BDS-3 multi-frequency cycle slip detection and repair method and system based on GCV regularization

By using the multi-frequency round-hop detection and repair method of GCV regularization assisted LAMBDA algorithm in GNSS system, the problem of perimeter jump detection in multi-frequency signal environment is solved, and the problem of perimeter jump detection being susceptible to pseudorange noise and being insensitive to small round-hop is improved. Detection sensitivity and repair accuracy provide stronger positioning and solution reliability.

CN118938265APending Publication Date: 2024-11-12ANHUI UNIV OF SCI & TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411118847.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-15
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

The existing GNSS cycle jump detection technology is susceptible to pseudorange noise and is not sensitive enough to small cycle jumps. It has serious pathological problems, especially in multi-frequency signal environments, which affects the accuracy of positioning solutions.

Method used

The BDS-3 multi-frequency round-hop detection and repair method based on GCV regularization is adopted to construct the optimal multi-frequency combined round-hop detection by combining multi-frequency signal observation data, and the weekly jump repair is carried out using GCV regularization auxiliary LAMBDA algorithm. Reduce the impact of pathological equations, improve detection sensitivity and repair accuracy.

Benefits of technology

It significantly improves the sensitivity of the round-hop detection and the reliability of repair, reduces repair errors, provides more accurate observation data, and provides reliable data support for subsequent baseline solution and positioning solution processing.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118938265A_ABST
    Figure CN118938265A_ABST
Patent Text Reader

Abstract

The invention discloses a BDS-3 multi-frequency cycle slip detection and restoration method and system based on GCV regularization. The method comprises the steps that multi-frequency pseudo-range observation data and carrier phase observation data of BDS-3 are acquired; the pseudo-range observation data and the carrier phase observation data are combined, a multi-frequency combined cycle slip detection quantity is constructed, and an optimal multi-frequency combined coefficient is optimized; performing difference on the optimal multi-frequency combined cycle slip detection quantity between adjacent epochs to construct a cycle slip detection equation, and setting a cycle slip judgment threshold to perform cycle slip detection per epoch; and repairing the detected cycle slip by using a GCV regularization assisted LAMBDA algorithm. According to the method, the selection thought of the multi-frequency optimal combination coefficient is expanded, regularization processing instead of simple subjective selection is directly carried out on ill-conditioned problems, selectable combinations are obviously more than those in the prior art, the cycle slip repairing speed is high, and therefore the reliability of multi-frequency cycle slip detection and repairing is guaranteed.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of GNSS high-precision positioning, and in particular to a BDS-3 multi-frequency cycle slip detection and repair method and system based on GCV regularization. Background Art

[0002] Carrier phase observation is the key to GNSS high-precision positioning and applications. Carrier phase observation may experience cycle slips due to the receiver's own reasons or the influence of physical shielding on GNSS signals. Therefore, when carrier phase observations are involved in GNSS high-precision positioning and related applications, effective cycle slip detection and repair are required.

[0003] Existing research on cycle slip detection and repair mainly includes three aspects: "time series analysis", "observation residuals" and "observation linear combination". Among them, the time series analysis method has poor ability to detect small cycle slips; although the observation residual method is more sensitive to small cycle slips, the actual operation is not simple enough and the parameter selection is more complicated; therefore, the cycle slip detection method based on the linear combination of observations has received attention.

[0004] However, the current linear combination method of observations is mainly aimed at GNSS single-frequency / dual-frequency observations, which requires less data, but is susceptible to pseudo-range noise or insensitive to small cycle slips. Especially with the gradual application of multi-frequency signals in the GNSS field in recent years, the serious pathological problem of multi-frequency cycle slip repair needs to be solved in a better way. Summary of the invention

[0005] In view of the problems existing in the above-mentioned prior art, the present invention provides a BDS-3 multi-frequency cycle slip detection and repair method and system based on GCV regularization, aiming to solve the problem that the existing cycle slip detection technology is easily affected by pseudorange noise or is not sensitive enough to small cycle slips, thereby improving the sensitivity of cycle slip detection; the cycle slip repair algorithm of the LAMBDA algorithm assisted by GCV regularization effectively eliminates the susceptibility of the repair to the influence of the ill-conditioned equation of the normal equation, and the cycle slip repair error is significantly smaller, thereby ensuring the stronger reliability and accuracy of the cycle slip detection and repair technology, and can provide accurate observation data for subsequent baseline solution and positioning solution processing.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization, comprising:

[0008] Obtain multi-frequency signal observation data of BDS-3, including pseudorange observation data and carrier phase observation data;

[0009] Combining the pseudorange observation data and the carrier phase observation data to construct a multi-frequency combination cycle slip detection amount, and selecting the optimal multi-frequency combination coefficient;

[0010] The cycle slip detection equation is constructed by differentially analyzing the optimal multi-frequency combination cycle slip detection quantities between adjacent epochs, and the cycle slip judgment threshold is set to perform cycle slip detection epoch by epoch.

[0011] The GCV regularization-assisted LAMBDA algorithm is used to repair the detected cycle slips.

[0012] Further, the multi-frequency signal observation data includes observation data of four-frequency signals and five-frequency signals; the multi-frequency combination cycle slip detection amount is composed of a phase geometry-free combination and a pseudo-range phase geometry-free ionosphere-free combination;

[0013] When constructing the multi-frequency combination cycle slip detection quantity, one four-frequency pseudo-range phase geometry-free ionosphere combination and three four-frequency phase geometry-free combinations are selected to construct the four-frequency combination cycle slip detection quantity;

[0014] One five-frequency pseudorange phase geometry-free ionosphere-free combination and four five-frequency phase geometry-free combinations are selected to construct the five-frequency combination cycle slip detection quantity.

[0015] Further, the step of selecting the optimal multi-frequency combination coefficient includes:

[0016] Set the search interval of multi-frequency combination coefficients to [-10,10];

[0017] For the phase geometry-free combination, the optimal condition is: considering the constraints that the combination noise and the ionospheric delay amplification factor are both relatively small, the combination with high data richness among the multiple phase geometry-free combinations is preferred;

[0018] For the pseudorange phase geometry-free and ionosphere-free combination, the optimal conditions are: the optimal combination is selected based on the long wavelength as the primary condition, and the relatively small combination noise and ionospheric delay amplification factor are considered secondly;

[0019] Based on the preferred conditions, the optimal multi-frequency combination cycle slip detection amount is selected for the four-frequency signal and the five-frequency signal in the search interval.

[0020] Furthermore, after the cycle slip is detected, it also includes establishing a multi-frequency cycle slip estimation equation:

[0021] A k ΔN k =L k

[0022] Where k represents the number of frequencies; A k =[X GF11 ,…,X GFnk ,X GFIF] T represents the multi-frequency combination coefficient matrix; X GFnk =p mnk λ n , p mnk represents the combination coefficient of the nth frequency point of the mth single phase geometry-free combination of the k-frequency combination, λ n Indicates the wavelength of the nth frequency point; N GFIF represents the combination coefficient of pseudorange phase without geometry and ionosphere; L k is the k-frequency combination cycle slip detection matrix; ΔN k represents the k-frequency cycle slip;

[0023] Based on the multi-frequency cycle slip estimation equation, the GCV regularization-assisted LAMBDA algorithm is used to repair the detected cycle slips.

[0024] Furthermore, the method of using the GCV regularization-assisted LAMBDA algorithm to repair the detected cycle slips includes:

[0025] Obtain the minimum value of the GCV curve to obtain the regularization parameter;

[0026] Select the unit matrix as the regularization matrix;

[0027] Constructing a GCV correction term based on the regularization parameter and the regularization matrix;

[0028] Using the optimal multi-frequency combination cycle slip detection amount, performing least square correction through the GCV correction term, obtaining an estimated value of the cycle slip between epochs, and obtaining a floating-point solution for the cycle slip;

[0029] Modifying the variance-covariance matrix of the LAMBDA algorithm based on the GCV correction term;

[0030] The floating point solution of the cycle slip and the corrected variance-covariance matrix are input into the LAMBDA algorithm, and the floating point solution of the cycle slip is fixed by using the LAMBDA algorithm to obtain the cycle slip repair parameter solution of the epoch.

[0031] Further, the minimum value of the GCV curve is expressed as:

[0032]

[0033] in, k represents the number of frequencies, α represents the regularization parameter, tr() represents the trace of the matrix, T represents the transpose of the matrix, and I represents the identity matrix;

[0034] The GCV correction term is: α·R, where α represents a regularization parameter and R is a unit matrix selected as a regularization matrix.

[0035] Further, the inter-epoch cycle slip estimation value is expressed as:

[0036]

[0037] in, For the power matrix, Q k is the observation matrix L k The variance-covariance matrix of , R is the regularization matrix;

[0038] Using Cholesky decomposition, the weight matrix P k Unitize to get the new combination coefficient matrix A GCV =G*A k , G is an upper triangular matrix;

[0039] Then the variance-covariance matrix of the LAMBDA algorithm modified based on the GCV correction term is:

[0040]

[0041] In a second aspect, the present invention provides a BDS-3 multi-frequency cycle slip detection and repair system based on GCV regularization, comprising:

[0042] Data acquisition module, used to obtain multi-frequency signal observation data of BDS-3, including pseudo-range observation data and carrier phase observation data;

[0043] A combination optimization module is used to combine the pseudorange observation data and the carrier phase observation data to construct a multi-frequency combination cycle slip detection amount and select the optimal multi-frequency combination coefficient;

[0044] A cycle slip detection module is used to construct a cycle slip detection equation by differentially determining the optimal multi-frequency combination cycle slip detection quantity between adjacent epochs, and to set a cycle slip judgment threshold to perform cycle slip detection epoch by epoch;

[0045] The cycle slip repair module is used to repair the detected cycle slips using the GCV regularization-assisted LAMBDA algorithm.

[0046] In a third aspect, the present invention provides an electronic device, comprising:

[0047] A memory for storing executable instructions;

[0048] The processor is used to implement a BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization as described above when running the executable instructions stored in the memory.

[0049] In a fourth aspect, the present invention provides a computer-readable storage medium storing executable instructions, characterized in that when the executable instructions are executed by a processor, a BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization as described above is implemented.

[0050] The BDS-3 multi-frequency cycle slip detection and repair method and system based on GCV regularization of the present invention have the following beneficial effects:

[0051] 1. The present invention uses multi-frequency (quad / quintuple frequency) observation data, based on the problem of pseudorange noise and small cycle slips being not sensitive enough, to construct the optimal multi-frequency combination cycle slip detection of linearly independent GFIF combination and GF combination, giving priority to the wavelength of the combined observation value, comprehensively considering the three factors of the combined observation value noise, ionospheric delay amplification factor and insensitive cycle slip number, among which the wavelength is given priority, and the ability to resist pseudorange noise changes is better, and the cycle slip detection performance and reliability in real scenarios are significantly enhanced. This method expands the selection idea of ​​multi-frequency optimal combination coefficients, and the selected multi-frequency combination observation value can form a combination observation value with a longer wavelength, a smaller ionospheric delay and a smaller noise, thereby improving the performance of multi-frequency cycle slip detection and ensuring detection sensitivity.

[0052] 2. The present invention aims at the fact that the repair of multi-frequency cycle slips is easily affected by the ill-conditioned equation of the normal equation, and proposes a repair method of the GCV regularization-assisted LAMBDA algorithm. The floating-point solution of the GCV algorithm is first used to replace the floating-point solution of the least squares method, and then the variance-covariance matrix of the LAMBDA algorithm is corrected using the GCV correction term. There is no restriction on the number of conditions, the selection range is large, a large number of multi-frequency combinations are carefully discussed and fully utilized, and the interpretability of the optimal multi-frequency combination cycle slip detection quantity is strong; the regularized solution replaces the floating-point solution of the least squares method, the repair speed is fast, and the reliability of multi-frequency cycle slip detection and repair is guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a flow chart of a BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization of the present invention;

[0054] Figure 2 It is a flowchart of a multi-frequency cycle slip detection method provided by an embodiment of the present invention;

[0055] Figure 3 This is a result diagram of four-frequency cycle slip detection using a preferred optimal four-frequency combination cycle slip detection amount according to an embodiment of the present invention;

[0056] Among them, (a) is the detection result of the cycle slip [1, 0, 0, 0], (b) is the detection result of the cycle slip [1, 1, 0, 0], (c) is the detection result of the cycle slip [1, 1, 1, 0], (d) is the detection result of the cycle slip [1, 1, 1, 1], (e) is the detection result of random small cycle slips from (0, 0, 0, 0) to (6, 6, 6, 6);

[0057] Figure 4 This is a result diagram of five-frequency cycle slip detection using a preferred optimal five-frequency combination cycle slip detection amount according to an embodiment of the present invention;

[0058] Among them, (a) is the cycle slip detection result diagram of [1,0,0,0,0], (b) is the cycle slip detection result diagram of [1,1,0,0,0], (c) is the cycle slip detection result diagram of [1,1,1,0,0], (d) is the cycle slip detection result diagram of [1,1,1,1,0], (d) is the cycle slip detection result diagram of [1,1,1,1,1], (f) is the detection result diagram of random small cycle slips from (0,0,0,0,0) to (4,4,4,4,4);

[0059] Figure 5 It is a flow chart of using GCV regularization-assisted LAMBDA algorithm to repair multi-frequency cycle slips in one embodiment of the present invention;

[0060] Figure 6 It is a structural block diagram of a BDS-3 multi-frequency cycle slip detection and repair system based on GCV regularization of the present invention. DETAILED DESCRIPTION

[0061] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0062] To facilitate understanding of this embodiment, a BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization disclosed in an embodiment of the present invention is first introduced in detail.

[0063] The embodiment of the present invention provides a BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization, such as Figure 1 As shown, including:

[0064] Obtain multi-frequency signal observation data of BDS-3, including pseudorange observation data and carrier phase observation data;

[0065] Combining the pseudorange observation data and the carrier phase observation data to construct a multi-frequency combination cycle slip detection amount, and selecting the optimal multi-frequency combination coefficient;

[0066] The cycle slip detection equation is constructed by differentially analyzing the optimal multi-frequency combination cycle slip detection quantities between adjacent epochs, and the cycle slip judgment threshold is set to perform cycle slip detection epoch by epoch.

[0067] The GCV regularization-assisted LAMBDA algorithm is used to repair the detected cycle slips.

[0068] In view of the problem of BDS-3 multi-frequency cycle slip detection and repair, the present invention combines a set of geometry-free ionosphere-free phase GFIF combinations and multiple sets of linearly independent geometry-free phase GF combinations, gives priority to the wavelength of the combined observation value, comprehensively considers the three factors of the combined observation value noise, ionospheric delay amplification factor and insensitive cycle slip number, selects the optimal combination coefficient, and performs cycle slip detection based on the optimal combination coefficient. In order to reduce the interference of the ill-conditioned problem of the normal equation, it is proposed to use the GCV regularization-assisted LAMBDA algorithm for cycle slip repair, which has significantly smaller cycle slip repair errors and stronger reliability and accuracy.

[0069] The following is a detailed description of the cycle slip detection and repair process:

[0070] 1. Obtaining data

[0071] The BDS-3 system, namely the BeiDou-3 global satellite navigation system, can provide public signals of five frequencies, namely B1C (1575.42MHz), B1I (1561.098MHz), B3I (1268.52MHz), B2b (1207.14MHz) and B2a (1176.45MHz). Therefore, in this embodiment, the observation data of the four-frequency signal and the five-frequency signal are selected as the observation data for cycle slip detection and repair. It should be understood that the multi-frequency combined observation value can form a combined observation value with a longer wavelength, a smaller ionospheric delay and less noise, thereby effectively improving the performance of cycle slip detection.

[0072] Among them, the four frequencies of the four-frequency selected in this embodiment are generated by removing the B2b frequency from the five frequencies of the selected five-frequency. Taking the phase-free geometry combination [-1,1,0,0] as an example, when the five-frequency GF combination [-1,1,0,0,0] removes the B2b frequency, a four-frequency GF combination [-1,1,0,0] is formed. Therefore, in a broad sense, the five-frequency cycle slip detection and repair and the four-frequency cycle slip detection and repair are inclusive. Compared with the four-frequency signal, the five-frequency signal searches for the optimal coefficient in the same range, but the search process of the five-frequency signal nests the cycle of the fifth frequency. Relatively speaking, the five-frequency combination is easier to search for a better combination coefficient within a lower coefficient value range. However, the fifth frequency of the five-frequency signal also brings in corresponding noise, so that the five-frequency signal is not necessarily better than the four-frequency signal. Therefore, the present invention uses four-frequency and five-frequency signals to explore cycle slip detection and repair.

[0073] First, the pseudorange observation data P i It can be expressed as:

[0074]

[0075] Carrier phase observation data It can be expressed as:

[0076]

[0077] Where i represents the frequency, ρ is the satellite-to-ground distance, and dt s is the satellite clock error, dt r is the receiver clock error, d trp is the tropospheric delay, I1 represents the ionospheric delay, f i represents the frequency of the i-th frequency point, I1 is the ionospheric delay error corresponding to the frequency f1, and λ i is the wavelength of the ith frequency point, N is the integer ambiguity, ε Pi and are the observation noise of pseudorange and carrier phase respectively.

[0078] After obtaining the observation data, a multi-frequency combination cycle slip detection amount is constructed. In this embodiment, a geometry-free carrier phase (hereinafter referred to as GF in this specification) and a pseudorange phase geometry-free and ionosphere-free combination (GFIF) are combined. The GF combination is used to reduce the influence of pseudorange noise, and the GFIF combination is used to solve the problem of insufficient sensitivity to small cycle slips, so as to effectively improve the reliability of cycle slip detection.

[0079] When constructing the four-frequency combination cycle slip detection quantity, one four-frequency pseudorange phase geometry-free ionosphere combination and three four-frequency phase geometry-free combinations are selected;

[0080] When constructing the five-frequency combination cycle slip detection quantity, one five-frequency pseudorange phase geometry-free ionosphere-free combination and four five-frequency phase geometry-free combinations are selected.

[0081] (II) Cycle slip detection

[0082] See also Figure 2 , the specific steps of the multi-frequency cycle slip detection method in this embodiment are given, and are explained step by step below.

[0083] 1. Determine the optimal multi-frequency combination cycle slip detection quantity

[0084] (1) First, obtain the optimal GF combination. Combining the above pseudorange observation data and carrier phase observation data, a single multi-frequency GF combination observation value can be obtained: It is expressed as:

[0085]

[0086] In order to select the GF combination with low observation noise and weak ionospheric delay, the constraints for constructing the GF combination observation value are:

[0087]

[0088] In the formula, k represents the number of frequencies, which is 4 or 5 in this embodiment. The four frequencies selected in this embodiment are B1C, B1I, B3I and B2a, and the five frequencies are B1C, B1I, B3I, B2b and B2a.

[0089] is the carrier phase observation value of the frequency corresponding to the quad-frequency / quintuple-frequency (in cycles); η GF(k) The ionospheric delay amplification factor of the four-frequency / five-frequency frequency is δI1=I1 / λ1, which is the ionospheric delay of the first frequency (in weeks); ε GF(k) is the noise of the four-frequency / five-frequency GF combination observation, p1, p2, …, p k is the GF combination coefficient.

[0090] The phase GF combination does not use the pseudorange observations with lower accuracy but is based on the carrier phase observations with higher accuracy. Therefore, the use of the GF combination can greatly improve the accuracy of cycle slip detection.

[0091] In this embodiment, the accuracy of the pseudorange observation value is 0.3 meters, and the accuracy of the carrier phase observation value is 0.01 cycles. According to the above GF combination observation value constraint equation, a four-frequency / five-frequency GF combination with an observation value coefficient in the range of [-10,10] can be obtained.

[0092] However, within the range of 80 cycles, a single GF phase combination has many insensitive cycle slips: a single four-frequency GF combination has hundreds of insensitive cycle slips, and a single five-frequency GF combination has thousands of insensitive cycle slips; the maximum number of insensitive cycle slips of a single GF combination reaches 1088 and 11978 respectively, and the minimum value is 164 and 1990.

[0093] Therefore, a single multi-frequency GF combination cannot be directly used for cycle slip detection. To overcome the insensitive cycle slip problem of a single GF combination, multiple GF combinations are combined to perform cycle slip detection simultaneously to reduce the number of insensitive cycles. The following table 1 provides a representative GF combination, where η (4) represents the ionospheric delay coefficient, Indicates the noise in cycles.

[0094] Table 1 Multiple geometry-free phase GF combinations

[0095]

[0096] After obtaining Table 1, the optimal GF combination can be selected. According to the expression of the combined observation value, the smaller the combined noise and ionospheric delay amplification factor of the combined observation value, the more conducive it is to improving the accuracy and reliability of cycle slip detection.

[0097] On the one hand, the GF combination has a relatively high accuracy, and it is necessary to use as many GF combinations as possible for multi-frequency cycle slip detection; on the other hand, in order to reduce the impact of insensitive cycle slips, the selected GF combinations need to be linearly independent. From the knowledge of linear algebra, we know that n+1 n-dimensional vectors are always linearly related, so at most n GF combinations are combined. Considering that the multi-frequency combination cycle slip detection also includes the GFIF combination, n-1 GF combinations are selected (n is the number of frequencies). At the same time, it is also necessary to pay attention to the data richness of the selected combination, that is, to calculate the total number of non-zero coefficients of each frequency signal in the combination. The total number of non-zero coefficients should be as large as possible to make full use of the data of each frequency signal and reduce the possibility of collinearity between multiple GF combinations. As indicated in bold in Table 1, for four-frequency cycle slip detection, when the maximum noise is close, this embodiment selects the fourth group of four-frequency combinations with better ionospheric delay coefficients.

[0098] [-1,1,0,0][-1,1,1,-1][-1,2,-1,0]. Similarly, the five-frequency cycle slip detection selects [-1,1,0,0,0][0,0,-1,1,0][0,0,-1,0,1][0,-1,1,1,-1] which has less noise but better ionospheric delay coefficient. Compared with the selected optimal combination, the data richness of the first group of coefficients is obviously worse, which is consistent with the set optimal combination coefficient selection standard.

[0099] After obtaining the optimal three linearly independent GF combinations, the GF combination observations of adjacent epochs are differentiated to construct the cycle slip detection equation, and the cycle slip judgment threshold is set to perform cycle slip detection epoch by epoch, specifically:

[0100] By taking the difference between the single GF combination observations between epochs, we can get the change of the combined observations between the four-frequency / five-frequency epochs:

[0101]

[0102] in It is the GF combination after the difference of adjacent epochs.

[0103] The cycle skip detection measurement can be obtained as follows:

[0104]

[0105] Assume that the accuracy of each four-frequency / five-frequency carrier phase observation value is the same, that is, the standard deviation Then the standard deviation of the GF combined observations is It can be expressed as:

[0106]

[0107] (2) Secondly, obtain the optimal GFIF combination.

[0108] From the above analysis, it can be seen that k-frequency observations can only construct k-1 linearly independent GF combination observations at most, and it is impossible to solve k-frequency cycle slips. It is necessary to combine 1 GFIF combination to construct k linearly independent combination observations, further increasing the ability to detect cycle slips and deal with insensitive cycle slips.

[0109] Combining the above pseudorange observation data with the carrier phase observation data can obtain a single multi-frequency pseudorange / phase GFIF combined observation value It is expressed as:

[0110]

[0111]

[0112] In order to eliminate the geometric distance term and improve the sensitivity of cycle slip detection, the constraints for constructing the GF combined observation value are:

[0113] Where q1, q2, …, q k Represents the GFIF combined carrier phase combined observation coefficient, a1, a2, ..., a k represents the coefficient of GFIF combined pseudorange combined observation value, P i is the pseudo-range observation data of the ith frequency point, is the carrier phase observation data of the i-th frequency point, λ GFIF(k) is the combined wavelength of the GFIF combination, ε GFIF(k) is the noise of the four-frequency / five-frequency GFIF combination observation.

[0114] As shown in Table 2, high-quality multi-frequency pseudorange / phase GFIF combinations in the range of [-10,10] are given.

[0115] Table 2 Carrier phase combination coefficients of multi-frequency pseudorange phase geometry-free ionosphere-free GFIF combination

[0116]

[0117] Most of the existing multi-frequency related studies only pursue minimum noise, and simply select the five-frequency [1,-1,0,0,0] combination, without considering or trying GFIF combinations with larger combined wavelengths. Long wavelengths have strong penetration capabilities and better ability to resist pseudorange noise changes, and the cycle slip detection performance and reliability in real scenarios are significantly enhanced. Therefore, in this embodiment, for the GFIF combination, the optimal combination selection is performed with wavelength as the core. It can be seen from the data on the right side of Table 2 that the maximum combination of the five-frequency GFIF combination has no significant increase in noise relative to the five-frequency combination with the lowest noise, but the wavelength is expanded by 7 times. Therefore, [-2,2,1,-2,1] is selected as the five-frequency optimal GFIF combination. Considering that the five-frequency combination has higher data richness, it can search for long wavelength and low noise combination coefficients, and the search effect of the four-frequency is limited. Therefore, for the four-frequency GFIF combination, the longest wavelength combination [-1,2,-4,3] is still retained as the four-frequency optimal GFIF combination corresponding to the five-frequency optimal GFIF combination.

[0118] Then, similar to the GF combination, the inter-epoch difference can be obtained as follows: GFIF(k) :

[0119]

[0120] in It is the GFIF combination after the difference of adjacent epochs.

[0121] The standard deviation of a single pseudorange / phase GFIF combination observation can be expressed as:

[0122]

[0123] 2. Cycle slip detection at each epoch

[0124] The combined observations are differentiated, and the 99.9% confidence level under the normal distribution assumption is set to set the cycle slip judgment threshold as four times the standard deviation of the combined coefficient. If the single-difference combined observations do not reach the cycle slip detection threshold, cycle slip detection is performed epoch by epoch until the combined observations of a certain epoch are greater than the cycle slip detection threshold, and it is considered that a cycle slip has occurred in that epoch. Continue to judge the next epoch until the last epoch, and complete the cycle slip detection.

[0125] For the GF combination, the conditions for detecting a cycle slip are:

[0126]

[0127] For the GFIF combination, the conditions for detecting a cycle slip are:

[0128]

[0129] In order to verify the effectiveness of the disclosed cycle slip detection and repair method, the 1-second sampling rate data measured at the 7200 epoch of the SGOC station was selected for experimental verification on the C24 satellite, with 0:0:00 in 2023 as the starting epoch. Multiple special cycle slips (small cycle slips of 1 to 2 cycles occurring on a single frequency, or small cycle slips of the same jump occurring at different frequencies) were added to the four-frequency / five-frequency carrier phase observations, and a qualitative analysis was performed on the detection quantities of each combination of the four-frequency / five-frequency; general cycle slips (cycle slips of 1 to 10 cycles that can be detected in multiple combinations) were added to conduct a quantitative analysis of cycle slip detection. The verification results are as follows: Figure 3 and Figure 4 As shown in the figure, it can be seen that only one combination is needed to detect the cycle slip to complete the entire cycle slip detection process. The method of this embodiment can significantly reduce the number of insensitive cycle slips, greatly reduce the detection blind area, and effectively improve the success rate of cycle slip detection.

[0130] (III) Cycle slip repair

[0131] After the cycle slip is detected, the multi-frequency cycle slip estimation equation is established:

[0132] A k ΔN k =L k

[0133] Where k represents the number of frequencies; A k =[X GF11 ,…,X GFnk ,X GFIF ] T represents the multi-frequency combination coefficient matrix; X GFnk =p mnk λ n , p mnk represents the combination coefficient of the nth frequency point of the mth single phase geometry-free combination of the k-frequency combination, λn Indicates the wavelength of the nth frequency point; X GFIF represents the combination coefficient of pseudorange phase without geometry and ionosphere; L k is the k-frequency combination cycle slip detection matrix; ΔN k represents the k-frequency cycle slip;

[0134] Then, based on the multi-frequency cycle slip estimation equation, the floating point solution of the cycle slip is estimated, and further processed to obtain the integer solution of the cycle slip.

[0135] In the repair of quad-frequency / quintuple-frequency cycle slips, serious ill-conditioned problems must be considered, and the condition number is an important indicator for evaluating the ill-conditioned nature of the matrix. Usually, the condition number is used to measure the sensitivity of the matrix to small changes in the input data, thereby evaluating the relative stability of the solution.

[0136] It is generally believed that when the condition number is greater than 100, there is a moderate or strong pathology, and when it is greater than 1000, there is a serious pathology. However, the condition numbers of the four-band combinations are all above 10,000, and the condition numbers of the five-band combinations are mostly above 10,000. At the same time, the number of five-band combinations with condition numbers greater than 1000 reached 84.2%, and the number greater than 10,000 reached 68.4%, indicating a very serious pathology problem.

[0137] In the existing multi-frequency cycle slip detection and repair research, the combination with the smallest condition number is tended to be selected as the optimal combination. Although the pathological problem is avoided by actively selecting coefficients with smaller condition numbers, it also results in a waste of many combination coefficients, thus losing the possibility of selecting a better coefficient combination. Especially in five-frequency cycle slip detection and repair, simply selecting five-frequency data with smaller condition numbers will inevitably ignore many combinations with excellent performance.

[0138] This embodiment takes this problem into consideration. Figure 5 , the generalized cross validation (GCV) algorithm is used to assist the LAMBDA algorithm to repair the detected cycle slips, including:

[0139] The regularization parameter is obtained by finding the minimum value of the GCV curve, that is, solving the problem of minimum post-test unit weighted mean error. The post-test unit weighted mean error represents the average deviation between each observation and the true value under the unit weight condition.

[0140]

[0141] in, k represents the number of frequencies; α represents the regularization parameter, which is a floating point number and is calculated once per epoch; tr() represents the trace of the matrix, T represents the transpose of the matrix, and I represents the identity matrix;

[0142] The regularization parameter is determined based on whether the GCV curve has a minimum value. If the GCV curve has a minimum value, the corresponding horizontal axis value is the required regularization parameter, so regularization is performed. Each epoch has a corresponding specific regularization parameter. This method does not involve derivation, and the amount of calculation is small when seeking the optimal parameter.

[0143] Select the unit matrix as the regularization matrix;

[0144] A GCV correction term is constructed based on the regularization parameter and the regularization matrix: α·R, where α represents the regularization parameter and R is the unit matrix selected as the regularization matrix.

[0145] In order to speed up the solution of the cycle slip repair algorithm, the floating-point solution obtained based on the GCV algorithm in this embodiment replaces the floating-point solution of the least squares method. That is, the optimal multi-frequency combination cycle slip detection amount is used, and the cycle slip estimation value between epochs is represented by the GCV correction term to obtain the floating-point solution of the cycle slip:

[0146]

[0147] in, For the power matrix, Q k is the observation matrix L k The variance-covariance matrix of , R is the regularization matrix;

[0148] Using Cholesky decomposition, the weight matrix P k Unitization, Cholesky decomposition is the decomposition of a symmetric positive definite matrix represented as the product of a lower triangular matrix L and its transpose, that is, P k =G′*G, G is an upper triangular matrix;

[0149] Get the new combination coefficient matrix A GCV =G*A k ,

[0150] Then, the variance-covariance matrix of the LAMBDA algorithm is modified based on the GCV correction term:

[0151]

[0152] The floating point solution of the cycle slip and the corrected variance-covariance matrix are input into the LAMBDA algorithm, and the floating point solution of the cycle slip is fixed by the LAMBDA algorithm to obtain the cycle slip repair parameter solution of the epoch. Continue to judge the next epoch, return to the step of solving the regularization parameter, and process each epoch until the last epoch is processed to complete the cycle slip repair.

[0153] The essence of GCV is to obtain an error correction term to correct the variance-covariance matrix (Q matrix) in the least squares and LAMBDA algorithms. After correcting the Q matrix, the LAMBDA algorithm performs a spatial search to find the optimal parameter solution.

[0154] In order to verify the superiority of the cycle slip detection and repair method of this embodiment, Algorithm A, Algorithm B and Algorithm C are introduced as control experiments, and the cycle slip repair success rate, Ratio value and average repair error are selected as evaluation indicators of the cycle slip repair algorithm. The results are shown in Tables 3 and 4.

[0155] Algorithm A: LS+Rounding cycle slip repair, the floating point solution is obtained by the least squares algorithm, the floating point solution given by the least squares is directly rounded, and the rounded value is regarded as the cycle slip estimate;

[0156] Algorithm B: GCV+Rounding cycle slip repair. The floating point solution is obtained by the GCV algorithm. The floating point solution given by the GCV algorithm is directly rounded, and the rounded value is regarded as the cycle slip estimate.

[0157] Algorithm C: LS+LAMBDA cycle slip repair, the floating point solution is obtained by the least squares algorithm, and the floating point solution is fixed by the LAMBDA algorithm. The output value of the LAMBDA algorithm is regarded as the cycle slip estimation value;

[0158] Algorithm D: GCV+LAMBDA cycle slip repair. The floating point solution is obtained by the GCV algorithm, and the LAMBDA algorithm is corrected by GCV. The output value of the corrected LAMBDA algorithm is regarded as the cycle slip estimation value.

[0159] Table 3. Success rate of four-frequency cycle slip repair

[0160]

[0161] Table 4 Five-frequency cycle slip repair success rate

[0162]

[0163] According to Table 3 and Table 4, when the optimal combination coefficient of four-frequency / five-frequency selected in this embodiment is used to repair the cycle slip, the success rate of the cycle slip repair is 99.27% ​​and 99.94% respectively when the pathological problem is very serious. Compared with the least squares rounding method, the success rate of the four-frequency / five-frequency repair is increased by 95.15% and 91.41% respectively. The experiment shows that the Ratio value of the proposed algorithm is relatively higher, and the cycle slip repair error is significantly smaller, so it has stronger reliability and accuracy.

[0164] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0165] Based on the same inventive concept, the embodiment of the present application also provides a system for implementing the BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization. The implementation scheme for solving the problem provided by the system is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiment of the BDS-3 multi-frequency cycle slip detection and repair system based on GCV regularization provided below can be referred to the limitations of the BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization above, and will not be repeated here.

[0166] like Figure 6 As shown, the present invention also provides a BDS-3 multi-frequency cycle slip detection and repair system based on GCV regularization, comprising:

[0167] Data acquisition module, used to obtain multi-frequency signal observation data of BDS-3, including pseudo-range observation data and carrier phase observation data;

[0168] A combination optimization module is used to combine the pseudorange observation data and the carrier phase observation data to construct a multi-frequency combination cycle slip detection amount and select the optimal multi-frequency combination coefficient;

[0169] A cycle slip detection module is used to construct a cycle slip detection equation by differentially determining the optimal multi-frequency combination cycle slip detection quantity between adjacent epochs, and to set a cycle slip judgment threshold to perform cycle slip detection epoch by epoch;

[0170] The cycle slip repair module is used to repair the detected cycle slips using the GCV regularization-assisted LAMBDA algorithm.

[0171] It should be noted that: the BDS-3 multi-frequency cycle slip detection and repair system based on GCV regularization provided in this embodiment only uses the division of the above-mentioned functional modules as an example when performing cycle slip detection and repair. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the system can be divided into different functional modules. Each functional module can be composed of a single execution unit, or it can be integrated into a functional module by two or more execution units to realize all the functions of the functional module.

[0172] Those skilled in the art will appreciate that the above modules may be implemented in whole or in part by software, hardware, or a combination thereof. The above modules may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a memory in a computer device in the form of software, so that the processor may call and execute operations corresponding to the above modules.

[0173] An embodiment of the present invention further provides an electronic device, including:

[0174] A memory for storing executable instructions;

[0175] The processor is used to implement a BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization as described above when running the executable instructions stored in the memory. The processor can be a central processing unit, other general-purpose processors, digital signal processors, application-specific integrated circuits, field programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., or a combination of the above chips.

[0176] The specific details of the cycle slip detection and repair performed by the above-mentioned computer device can be understood by referring to the relevant description and effects corresponding to the above-mentioned method, and will not be repeated here.

[0177] Another embodiment of the present invention further provides a computer-readable storage medium storing executable instructions, characterized in that when the executable instructions are executed by a processor, a BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization as described above is implemented.

[0178] The storage medium may be a USB flash drive, a mobile hard disk, a read-only memory, a magnetic disk, or an optical disk, or any other computer-readable storage medium that can store program codes.

[0179] The present invention is not limited to the above-mentioned specific implementation modes. Various changes made by ordinary technicians in this field based on the above-mentioned concepts without creative work are all within the protection scope of the present invention.

Claims

1. A BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization, characterized in that: The steps include: Obtain multi-frequency signal observation data of BDS-3, including pseudorange observation data and carrier phase observation data; Combining the pseudorange observation data and the carrier phase observation data to construct a multi-frequency combination cycle slip detection amount, and selecting the optimal multi-frequency combination coefficient; The cycle slip detection equation is constructed by differentially analyzing the optimal multi-frequency combination cycle slip detection quantity between adjacent epochs, and the cycle slip judgment threshold is set to perform cycle slip detection epoch by epoch. The GCV regularization-assisted LAMBDA algorithm is used to repair the detected cycle slips.

2. The BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization according to claim 1, characterized in that: The multi-frequency signal observation data includes observation data of four-frequency signals and five-frequency signals; the multi-frequency combination cycle slip detection amount is composed of a phase geometry-free combination and a pseudo-range phase geometry-free ionosphere-free combination; When constructing the multi-frequency combination cycle slip detection quantity, one four-frequency pseudo-range phase geometry-free ionosphere combination and three four-frequency phase geometry-free combinations are selected to construct the four-frequency combination cycle slip detection quantity; One five-frequency pseudorange phase geometry-free ionosphere-free combination and four five-frequency phase geometry-free combinations are selected to construct the five-frequency combination cycle slip detection quantity.

3. The BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization according to claim 2, characterized in that: The step of selecting the optimal multi-frequency combination coefficient comprises: Set the search interval of multi-frequency combination coefficients to [-10,10]; For the phase geometry-free combination, the preferred condition is: taking into account the constraints that the combination noise and the ionospheric delay amplification factor are both relatively small, the combination with high data richness among the multiple phase geometry-free combinations connected together is preferred; the data richness is the total number of non-zero coefficients of each frequency signal of the phase geometry-free combination; For the pseudorange phase geometry-free and ionosphere-free combination, the optimal conditions are: the optimal combination is selected based on the long wavelength as the primary condition, and the relatively small combination noise and ionospheric delay amplification factor are considered secondly; Based on the preferred conditions, the optimal multi-frequency combination cycle slip detection amount is selected for the four-frequency signal and the five-frequency signal in the search interval.

4. The BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization according to claim 3, characterized in that: After the cycle slip is detected, the multi-frequency cycle slip estimation equation is established: A k ·ΔN k =L k Where k represents the number of frequencies; A k =[X GF11 ,…,X GFnk ,X GFIF ] T represents the k-frequency combination coefficient matrix; X GFnk =p mnk λ n , p mnk represents the combination coefficient of the nth frequency point of the mth single phase geometry-free combination of the k-frequency combination, λ n Indicates the wavelength of the nth frequency point; X GFIF L represents the combination coefficient of pseudorange phase without geometry and ionosphere; k is the k-frequency combination cycle slip detection matrix; ΔN k represents the k-frequency cycle slip; Based on the multi-frequency cycle slip estimation equation, the GCV regularization-assisted LAMBDA algorithm is used to repair the detected cycle slips.

5. The BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization according to claim 4, characterized in that: The method of using the GCV regularization-assisted LAMBDA algorithm to repair the detected cycle slips includes: Obtain the minimum value of the GCV curve to obtain the regularization parameter; Select the unit matrix as the regularization matrix; Constructing a GCV correction term based on the regularization parameter and the regularization matrix; Using the optimal multi-frequency combination cycle slip detection amount, performing least square correction through the GCV correction term, obtaining an estimated value of the cycle slip between epochs, and obtaining a floating-point solution for the cycle slip; Modifying the variance-covariance matrix of the LAMBDA algorithm based on the GCV correction term; The floating point solution of the cycle slip and the corrected variance-covariance matrix are input into the LAMBDA algorithm, and the floating point solution of the cycle slip is fixed by using the LAMBDA algorithm to obtain the cycle slip repair parameter solution of the epoch.

6. The BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization according to claim 5, characterized in that: The minimum value of the GCV curve is expressed as: in, k represents the number of frequencies, α represents the regularization parameter, tr() represents the trace of the matrix, T represents the transpose of the matrix, and I represents the identity matrix; The GCV correction term is: α·R, where α represents a regularization parameter and R is a unit matrix selected as a regularization matrix.

7. The BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization according to claim 6, characterized in that: The inter-epoch cycle slip estimate is expressed as: in, For the power matrix, Q k is the observation matrix L k The variance-covariance matrix of , R is the regularization matrix; Using Cholesky decomposition, the weight matrix P k Unitize to get the new combination coefficient matrix A GCV =G*A k , G is an upper triangular matrix; Then the variance-covariance matrix of the LAMBDA algorithm modified based on the GCV correction term is:

8. A BDS-3 multi-frequency cycle slip detection and repair system based on GCV regularization, characterized in that: include: Data acquisition module, used to obtain multi-frequency signal observation data of BDS-3, including pseudo-range observation data and carrier phase observation data; A combination optimization module is used to combine the pseudorange observation data and the carrier phase observation data to construct a multi-frequency combination cycle slip detection amount and select the optimal multi-frequency combination coefficient; A cycle slip detection module is used to construct a cycle slip detection equation by differentially determining the optimal multi-frequency combination cycle slip detection quantity between adjacent epochs, and to set a cycle slip judgment threshold to perform cycle slip detection epoch by epoch; The cycle slip repair module is used to repair the detected cycle slips using the GCV regularization-assisted LAMBDA algorithm.

9. An electronic device, characterized in that: The electronic device comprises: A memory for storing executable instructions; A processor, configured to implement a BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization as described in any one of claims 1 to 7 when running the executable instructions stored in the memory.

10. A computer-readable storage medium storing executable instructions, characterized in that: When the executable instructions are executed by the processor, the BDS-3 multi-frequency cycle slip detection and repair method based on GCV regularization as described in any one of claims 1 to 7 is implemented.