence of four base stations or four baselines to fix the wide-lane ambiguity of the same satellite constellation. This is to prevent misfixation of the wide-lane ambiguity itself, which is used to validate atmospheric models with extra observational data and to avoid misfixation of the ambiguity around the entire wide-lane circumference used in the modeling.

[0032] Step 3: Calculate the wide-lane ionospheric delay corresponding to the wide-lane ambiguity of each collected group, and construct a regional ionospheric delay model.

[0033] First, the wide-lane ambiguity is processed at a fixed baseline, and the approximate ionospheric delay, i.e., the wide-lane ionospheric delay ∇ΔI, is extracted based on the wide-lane ambiguity at that baseline.WL The formula for obtaining this value is as follows:

number

[0034] Using the above formula, we extract the wide-lane ionospheric delay corresponding to a fixed baseline for wide-lane ambiguity, construct a regional ionospheric delay model, and the formula is as follows: V = BX - L In the equation, V represents the residual matrix between the wide-lane ionospheric delay model and the measured observed values, B is the model design matrix, X is the parameter matrix of the model to be estimated, L is the matrix of the measured observed values, ΔX1 and ΔY1 are the distances from each point in the reference station center coordinate system of Baseline 1 to the east and north of the reference point, ΔX2 and ΔY2 are the distances from each point in the reference station center coordinate system of Baseline 2 to the east and north of the reference point, and ΔX n ΔY n This is the distance from each point in the reference station center coordinate system of baseline n to the east and north directions of the reference point, satisfying the following equation:

number

number

[0035] Step 4: Based on the model, determine whether there is misfixation of wide lane ambiguity. If misfixation exists, remove the misfixed wide lane ambiguity. Determine whether the remaining data consists of four or more groups. If there are four or more groups, repeat Step 3. If there are fewer than four groups, terminate the calculation flow.

[0036] After the regional ionospheric delay model was constructed, the residuals of the observational model were validated due to the existence of extraneous observations, and the validation indices are as follows.

number

[0037] If the residuals exceed the acceptable threshold, the maximum residual value is removed and the model is rebuilt until residual validation passes or the number of available baselines is less than four.

[0038] Step 5: If no misfixation was found in Step 4 during the process of determining the misfixation of wide lane ambiguity, the system continues to determine whether or not there is unfixed ambiguity within the current ambiguity calculation unit. If unfixed ambiguity exists, the system uses the regional ionospheric delay model from Step 3 to interpolate the wide lane ionospheric delay corresponding to the unfixed ambiguity, and calculates candidate values ​​for wide lane ambiguity and the corresponding wide lane ionospheric delay.

[0039] After the residual check is passed, the wide-lane ionospheric interpolation values ​​for the remaining unfixed baseline are calculated using the following formula. ∇ΔIWL-IPν =a1·ΔX ν +a2·ΔY ν In the formula, ∇ΔI WL-IPν ΔX is the unfixed double differential ionospheric delay at baseline v. ν ΔY ν This represents the distance from each point in the reference station's central coordinate system of baseline v to the east and north directions of the reference point, respectively. A clause-by-clause check and judgment is performed on the baseline with unfixed wide-lane ambiguity. The integer closest to the floating solution of wide-lane ambiguity is used as the center, and the system is extended by 2 or 3 cycles at each end to obtain several candidate integer solutions. These solutions are substituted into Step 3 to calculate the corresponding wide-lane ionospheric delay. If the wide-lane ionospheric delay of a candidate point satisfies the judgment conditions in the following equation, the ambiguity can be directly fixed. |∇ΔI WL-IPν -∇ΔI WLν |<0.3·μ In the formula, ∇ΔI WLν This is the double-differential ionospheric delay calculated based on candidate integer solutions.

[0040] Since μ is approximately 0.65 m, the order of magnitude of errors due to multipath and unmodeled errors in wide-lane ionospheric delay is generally at the centimeter level and therefore does not affect the verification. On the other hand, because the size of the area used is limited by this method, the interpolation effect of the ionospheric region between overlapping puncture points is good, and the interpolation error is unlikely to exceed 10 cm. Therefore, theoretically, the accuracy of the judgment using the following formula is very high.

[0041] Step 6: Check whether the candidate values ​​for wide lane ambiguity calculated in Step 5 meet the criteria. If YES, add to Step 3 and iteratively construct the regional ionospheric model. If NO, return to Step 5 and perform the calculation again to fix the wide lane ambiguity.

[0042] After all baselines have been checked in sequence, the model is rebuilt by substituting the newly fixed baselines to improve the accuracy of the ionospheric model. This process is repeated until all baselines are fixed or no new baselines are fixed.

[0043] The fixing of L1 ambiguity and the generation of subsequent correction values ​​both depend on the accurate fixing of wide lane ambiguity. While general wide lane ambiguity resolution methods can fix wide lane ambiguity in GNSS regional augmentation positioning, it is not possible to ensure that all wide lane ambiguity is fixed accurately within a relatively short time. On the other hand, unfixed or misfixed wide lane ambiguity affects the flow of all subsequent data processing, leading to a reduction in the number of satellites available for correction values ​​and degrading the user's positioning effectiveness. This invention proposes a method, apparatus, and system for fixing wide-lane ambiguities across an entire network based on ionospheric relationships, by linking the spatial relationships of atmospheric delays in GNSS regional augmentation positioning. This method utilizes several fixed wide-lane ambiguities within a CORS region to calculate the corresponding ionospheric delays, interpolates the ionospheric delays corresponding to unfixed wide-lane ambiguities using the spatial relationships of ionospheric delays, fixes the remaining unfixed wide-lane ambiguities using wavelengths with relatively large wide-lane observation values, and can also verify whether the fixed wide-lane ambiguities have been correctly fixed.

[0044] The ionospheric relationship-based wide-lane ambiguity network fixing device divides all base stations in the CORS network into multiple independent ambiguity calculation units using overlapping ionospheric puncture points. For each ambiguity calculation unit, it provides a decision module for collecting four or more groups of fixed wide-lane ambiguities within the unit, and calculates the wide-lane ionospheric delay corresponding to each group of wide-lane ambiguities collected. It constructs a regional ionospheric delay model, determines whether there are misfixed wide-lane ambiguities based on the model, removes the misfixed wide-lane ambiguities if they exist, determines whether the remaining data consists of four or more groups, and if there are four or more groups, The steps described above are repeated, and if there are fewer than four groups, the calculation flow is terminated. If no misfixation is found in the process of determining misfixation of wide lane ambiguity, the process continues to determine whether there is unfixed ambiguity within the current ambiguity calculation unit. If there is, a fixing module is used to interpolate the wide lane ionospheric delay corresponding to the unfixed ambiguity using a regional ionospheric delay model to calculate candidate values ​​for wide lane ambiguity and the corresponding wide lane ionospheric delay. A modeling module is also included to check whether the calculated candidate values ​​satisfy the criteria, and if YES, iteratively construct the regional ionospheric model; if NO, iteratively perform the calculation to fix the wide lane ambiguity.

[0045] A fixed system device for a wide-lane ambiguity network based on ionospheric relationships, comprising a processor, memory, and a fixed program for a wide-lane ambiguity network based on ionospheric relationships stored in the memory and executable by the processor, wherein when the above-described fixed program for a wide-lane ambiguity network based on ionospheric relationships is executed by the processor, the steps of the above-described method for fixing a wide-lane ambiguity network based on ionospheric relationships are realized.

[0046] (Examples) Using data from nine GNSS reference stations that meet the requirement of overlapping puncture points within a CORS network in a certain region, a simulation test was conducted in a low-latitude region of southern China. The average distance between stations was 48 km, and the distribution of measurement points is shown in Figure 1.

[0047] Experiment 1: Verification of outlier detection Conventional methods for fixing wide-lane ambiguities cannot guarantee perfectly accurate ambiguity fixing. Even a single cycle of misfixation of a wide-lane ambiguity can result in a very large error of approximately 0.65 m in the subsequent extraction, modeling, and interpolation of ionospheric delay. Furthermore, because this method performs fixing across the entire network of wide-lane ambiguities based on spatial relationships, the failure to detect misfixed wide-lane ambiguities will have a significant impact on other unfixed wide-lane ambiguities within the unit. Therefore, it is crucial to verify the outlier detection capability of the new method.

[0048] To check the detection capability of the new method when an outlier is present, by adding a one-cycle outlier to a random, fixed wide-lane ambiguity within the unit, and to verify whether the amount of data used for solving affects detection, nine CORS base stations are selected, and the post-validation residuals and post-validation accuracy are compared to verify whether there is any change when the quantity of observations used for modeling differs.

[0049] This experiment is a validation experiment for the GPS PRN06 satellite. A one-cycle outlier was added to the integer solution of the GPS PRN06 wide-lane ambiguity in the baseline GT14_GT18. As can be seen in Figures 2 and 3, the residual of GPS PRN06 fluctuates significantly to 0.4m due to the cause of the integrated adjustment, making the outlier change easily detectable. Correspondingly, the mean squared error of the wide-lane ionospheric region across the entire GPS PRN06 network also fluctuates significantly from approximately 0.03m to 0.13m, demonstrating a significant impact of the wide-lane ambiguity outlier on the posterior accuracy of the wide-lane ionospheric model. The drops at both ends of the figures indicate that the GPS PRN06 in the GT14_GT18 baseline was not conspicuous during the initial and final arc segments and was not used for adjustment, resulting in near agreement between the results including the outlier and the correct results.

[0050] Based on the above, the average inter-station distance in China's CORS network is often distributed within the range of 30 to 100 km. Therefore, the number of base stations within a single computing unit is limited, mostly 8 to 10. Considering the experimental results above, by setting the post-validation residual threshold for a single observation to 0.3 times μ and the overall post-existing accuracy to 0.2 times μ, it is possible to effectively determine the misfixation of wide lane ambiguity.

[0051] Experiment 2: Verification experiment for fixing wide lane ambiguity After determining the detection outlier threshold, the CORS data is resolved to fix the wide-lane ambiguity. To verify the ambiguity fixing performance, regardless of whether the wide-lane ambiguity in the epoch is completely fixed or not, only the four groups of fixed wide-lane ambiguities are selected, and the remaining wide-lane ambiguities are fixed across the entire network of wide-lane ambiguities based on ionospheric relationships. Furthermore, information such as the fixing accuracy is statistically calculated using post-hoc true values. Since this method does not involve relationships between epochs and is actually a fixing method for a single epoch, the data results for each epoch can be considered independent. The results are summarized in the table below. [Table 1]

[0052] As can be seen from the table above, the new method according to the present invention exhibits extremely good performance in terms of fixation rate and accuracy, and no misestimation occurred in approximately 400,000 groups of wide lane ambiguities. It has a relatively good ability to detect and correct ambiguity misfixation in conventional wide lane ambiguity fixing methods, and the extremely high fixation rate of the new method means that if the wide lane ambiguities of four baselines or four base stations are fixed in this computing unit, it is possible to fix almost all wide lane ambiguities in this epoch.

[0053] Experiment 3: Positioning Verification To further verify the effectiveness of the present invention's method and the improvement of user positioning effectiveness by maximizing and fixing wide lane ambiguity, the present invention simulates a GT20 observation station in CORS as a user. For convenience of comparison, the method without employing the present invention's method is referred to as Method 1, and the method employing the present invention and fixing the wide lane ambiguity of the CORS network is referred to as Method 2. The mobile station's positioning resolution mode is unified to the ambiguity fixing mode in a dynamic single epoch, representing the actual performance of rapid ambiguity fixing at the user terminal.

[0054] Figures 4 to 9 are time-series diagrams of positioning results for a single BDS, a single GPS, and a GC (a combination of GPS and BDS) corresponding to Methods 1 and 2. It can be seen that this method significantly improves both the positioning accuracy and fixation rate of all three combinations. The improvements in fixation rates for BDS, GPS, and GC were 48.7%, 8.8%, and 12.8%, respectively, and the accuracy in the elevation direction improved by 32.3%, 3.8%, and 17.4%, respectively, showing a remarkable improvement in the positioning effectiveness of the BDS system.

[0055] Referring to the conventional method in Table 2, the correction values ​​distributed by network RTK include only an average of 4.4 groups of BDS observations, which is far fewer than the 6.6 groups obtained when the method of the present invention is applied. The main reason for this is that the BDS is significantly affected by multipath effects from satellites, resulting in a decrease in the fixation rate of BDS wide lane ambiguity. [Table 2]

[0056] As described above, the novel method for fixing wide lane ambiguity based on ionospheric relationships according to the present invention can accurately detect mis-fixed wide lane ambiguity and fix regional wide lane ambiguity almost completely and accurately, thereby removing the constraints on subsequent data processing of network RTK caused by wide lane ambiguity fixing. Furthermore, the accuracy of this method was verified through positioning experiments, and it was confirmed that when wide lane ambiguity is sufficiently and accurately fixed, it significantly improves indicators such as the fixation rate and positioning accuracy in positioning using three types of combinations of GPS, BDS, and GC. Here, the greatest improvement was observed in the BDS system, with the overall fixation rate improving by approximately 35% and vertical positioning accuracy improving by more than 30%.

[0057] (Note) (Note 1) Step 1 involves using overlapping ionospheric puncture points to divide all base stations within the CORS network into multiple independent ambiguity computing units, Step 2 involves collecting four or more fixed wide-lane ambiguities within each ambiguity calculation unit. Step 3 involves calculating the wide-lane ionospheric delay corresponding to the wide-lane ambiguity of each collected group and constructing a regional ionospheric delay model. Step 4 determines whether there is misfixation of wide lane ambiguity based on the regional ionospheric delay model. If misfixation exists, it removes the misfixed wide lane ambiguity. It then determines whether there are four or more groups of remaining wide lane ambiguity data. If there are four or more groups, Step 3 is repeated. If there are fewer than four groups of remaining wide lane ambiguity data, the calculation flow is terminated. Step 4 determines whether a wide-lane ambiguity is misfixed. If no misfixation is found in the process of determining the misfixation of the wide-lane ambiguity, Step 5 determines whether an unfixed wide-lane ambiguity exists within the current ambiguity calculation unit. If an unfixed wide-lane ambiguity exists, Step 5 uses the regional ionospheric delay model from Step 3 to interpolate the wide-lane ionospheric delay corresponding to the unfixed ambiguity, and calculates candidate values ​​for the wide-lane ambiguity and the corresponding wide-lane ionospheric delay. A method for fixing wide lane ambiguity across an entire network based on ionospheric relationships, characterized by including Step 6, which involves checking whether the candidate values ​​of wide lane ambiguity calculated in Step 5 satisfy the judgment conditions, and if YES, iteratively constructing a regional ionospheric delay model in addition to Step 3, and if NO, returning to Step 5 to perform calculations again to fix the wide lane ambiguity.

[0058] (Note 2) The specific process of Step 1 is as follows: The specific formula for determining the ionospheric puncture point is as follows: Formula (1)

number

number

number

[0059] (Note 3) The specific process for Step 2 is as follows: The method for fixing wide-lane ambiguities in an entire network based on ionospheric relationships, as described in Appendix 2, is characterized by first attempting to fix ambiguities for each satellite-base station combination in the ambiguity calculation unit, increasing the number of fixed ambiguities by fixing some ambiguities if it is not possible to fix all ambiguities, then calculating the residuals and variances of the observational model of the satellite-base station with fixed ambiguities and assigning the residuals and variances to each other variable part including the parameter to be estimated according to the law of error propagation, extracting the residuals and variances of the ionospheric-related part, and finally selecting and sorting the residuals of the ionospheric-related part, selecting and fixing the fixed wide-lane ambiguities with the smallest variance in order, and further using them in subsequent calculations.

[0060] (Note 4) The specific process for Step 3 is as follows: First, the wide-lane ambiguity is processed at a fixed baseline, and the approximate ionospheric delay, i.e., the wide-lane ionospheric delay ∇ΔI, is extracted based on the wide-lane ambiguity at that baseline. WL The calculation formula is as follows: Formula (2)

number

number

number

[0061] (Note 5) The specific process for Step 4 is as follows: After the regional ionospheric delay model was constructed, the residuals of the observational model were validated due to the existence of extraneous observations, and the validation indices were as follows:

number

[0062] (Note 6) In Step 5, after the residual check is passed, the remaining unfixed baseline wide lane ionospheric interpolation value is calculated using the following formula: ∇ΔI WL-IPν =a1·ΔX ν +a2·ΔYν In the formula, ∇ΔI WL-IPν ΔX is the unfixed double differential ionospheric delay at baseline v. ν ΔY ν This represents the distance from each point in the reference station's central coordinate system of baseline ν to the east and north of the reference point, respectively. A clause-by-clause check and judgment is performed on the baseline with unfixed wide-lane ambiguity. The integer closest to the floating solution of wide-lane ambiguity is used as the center, and the system is extended by 2 or 3 cycles at each end to obtain several candidate integer solutions. These solutions are substituted into Step 3 to calculate the corresponding wide-lane ionospheric delay. If the wide-lane ionospheric delay of a candidate point satisfies the judgment conditions in the following equation, the ambiguity is directly fixed. |∇ΔI WL-IPν -∇ΔI WLν |<0.3·μ In the formula, ∇ΔI WLν A method for fixing a wide-lane ambiguity network based on ionospheric relationships, as described in Appendix 5, characterized in that the delay is a double-difference ionospheric delay calculated based on candidate integer solutions.

[0063] (Note 7) Step 6 is characterized by collecting ionospheric delay observation values ​​with wide lane ambiguity fixed for all measurement points in the current epoch, calculating model calculation values ​​for the regional ionospheric delay model at the above observation points, calculating the mean error and standard difference of the current model using the difference between the model calculation output and the observation values, and if the candidate value satisfies the judgment conditions, first calculating the ionospheric delay value for the region where the candidate value is located using the existing regional ionospheric delay model, and finally determining whether the difference between the candidate value and the regional ionospheric delay model value is less than a standard difference of 3, and if YES, constructing the regional ionospheric model iteratively, and if NO, returning to Step 5 and performing the calculation to fix the wide lane ambiguity again, as described in Appendix 6.

[0064] (Note 8) A method for fixing a wide-lane ambiguity network based on ionospheric relationships as described in Appendix 7, characterized in that, after Step 6 is completed, if fluctuations occur in the existing independent ambiguity calculation units due to satellite motion, causing two independent ambiguity calculation units to merge with each other, or if a puncture point within a single ambiguity calculation unit splits because it does not satisfy the constraints of Step 1, the method involves iteratively recalculating the regional ionospheric delay model according to Steps 1 to 6 using the existing ionospheric delay model.

[0065] (Note 9) By utilizing overlapping ionospheric puncture points, all base stations within the CORS network are divided into multiple independent ambiguity computing units, and each ambiguity computing unit is provided with a decision module for collecting four or more fixed wide-lane ambiguities within the ambiguity computing unit. The wide-lane ionospheric delay corresponding to each collected group of wide-lane ambiguities is calculated, a regional ionospheric delay model is constructed, and based on the regional ionospheric delay model, it is determined whether or not there are misfixed wide-lane ambiguities. If misfixation exists, the misfixed wide-lane ambiguities are removed, and it is determined whether or not there are four or more groups of remaining data. If there are four or more groups, the above steps are repeated. If there are fewer than four groups, the calculation flow is terminated. If no misfixation is found in the process of determining the misfixation of wide-lane ambiguities, it is further determined whether or not there are unfixed ambiguities within the current ambiguity calculation unit. If there are, the wide-lane ionospheric delay corresponding to the unfixed ambiguity is interpolated using the regional ionospheric delay model, and a fixing module is created to calculate candidate values ​​for wide-lane ambiguities and the corresponding wide-lane ionospheric delays. An apparatus for using a method for fixing wide-lane ambiguity across an entire network based on ionospheric relationships as described in Appendix 8, characterized by including a modeling module for checking whether the calculated candidate values ​​satisfy the judgment conditions, iteratively constructing a regional ionospheric model if YES, and for redoing calculations to fix wide-lane ambiguity if NO.

[0066] (Note 10) A system using the method for fixing a wide-lane ambiguity network based on ionospheric relationships as described in Appendix 8, characterized in that the system equipment includes a processor, memory, and a fixing program for the entire network of ionospheric relationships based on wide-lane ambiguity that is stored in the memory and executable by the processor.

Claims

1. Step 1 involves using overlapping ionospheric puncture points to divide all base stations within the CORS network into multiple independent ambiguity computing units, Step 2 involves collecting four or more fixed wide-lane ambiguities within each ambiguity calculation unit, Step 3 involves calculating the wide-lane ionospheric delay corresponding to the wide-lane ambiguity of each collected group and constructing a regional ionospheric delay model. Step 4 determines whether there is misfixation of wide lane ambiguity based on the regional ionospheric delay model. If misfixation exists, it removes the misfixed wide lane ambiguity. It then determines whether there are four or more groups of remaining wide lane ambiguity data. If there are four or more groups, Step 3 is repeated. If there are fewer than four groups of remaining wide lane ambiguity data, the calculation flow is terminated in Step 4. Step 4 determines whether a wide-lane ambiguity is misfixed. If no misfixation is found in the process of determining the misfixation of the wide-lane ambiguity, Step 5 determines whether an unfixed wide-lane ambiguity exists within the current ambiguity calculation unit. If an unfixed wide-lane ambiguity exists, Step 5 uses the regional ionospheric delay model from Step 3 to interpolate the wide-lane ionospheric delay corresponding to the unfixed ambiguity, and calculates candidate values ​​for the wide-lane ambiguity and the corresponding wide-lane ionospheric delay. A method for fixing wide lane ambiguity in an entire network based on ionospheric relationships, characterized by including Step 6, which involves checking whether the candidate values ​​of wide lane ambiguity calculated in Step 5 satisfy the judgment conditions, and if YES, adding to Step 3 to iteratively construct a regional ionospheric delay model, and if NO, returning to Step 5 to perform calculations again to fix the wide lane ambiguity.

2. The specific process of Step 1 is as follows: The specific formula for determining the ionospheric puncture point is as follows: Formula (1) [Math 1] During the ceremony, [Math 2] and [Math 3] The method for fixing wide-lane ambiguity based on ionospheric relationships across the entire network, as described in claim 1, is characterized in that, by determining whether or not the ionospheric puncture points are overlapping, all base stations in the CORS network are divided into multiple independent ambiguity calculation units, and it is ensured that the ionospheric puncture points of each base station within the same calculation unit all overlap.

3. The specific process of Step 2 is as follows: A method for fixing wide-lane ambiguities in an entire network based on ionospheric relationships, as described in 2, characterized in that: first, an attempt is made to fix ambiguities for each satellite-base station combination in the ambiguity calculation unit; if it is not possible to fix all ambiguities, the number of ambiguities to be fixed is increased by fixing some of the ambiguities; then, the residuals and variances of the observational value model of the satellite-base station with fixed ambiguities are calculated and assigned to each other variable part including the parameter to be estimated according to the law of error propagation; the residuals and variances of the part related to the ionosphere are extracted; finally, the residuals of the part related to the ionosphere are selected and sorted; the fixed wide-lane ambiguities with the smallest variance are selected in order and fixed, and used in subsequent calculations.

4. The specific process of Step 3 is as follows: First, a baseline with fixed wide-lane ambiguity is processed, and the approximate ionospheric delay, i.e., the wide-lane ionospheric delay ∇ΔI, is extracted based on the wide-lane ambiguity at that baseline. ＷＬ The calculation formula is as follows: Formula (2) [Math 4] where λ ｉ ＷＬ represents the wavelength of the observed value of the wide-lane carrier, and ∇Δφ ｉｊ ＷＬ，ｒｍ represents the observed value of the double-difference wide-lane carrier, and ∇ΔN ｉｊ ＷＬ，ｒｍ is the fixed solution of the known double-difference wide-lane ambiguity, and ∇Δρ ｉｊ ｒｍ is the double-difference satellite-to-ground distance, obtained by calculation, f １ and f ２ represent the frequencies of the first frequency point and the second frequency point, ∇ΔI is the double-difference ionospheric delay of the observed value of the L1 carrier, L1 is the first frequency point of the observed value of the GPS system, m ｉ ｒ , m ｊ ｒ , m ｉ ｍ , m ｊ ｍ respectively represent the projection function values of the tropospheric delay errors between the observation stations r and m and the satellites i and j, T ｒ and T ｍ represent the tropospheric delays above the observation stations r and m, and ∇Δγ ｉｊ ＷＬ，ｒｍ represents the hardware delay error of the double-difference carrier, and ε(φ ｉｊ ｒｍ ) represents the noise of the observed value of the carrier, Equation (2) extracts the wide-lane ionospheric delay corresponding to a baseline with fixed wide-lane ambiguity, and constructs a regional ionospheric delay model, the equation of which is as follows: V = BX - L In the equation, V represents the residual matrix between the wide-lane ionospheric delay model and the observed values, B is the model design matrix, X is the parameter matrix of the model to be estimated, L is the matrix of observed values, and ΔX １ ΔY １ ΔX is the distance from each point in the reference station center coordinate system of baseline 1 to the east and north directions of the reference point. ２ ΔY ２ ΔX is the distance from each point in the reference station center coordinate system of baseline 2 to the east and north directions of the reference point. ｎ ΔY ｎ This is the distance from each point in the reference station center coordinate system of baseline n to the east and north directions of the reference point, and satisfies the following equation: [Math 5] a １ a ２ These are the corresponding interpolation coefficients, [Math 6] The method for fixing a wide lane ambiguity network based on ionospheric relationships as described in claim 3.

5. The specific process of Step 4 is as follows: After the regional ionospheric delay model was constructed, the residuals of the observational model were validated due to the existence of extraneous observations, and the validation indices were as follows: [Number 7] In the formula, V ｉ represents the i-th residual, λ ＷＬ This represents the wavelength of the observed wide lane carrier, V Ｔ μ is the transpose vector of the residual matrix, and μ is the delayed change in the wide lane ionosphere due to the misfixation of one cycle of wide lane ambiguity, which can be derived from the equation in Step 3. A method for fixing a wide-lane ambiguity network based on ionospheric relationships, according to claim 4, characterized in that if the residuals exceed an acceptable threshold, the maximum residual value is removed and the model is rebuilt until the residual validation passes or the number of available baselines is less than four.

6. In Step 5, after the residual check is passed, the wide-lane ionospheric interpolation value of the remaining unfixed baseline is calculated using the following formula. ∇ΔI ＷＬ－ＩＰν =a １ ・ΔX ν +a ２ ・ΔY ν In the formula, ∇ΔI ＷＬ－ＩＰν ΔX is the unfixed double differential ionospheric delay at baseline v. ν ΔY ν This represents the distance from each point in the reference station's central coordinate system of baseline ν to the east and north of the reference point, respectively. A clause-by-clause check and judgment is performed on the baseline with unfixed wide-lane ambiguity. The integer closest to the floating solution of wide-lane ambiguity is used as the center, and two or three cycles are extended to each end to obtain several candidate integer solutions. These are substituted into Step 3 to calculate the corresponding wide-lane ionospheric delay. If the wide-lane ionospheric delay of a certain candidate point satisfies the judgment conditions in the following equation, the ambiguity is directly fixed. |∇ΔI ＷＬ－ＩＰν -∇ΔI ＷＬν |<0.3・μ In the formula, ∇ΔI ＷＬν The method for fixing a wide-lane ambiguity network based on ionospheric relationships according to claim 5, characterized in that is a double-difference ionospheric delay calculated based on candidate integer solutions.

7. Step 6 is characterized in that, in Step 6, ionospheric delay observation values ​​with wide lane ambiguity fixed for all measurement points in the current epoch are collected, model calculation values ​​of the regional ionospheric delay model at the above observation points are calculated, the mean error and standard difference of the current model are calculated using the difference between the model calculation output and the observation values, and if the candidate value satisfies the judgment conditions, first the ionospheric delay value of the region where the candidate value is located is calculated using the existing regional ionospheric delay model, and finally it is determined whether the difference between the candidate value and the regional ionospheric delay model value is less than a standard difference of 3, and if YES, the regional ionospheric model is constructed iteratively, and if NO, the calculation to fix the wide lane ambiguity is performed again, returning to Step 5.

8. The method for fixing the entire network of ionospheric relationships-based wide-lane ambiguity according to claim 7, characterized in that, after Step 6 is completed, fluctuations occur in the existing independent ambiguity calculation units due to satellite motion, causing two independent ambiguity calculation units to merge with each other, or causing a puncture point within a single ambiguity calculation unit to split because it does not satisfy the constraints of Step 1, the existing ionospheric delay model is used to perform iterative recalculations of the regional ionospheric delay model according to Steps 1 to 6.

9. By utilizing overlapping ionospheric puncture points, all base stations within the CORS network are divided into multiple independent ambiguity computing units, and each ambiguity computing unit is provided with a decision module for collecting four or more fixed wide-lane ambiguities within the ambiguity computing unit. The wide-lane ionospheric delay corresponding to the wide-lane ambiguity of each collected group is calculated, a regional ionospheric delay model is constructed, and based on the regional ionospheric delay model, it is determined whether or not there are misfixed wide-lane ambiguities. If misfixation exists, the misfixed wide-lane ambiguities are removed, and it is determined whether or not there are four or more groups of remaining data. If there are four or more groups, the above steps are repeated. If there are fewer than four groups, the calculation flow is terminated. If no misfixation is found in the process of determining the misfixation of wide-lane ambiguities, it is further determined whether or not there are unfixed ambiguities within the current ambiguity calculation unit. If there are, the regional ionospheric delay model is used to interpolate the wide-lane ionospheric delay corresponding to the unfixed ambiguity, and a fixing module is used to calculate candidate values ​​for wide-lane ambiguities and the corresponding wide-lane ionospheric delays. An apparatus for using a method for fixing wide-lane ambiguity across an entire network based on ionospheric relationships, as described in 8, comprising a modeling module for checking whether the calculated candidate values ​​satisfy a judgment condition, iteratively constructing a regional ionospheric model if YES, and for redoing calculations to fix wide-lane ambiguity if NO.

10. A system using the method for fixing a wide-lane ambiguity network based on ionospheric relationships according to claim 8, wherein the system equipment includes a processor, memory, and a program for fixing a wide-lane ambiguity network based on ionospheric relationships stored in the memory and executable by the processor.

Images