A quality control based ambiguity transfer method, system, device and medium

Abstract

The application discloses a quality control-based ambiguity transfer method, system, device and medium, and relates to the technical field of satellite navigation and positioning. The method comprises the following steps: acquiring a floating-point ambiguity; checking the decimal part of the floating-point ambiguity; if the decimal part is greater than a set threshold, the floating-point ambiguity is not fixed; otherwise, the floating-point ambiguity is fixed, and the obtained candidate fixed solution is subjected to Ratio checking; if the candidate fixed solution is less than the set threshold, the epoch retains the floating-point solution; otherwise, the fixed ambiguity is subjected to quality control; the fixed ambiguity after quality control is checked; if the number of the fixed ambiguity is less than the set threshold, the epoch retains the floating-point solution; otherwise, a pseudo-observation equation is established according to the fixed ambiguity, filtering is solved, the current epoch ambiguity parameter is updated, and the epoch is subsequently inherited to a subsequent continuous arc segment without cycle slip. The method aims to solve the problem of decimeter-level positioning error caused by the transfer of the wrong fixed ambiguity in the traditional FH mode.

Description

Technical Field

[0001] This invention relates to the field of satellite navigation and positioning technology, specifically to a method, system, device, and medium for ambiguity transfer based on quality control. Background Technology

[0002] Precise Point Positioning (PPP) is a high-precision positioning technology based on the Global Navigation Satellite System (GNSS). PPP eliminates most of the error sources in traditional point positioning (such as satellite orbit errors and clock errors) by using precise satellite orbit and clock bias products, thus achieving high-precision positioning. Traditional PPP uses floating-point solutions, which have the disadvantage of slow convergence speed, limiting its application range. Compared with ambiguity-based floating-point solutions, PPP with fixed ambiguity solutions can significantly shorten the convergence time and improve positioning accuracy to a certain extent. The key is whether the integer ambiguity can be fixed quickly and reliably. Only when the ambiguity is correctly fixed can the phase observation become an absolute observation similar to pseudorange, but with ranging accuracy reaching the centimeter or even millimeter level.

[0003] Ambiguity fixing (AR) has thus become a breakthrough in accelerating PPP convergence, and numerous scholars both domestically and internationally have conducted in-depth research on it. First, the integer characteristic of ambiguity is ignored, and a floating-point solution for PPP is executed. Then, integer ambiguity fixing is divided into two steps: ambiguity resolution and verification. The least squares ambiguity decorrelation adjustment (LAMBDA) method is the most commonly used ambiguity resolution method and can be used to search for integer candidate solutions. Ambiguity verification methods mainly include fixed threshold ratio-test, bootstrapping success rate test, and fixed failure rate ratio-test. Due to the simplicity and effectiveness of the fixed threshold ratio-test, most scholars and software primarily use it for verification. After passing the ambiguity verification, the fixed solution update formula can be used, or the fixed solution for parameters of interest such as position can be obtained by back-substituting integer ambiguity.

[0004] For widely used real-time high-precision PPP-AR solutions, how to use and inherit the ambiguity fixed solution information of the current epoch is also a key step, significantly affecting the fixation rate and accuracy of the fixed solution. In the commonly used Fixand Hold (FH) mode, integer ambiguities can be used to impose strict constraints on floating-point ambiguity estimates. The fixed ambiguities are maintained and inherited in subsequent continuous arcs without cycle slips, improving the continuity and reliability of positioning. This method is mostly used in applications with high requirements for positioning accuracy and stability, such as high-precision measurement and precision navigation. According to the research of Takasu and Yasuda in 2010, in this mode, the E / N / U direction errors are 1.1 cm, 1.9 cm, and 3.5 cm, respectively. The fixation rate is improved to 99.3%, and this mode is effective in improving initialization performance and ambiguity rate. However, in the FH ambiguity fixed mode, there is a situation where the subsequent positioning results diverge due to an ambiguity fixing error in a certain epoch, resulting in poor positioning accuracy. Therefore, some scholars have designed a dual-horizontal-line processing line, including an "updating" processing line and an FH processing line. The "updating" processing line is responsible for updating a copy of the filter state using all successfully fixed ambiguities. Its primary purpose is to output navigation and positioning parameters in each cycle, while the FH processing line is designed to update the filter state directly with the "correctly" fixed ambiguities, passing this tight constraint to the next cycle.

[0005] In summary, the existing FH method cannot completely distinguish incorrect fixed ambiguities. Incorrect fixed ambiguities will lead to positioning deviations at the decimeter level or even larger, and the solution results are even worse than the floating-point solution of ambiguity, thereby reducing the positioning accuracy of the satellite system. Summary of the Invention

[0006] To address the issue that existing technologies cannot completely distinguish incorrectly fixed ambiguities, which can lead to positioning errors at the decimeter level or even larger, this invention proposes a quality-controlled ambiguity transfer method, system, device, and medium. By implementing a quality control strategy, low-precision fixed ambiguities are effectively screened out, ensuring that only high-quality ambiguities are transferred to the next epoch, thereby greatly improving the problems existing in the prior art.

[0007] A fuzzy resolution method based on quality control includes the following steps:

[0008] Receive satellite system observation data, construct a precise single-point positioning (PPP) observation model, solve the observation model to obtain floating-point ambiguity, and fix the floating-point ambiguity to obtain fixed ambiguity;

[0009] The cutoff elevation angle of the satellite corresponding to the fixed ambiguity is checked. If the cutoff elevation angle is less than a set threshold, the fixed ambiguity of the satellite is marked as unreliable. Otherwise, the signal-to-noise ratio (SNR) of the L1 and L2 carrier frequencies of the satellite is checked. If the SNR of the L1 and L2 carrier frequencies is lower than a set threshold, the fixed ambiguity of the satellite is marked as unreliable. Otherwise, the fixed ambiguities corresponding to the satellites marked as downweighted during the observation model solution are filtered. The phase residual of the filtered satellite during the filtered solution is checked. If the phase residual of the satellite is greater than a set threshold, the fixed ambiguity is excluded. Otherwise, it is determined whether the number of consecutive fixed epochs of the satellite is greater than a set threshold. If the number of consecutive fixed epochs is less than a set threshold, the fixed ambiguity of the satellite is marked as unreliable. After excluding all fixed ambiguities marked as unreliable, the remaining fixed ambiguities are obtained.

[0010] The pseudo-observation equation is established based on the remaining fixed ambiguity to update the ambiguity parameters of the current epoch, and then inherited to the epochs of the continuous arc segments of the satellite system that have not experienced cycle slips.

[0011] Furthermore, the process of receiving satellite system observation data, constructing a precise single-point positioning (PPP) observation model, solving the observation model to obtain floating-point ambiguity, and fixing the floating-point ambiguity to obtain fixed ambiguity specifically includes the following steps:

[0012] Receive satellite system observation data and construct a precise single-point positioning PPP observation model;

[0013] The Kalman filtering method is used to solve the observation model to obtain the floating-point ambiguity;

[0014] The decimal part of the floating-point ambiguity is checked. If the decimal part is greater than or less than a set threshold, the ambiguity is not fixed. Otherwise, the floating-point ambiguity is fixed by using the least squares ambiguity decorrelation adjustment method LAMBD A to obtain the fixed ambiguity.

[0015] Furthermore, the received satellite system observation data is used to construct a precise single-point positioning (PPP) observation model, which is expressed as follows:

[0016]

[0017] in, dt is the geometric distance between the satellite and the receiver, measured in meters; c is the speed of light; dt r and dt s These are the receiver and satellite clock errors, respectively, in seconds; λ is the tropospheric delay, measured in meters. IF B represents the carrier wavelength of the ionospheric-free view combination; r,IF and These are the carrier phase fractional deviations (UPD) at the receiver and satellite ends, respectively. It is the whole-cycle blur; b r,IF It is the hardware delay of the code pseudorange between the receiver antenna and the signal correlator; It is the hardware delay of the code pseudorange between the satellite signal transmitter and the satellite antenna; Indicates pseudorange measurement error; Indicates carrier phase measurement error; and These represent pseudorange observations and carrier phase observations, respectively.

[0018] Furthermore, the floating-point ambiguity is fixed using the least squares ambiguity decorrelation adjustment method LAMBDA, and the resulting fixed ambiguity is specifically expressed as follows:

[0019]

[0020] in, Indicates the floating-point ambiguity of the IF combination; D nl The variance representing the NL ambiguity; D represents the fixed width ambiguity (WL); IF is the non-differenced IF combined floating-point ambiguity covariance matrix; f1 and f2 represent the frequencies of carriers L1 and L2, respectively.

[0021] Furthermore, the pseudo-observation equation y established based on the remaining fixed ambiguity is expressed as:

[0022] y=Hx+vv~N(0,R s )

[0023] in,

[0024]

[0025] R s =diag(δ s 2 ,δ s 2 ,…)

[0026] In the formula, H represents the coefficient matrix of the pseudo-observation equation; x represents the fixed satellite pair ambiguity vector, x1 represents the reference satellite ambiguity, and x... n Represents the nth fixed satellite ambiguity; v represents the observation noise vector, δ s 2 The observed noise follows a pattern with a mean of 0 and a variance-covariance matrix of R. s The normal distribution, R s This represents the variance of the ambiguity retention error of the s-satellite system.

[0027] The present invention also includes a fuzzy resolution system based on quality control, comprising:

[0028] The fixed ambiguity acquisition module is used to receive satellite system observation data, construct a precise single-point positioning PPP observation model, solve the observation model to obtain floating-point ambiguity, and fix the ambiguity of the floating-point ambiguity to obtain fixed ambiguity;

[0029] The quality control module is used to verify the cutoff elevation angle of the satellite corresponding to the fixed ambiguity. If the cutoff elevation angle is less than a set threshold, the fixed ambiguity of the satellite is marked as unreliable. Otherwise, it verifies the signal-to-noise ratio (SNR) of the L1 and L2 carrier frequencies of the satellite. If the SNR of the L1 and L2 carrier frequencies is lower than a set threshold, the fixed ambiguity of the satellite is marked as unreliable. Otherwise, it filters the fixed ambiguities corresponding to the satellites marked as downweighted during the observation model solution. It then verifies the phase residual of the filtered satellite during the filtered solution. If the phase residual of the satellite is greater than a set threshold, the fixed ambiguity is excluded. Otherwise, it determines whether the number of consecutive fixed epochs of the satellite is greater than a set threshold. If the number of consecutive fixed epochs is less than a set threshold, the fixed ambiguity of the satellite is marked as unreliable. After excluding all fixed ambiguities marked as unreliable, the remaining fixed ambiguities are obtained.

[0030] The transfer module is used to establish a pseudo-observation equation based on the remaining fixed ambiguity to update the ambiguity parameters of the current epoch, and to inherit them to the epochs of the continuous arc segments of the satellite system that have not experienced cycle slips.

[0031] The present invention also includes a quality control-based ambiguity transfer computer device, comprising: a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the quality control-based ambiguity transfer method.

[0032] The present invention also includes a readable storage medium storing a computer program, the computer program including program instructions, which, when executed by a processor, are used to perform the steps of the ambiguity transfer method based on quality control.

[0033] This invention provides a fuzzy resolution method based on quality control, which has the following advantages:

[0034] This invention addresses the issue of fixed ambiguity caused by the propagation of erroneous fixed ambiguities in traditional FH mode. It considers that signals from low-elevation satellites are typically more susceptible to multipath effects and higher noise levels. Therefore, it verifies the cutoff elevation angle of the satellite corresponding to the fixed ambiguity. Furthermore, it examines the signal-to-noise ratio (SNR) of satellite measurements with low signal-to-noise ratios (SNR) due to interference or severe noise levels, resulting in low-precision fixed ambiguities. Since the accuracy of the PPP floating-point solution significantly impacts the performance of fixed ambiguities, highly reliable measurements are prioritized during subset selection. Considering that carrier phase multipath effects and undetected cycle slips are reflected in the posterior phase residual, fixed ambiguities with large posterior phase residuals are excluded. This method aims to solve the decimeter-level positioning error problem caused by the propagation of erroneous fixed ambiguities in traditional FH mode. By implementing a more rigorous preprocessing strategy, it effectively filters out low-precision fixed ambiguities, preventing them from contaminating the ambiguity preservation process and improving the positioning accuracy of the satellite system. Attached Figure Description

[0035] Figure 1 This is a flowchart of the ambiguity fixing method in an embodiment of the present invention;

[0036] Figure 2 This is a flowchart of FH ambiguity transfer based on quality control in an embodiment of the present invention;

[0037] Figure 3 This is a user station distribution diagram in an embodiment of the present invention;

[0038] Figure 4 This is a histogram of the positioning error statistics in the east, north, and sky directions in the static mode of this invention embodiment;

[0039] Figure 5 This is a cumulative distribution diagram of positioning errors in static mode in an embodiment of the present invention;

[0040] Figure 6 This is a coordinate sequence diagram of the KRGG station in the east, north, and sky directions in the simulated dynamic mode of the present invention, for the year 075 of 2024.

[0041] Figure 7 This is a histogram of positioning statistics in the east, north, and sky directions in the simulated dynamic mode of this invention embodiment;

[0042] Figure 8 This is a cumulative distribution diagram of positioning error in the simulated dynamic mode of this invention. Detailed Implementation

[0043] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0044] This invention proposes a partial ambiguity preservation strategy based on quality control. This strategy aims to solve the decimeter-level positioning error problem caused by the propagation of incorrect fixed ambiguities in traditional FH mode, while also overcoming the shortcomings of existing ambiguity preservation subset selection methods that do not fully consider various factors affecting ambiguity accuracy. By cleverly integrating the idea of ​​PAR subset selection into FH mode and implementing a more rigorous preprocessing strategy, low-precision fixed ambiguities are effectively screened out, thus avoiding their contamination of the ambiguity preservation process. Furthermore, the fixed ambiguities considered correct are used as pseudo-observations and participate in the filtering process of the corresponding floating-point solution, ensuring that only high-quality ambiguities are propagated to the next epoch. This not only realizes the partial ambiguity preservation subset selection scheme but also verifies its practical application effect in PPP-AR technology. Specifically, it includes the following steps:

[0045] S1. Receive GNSS observation data and construct a Precise Point Positioning (PPP) observation model. An ionospherically incompatible (IF) combination model is used as the observation method. Since both carriers used by GPS satellites are located in the L-band of microwaves, they are referred to as L1 carrier and L2 carrier, respectively. IF combination is performed on the GPS using L1 and L2 carriers. IF combination is one of the most commonly used observation combination models in PPP. This model can eliminate the first-order ionospheric delay in pseudorange and carrier phase measurements through dual-frequency linear combination. The pseudorange and phase observation equations for ionospherically incompatible combination can be expressed as:

[0046]

[0047] here dt is the geometric distance between the satellite and the receiver, measured in meters; c is the speed of light; dt r and dt s These are the receiver and satellite clock errors, respectively, in seconds; λIF represents the tropospheric delay, measured in meters; λIF represents the carrier wavelength of the ionospheric-free view combination; B r,IF and These are the carrier phase fractional deviation (UPD) at the receiver end and the satellite end, respectively. Only stable and reliable UPD products can achieve subsequent PPP ambiguity fixing. It is the whole-cycle blur; b r,IF It is the hardware delay of the code pseudorange between the receiver antenna and the signal correlator; It is the hardware delay of the code pseudorange between the satellite signal transmitter and the satellite antenna; Indicates pseudorange measurement error; Indicates carrier phase measurement error. and These represent carrier phase observations and pseudorange observations, respectively.

[0048] In practice, because precision satellite clock corrections are estimated using ionospherically unaffected combined observations, the hardware delay of the satellite-end IF combined pseudorange is significant when using precision products. Eliminated by satellite clock bias products; since pseudorange observations provide the absolute reference for receiver clock bias, the receiver clock bias correction c·dt r Absorbing receiver-side ionospheric combination pseudorange hardware delay b r,IF In ambiguity floating-point PPP data processing, since the phase delay is linearly related to the ambiguity parameter, the ambiguity parameter absorbs the phase delay.

[0049] After partial error correction using precision products, equation (1) can be expressed as:

[0050]

[0051] In the formula and These represent the reparameterized receiver clock error and ambiguity parameters, respectively:

[0052]

[0053]

[0054] According to equation (2), the ambiguity term in the ionospheric linear combination observations is not an integer, so the ambiguity is generally estimated as a real unknown parameter. In order to obtain an effective PPP fixed solution, after the initial phase and hardware delay are eliminated, the ionospheric ambiguity parameter can be decomposed into a combination of wide-lane (WL) and narrow-lane (NL) ambiguities. Thus, the problem of finding integer solutions to PPP ambiguities is transformed into finding integer solutions to wide-lane and narrow-lane ambiguities. Once the wide-lane and narrow-lane ambiguities of the carrier phase are correctly fixed, they are recombine as known quantities to form the IF combined ambiguity, and the resulting ambiguity is the fixed value of the IF combined ambiguity parameter.

[0055] S2. The observation model is solved using the Kalman filtering method to obtain the floating-point ambiguity.

[0056] S3. After the decimal deviation correction, the floating-point ambiguity has recovered its integer characteristics. The ambiguity with recovered integer characteristics is checked for its decimal part. If the decimal part is greater than the set threshold, it is not fixed; otherwise, the ambiguity is fixed.

[0057] S4. The method of fixing the ambiguity of the wide lane (WL) and narrow lane (NL) ambiguities step by step is adopted to achieve the ionosphere-free ambiguity fixation of PPP. Among them, the wide lane ambiguity fixation directly uses the rounding method, while the narrow lane ambiguity fixation adopts the least-squares ambiguity decorrelation adjustment (LAMBDA) method.

[0058] PPP ambiguity fixing; PPP-AR ambiguity fixing is a crucial step in high-precision positioning. It involves correcting real-number ambiguities that have lost their integer properties to restore their integer characteristics. When the real-number ambiguities converge to a certain accuracy, ambiguity fixing can be performed, thereby obtaining high-precision positioning results. The flowchart for ambiguity fixing is as follows: Figure 1 As shown.

[0059] Because the ambiguity of the PPP non-ionospheric combination method does not possess integer properties, PPP ambiguity fixing is typically decomposed into attempting to fix the wide-lane and narrow-lane ambiguities sequentially. (IF combination ambiguity) The wide-lane ambiguity and the narrow-lane ambiguity can be expressed as:

[0060]

[0061] In the formula, N wl and λ wl These are the wide-lane ambiguity and its wavelength; N1 and λ. nl These are the narrow lane ambiguity and its wavelength. When fixing the combined ambiguity of wide lane, narrow lane, and ionosphere-free (IF), users need to perform inter-satellite single-difference processing to eliminate the influence of receiver-side UPD.

[0062] Wide-lane ambiguity is typically fixed using a combination of wide-lane phase minus narrow-lane pseudorange (MW). However, due to the influence of observation noise and multipath effects, the MW combination requires multi-epoch smoothing before the wide-lane ambiguity is fixed.

[0063]

[0064] In the formula, Indicates the combined observation values ​​of MW; and Let $\mathbf{L}$ represent floating-point single-difference WL ambiguity and single-difference WL ambiguity with integer characteristics, respectively. From the above formula, it can be seen that to obtain integer WL ambiguity, the receiver-side and satellite-side $\mathbf{D}$ values ​​are required. r,wl and UPD at the receiver end is eliminated by inter-satellite single difference, and UPD at the satellite end is corrected by analyzing the wide-lane ambiguity product broadcast by the analysis center. WL ambiguity is less affected by measurement noise and observation error due to its longer wavelength, and can achieve high accuracy after smoothing for several epochs. Therefore, WL ambiguity is fixed and directly uses the rounding method.

[0065] When a fixed WL ambiguity is obtained, combining it with the ionosphere-free combined floating-point ambiguity yields an NL ambiguity with integer characteristics and its variance:

[0066]

[0067] In the formula, Indicates the floating-point ambiguity of the IF combination; This indicates a fixed WL ambiguity; f1 and f2 represent the frequencies of the L1 and L2 carriers, respectively, with magnitudes of 1575.42MHz and 1227.60MHz; D IF This is the non-differential, ionosphere-free floating-point ambiguity covariance matrix. After phase deviation correction, the NL ambiguity recovers its integer characteristics.

[0068] Since the single-difference NL ambiguities in PPP have strong correlation, the NL ambiguities are fixed using the least squares ambiguity decorrelation adjustment (LAMBDA) method. After obtaining the fixed wide-lane and narrow-lane ambiguities, they are substituted into equation (7) to obtain the fixed IF combined ambiguities. The PPP-AR function model can be simplified as follows:

[0069] E(y)=Aa+Bb+ε,D(y)=Q y (11)

[0070] In the formula, E(·) and D(·) represent the expectation and variance, respectively; y is the carrier phase observation vector; a is the integer ambiguity parameter vector; b is the coordinates and other parameters independent of ambiguity; A and B are the corresponding coefficient matrices; ε is the observation residual vector; Q y Let be the variance-covariance matrix of the observed values.

[0071] According to the least squares theory,

[0072]

[0073] Estimate the unknowns a and b under the given conditions. The least squares theory when some unknowns are limited to integers is called the integer least squares theory. In the integer least squares estimation algorithm, equation (12) cannot be solved directly, so it needs to be solved in three steps:

[0074] ① Ignore integer parameters; use traditional least squares estimation methods to solve for position parameters. With fuzzyness parameter floating-point solution And the corresponding covariance matrix:

[0075]

[0076] Then, the NL ambiguity for recovering integer properties is obtained according to the method in step S2.

[0077] ② Considering the integer characteristics of the ambiguity parameter, the LAMBDA algorithm proposed by Professor Teunissen is used to fix the NL ambiguity. The LAMBDA algorithm used to fix the NL ambiguity includes two parts: first, ambiguity downcorrelation processing, which is integer Z-transform, is performed, and then integer ambiguity search is performed.

[0078] The phase-biased NL ambiguity is transformed by an integer Z-transform to obtain a new ambiguity vector:

[0079]

[0080] In the formula, and Represents the original ambiguity and the variance-covariance matrix of the original ambiguity; and Z represents the ambiguity vector and ambiguity variance-covariance matrix after decorrelation; Z is an invertible integer transformation matrix.

[0081] Search for integer candidate solutions to the floating-point ambiguity solution and its variance-covariance matrix after downcorrelation processing in order to obtain the optimal solution with fixed ambiguity parameters. and its variance Therefore, based on step one, consider the optimal estimation criterion:

[0082]

[0083] In the formula, for Integer vector solutions.

[0084] Finally, the integer ambiguities of the newly searched space are transformed inversely to obtain the real ambiguities of the original space.

[0085] Based on fuzzy parameter fixed solution and its variance and covariance matrix The position parameters are updated to obtain a fixed solution for the position parameters. and its variance and covariance matrix

[0086]

[0087] In the formula, and These represent the floating-point solutions of the position parameters and their variance-covariance matrices, respectively. This represents the floating-point solution for ambiguity.

[0088] ③ Substitute the integer fuzziness obtained from the search into the normal equations to solve for the fixed solution again.

[0089] S5. After searching using the LAMBDA method, multiple candidate NL fixed solutions can be obtained, and the Ratio test is performed on the candidate NL fixed ambiguity.

[0090] The most commonly used Ratio test is a data-driven indicator of the reliability of fixed ambiguity. It refers to the ratio of the sum of squared errors between the second-best and best ambiguity integers. The larger the ratio, the stronger the reliability of the candidate best integer value for ambiguity. The specific formula is as follows:

[0091]

[0092] In the formula, and The floating-point ambiguity vector and variance have integer ambiguity characteristics; and represents the candidate vectors for the optimal and suboptimal ambiguity integer solutions, respectively; k is the threshold, a constant given empirically, often taking the value of 2 or 3.

[0093] S6. Perform quality control on the obtained fixed ambiguity. If the fixed ambiguity passes all quality control measures, it is considered to be correctly fixed; otherwise, it is marked as an unreliable ambiguity.

[0094] In PPP-AR, effectively utilizing fixed ambiguities is key to improving fixation performance and positioning accuracy. However, the traditional FH ambiguity fixation mode carries the risk of maintaining incorrectly fixed ambiguities. The traditional FH fixation mode uses the fixed solutions of all ambiguities in a given epoch as measurements, the calculated floating-point ambiguity solutions as the state, performs Kalman filtering, and uses the fused ambiguity as the current ambiguity state. A pseudo-observation equation y is formed using the fixed ambiguities corresponding to the fixed satellites:

[0095] y=Hx+vv~N(0,R s (18)

[0096] here:

[0097]

[0098] R s =diag(δ s 2 ,δ s 2 ,…) (twenty one)

[0099] In the formula, H represents the coefficient matrix of the pseudo-observation equation; x represents the fixed satellite pair ambiguity vector, x1 represents the reference satellite ambiguity, and x... n Represents the ambiguity of the nth fixed satellite; v represents the observation noise vector, δ s 2 This indicates that the observed noise follows a pattern with a mean of 0 and a variance-covariance matrix of R. s The normal distribution, R s This represents the variance of the ambiguity retention error of the s-satellite system.

[0100] In FH mode, integer ambiguity is used to constrain floating-point ambiguity, which improves the fixed accuracy of ambiguity. However, there is a possibility of using incorrectly fixed ambiguity for constraint. In this mode, if an incorrect fixed ambiguity is maintained, it will result in a worse effect than when no floating-point ambiguity constraint is applied, resulting in positioning accuracy at the decimeter or even meter level.

[0101] To optimize this method, this invention proposes an FH mode based on strict quality control. Compared to the traditional mode, this mode implements a more stringent preprocessing strategy for the preserved ambiguities, applying constraints to a subset of fixed ambiguities. The thresholds involved in this strategy are all empirically set. Specifically, they include:

[0102] (1) Altitude check. Generally speaking, the lower the satellite elevation angle, the more susceptible its observations are to the multipath effect. Therefore, in this invention, satellites with elevation angles below 10° are excluded.

[0103] Considering that signals from low-elevation satellites are usually more affected by multipath effects and have greater noise, the cutoff elevation angle of the satellite corresponding to the fixed ambiguity is checked. If the cutoff elevation angle is less than 10°, the fixed ambiguity of the satellite is not subject to the constraint of the floating-point ambiguity; otherwise, the next quality control index is judged.

[0104] (2) Signal-to-noise ratio (SNR) check. If the SNR of the satellite signal is too low, it may mean that the signal is interfered with or severely attenuated, which will affect the accurate determination of ambiguity. Therefore, for a satellite, if the SNR of the first and second frequencies is lower than a given value (the empirical value of this invention is 35 dB-Hz), they will not participate in ambiguity preservation.

[0105] Considering that satellite measurements with low signal-to-noise ratio may have interference or severe noise levels, resulting in low-precision fixed ambiguity, the signal-to-noise ratio of the first and second frequencies of the satellite that passed the test (1) is tested. If the signal-to-noise ratio at the first or second frequency is lower than 35dB-Hz, the corresponding fixed ambiguity will not be used as a constraint condition for subsequent epochs; otherwise, the next quality control index will be judged.

[0106] (3) Floating-point solution accuracy. The accuracy of the floating-point solution has a significant impact on the fixation of narrow-lane ambiguity. Therefore, satellite observations with higher weights and greater reliability were prioritized, and satellites whose floating-point solutions were downweighted were screened to ensure data quality.

[0107] Since the accuracy of the PPP floating-point solution has a significant impact on the performance of fixed ambiguities, highly reliable measurements are prioritized during subset selection. In the PPP floating-point solution calculation in step one, GEO satellites and satellites with large prefit residuals are weighted and marked. If a marked satellite is detected, the corresponding fixed ambiguity is filtered out; otherwise, the next quality control indicator is evaluated.

[0108] (4) Posterior carrier phase residual. Considering that factors such as multipath effect and ionospheric delay may increase the floating-point solution phase residual, thus affecting positioning accuracy, satellites with large phase residuals were also filtered out.

[0109] Carrier phase multipath effects and undetected cycle slips are both reflected in the posterior phase residual. Therefore, if the fixed ambiguity has a large posterior phase residual, it is excluded; otherwise, the final step of quality control is performed.

[0110] (5) Continuous fixed epoch number. During the positioning convergence process, there may be occasional situations where the decimal part of the ambiguity is just close to 0 and is fixed. Such fixation is not very reliable. Therefore, only satellites that have been fixed for more than a threshold number of consecutive epochs will maintain their fixed ambiguity (4 epochs in this invention) in order to avoid transmitting unreliable ambiguities fixed during the convergence process.

[0111] To ensure that the ambiguity accuracy of most satellites has converged, only satellites that meet all the above quality control criteria are marked as fixed satellites. In this invention, maintaining fixed ambiguity is only allowed when the number of fixed satellites is no less than four. Otherwise, the fixed ambiguity is marked as unreliable.

[0112] S7. Exclude all fixed ambiguities marked as unreliable. Check the number of correct fixed ambiguities. If the number of correct fixed ambiguities is less than a set threshold, the PPP floating-point solution is retained for this epoch, and the fixed ambiguities are not used to constrain the parameters to be estimated. Otherwise, filter the ambiguity based on the correct fixed solution and inherit the integer ambiguity solution into the parameter vector to be estimated.

[0113] S8. The fixed ambiguity obtained through quality control is used as a pseudo-observation equation to update the parameters to be estimated. At the same time, the fixed solution ambiguity is used to update the ambiguity parameters of the current epoch, and then inherited to the epochs of subsequent continuous arc segments that have not experienced cycle slips.

[0114] Experimental proof and results analysis:

[0115] To verify and evaluate the proposed quality control-based partial ambiguity preservation strategy, experiments were conducted in both static and pseudo-dynamic modes. Both experiments described the dataset, processing methods, and evaluation metrics (ambiguity fixation performance and positioning accuracy). The ambiguity fixation performance analysis focused on two dimensions: epoch fixation rate and correct fixation rate. The epoch fixation rate is the ratio of fixed solution epochs to the total number of epochs; the correct fixation rate is the ratio of fixed solution epochs with a 3D positioning error less than 1 dm to the total number of fixed epochs. In both static and pseudo-dynamic modes, three control groups were set up, and the specific experimental design is shown in Table 1.

[0116] Table 1 Experimental Design Scheme

[0117]

[0118] Experimental Data and Processing Strategy: The data for this invention were selected from 10 days of 2024 DOY (DoY) 70-79, with a sampling interval of 30 seconds, covering observational data from 50 MGEX stations roughly evenly distributed globally. The distribution of user stations is as follows: Figure 3 As shown. The precise orbit and clock bias products used were post-hoc multi-system precise orbit and clock bias products provided by the Center for Orbit Determination in Europe (CODE). The server-side employed 100 globally evenly distributed MGEX stations, with self-estimated post-hoc UPD products to achieve rapid ambiguity fixation. The user-side used forward Kalman filtering to simulate real-time processing of the IGS MGEX station tracking dataset based on IF combinations. The 24-hour static post-processed PPP was used as the true reference value for the station coordinates.

[0119] To mitigate the potential impact of receiver-side UPD on observations, the satellite with the largest elevation angle was selected as the reference benchmark, and inter-satellite single-difference calculations were performed. When ambiguity was fixed, the satellite cutoff elevation angle was set to 15°; the GPS error variance was set to 0.03 weeks. Furthermore, the PPP-AR resolution strategy remained the same for different ambiguity-fixed modes, with no special settings. During data processing in static mode, each observation file was divided into 3-hour segments, totaling 8 observation arcs per day, with a total of 4000 observation arcs in the experiment. Detailed data resolution strategies are shown in Table 2.

[0120] Table 2 Solution Processing Strategy

[0121]

[0122] 3-hour Static PPP Result Analysis and Comparison: Table 3 lists the statistical results of epoch fixation rate and correct fixation rate in the static mode. It is clear from Table 3 that, in the static mode, the ambiguity fixation rates of Experiment II and Experiment III both exceeded 90%. However, the ambiguity fixation rate is insufficient to fully reflect the reliability of the fixed solution; therefore, a further comparative analysis of the ambiguity correct fixation rate is conducted. The correct fixation rate of Experiment III is significantly higher than that of Experiment II, reaching 98.81%. This result clearly reveals that Experiment III possesses higher accuracy and reliability in fixing ambiguities. This indicates that the algorithm presented in this study exhibits stable fixation performance in the static mode and demonstrates a certain superiority over traditional methods in terms of correct fixation rate.

[0123] Table 3. Statistics on epoch fixation rate and correct fixation rate of the two groups of experiments under static mode.

[0124]

[0125] Figure 4 and Figure 5 Figures (a), (b), and (c) respectively show the histograms and cumulative distribution (CDF) plots of positioning error statistics for each arc segment in static mode, the fixed solution under the traditional ambiguity-preserving fixed algorithm, and the fixed solution under the partially ambiguity-preserving fixed algorithm proposed in this invention, in the East, North, and Up directions. It can be observed from the figures that the fixed solution positioning deviation RMS of Experiment III achieves significant reductions of approximately 50%, 17%, and 5% in the East, North, and Up directions, respectively, compared to the floating solution. Furthermore, Experiment III has a high percentage of deviations with absolute values ​​less than 2 cm in the three directions (94.8%, 98.4%, and 70.5%), which fully demonstrates that the PPP fixed solution maintains its superior positioning accuracy. Moreover, Experiment III performs particularly well in terms of positioning deviation at the 95th percentile, with the smallest value among the three groups of experiments, clearly indicating that the method of this invention can guarantee small positioning errors in most cases, thus demonstrating higher positioning accuracy. In Experiment III, the percentages of positioning deviations greater than 1 dm in the East, North, and Sky directions were 0.4%, 0.1%, and 0.8%, respectively. Compared to Experiment II, no data point with a deviation exceeding 1 dm was found to have a higher percentage than the floating-point solution. This indicates that the algorithm effectively reduces the risk of incorrectly fixing ambiguities while maintaining high accuracy of the fixed solution, thus demonstrating superior stability and reliability. Figure 5 It can be seen that the CDF curves of the three sets of experiments are quite similar, with no significant differences. This is because in the static solution, the convergence of each estimated parameter is good, and the positioning accuracy of each method is comparable after convergence. However, the method proposed in this invention maintains the fastest ascent speed at smaller error values, meaning that this method has a higher probability of obtaining smaller errors in all directions, thereby ensuring higher positioning accuracy.

[0126] Comparison of results of simulated dynamic PPP fixed solution: In order to verify the positioning performance of the algorithm in different scenarios, a simulated dynamic PPP solution experiment was conducted. The coordinates of the user station were estimated every 30 seconds for 24 hours. Other processing strategies were the same as those of the static solution mode.

[0127] Figure 6 Figures (a), (b), and (c) show the coordinate sequence diagrams in three directions for the KRGG station in simulated dynamic mode (floating-point solution), Hold mode (fixed solution), and the fixed solution using the method proposed in this invention, for a day with an annual day of 75 in 2024. As can be seen from the figures, after approximately 8 hours, the Hold mode experiences significant fluctuations in positioning deviation in the three directions due to a fixed ambiguity error. This error not only propagates rapidly but also causes the positioning in Hold mode to diverge, resulting in a sharp deterioration in accuracy. The maximum deviation in the three directions even reaches the meter level, and the RMS values ​​are as high as 16.8 cm, 19.0 cm, and 25.2 cm, respectively. In contrast, the method of this invention, through its more rigorous quality control screening, avoids the problem of erroneous ambiguity propagation, thereby maintaining the stability and high accuracy of positioning. Compared with the floating-point solution, this method improves the positioning accuracy in the three directions by 60.7%, 23.8%, and 21.2%, respectively. This preliminary result clearly demonstrates that the method proposed in this invention can effectively compensate for the shortcomings of traditional methods in identifying ambiguities in fixed errors, fully leverage the advantages of fixed solutions, and provide strong assurance for the accuracy and reliability of positioning.

[0128] Table 4 provides statistics on epoch fixation rate and correct fixation rate under the simulated dynamic mode. Analysis of the data in the table reveals that Experiment III exhibits significant advantages over Experiment II in both epoch fixation rate and correct fixation rate, improving by 3.59% and 4.45%, respectively. This result demonstrates that, under the simulated dynamic mode, the algorithm proposed in this invention can provide more stable and reliable positioning results.

[0129] Table 4. Statistics on epoch fixation rate and correct fixation rate of the two groups of experiments under the simulated dynamic mode.

[0130]

[0131] Figure 7Images (a), (b), and (c) respectively show the error statistics histograms for the floating-point solution in the simulated dynamic mode, the fixed solution under the traditional ambiguity-preserving fixed algorithm, and the fixed solution under the partially ambiguity-preserving fixed algorithm proposed in this invention, in the East, North, and Sky directions. Theoretically, the accuracy of the fixed solution should be higher than that of the floating-point solution. However, the RMS of the fixed solution in Experiment II is abnormally higher than that of the floating-point solution in the three directions. The proportion of positioning errors exceeding 1 dm in Experiment II is significantly higher than in the other two groups of experiments, reflecting that the incorrect ambiguity fixation is the most serious in this experiment. In contrast, the fixed solution in Experiment III shows the lowest RMS error, demonstrating its better positioning accuracy, specifically improved by 32%, 15%, and 8% in the three directions, respectively. In Experiment III, the proportions of positioning deviations with absolute values ​​less than 2 cm in the three directions reached 86.4%, 88.3%, and 49%, respectively, which are 2.8, 3.5, and 1.2 percentage points higher than those in Experiment II. At the 95th percentile, the positioning errors in all three directions of Experiment III were significantly smaller than those in Experiment II, especially in the east direction, where the error reduction exceeded 61%. This indicates that the positioning performance of the algorithm of this invention is superior under extreme conditions. Furthermore, Experiment III exhibited the lowest proportion of errors exceeding 1 decimeter, which not only further confirms the significant effect of the algorithm of this invention in improving positioning accuracy but also shows a more concentrated error distribution, effectively avoiding a large proportion of excessively large positioning errors, thus providing users with more stable and accurate positioning results.

[0132] Figure 8 Figures (a), (b), and (c) show the CDF plots of the positioning errors in the East, North, and Sky directions under the simulated dynamic mode, respectively. In the simulated dynamic mode, the Hold mode can correctly fix most data ambiguities, achieving positioning accuracy superior to the floating-point solution; its error is better than the floating-point solution within approximately 80% cumulative probability. However, a few data fixing errors in the Hold mode lead to larger positioning deviations, thus its error is higher than the floating-point solution when the cumulative probability is above 90%. Experiment III generally shows smaller positioning errors in all directions, and its CDF curve is above that of Experiment II for most of the error range. This indicates that the method proposed in this invention can achieve smaller positioning errors with a higher probability, thereby significantly improving positioning accuracy and reliability. In contrast, in the static mode, the CDF curves of the three sets of experiments are relatively similar, with no significant differences, but the method proposed in this invention still shows the best positioning accuracy among the three sets of experiments. In the simulated dynamic PPP experiment, due to the complexity of environmental factors and the influence of motion, the positioning accuracy is often difficult to match that of the static experiment. However, the algorithm proposed in this invention can still demonstrate excellent positioning performance under dynamic conditions. This discovery provides strong support for the promotion and application of the algorithm in dynamic application scenarios.

[0133] This invention proposes a quality control-based ambiguity preservation strategy to optimize the ambiguity propagation process. Compared to the traditional FH method, this strategy implements rigorous quality control measures to identify potential erroneous ambiguities and then selects a subset of high-confidence fixed ambiguities. This subset is then used to constrain the floating-point solutions of the ambiguities, thereby significantly improving the correct ambiguity fixation rate and effectively mitigating the negative impact of erroneous fixed ambiguities on subsequent positioning results.

[0134] To verify the effectiveness of this strategy, this invention selected data from 50 globally evenly distributed MGEX stations during the DOY period (070-079) of 2024, and conducted three sets of experiments in static and pseudo-dynamic modes: PPP floating-point solution, traditional FH mode fixed solution, and the fixed solution of the method of this invention. Experimental results show that in static mode, the fixed solution exhibited by the algorithm of this invention has a significant advantage in positioning accuracy compared to the floating-point solution, especially in the eastward direction, where RMS is reduced by 50%. Compared with the FH mode, the proportion of positioning deviations greater than 1 dm is reduced. This algorithm not only leverages the high accuracy advantage of the fixed solution but also overcomes, to some extent, the contamination of subsequent epochs by incorrectly fixed ambiguities. In the pseudo-dynamic experiments, the advantages of this algorithm are even more prominent. Its correct ambiguity fixing rate increases from 91.0% to 95.5% compared to the traditional algorithm, and the proportion of deviations less than 1 dm is reduced by up to 4.2 percentage points, without any decrease in positioning accuracy compared to the floating-point solution. These experimental results strongly demonstrate the feasibility and practicality of the algorithm of this invention.

[0135] Based on the same inventive concept, this invention also proposes a fuzzy resolution system based on quality control, comprising:

[0136] The fixed ambiguity acquisition module is used to receive GNSS observation data, construct a precise single-point positioning (PPP) observation model, and use the Kalman filtering method to solve the observation model to obtain floating-point ambiguity. The decimal value of the floating-point ambiguity is checked. If the decimal value is greater than a set threshold, the ambiguity is not fixed. Otherwise, the floating-point ambiguity is fixed by using the least squares ambiguity decorrelation adjustment method LAMBDA.

[0137] The quality control module performs a ratio check on the fixed ambiguities obtained through LAMBDA search. If the fixed ambiguity is less than a set threshold, the PPP floating-point solution is retained for that epoch; otherwise, quality control is performed on the fixed ambiguity. This quality control specifically includes: checking the cutoff elevation angle of the satellite corresponding to the fixed ambiguity; if the cutoff elevation angle is less than a set threshold, the fixed ambiguity of that satellite is marked as unreliable; otherwise, checking the signal-to-noise ratio (SNR) of the first and second frequencies of the satellite; if the SNR at either the first or second frequency is lower than a set threshold, the fixed ambiguity of that satellite is marked as unreliable; otherwise, filtering the fixed ambiguities corresponding to satellites marked as downweighted during filter parameter calculation; otherwise, checking the phase residual during the filter calculation of the satellite; if the satellite residual is greater than a set threshold, the fixed ambiguity is excluded; otherwise, determining whether the number of consecutive fixed epochs is greater than a set threshold; if the number of consecutive fixed epochs is less than a set threshold, the fixed ambiguity of that satellite is marked as unreliable.

[0138] The transfer module is used to examine the remaining fixed ambiguities after excluding all fixed ambiguities marked as untrusted. If the number of remaining fixed ambiguities is less than a set threshold, the PPP floating-point solution is retained for that epoch. Otherwise, a pseudo-observation equation is established based on the remaining fixed ambiguities to update the ambiguity parameters of the current epoch, and then inherited to subsequent epochs of continuous arc segments that have not experienced cycle slips.

[0139] The present invention also proposes a fuzzy resolution computer device based on quality control, comprising: a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the fuzzy resolution method based on quality control.

[0140] The present invention also proposes a readable storage medium storing a computer program, the computer program including program instructions, which, when executed by a processor, are used to perform steps of a quality control-based ambiguity transfer method.

[0141] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A fuzzy resolution method based on quality control, characterized in that, Includes the following steps: The process involves receiving satellite system observation data, constructing a precise point positioning (PPP) observation model, solving the observation model to obtain floating-point ambiguity, and then fixing the ambiguity to obtain fixed ambiguity. Specifically, the steps include: receiving satellite system observation data and constructing a precise point positioning (PPP) observation model; solving the observation model using the Kalman filtering method to obtain floating-point ambiguity; checking the decimal value of the floating-point ambiguity; if the decimal value is greater than or less than a set threshold, ambiguity fixing is not performed; otherwise, the floating-point ambiguity is fixed using the least squares ambiguity decorrelation adjustment (LAMBDA) method to obtain fixed ambiguity. The cutoff elevation angle of the satellite corresponding to the fixed ambiguity is checked. If the cutoff elevation angle is less than a set threshold, the fixed ambiguity of the satellite is marked as unreliable. Otherwise, the signal-to-noise ratio (SNR) of the L1 and L2 carrier frequencies of the satellite is checked. If the SNR of the L1 and L2 carrier frequencies is lower than a set threshold, the fixed ambiguity of the satellite is marked as unreliable. Otherwise, the fixed ambiguities corresponding to the satellites marked as downweighted during the observation model solution are filtered. The phase residual of the filtered satellite during the filtered solution is checked. If the phase residual of the satellite is greater than a set threshold, the fixed ambiguity is excluded. Otherwise, it is determined whether the number of consecutive fixed epochs of the satellite is greater than a set threshold. If the number of consecutive fixed epochs is less than a set threshold, the fixed ambiguity of the satellite is marked as unreliable. After excluding all fixed ambiguities marked as unreliable, the remaining fixed ambiguities are obtained. The pseudo-observation equation is established based on the remaining fixed ambiguity to update the ambiguity parameters of the current epoch, and then inherited to the epochs of the continuous arc segments of the satellite system that have not experienced cycle slips.

2. The fuzzy resolution method based on quality control according to claim 1, characterized in that, The received satellite system observation data is used to construct a precise single-point positioning (PPP) observation model, which is represented as follows: in, It is the geometric distance between the satellite and the receiver, measured in meters; It's the speed of light; and These are the receiver and satellite clock errors, respectively, in seconds; It is the tropospheric delay, measured in meters; The carrier wavelength represents the combination of ionospheric views; and These are the carrier phase fractional deviations (UPD) at the receiver and satellite ends, respectively. It refers to the overall blurriness. It is the hardware delay of the code pseudorange between the receiver antenna and the signal correlator; It is the hardware delay of the code pseudorange between the satellite signal transmitter and the satellite antenna; Indicates pseudorange measurement error; Indicates carrier phase measurement error; and These represent pseudorange observations and carrier phase observations, respectively.

3. The fuzzy resolution method based on quality control according to claim 1, characterized in that, The floating-point ambiguity is fixed by using the least squares ambiguity decorrelation adjustment method (LAMBDA). The resulting fixed ambiguity is specifically represented as follows: in, Indicates the floating-point ambiguity of the IF combination; The variance representing the NL ambiguity; This indicates a fixed width of the alleyway (WL) with ambiguity. The non-differenced IF combined floating-point ambiguity covariance matrix; and These represent the frequencies of the L1 and L2 carriers, respectively.

4. The fuzzy resolution method based on quality control according to claim 1, characterized in that, The pseudo-observation equation is established based on the remaining fixed ambiguity. , is represented as: in, In the formula, The coefficient matrix representing the pseudo-observation equation; This represents a fixed satellite pair ambiguity vector. Indicates the ambiguity of the reference satellite. Indicates the first A fixed satellite ambiguity; Represents the observation noise vector. The observed noise follows a pattern with a mean of 0 and a variance-covariance matrix of... The normal distribution express The variance of the ambiguity error of the satellite system.

5. A fuzzy resolution system based on quality control, characterized in that, include: The fixed ambiguity acquisition module is used to receive satellite system observation data, construct a precise point positioning (PPP) observation model, solve the observation model to obtain floating-point ambiguity, and fix the ambiguity of the floating-point ambiguity to obtain fixed ambiguity. Specifically, it includes the following steps: receiving satellite system observation data and constructing a precise point positioning (PPP) observation model; solving the observation model using the Kalman filtering method to obtain floating-point ambiguity; checking the decimal value of the floating-point ambiguity; if the decimal value is greater than or less than a set threshold, ambiguity fixing is not performed; otherwise, the floating-point ambiguity is fixed using the least squares ambiguity decorrelation adjustment (LAMBDA) method to obtain fixed ambiguity. The quality control module is used to verify the cutoff elevation angle of the satellite corresponding to the fixed ambiguity. If the cutoff elevation angle is less than a set threshold, the fixed ambiguity of the satellite is marked as unreliable. Otherwise, it verifies the signal-to-noise ratio (SNR) of the L1 and L2 carrier frequencies of the satellite. If the SNR of the L1 and L2 carrier frequencies is lower than a set threshold, the fixed ambiguity of the satellite is marked as unreliable. Otherwise, it filters the fixed ambiguities corresponding to the satellites marked as downweighted during the observation model solution. It then verifies the phase residual of the filtered satellite during the filtered solution. If the phase residual of the satellite is greater than a set threshold, the fixed ambiguity is excluded. Otherwise, it determines whether the number of consecutive fixed epochs of the satellite is greater than a set threshold. If the number of consecutive fixed epochs is less than a set threshold, the fixed ambiguity of the satellite is marked as unreliable. After excluding all fixed ambiguities marked as unreliable, the remaining fixed ambiguities are obtained. The transfer module is used to establish a pseudo-observation equation based on the remaining fixed ambiguity to update the ambiguity parameters of the current epoch, and to inherit them to the epochs of the continuous arc segments of the satellite system that have not experienced cycle slips.

6. A fuzzy resolution computer device based on quality control, characterized in that, include: A memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the ambiguity transfer method based on quality control as described in any one of claims 1-4.

7. A readable storage medium, characterized in that, The readable storage medium stores a computer program, which includes program instructions that, when executed by a processor, perform the steps of the ambiguity transfer method based on quality control as described in any one of claims 1-4.

Images