GNSS partial ambiguity fixed acceleration method based on isomorphic multiplexing

Abstract

The invention provides a GNSS partial ambiguity fixed acceleration method based on isomorphic multiplexing, and the method comprises the steps: carrying out the first complete decorrelation calculation of a covariance matrix of a first ambiguity subset, and systematically recording the decorrelation transformation information of the first ambiguity subset; when a subsequent ambiguity subset is processed, pre-transforming the covariance matrix of the current subset into a state closer to a decorrelation characteristic by utilizing the structure isomorphism between the covariance matrix of the current subset and the covariance matrix of the previous subset and multiplexing the recorded transformation information, and taking the state as an efficient starting point of iterative optimization; therefore, completely independent and time-consuming correlation reduction operation is prevented from being carried out on each subset covariance matrix from zero. According to the method, the overall calculation burden of sequential drop correlation in the partial ambiguity fixing process is remarkably reduced, the calculation efficiency is greatly improved, and the method is particularly suitable for a low-cost GNSS receiver and other high-real-time application platforms which have strict real-time requirements or are limited in calculation power.

Description

Technical Field

[0001] This invention relates to the field of navigation and positioning technology, and in particular to a method for fixing and accelerating partial ambiguity in GNSS based on isomorphic reuse. Background Technology

[0002] In high-precision GNSS carrier phase relative positioning, accurate fixation of integer ambiguities is crucial for achieving centimeter-level or even millimeter-level positioning accuracy. Its mathematical model is typically expressed as a mixed-integer least squares problem. Currently, the most mainstream ambiguity fixation method is the LAMBDA (Least-squares Ambiguity Decorrelation Adjustment) method proposed by Teunissen and its variants.

[0003] Furthermore, the core process of this method includes two main steps: First, the variance-covariance matrix of ambiguity is "decorrelation" or "lattice basis reduction" is performed through Integer Gaussian Transformations (IGTs) and conditional variance remutation to weaken the correlation between ambiguity parameters, thereby compressing the search space; then, an efficient integer search (e.g., SE-VB search) is performed in the transformed space to determine the optimal integer ambiguity vector.

[0004] With the development of multi-frequency, multi-system GNSS, the number of available satellites has increased significantly. While this improves observation redundancy, it also increases the ambiguity dimension. In complex observation environments (such as multipath effects and ionospheric disturbances), the success rate and reliability of simultaneously fixing all ambiguities may decrease. Therefore, the Partial Ambiguity Resolution (PAR) strategy has emerged. PAR improves the overall fixing success rate and positioning robustness by selecting and fixing a reliable subset of ambiguities and discarding those that are difficult to fix. Among these strategies, iteratively eliminating "unreliable" ambiguities within the original ambiguity domain (e.g., based on criteria such as ratio test value growth, ADOP changes, or residual analysis) and repeatedly executing the fixing process is a physically meaningful and widely adopted mainstream PAR strategy.

[0005] However, while this iterative PAR strategy improves robustness, it introduces a significant computational burden. In each iteration, after removing an ambiguity, the complete LAMBDA decorrelation and search process needs to be re-executed on the variance-covariance matrix corresponding to the remaining subset of ambiguities, which has been reduced in dimensionality. Existing research shows that, under the common processing dimensions of GNSS (10-40 dimensions), the computational time for decorrelation (lattice basis reduction) is much greater than that for integer search. Current research, such as the MLAMBDA algorithm proposed by Chang et al., and many subsequent improvements, mainly focuses on how to accelerate the decorrelation process of a single given high-dimensional covariance matrix.

[0006] Despite their significant achievements, these methods suffer from a fundamental efficiency bottleneck when applied to the iterative PAR scenario described above: during the continuous ambiguity removal and generation of a series of nested ambiguity subsets, the covariance matrices of consecutive subsets exhibit high structural similarity (isomorphism). Current decorrelation algorithms treat the covariance matrix of the current subset as a completely new and independent matrix in each iteration, ignoring this inherent connection with the (parent) matrix processed in the previous iteration, resulting in a large amount of repetitive and unnecessary computation. This redundant computation severely restricts the application of PAR methods on platforms with high real-time requirements (such as autonomous driving and drones) or on low-cost GNSS receivers with limited computing power.

[0007] Therefore, how to fully utilize the structural similarity between the covariance matrices of adjacent subsets in the PAR process, and design a sequential decorrelation method that can reuse historical decorrelation information, avoiding the need to perform independent and complete decorrelation calculations for each subset matrix from scratch, has become a key issue in improving the overall efficiency of iterative PAR and promoting its wider application. Summary of the Invention

[0008] This invention provides a method for accelerating GNSS partial ambiguity fixing based on isomorphic reuse. It addresses the high latency caused by iterative ambiguity fixing in existing high-precision real-time GNSS solutions, which hinders real-time applications. Specifically, it addresses the issue of excessive computational redundancy caused by existing LAMBDA and its improved algorithms performing completely independent, zero-based decorrelation calculations on the covariance matrix of each nested ambiguity subset during sequential processing, neglecting the significant computational redundancy resulting from the high structural similarity (isomorphism) between the covariance matrices of adjacent subsets. This invention provides a method and apparatus for accelerating GNSS partial ambiguity fixing and decorrelation based on isomorphic reuse. The aim of this invention is to significantly reduce the computational complexity of the overall PAR process and improve the execution efficiency of the decorrelation step, thereby promoting the widespread application of robust PAR technology in real-time high-precision positioning and low-cost GNSS equipment.

[0009] In a first aspect, the present invention provides a GNSS partial ambiguity fixing acceleration method based on isomorphic reuse, comprising: An observation equation is established using GNSS observations, and the floating-point estimate of the corresponding carrier phase ambiguity and the corresponding original covariance matrix are calculated using the least squares principle. Perform the first decorrelation calculation on the original covariance matrix to obtain the parent set decorrelation transformation matrix, and make the matrix obtained by multiplying the parent set decorrelation transformation matrix, the original covariance matrix and the transpose of the parent set decorrelation transformation matrix close to a diagonal matrix; According to the PAR strategy, any ambiguity is removed from the current ambiguity set to form a subset of ambiguities. A sequential downcorrelation acceleration operation is then performed on the subset of ambiguities to obtain a new downcorrelation transformation matrix for the subset of ambiguities. Using the ambiguity subset as the new parent set, the sequential decorrelation acceleration operation based on isomorphism is repeatedly executed using the PAR strategy based on the new decorrelation transformation matrix until the preset stopping condition of PAR is met, generating multiple ambiguity subsets with successively decreasing dimensions.

[0010] According to the present invention, a GNSS partial ambiguity fixed acceleration method based on isomorphic reuse is provided. This method removes any ambiguity from the current ambiguity set according to a partial ambiguity fixed PAR strategy to form an ambiguity subset. A sequential downcorrelation acceleration operation is then performed on the ambiguity subset to obtain a new downcorrelation transformation matrix for the ambiguity subset. The method includes: Based on the parent set downcorrelation transformation matrix The intermediate transformation matrix is ​​generated through elementary row and column transformations. , The subset covariance matrix corresponding to the subset after removing the i-th ambiguity. Transformed into an intermediate matrix that approximates a diagonal matrix ; With the intermediate matrix Perform a decorrelation calculation as the starting point of the iteration to obtain the covariance matrix of the subset. The new decorrelation transformation matrix .

[0011] According to the present invention, a GNSS partial ambiguity fixed acceleration method based on isomorphism reuse is provided, which is based on the parent set decorrelation transformation matrix. The intermediate transformation matrix is ​​generated through elementary row and column transformations. ,include: The parent set decorrelation transformation matrix Multiply by row and column permutation operations The inverse matrix is ​​obtained by finding the matrix. ,in , This represents the transformation matrix that swaps rows i and n. For matrix The last column undergoes Gaussian integer elimination. Let the single-mode transformation matrix be... , so that satisfaction go through The transformation achieves a last column of 0s, with the last element of the last column being 1. Take the transformation matrix The former The block matrix is .

[0012] According to the present invention, a GNSS partial ambiguity fixed acceleration method based on isomorphic reuse is provided, using the intermediate matrix. As the starting point for iteration, it includes: Calculate the intermediate matrix The intermediate matrix will be calculated The LAMBDA method is used to fix the ambiguity. Among them, the calculated intermediate matrix Compared to the untransformed subset covariance matrix It exhibits weaker off-diagonal correlation, reducing the number of integer Gaussian transformations and conditional variance reorderings required for subsequent decorrelation iterations.

[0013] According to the present invention, a GNSS partial ambiguity fixed acceleration method based on isomorphism reuse is provided, wherein the single-mode transformation matrix is: A matrix satisfies the condition that its absolute determinant is 1 and all its elements are integers, and is obtained by applying the matrix... The last column is constructed using integer Gaussian elimination, resulting in... The matrix is ​​preserved as an integer monomodal matrix.

[0014] According to the present invention, a GNSS partial ambiguity fixed acceleration method based on isomorphism reuse takes the ambiguity subset as a new parent set and repeatedly performs the isomorphism-based sequential decorrelation acceleration operation using the PAR strategy based on the new decorrelation transformation matrix, including: To sequentially process multiple subsets of ambiguity formed by the successive removal of ambiguity; For the (i+1)th subset, its sequential decorrelation acceleration operation reuses the decorrelation transformation matrix information of the i-th subset or the parent set.

[0015] Secondly, the present invention also provides a GNSS partial ambiguity fixed acceleration system based on isomorphic reuse, comprising: A module is established to build observation equations from GNSS observations and use the least squares principle to calculate the floating-point estimate of the corresponding carrier phase ambiguity and the corresponding original covariance matrix. The calculation module is used to perform the first decorrelation calculation on the original covariance matrix to obtain the parent set decorrelation transformation matrix, so that the matrix obtained by multiplying the parent set decorrelation transformation matrix, the original covariance matrix and the transpose of the parent set decorrelation transformation matrix is ​​close to a diagonal matrix. The acceleration module is used to remove any ambiguity from the current ambiguity set according to the PAR strategy to form an ambiguity subset, and to perform a sequential decorrelation acceleration operation on the ambiguity subset to obtain a new decorrelation transformation matrix of the ambiguity subset. The iterative module is used to take the ambiguity subset as a new parent set, and repeatedly execute the isomorphism-based sequential decorrelation acceleration operation based on the new decorrelation transformation matrix using the PAR strategy until the preset stopping condition of PAR is met, generating multiple ambiguity subsets with successively decreasing dimensions.

[0016] Thirdly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the GNSS partial ambiguity fixing acceleration method based on isomorphism reuse as described above.

[0017] Fourthly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the GNSS partial ambiguity fixing acceleration method based on isomorphism reuse as described above.

[0018] The GNSS partial ambiguity fixing acceleration method based on isomorphic reuse provided by this invention has the following beneficial effects: (1) Significantly improved computational efficiency: This invention innovatively utilizes the structural isomorphism of the covariance matrices of adjacent subsets in the PAR process. By reusing historical decorrelation information, it provides a high-quality starting point close to the final state for subsequent decorrelation calculations of subsets. This avoids the large number of repetitive iterations (IGTs and Permutations) generated by performing completely independent decorrelation calculations on the covariance matrix of each subset from scratch in the prior art, thereby reducing the average time consumption of each subsequent decorrelation calculation and significantly accelerating the entire PAR process.

[0019] (2) Ensuring equivalence of results: All transformations (permutations, integer Gaussian elimination) in this invention employ integer single-mode transformations, strictly preserving the mathematical properties of the correlation reduction process. Ultimately, the correctness and optimality of the results are ensured through standard correlation reduction iterations. Therefore, this invention is a lossless acceleration technique that improves speed without sacrificing the success rate of fixed ambiguity or the accuracy of fixed solutions.

[0020] (3) Universality and ease of integration: The sequential acceleration framework proposed in this invention is independent of specific decorrelation algorithm kernels (such as LAMBDA, MLAMBDA, etc.). It can be used as a preprocessing or acceleration module, seamlessly integrated with existing mainstream decorrelation algorithms, and enhances their performance in iterative PAR scenarios without modifying the search and verification parts of the algorithm core. It is easy to integrate into existing GNSS high-precision positioning software.

[0021] (4) Applicable to high real-time and low-cost platforms: By significantly reducing the computational load of the most time-consuming decorrelation step in the PAR process, this invention enables computationally intensive, robust PAR strategies to operate more efficiently. This is particularly beneficial for applications with extremely high real-time requirements (such as autonomous driving, UAV navigation, and dynamic deformation monitoring), and also makes it possible to achieve reliable, high-precision positioning on low-cost GNSS chips or receivers with limited computing power. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0023] Figure 1 This is one of the flowcharts of the GNSS partial ambiguity fixing acceleration method based on isomorphic reuse provided by the present invention; Figure 2 This is the second flowchart of the GNSS partial ambiguity fixing acceleration method based on isomorphic reuse provided by the present invention; Figure 3 This is a schematic diagram of the original covariance submatrix and transformation matrix provided by the present invention; Figure 4 This is a flowchart of the sequential decorrelation based on isomorphism provided by the present invention; Figure 5 This is a schematic diagram of the structure of the GNSS partial ambiguity fixed acceleration system based on isomorphic reuse provided by the present invention; Figure 6 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0025] To address the problems existing in the prior art, this invention proposes a fixed acceleration method for GNSS partial ambiguity based on isomorphic reuse. The core idea is to fully utilize the structural isomorphism between the covariance matrices of the two subsets when sequentially processing ambiguity subsets. By reusing the effective transformation information generated by the decorrelation process of the previous subset (parent set), a high-quality "starting point" closer to the final goal is provided for the decorrelation calculation of the current subset, thereby avoiding completely repetitive calculations and accelerating the decorrelation process.

[0026] Figure 1 This is one of the flowcharts illustrating the GNSS partial ambiguity fixing acceleration method based on isomorphic reuse provided in this embodiment of the invention, as shown below. Figure 1 As shown, it includes: Step 100: Establish observation equations using GNSS observations, and calculate the floating-point estimate of the corresponding carrier phase ambiguity and the corresponding original covariance matrix using the least squares principle; Step 200: Perform the first decorrelation calculation on the original covariance matrix to obtain the parent set decorrelation transformation matrix, so that the matrix obtained by multiplying the parent set decorrelation transformation matrix, the original covariance matrix, and the transpose of the parent set decorrelation transformation matrix is ​​close to a diagonal matrix; Step 300: Remove any ambiguity from the current ambiguity set according to the PAR strategy to form an ambiguity subset, and perform a sequential downcorrelation acceleration operation on the ambiguity subset to obtain a new downcorrelation transformation matrix for the ambiguity subset; Step 400: Using the ambiguity subset as the new parent set, and based on the new decorrelation transformation matrix, repeatedly execute the isomorphism-based sequential decorrelation acceleration operation using the PAR strategy until the PAR preset stopping condition is met, generating multiple ambiguity subsets with sequentially decreasing dimensions.

[0027] Specifically, the present invention includes the following steps: S1. Initialization and first complete descent are related.

[0028] Obtain the original floating-point solution vector of ambiguity containing n ambiguity parameters and its corresponding variance-covariance matrix. For the matrix Perform a complete, standard decorrelation calculation (e.g., perform a complete LAMBDA decorrelation procedure). This calculation produces a parent set integer single-mode transformation matrix. The transformed matrix The off-diagonal elements (i.e., the correlation between ambiguities) are greatly weakened, making it closer to a diagonal matrix. Record the complete decorrelation results, especially the transformation matrix. .

[0029] S2. Accelerated sequential decorrelation based on isomorphism.

[0030] When the PAR strategy requires removing the i-th ambiguity from the current ambiguity set to form a new subset containing n-1 ambiguities, the covariance matrix of the new subset is not directly calculated. Instead of performing a completely new de-correlation process, the following acceleration process is executed: S21. Information reuse through isomorphic transformation.

[0031] Using the parent set decorrelation transformation matrix Z obtained in step S1, an intermediate transformation matrix for the current subset is constructed through a series of low-complexity elementary row and column transformations. Specifically: a) For the matrix Right multiplication of row and column permutation operations The inverse of the matrix is ​​obtained. .

[0032] b) For the matrix The last column undergoes integer Gaussian elimination to generate a specific single-mode transformation matrix. This operation only involves processing a single column, and its computational complexity is much lower than that of reducing correlation for the complete matrix.

[0033] c) Obtain the intermediate transformation matrix Top left block is The matrix It preserves the properties of integer single-modulus matrices.

[0034] S22. Generate a high-quality starting point for iteration.

[0035] Using the constructed intermediate transformation matrix covariance matrix of the current subset Perform the transformation: .

[0036] because Inherited the parent set's decorrelation transformation matrix The "decorrelated" property of the matrix is ​​utilized, and elementary transformations are applied to adjust for the subset structure, thus the transformed matrix... Compared to the original subset covariance matrix Its off-diagonal elements are weaker, and it is closer to a reduced-correlation matrix (i.e., closer to a diagonal matrix). (See the attached diagram provided by the user) This visually demonstrates that after this transformation, the off-diagonal correlation of the matrix is ​​significantly reduced.

[0037] S23. Efficiently reduce correlation.

[0038] by As the initial matrix for iterative optimization, rather than the original matrix. Then, execute a decorrelation algorithm (such as the core iterative process of LAMBDA). Due to the starting point... Having already acquired excellent "quasi-diagonal" properties, the number of iterations required for integer Gaussian transforms (IGTs) and conditional variance reordering (Permutation) in the algorithm will be significantly reduced, thus enabling rapid convergence to obtain the final decorrelation transformation matrix for the current subset. .

[0039] S3. Iterative loop.

[0040] Treating the current subset as the new "parent set", its decorrelation transformation matrix This serves as a new basis for reuse. When the PAR strategy decides to further eliminate ambiguity and generate the next subset, step S2 is repeated. This process continues sequentially until the PAR stopping condition is met (such as the ratio value reaching a certain threshold or the preset subset size being reached).

[0041] Through the above process, for a series of fuzzy subsets with decreasing dimensions generated during the PAR process, except for the first subset (or parent set) which needs to undergo a complete standard downcorrelation, the downcorrelation calculation of each subsequent subset starts from a high-quality approximate solution, avoiding a large number of repetitive iterative calculations starting from zero. Thus, while ensuring the mathematical equivalence of the solution results, a significant improvement in computational efficiency is achieved.

[0042] like Figure 3 As shown, the left subplot illustrates the 14-dimensional subset covariance matrix corresponding to the original 15-dimensional ambiguity variance matrix after removing one ambiguity. (Before transformation). The right-hand sub-figure shows the intermediate matrix obtained after transformation in step S22 of this invention. The comparison shows that... The magnitude of the off-diagonal elements is significantly reduced, and the matrix is ​​closer to the diagonal form, proving that the starting point of the reused information generation has excellent "quasi-decreasing correlation" characteristics, which lays the foundation for subsequent rapid convergence.

[0043] The following specific example illustrates how to utilize the reduced correlation information of the previous ambiguity subset (parent set) to accelerate the processing of the current ambiguity subset. The core of this method lies in adapting the reduced correlation transformation matrix of the parent set to the current subset through a series of low-complexity integer transformations. This provides a highly approximate, high-quality starting point for the reduced correlation calculation of the current subset, avoiding iteration from scratch and significantly improving computational efficiency.

[0044] Suppose that in a certain iteration with partially fixed ambiguities (PAR), standard Lambda decorrelation has been performed on a parent set containing n ambiguities. Let the variance-covariance matrix of the parent set ambiguities be... By reducing correlation, a parent set integer single-mode transformation matrix was obtained. The transformed matrix The matrix is ​​approximately diagonal, meaning the correlation between ambiguity parameters is significantly weakened. Now, according to the PAR strategy (e.g., based on ratio test value growth), the i-th ambiguity needs to be removed from the parent set to form a new subset containing n-1 ambiguities. The covariance matrix corresponding to the subset is... The goal of this embodiment is to quickly obtain the subset. The decorrelation transformation matrix The specific steps are as follows: Figure 4 As shown: Step S101: Construct a permutation matrix and rearrange the ambiguity order. First, construct a permutation matrix. Its function is to swap the row and column corresponding to the i-th ambiguity to be removed with the row and column corresponding to the last (n-th) ambiguity. Applying the inverse of this permutation to the descending correlation matrix Z of the parent set yields the intermediate matrix. Accordingly, the parent set covariance matrix can be represented in blocks after the permutation as follows: (1) in, That is, the (n-1)-dimensional subset covariance matrix that needs to be processed. For the corresponding cross-covariance vector. For the first The variance of the ambiguity.

[0045] Step S102: Perform integer Gaussian elimination on the last column of the intermediate matrix.

[0046] The matrix obtained in step S101 The last column (corresponding to the ambiguities to be removed) is subjected to Integer Gaussian Transform (IGTs). Because... Given an integer single-mode matrix, the greatest common divisor of the elements in its last column is 1. Therefore, a corresponding single-mode transformation matrix P can always be found such that: (2) Matrix B has a specific structure: its last column contains only 0 elements, with the last element being 1. This step involves operations on only a single column, with a computational complexity of O(n log n). This is far lower than the complete decorrelation calculation.

[0047] Step S103: Derive the approximate transformation form of the subset covariance matrix.

[0048] The transformation relation obtained in step S102 is applied to the permuted covariance matrix. Substituting equation (2) into relation (1) yields... (3) Examine the top left corner From the blocks, we can obtain: (4) matrix and Perform the corresponding block division, and set , Then the right side of equation (4) can be expanded as follows: (5) Step S104: Analyze the transformation effect to obtain a high-quality iteration starting point.

[0049] because It is a matrix obtained by fully reducing the correlation of the parent set, and its off-diagonal elements are very weak, i.e., vectors. Approaching the zero vector, and Since it is the minimum conditional variance, it is a relatively small value. Therefore, the last three terms on the right side of equation (5) are... The overall effect of the matrix is ​​limited. Although by The introduced elementary row and column transformations will cause some perturbations, but the matrix Compared to the original Its off-diagonal correlation is greatly weakened, becoming closer to a matrix with reduced correlation. For example... Figure 4 As shown, by reusing information from the parent set, only after... Matrix operations with complexity of O(log n) are then... A high-quality preprocessing matrix was obtained. Using this as the starting point for the next standard reduction iteration can significantly reduce the number of iterations.

[0050] Step S105: Complete the final correlation reduction with a high-quality starting point.

[0051] Will As input, perform a standard LAMBDA decorrelation process once. (Due to the starting point...) Having already undergone a high degree of "quasi-diagonalization", the current reduction correlation will converge rapidly, yielding a new (n-1)-dimensional reduction correlation matrix. ,satisfy: (6) in A matrix with a higher degree of diagonalization. Combining equations (5) and (6), we finally obtain the covariance matrix for the original subset. Complete decorrelation transformation matrix : (7) matrix That is, what is needed, can An effective decorrelated integer single-mode transformation matrix.

[0052] This completes the process for the current subset. The decorrelation calculation is accelerated. Throughout the process, all operations (permutation, integer Gaussian elimination, matrix multiplication) are integer single-mode transformations, fully preserving the mathematical properties required for decorrelation. The advantage of this invention is that, except for the initial standard decorrelation calculation performed on the complete parent set, subsequent processing of each subset starts from a highly optimized starting point by reusing parent set information, thus avoiding a large number of repetitive iterative calculations. In the fixed GNSS ambiguity process, this greatly improves computational efficiency, reduces the correlation between ambiguities, and alleviates latency issues on real-time GNSS application platforms.

[0053] The following describes the GNSS partial ambiguity fixing acceleration system based on isomorphic reuse provided by the present invention. The GNSS partial ambiguity fixing acceleration system based on isomorphic reuse described below can be referred to in correspondence with the GNSS partial ambiguity fixing acceleration method based on isomorphic reuse described above.

[0054] Figure 5 This is a schematic diagram of the structure of a GNSS partial ambiguity fixed acceleration system based on isomorphic reuse provided in an embodiment of the present invention, as shown below. Figure 5 As shown, it includes: a setup module 51, a calculation module 52, an acceleration module 53, and an iteration module 54, wherein: Module 51 is used to establish observation equations using GNSS observations and calculate the floating-point estimate of the corresponding carrier phase ambiguity and the corresponding original covariance matrix using the least squares principle. The calculation module 52 is used to perform the first decorrelation calculation on the original covariance matrix to obtain the parent set decorrelation transformation matrix, so that the matrix obtained by multiplying the parent set decorrelation transformation matrix, the original covariance matrix and the transpose of the parent set decorrelation transformation matrix is ​​close to a diagonal matrix. Acceleration module 53 is used to remove any ambiguity from the current ambiguity set according to the PAR strategy to form an ambiguity subset, and perform a sequential decorrelation acceleration operation on the ambiguity subset to obtain a new decorrelation transformation matrix of the ambiguity subset; The iteration module 54 is used to take the ambiguity subset as a new parent set, and repeatedly execute the sequential decorrelation acceleration operation based on isomorphism using the PAR strategy based on the new decorrelation transformation matrix until the PAR preset stopping condition is met, generating multiple ambiguity subsets with successively decreasing dimensions.

[0055] Figure 6 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 6 As shown, the electronic device may include: a processor 610, a communication interface 620, a memory 630, and a communication bus 640, wherein the processor 610, the communication interface 620, and the memory 630 communicate with each other through the communication bus 640. The processor 610 can call logic instructions in the memory 630 to execute a GNSS partial ambiguity fixed acceleration method based on isomorphism reuse. This method includes: establishing an observation equation using GNSS observations; calculating the floating-point estimate of the corresponding carrier phase ambiguity and the corresponding original covariance matrix using the least squares principle; performing an initial downcorrelation calculation on the original covariance matrix to obtain a parent set downcorrelation transformation matrix, such that the matrix obtained by multiplying the parent set downcorrelation transformation matrix, the original covariance matrix, and the transpose of the parent set downcorrelation transformation matrix is ​​close to a diagonal matrix; removing any ambiguity from the current ambiguity set according to the PAR strategy to form an ambiguity subset; performing a sequential downcorrelation acceleration operation on the ambiguity subset to obtain a new downcorrelation transformation matrix for the ambiguity subset; using the ambiguity subset as a new parent set, repeatedly performing the isomorphism-based sequential downcorrelation acceleration operation based on the new downcorrelation transformation matrix using the PAR strategy until the PAR preset stopping condition is met, generating multiple ambiguity subsets with progressively decreasing dimensions.

[0056] Furthermore, the logical instructions in the aforementioned memory 630 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0057] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the GNSS partial ambiguity fixed acceleration method based on isomorphic reuse provided by the above methods. This method includes: establishing an observation equation through GNSS observations; calculating the floating-point estimate of the corresponding carrier phase ambiguity and the corresponding original covariance matrix using the least squares principle; performing a first decorrelation calculation on the original covariance matrix to obtain the parent set decorrelation transformation matrix, so that... The matrix obtained by multiplying the parent set downcorrelation transformation matrix, the original covariance matrix, and the transpose of the parent set downcorrelation transformation matrix is ​​close to a diagonal matrix. According to the PAR strategy, any ambiguity is removed from the current ambiguity set to form an ambiguity subset. A sequential downcorrelation acceleration operation is performed on the ambiguity subset to obtain a new downcorrelation transformation matrix for the ambiguity subset. Using the ambiguity subset as a new parent set, the sequential downcorrelation acceleration operation based on isomorphism is repeatedly performed using the PAR strategy based on the new downcorrelation transformation matrix until the preset PAR stopping condition is met, generating multiple ambiguity subsets with progressively decreasing dimensions.

[0058] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the GNSS partial ambiguity fixing acceleration method based on isomorphic reuse provided by the methods described above. This method includes: establishing an observation equation using GNSS observations; calculating a floating-point estimate of the corresponding carrier phase ambiguity and the corresponding original covariance matrix using the least squares principle; performing a first downcorrelation calculation on the original covariance matrix to obtain a parent set downcorrelation transformation matrix, such that the parent set downcorrelation transformation matrix and the original covariance matrix... The matrix obtained by multiplying the initial covariance matrix and the transpose of the parent set's downcorrelation transformation matrix is ​​close to a diagonal matrix. According to the PAR strategy, any ambiguity is removed from the current ambiguity set to form an ambiguity subset. A sequential downcorrelation acceleration operation is performed on this subset to obtain a new downcorrelation transformation matrix. Using this ambiguity subset as the new parent set, the PAR strategy is used to repeatedly execute the isomorphism-based sequential downcorrelation acceleration operation based on the new downcorrelation transformation matrix until the preset PAR stopping condition is met, generating multiple ambiguity subsets with progressively decreasing dimensions.

[0059] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0060] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0061] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A GNSS partial ambiguity fixed acceleration method based on isomorphic reuse, characterized in that, include: An observation equation is established using GNSS observations, and the floating-point estimate of the corresponding carrier phase ambiguity and the corresponding original covariance matrix are calculated using the least squares principle. Perform the first decorrelation calculation on the original covariance matrix to obtain the parent set decorrelation transformation matrix, and make the matrix obtained by multiplying the parent set decorrelation transformation matrix, the original covariance matrix and the transpose of the parent set decorrelation transformation matrix close to a diagonal matrix; According to the partial ambiguity fixed PAR strategy, any ambiguity is removed from the current ambiguity set to form a subset of ambiguities. A sequential downcorrelation acceleration operation is performed on the subset of ambiguities to obtain a new downcorrelation transformation matrix of the subset of ambiguities. Using the ambiguity subset as the new parent set, the sequential decorrelation acceleration operation based on isomorphism is repeatedly executed using the PAR strategy based on the new decorrelation transformation matrix until the preset stopping condition of PAR is met, generating multiple ambiguity subsets with successively decreasing dimensions.

2. The GNSS partial ambiguity fixing acceleration method based on isomorphic reuse according to claim 1, characterized in that, Based on a partial ambiguity fixed PAR strategy, any ambiguity is removed from the current ambiguity set to form a subset of ambiguities. A sequential downcorrelation acceleration operation is then performed on this subset to obtain a new downcorrelation transformation matrix, including: Based on the parent set downcorrelation transformation matrix The intermediate transformation matrix is ​​generated through elementary row and column transformations. , The subset covariance matrix corresponding to the subset after removing the i-th ambiguity. Transformed into an intermediate matrix that approximates a diagonal matrix ; With the intermediate matrix Perform a decorrelation calculation as the starting point of the iteration to obtain the covariance matrix of the subset. The new decorrelation transformation matrix .

3. The GNSS partial ambiguity fixing acceleration method based on isomorphic reuse according to claim 2, characterized in that, Based on the parent set downcorrelation transformation matrix The intermediate transformation matrix is ​​generated through elementary row and column transformations. ,include: The parent set decorrelation transformation matrix Multiply by row and column permutation operations The inverse matrix is ​​obtained by finding the matrix. ,in , This represents the transformation matrix that swaps rows i and n. For matrix The last column undergoes Gaussian integer elimination. Let the single-mode transformation matrix be... , so that satisfaction go through The transformation achieves a last column of 0s, with the last element of the last column being 1. Take the transformation matrix The former The block matrix is .

4. The GNSS partial ambiguity fixing acceleration method based on isomorphic reuse according to claim 2, characterized in that, With the intermediate matrix As the starting point for iteration, it includes: Calculate the intermediate matrix The intermediate matrix will be calculated The LAMBDA method is used to fix the ambiguity. Among them, the calculated intermediate matrix Compared to the untransformed subset covariance matrix It exhibits weaker off-diagonal correlation, reducing the number of integer Gaussian transformations and conditional variance reorderings required for subsequent decorrelation iterations.

5. The GNSS partial ambiguity fixing acceleration method based on isomorphic reuse according to claim 3, characterized in that, The single-mode transformation matrix is: A matrix satisfies the condition that its absolute determinant is 1 and all its elements are integers, and is obtained by applying the matrix... The last column is constructed using integer Gaussian elimination, resulting in... The matrix is ​​preserved as an integer monomodal matrix.

6. The GNSS partial ambiguity fixing acceleration method based on isomorphic reuse according to claim 1, characterized in that, Using the ambiguity subset as the new parent set, and based on the new decorrelation transformation matrix, the PAR strategy is used to repeatedly perform the isomorphism-based sequential decorrelation acceleration operation, including: To sequentially process multiple subsets of ambiguity formed by the successive removal of ambiguity; For the (i+1)th subset, its sequential decorrelation acceleration operation reuses the decorrelation transformation matrix information of the i-th subset or the parent set.

7. A GNSS partial ambiguity fixed acceleration system based on isomorphic reuse, characterized in that, include: A module is established to build observation equations from GNSS observations and use the least squares principle to calculate the floating-point estimate of the corresponding carrier phase ambiguity and the corresponding original covariance matrix. The calculation module is used to perform the first decorrelation calculation on the original covariance matrix to obtain the parent set decorrelation transformation matrix, so that the matrix obtained by multiplying the parent set decorrelation transformation matrix, the original covariance matrix and the transpose of the parent set decorrelation transformation matrix is ​​close to a diagonal matrix. The acceleration module is used to remove any ambiguity from the current ambiguity set according to the PAR strategy to form an ambiguity subset, and to perform a sequential decorrelation acceleration operation on the ambiguity subset to obtain a new decorrelation transformation matrix of the ambiguity subset. The iterative module is used to take the ambiguity subset as a new parent set, and repeatedly execute the isomorphism-based sequential decorrelation acceleration operation based on the new decorrelation transformation matrix using the PAR strategy until the preset stopping condition of PAR is met, generating multiple ambiguity subsets with successively decreasing dimensions.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the GNSS partial ambiguity fixing acceleration method based on isomorphic reuse as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the GNSS partial ambiguity fixing acceleration method based on isomorphic reuse as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the GNSS partial ambiguity fixing acceleration method based on isomorphic reuse as described in any one of claims 1 to 6.

Images