A method for correcting multi-path error of BDS considering creep displacement of monitoring points
By using the overall local mean decomposition method to distinguish noise and multipath signals, and reconstructing the multipath and creep displacement signals of monitoring points, the problem of not distinguishing low-frequency creep displacement signals in traditional methods is solved, thus improving monitoring accuracy and reliability.
Patent Information
- Application Number
- CN202611133160.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-29
- Publication Date
- 2026-11-10
AI Technical Summary
Traditional multipath error correction methods fail to effectively distinguish low-frequency creep displacement signals, resulting in a loss of accuracy in monitoring results and affecting the reliability and stability of BDS in high-precision monitoring projects.
The overall local mean decomposition method is adopted. By constructing a continuous average energy distribution map of PF components, the boundary point between noise signals and multipath signals is determined, multipath correction signals and creep displacement signals are reconstructed, high-frequency noise signals are eliminated, and low-frequency creep displacement information is retained.
It improves the accuracy of multipath effect correction, enhances the positioning accuracy of monitoring points, reduces monitoring errors, and strengthens the reliability of BDS in high-precision monitoring.
Smart Images

Figure CN123065334A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of monitoring data processing, and specifically to a BDS multipath error correction method that takes into account the creep displacement of monitoring points. Background Technology
[0002] With the rapid development of the BeiDou Navigation Satellite System (BDS), its application in my country's infrastructure sector is deepening. In monitoring projects such as slopes, roadbeds, and bridges, BDS, with its advantages of all-weather operation, high precision, and automation, is increasingly being used to acquire surface deformation information, providing crucial data support for safety status assessment and early disaster warning. However, BDS satellite signals are susceptible to reflection and obstruction, resulting in multipath effects. Multipath errors are difficult to eliminate effectively using conventional differential techniques, becoming a key bottleneck limiting the further improvement of BDS's reliability and stability in high-precision monitoring projects. Therefore, to improve monitoring accuracy, the impact of multipath errors must be eliminated.
[0003] Currently, some post-processing methods for multipath error correction data are widely used in practical monitoring projects due to their low cost. Low-frequency creep displacement signals at monitoring points are an important component in measuring the deformation trend of these points. However, traditional multipath error correction methods do not further distinguish low-frequency creep displacement signals, treating them as part of the multipath correction signal. This leads to a certain degree of accuracy loss in the monitoring results even after multipath error correction. Summary of the Invention
[0004] To address the problems existing in the prior art, this invention provides a BDS multipath error correction method that takes into account the creep displacement of the monitoring point, so as to accurately construct the multipath correction signal while also correctly distinguishing the low-frequency creep displacement signal of the monitoring point.
[0005] To achieve the above-mentioned objectives, the present invention adopts the following technical solution:
[0006] A BDS multipath error correction method that takes into account the creep displacement of monitoring points includes the following steps:
[0007] S1, obtain the BDS coordinate sequence of the benchmark point and the monitoring point, and use the double difference mode to solve it to obtain the original BDS coordinate residual sequence of the monitoring point in a certain direction;
[0008] S2, the original BDS coordinate residual sequence obtained in S1 is decomposed by the global local mean decomposition to obtain several PF component sequences and one residual component sequence;
[0009] S3, determine the boundary point between the noise signal component sequence and the multipath signal component sequence in the plurality of PF component sequences. ;
[0010] S4, based on the boundary point in S3, remove the high-frequency noise signal in the original BDS coordinate residual sequence, use the multipath signal component sequence to reconstruct the multipath correction signal sequence of the monitoring point, and use the residual component to reconstruct the creep displacement signal sequence of the monitoring point.
[0011] S5. Using the same method as step S1, obtain the original BDS coordinate residual sequence in the direction corresponding to the new time period of the monitoring point. Use the multipath correction signal sequence obtained in S4 to correct the original BDS coordinate residual sequence in the direction corresponding to the new time period of the monitoring point to obtain the BDS corrected coordinate residual sequence that takes into account the creep displacement of the monitoring point.
[0012] Step S1 above includes:
[0013] Let the BDS coordinate sequence of the monitoring point direction d be as follows: Where d represents any coordinate direction among X, Y, and Z, For direction d, the first BDS coordinates of each epoch. ;
[0014] The calculated coordinate residual sequence in this direction is as follows: Among them, coordinate residuals The calculation formula is as follows:
[0015] (1),
[0016] In the formula, Used as reference coordinates;
[0017] The mean-removing process is performed on the coordinate residual sequence to obtain the original BDS coordinate residual sequence along the direction d of the monitoring point. Among them, the original coordinate residual The calculation formula is as follows:
[0018] (2),
[0019] In the formula, The mean of the coordinate residual sequence. .
[0020] Step S2 above includes the following steps:
[0021] S21, Construction Grouped noisy coordinate residual sequences;
[0022] S22, perform local mean decomposition on the noisy coordinate residuals in each group of noisy coordinate residual sequences, decomposing them into... One PF component and one residual component;
[0023] S23, Construction The total local mean decomposition of the original BDS coordinate residual sequence is completed by using one PF component sequence and one residual component sequence.
[0024] In step S21 above:
[0025] Let the m-th set of noisy coordinate residuals along the direction d of the monitoring point be... , Among them, the noisy coordinate residual The calculation formula is:
[0026] (3),
[0027] In the formula, This is the white noise amplitude coefficient, with a value ranging from 0.1 to 0.3; The standard deviation of the original coordinate residual sequence of the BDS; This is the m-th group of white noise signals.
[0028] In step S22 above:
[0029] For each set of noisy coordinate residual sequences, the noisy coordinate residuals are respectively... Perform local mean decomposition, which decomposes the mean into There are one PF component and one residual component, and the formula is:
[0030] (4),
[0031] in, The noisy coordinate residual The PF components, order , The noisy coordinate residual The residual component.
[0032] Step S23 above includes:
[0033] By constructing the overall average PF component sequence:
[0034] Let the first The PF component sequence is , where PF component values The formula for calculating the overall average is as follows:
[0035] (5);
[0036] Construct the residual component sequence by overall averaging. :
[0037] Among them, residual component values The formula for calculating the overall average is as follows:
[0038] (6);
[0039] This completes the processing of the original coordinate residual sequence of the BDS. The overall local mean decomposition, where the original coordinate residuals After decomposition of the overall local mean, it is expressed as:
[0040] (7).
[0041] Step S3 above includes:
[0042] Calculate the average energy value of each PF component sequence to obtain a continuous average energy distribution map of the PF components; where the horizontal axis represents the order. Values, with the vertical axis representing the average energy value, where the first... Average energy of the PF component sequence The calculation formula is:
[0043] (8),
[0044] The boundary between the noise signal component and the multipath signal component The minimum point corresponding to the continuous average energy distribution map of the PF component before the global maximum point. The formula is:
[0045] (9),
[0046] In the formula, These represent the global maximum and global minimum points, respectively.
[0047] Order less than or equal to The PF component sequences are all high-frequency noise signal component sequences with an order greater than 1. The PF component sequences are all multipath signal component sequences.
[0048] In step S4 above:
[0049] The multipath correction signal sequence of the monitoring point is reconstructed using the multipath signal component sequence. The creep displacement signal sequence of the monitoring points was reconstructed using the residual components. ,in:
[0050] Multipath correction signal value The calculation formula is:
[0051] (10)
[0052] Creep displacement signal value The calculation formula is:
[0053] (11).
[0054] In step S5 above:
[0055] The BDS original coordinate residual sequence for the direction d corresponding to the monitoring point in the subsequent new time period is as follows: .
[0056] In step S5 above:
[0057] Use the multipath correction signal sequence obtained by S4 Regarding the After correction processing, a BDS corrected coordinate residual sequence that takes into account the creep displacement of the monitoring points is obtained. Among them, the corrected coordinate residuals The calculation formula is:
[0058] (12).
[0059] Compared with the prior art, the present invention has the following beneficial effects:
[0060] 1. This invention uses global local mean decomposition to decompose the original BDS coordinate residual sequence of monitoring points, calculates the average energy value of all PF components, and then calculates the boundary point between the high-frequency noise signal PF component and the low-frequency useful signal PF component based on the continuous average energy distribution map of the PF components. In this way, the multipath correction signal and the low-frequency creep displacement signal are reconstructed, which solves the problem of optimal calculation of the boundary point caused by the small difference in energy density of the high-order PF components.
[0061] 2. This invention utilizes the reconstructed multipath correction signal to correct the subsequent BDS original coordinate residual sequence, while retaining the low-frequency creep displacement information of the monitoring points, thereby improving the multipath effect correction accuracy and thus enhancing the positioning accuracy of the monitoring points. Attached Figure Description
[0062] Figure 1 A flowchart of a BDS multipath error correction method that takes into account the creep displacement of monitoring points according to the present invention;
[0063] Figure 2The original BDS coordinate residual sequence of the monitoring point in the X, Y, and Z directions on the 10th day (DAY10) in the embodiment of the present invention is shown in Figure (a) for the X direction, Figure (b) for the Y direction, and Figure (c) for the Z direction.
[0064] Figure 3 for Figure 2 The extraction results of the original BDS coordinate residual sequence in the X direction of the monitoring point are based on the global local mean decomposition. Among them, Figure (a) is the high-frequency noise signal sequence that needs to be removed, Figure (b) is the multipath correction signal sequence, and Figure (c) is the creep displacement signal sequence.
[0065] Figure 4 The original BDS coordinate residual sequence of the monitoring point in the X, Y, and Z directions on the 17th day (DAY17) in the embodiment of the present invention is shown in Figure (a) for the X direction, Figure (b) for the Y direction, and Figure (c) for the Z direction.
[0066] Figure 5 The following is a sequence of BDS-corrected coordinate residuals of the monitoring point in the X, Y, and Z directions on day 17 (DAY17) in an embodiment of the present invention. Figure (a) shows the X direction, Figure (b) shows the Y direction, and Figure (c) shows the Z direction. Detailed Implementation
[0067] The specific embodiments of the present invention will now be described in conjunction with the accompanying drawings and examples to enable those skilled in the art to understand the present invention.
[0068] Example
[0069] like Figure 1 As shown, a BDS multipath error correction method that takes into account the creep displacement of monitoring points includes the following steps:
[0070] S1. Obtain the BDS coordinate sequences of the benchmark point and monitoring point, and use the double-difference mode to solve for the original BDS coordinate residual sequence of the monitoring point. The specific steps are as follows:
[0071] Let the BDS coordinate sequence of the monitoring point direction d be as follows: Where d represents any coordinate direction among X, Y, and Z, For direction d, the first BDS coordinates of each epoch (observed epoch), .
[0072] The calculated coordinate residual sequence in this direction is as follows: Among them, coordinate residuals The calculation formula is as follows:
[0073] (1),
[0074] In the formula, The reference coordinates are obtained from the average value over a stable period.
[0075] The mean-removing process is performed on the coordinate residual sequence to obtain the original BDS coordinate residual sequence along the direction d of the monitoring point. Among them, the original coordinate residual The calculation formula is as follows:
[0076] (2),
[0077] In the formula, The mean of the coordinate residual sequence. .
[0078] In this embodiment, as Figure 2 As shown, the BDS original coordinate residual sequence of a certain monitoring point on the 10th day (DAY10) was obtained in the X, Y and Z directions.
[0079] S2, using global local mean decomposition, decomposes the original BDS coordinate residual sequence obtained in S1, resulting in several PF component sequences and one residual component sequence. The specific operation is as follows:
[0080] S21, Construction Group of Noisy Coordinate Residual Sequences: Let the m-th group of noisy coordinate residual sequences along the direction d of the monitoring point be... , Among them, the noisy coordinate residual The calculation formula is:
[0081] (3),
[0082] In the formula, This is the white noise amplitude coefficient, an empirical value ranging from 0.1 to 0.3. The standard deviation of the original coordinate residual sequence of the BDS; This is the m-th group of white noise signals.
[0083] S22, respectively, for each group of noisy coordinate residual sequences, the noisy coordinate residuals... Perform local mean decomposition, which decomposes the mean into There are one PF component and one residual component, and the formula is:
[0084] (4),
[0085] in, The noisy coordinate residual The PF components, order , The noisy coordinate residual The residual component.
[0086] S23, Construction Using one PF component sequence and one residual component sequence, the overall local mean decomposition of the original BDS coordinate residual sequence is completed. The specific operation is as follows:
[0087] By constructing the overall average PF component sequence:
[0088] Let the first The PF component sequence is , where PF component values The formula for calculating the overall average is as follows:
[0089] (5);
[0090] Construct the residual component sequence by overall averaging. :
[0091] Among them, residual component values The formula for calculating the overall average is as follows:
[0092] (6);
[0093] This completes the processing of the original coordinate residual sequence of the BDS. The overall local mean decomposition, where the original coordinate residuals After decomposition of the overall local mean, it is expressed as:
[0094] (7).
[0095] S3, determine the boundary point between the noise signal component sequence and the multipath signal component sequence in the plurality of PF component sequences. The specific steps are as follows:
[0096] The average energy value of each PF component sequence is calculated separately to obtain a continuous average energy distribution map (line graph) of the PF components. The horizontal axis represents the order. Values, with the vertical axis representing the average energy value, where the first... Average energy of the PF component sequence The calculation formula is:
[0097] (8),
[0098] The boundary between the noise signal component and the multipath signal component The minimum point corresponding to the continuous average energy distribution map of the PF component before the global maximum point. The formula is:
[0099] (9),
[0100] In the formula, These represent the global maximum and global minimum points, respectively.
[0101] Order less than or equal to The PF component sequences are all high-frequency noise signal component sequences with an order greater than 1. The PF component sequences are all multipath signal component sequences.
[0102] S4, based on the boundary points in S3, remove high-frequency noise signals from the original BDS coordinate residual sequence, and reconstruct the multipath correction signal sequence of the monitoring points using the remaining PF component sequences (multipath signal component sequences). The creep displacement signal sequence of the monitoring points was reconstructed using the residual components. ,in:
[0103] Multipath correction signal value The calculation formula is:
[0104] (10)
[0105] Creep displacement signal value The calculation formula is:
[0106] (11).
[0107] In this embodiment, the multipath correction signal sequence and creep displacement signal sequence of a certain monitoring point on the 10th day (DAY10) in the X, Y and Z directions were obtained.
[0108] Figure 3 Showing the Figure 2 The extraction results of the BDS original coordinate residual sequence in the X direction of the monitoring point in the data based on the global local mean decomposition; where, Figure 3 (a) is the sequence of high-frequency noise signals that need to be removed. Figure 3 (b) is the reconstructed multipath correction signal sequence. Figure 3 (c) is the reconstructed creep displacement signal sequence. Furthermore, Figure 3 The scatter plots in (c) represent the total station monitoring data in the X direction of the monitoring point. It can be seen that the creep displacement change at the monitoring point is about 1.5 mm, and the difference between the creep displacement change and the total station monitoring data is within 0.3 mm. Therefore, it can be proven that the reconstructed multipath correction signal sequence does not contain low-frequency creep displacement information.
[0109] S5, using the same method as step S1, obtain the original BDS coordinate residual sequence of the monitoring point in the direction d corresponding to the new time period in the subsequent time period. The multipath correction signal sequence obtained using S4 The original BDS coordinate residual sequence for the corresponding direction d in the subsequent new time period of the monitoring point is corrected to obtain the BDS corrected coordinate residual sequence that takes into account the creep displacement of the monitoring point. Among them, the corrected coordinate residuals The calculation formula is:
[0110] (12).
[0111] In this embodiment, as Figure 4 As shown, this is the BDS original coordinate residual sequence of the monitoring point in the X, Y, and Z directions on day 17 (DAY17). The multipath correction signal sequences of the monitoring point in the X, Y, and Z directions obtained from S4 on day 10 (DAY10) are used to correct the BDS original coordinate residual sequences of the monitoring point in the X, Y, and Z directions on day 17 (DAY17). The resulting BDS corrected coordinate residual sequences in the X, Y, and Z directions are shown below. Figure 5 As shown.
[0112] The root mean square (RMS) values of the original BDS coordinate residual sequences in the X, Y, and Z directions of the monitoring point on day 17 (DAY17) were calculated, as well as the root mean square (RMS) values of the corrected BDS coordinate residual sequences in the X, Y, and Z directions. The results show that the RMS values of the corrected coordinate residual sequences were reduced by 29.8%, 33.6%, and 28.5%, respectively. This demonstrates that the method of the present invention effectively corrects multipath errors while preserving the creep displacement information of the coordinate residual sequences of the monitoring points, thereby improving the monitoring accuracy.
Claims
1. A BDS multipath error correction method that takes into account the creep displacement of monitoring points, characterized in that, Includes the following steps: S1, obtain the BDS coordinate sequence of the benchmark point and the monitoring point, and use the double difference mode to solve it to obtain the original BDS coordinate residual sequence of the monitoring point in a certain direction; S2, the original BDS coordinate residual sequence obtained in S1 is decomposed by the global local mean decomposition to obtain several PF component sequences and one residual component sequence; S3, determine the boundary point between the noise signal component sequence and the multipath signal component sequence in the plurality of PF component sequences. ; S4, based on the boundary point in S3, remove the high-frequency noise signal in the original BDS coordinate residual sequence, use the multipath signal component sequence to reconstruct the multipath correction signal sequence of the monitoring point, and use the residual component to reconstruct the creep displacement signal sequence of the monitoring point. S5. Using the same method as step S1, obtain the original BDS coordinate residual sequence in the direction corresponding to the new time period of the monitoring point. Use the multipath correction signal sequence obtained in S4 to correct the original BDS coordinate residual sequence in the direction corresponding to the new time period of the monitoring point to obtain the BDS corrected coordinate residual sequence that takes into account the creep displacement of the monitoring point.
2. The BDS multipath error correction method according to claim 1, characterized in that, S1 includes: Let the BDS coordinate sequence of the monitoring point direction d be as follows: Where d represents any coordinate direction among X, Y, and Z, For direction d, the first BDS coordinates of each epoch. ; The calculated coordinate residual sequence in this direction is as follows: Among them, coordinate residuals The calculation formula is as follows: (1), In the formula, Used as reference coordinates; The mean-removing process is performed on the coordinate residual sequence to obtain the original BDS coordinate residual sequence along the direction d of the monitoring point. Among them, the original coordinate residual The calculation formula is as follows: (2), In the formula, The mean of the coordinate residual sequence. .
3. The BDS multipath error correction method according to claim 2, characterized in that, S2 includes the following steps: S21, Construction Grouped noisy coordinate residual sequences; S22, perform local mean decomposition on the noisy coordinate residuals in each group of noisy coordinate residual sequences, decomposing them into... One PF component and one residual component; S23, Construction The total local mean decomposition of the original BDS coordinate residual sequence is completed by using one PF component sequence and one residual component sequence.
4. The BDS multipath error correction method according to claim 3, characterized in that, In S21: Let the m-th set of noisy coordinate residuals along the direction d of the monitoring point be... , Among them, the noisy coordinate residual The calculation formula is: (3), In the formula, This is the white noise amplitude coefficient, with a value ranging from 0.1 to 0.3; The standard deviation of the original coordinate residual sequence of the BDS; This is the m-th group of white noise signals.
5. The BDS multipath error correction method according to claim 4, characterized in that, In S22: For each set of noisy coordinate residual sequences, the noisy coordinate residuals are respectively... Perform local mean decomposition, which decomposes the mean into There are one PF component and one residual component, and the formula is: (4), in, The noisy coordinate residual The PF components, order , The noisy coordinate residual The residual component.
6. The BDS multipath error correction method according to claim 5, characterized in that, S23 includes: By constructing the overall average PF component sequence: Let the first The PF component sequence is , where PF component values The formula for calculating the overall average is as follows: (5); Construct the residual component sequence by overall averaging. : Among them, residual component values The formula for calculating the overall average is as follows: (6); This completes the processing of the original coordinate residual sequence of the BDS. The overall local mean decomposition, where the original coordinate residuals After decomposition of the overall local mean, it is expressed as: (7)。 7. The BDS multipath error correction method according to claim 6, characterized in that, S3 include: Calculate the average energy value of each PF component sequence to obtain a continuous average energy distribution map of the PF components; where the horizontal axis represents the order. Values, with the vertical axis representing the average energy value, where the first... Average energy of the PF component sequence The calculation formula is: (8), The boundary between the noise signal component and the multipath signal component The minimum point corresponding to the continuous average energy distribution map of the PF component before the global maximum point. The formula is: (9), In the formula, These represent the global maximum and global minimum points, respectively. Order less than or equal to The PF component sequences are all high-frequency noise signal component sequences with an order greater than 1. The PF component sequences are all multipath signal component sequences.
8. The BDS multipath error correction method according to claim 7, characterized in that, In S4: The multipath correction signal sequence of the monitoring point is reconstructed using the multipath signal component sequence. The creep displacement signal sequence of the monitoring points was reconstructed using the residual components. ,in: Multipath correction signal value The calculation formula is: (10), Creep displacement signal value The calculation formula is: (11)。 9. The BDS multipath error correction method according to claim 8, characterized in that, In S5: The BDS original coordinate residual sequence for the direction d corresponding to the monitoring point in the subsequent new time period is as follows: .
10. The BDS multipath error correction method according to claim 9, characterized in that, In S5: Use the multipath correction signal sequence obtained by S4 Regarding the After correction processing, a BDS corrected coordinate residual sequence that takes into account the creep displacement of the monitoring points is obtained. Among them, the corrected coordinate residuals The calculation formula is: (12)。
