Beidou +5G GNSS data resolving method

Abstract

The invention discloses a Beidou + 5G GNSS (Global Navigation Satellite System) data resolving method, which comprises the following steps of: acquiring original data of each GNSS in a monitoring area and a server, performing system control and data processing by GNSS monitoring software HCMonitor, and adopting advanced nonlinear Kalman filtering double-difference solution and three-difference solution algorithms in the aspect of GNSS data processing. The static resolving algorithm is matched with the HCMonitor to carry out differential resolving on GNSS original observation values at the same moment (within one microsecond), the static resolving algorithm adopts an expanded dynamic nonlinear Kalman filtering algorithm to carry out differential resolving, the HCMonitor can adopt the expanded dynamic nonlinear Kalman filtering algorithm to carry out differential resolving, the requirement of the HCMonitor algorithm for hardware of a system is high, and the algorithm can be used for carrying out differential resolving on the GNSS original observation values at the same moment (within one microsecond). Generally, in a high-performance computer, an RTK algorithm is always provided by a GNSS receiver manufacturer and is solidified in a GNSS receiver. Static solution needs manual intervention, and a double-difference fixed solution mode is generally adopted.

Description

Technical Field

[0001] This invention relates to the field of communication and algorithm technology, specifically a GNSS data processing method using BeiDou + 5G. Background Technology

[0002] The Monitor manages the cluster state. Its primary task is to maintain the consistency of the cluster view. This consistency maintenance utilizes the Paxos algorithm (protocol), which is instantiated in the database for easy access later. The Monitor consists of three parts: key-value storage, Paxos, and PaxosService. The key-value storage serves as the persistent storage layer for Monitor data. Paxos provides consistent data access logic for the PaxosService layer. Each PaxosService represents a type of cluster state information. PaxosServer is actually an abstraction that includes: PGMonitor, OSDMonitor, MDSMonitor, AuthMonitor, LogMonitor, etc.

[0003] GNSS data processing is a core component of GNSS automated monitoring systems. The accuracy of the processing results directly affects our judgment and analysis of the stability of the monitored object and influences the decision-making of management personnel.

[0004] Traditional GNSS RTK methods do not require differential corrections to be synchronized with the rover's observation data. The typical reference station receiver's differential correction broadcast update rate is 1Hz. Therefore, differential corrections are generally delayed by 0.5 to 2 seconds. In special cases, the rover can allow differential corrections from one minute prior to be included in the calculation. RTK receivers achieve centimeter-level dynamic positioning accuracy. Typical RTK methods require communication between the reference station and the rover, as well as communication between the rover and the data center. RTK is commonly used in surveying and high-precision navigation. For RTK receivers, if the GNSS signal loses lock, the receiver needs to be reinitialized and the integer ambiguity solved, resulting in the inability to output centimeter-level positioning solutions within short time intervals.

[0005] Traditional static data processing requires manual intervention for calculation and network adjustment, and also requires long-term observation at each point in different time periods. If the monitoring point is in a moving state, it cannot calculate effective results.

[0006] Based on the problems mentioned above, this application proposes a new method for GNSS data processing. Summary of the Invention

[0007] The purpose of this invention is to provide a GNSS data processing method based on BeiDou + 5G to solve the problems mentioned in the background art.

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] The BeiDou + 5G GNSS data processing method includes the acquisition and server of raw GNSS data in the monitoring area, and the GNSS monitoring software HCMonitor for system control and data processing. In terms of GNSS data processing, advanced nonlinear Kalman filter double-difference and triple-difference algorithms are adopted. At the same time, an ionospheric correction model is added, supporting multi-reference station processing and real-time independent baseline network adjustment function. The specific accuracy is less than 2mm in the plane and 4mm in the elevation.

[0010] HCMonitor only requires receiving raw observation data from the GNSS receiver; communication between the software (server) and the GNSS receiver is one-way.

[0011] As a further improvement of the present invention: HCMonitor supports BDS single-frequency and dual-frequency computation;

[0012] If the environment permits (e.g., baseline <10km), BDS single-frequency solution can be selected; to improve the accuracy, stability and reliability of the solution results, GNSS dual-frequency solution can be selected.

[0013] As a further aspect of this invention: GNSS ambiguity fixation is the core of the software. With the addition of GNSS satellites, the number of ambiguity searches increases exponentially, and a unique partial ambiguity search algorithm is adopted.

[0014] As a further aspect of the present invention: GNSS supports mean filters and Kalman filters;

[0015] The static solution engine assumes that the slowly changing monitoring points are also moving. Based on this, a static solution filtering engine is built. This mode can reflect the displacement changes of each monitoring point in real time, and the accuracy also increases with the increase of observation time.

[0016] Filtering methods are used to eliminate random errors in GNSS dynamic positioning data, and the true state (positioning result) is estimated optimally in real time from various random disturbances.

[0017] The discrete state-space model for GNSS dynamic positioning is as follows:

[0018]

[0019] In the formula for dimensional state vector, For the state dimensional nonlinear function, for Dimensional output quantity, for dimensional system transition matrix, , They are respectively , A sequence of dimensional random vectors, satisfying the following statistical properties:

[0020]

[0021] for A positive semidefinite symmetric matrix. for A positive definite symmetric matrix;

[0022] Further define the initial state A random vector that satisfies the following statistical properties:

[0023] ;

[0024] When a nonlinear system is subjected to a Taylor expansion along its nominal trajectory, approximating its linear portion, and then state-estimated using a standard Kalman filter, the well-known extended Kalman filter is obtained as follows:

[0025] .

[0026] As a further aspect of the present invention: the static solution algorithm is combined with the GNSS raw observations of HCMonitor at the same time (within 1 microsecond) for differential calculation. The static solution algorithm adopts an extended dynamic nonlinear Kalman filter algorithm for differential calculation.

[0027] As a further aspect of the present invention: the static solution algorithm directly applies the raw data from the GNSS receiver, the observation data of the reference station and the rover station are kept strictly synchronized, and the common error caused by atmospheric delay is offset to the greatest extent.

[0028] As a further aspect of the present invention: HCMonitor only requires receiving the raw observation data from the GNSS receiver, so it should be required that the software (server) and the GNSS receiver only need to achieve one-way communication. In the HCMonitor mode, the raw data of the rover only needs one serial data communication and one network data communication to reach the data center, and the reference station can be directly connected to the data center server.

[0029] As a further aspect of this invention: the GNSS baseline is formed by synchronous observation data from two receivers. The HCMonitor software integrates real-time dynamic processing, quasi-dynamic processing algorithms, and different filtering and ionospheric correction models. Therefore, it is necessary to set specific data processing methods, correction models, etc., according to the actual needs of the monitoring project, and even include limits on satellite elevation angle and PDOP value. The reference station provides reference coordinates and carrier phase observation values ​​for relative positioning calculation in the entire system. By differential processing of the reference station data and the data of each monitoring station, the precise three-dimensional coordinates of each monitoring point are calculated.

[0030] As a further improvement of the present invention: the HCMonitor system adopts a single-frequency / dual-frequency hybrid solution method;

[0031] In GNSS positioning, one or more signals modulated on L1 and L2 are often used to calculate the baseline vector between the points to be determined and to determine the coordinates of the points. The system uses a single-frequency L1 or a single-frequency / dual-frequency L1 and L2 hybrid solution method.

[0032] 8. The GNSS data processing method for BeiDou + 5G according to claim 1, characterized in that the baseline processing step,

[0033] Baseline solution parameter control

[0034] Baseline optimization can be achieved by setting control parameters. These parameters mainly include the data sampling interval, cutoff angle, reference satellite and its ionosphere, and the solution model.

[0035] b Baseline solution self-check

[0036] Before baseline calculation, the software first sets the baseline calculation control parameters, observation data and ephemeris files, and starting coordinates;

[0037] c import ephemeris data

[0038] The ephemeris data portion of the raw data can also be extracted by downloading precise ephemeris data from relevant websites to improve the accuracy of the calculation;

[0039] d Read in observation data

[0040] Read the observation data of the raw data, including the observation data of the starting station and the ending station, the single-point positioning coordinates, the observation time, the C / A code pseudorange, the P code pseudorange, and the carrier phase;

[0041] e Tri-difference solution

[0042] Subtract the double-difference observations between epochs to form triple-difference observations, establish the observation equation, solve it, and obtain the triple-difference solution;

[0043] f-cycle slip repair

[0044] The key to baseline resolution is finding the correct integer ambiguity. The prerequisite for solving the integer ambiguity is that the receiver continuously tracks the carrier phase. The main problem that baseline processing software needs to solve is that the receiver continuously tracks the carrier phase. Obstruction, interference, etc. can cause the tracking of the carrier phase to be interrupted.

[0045] g performs double-difference floating-point calculations

[0046] If signals from N satellites are observed, the double-difference observation equation system will have N-1 more unknowns than the triple-difference observation equation system. The double-difference solution yields further unknown point coordinates and integer ambiguities expressed in floating-point numbers. The integer ambiguities should be integers.

[0047] h Integer fuzziness decomposition

[0048] The Fast Ambiguity Resolution Approach (FARA) and the LAMBDA method are employed.

[0049] i performs double-difference fixed solution

[0050] After the integer ambiguity is correctly fixed, double difference fixed solution is performed.

[0051] As a further solution of the present invention: GNSS real-time independent baseline network adjustment. The method of network adjustment for this monitoring system is independent baseline network adjustment, that is, when adjusting the same monitoring point, only the baseline related to the reference station is used.

[0052] Compared with the prior art, the beneficial effects of the present invention are:

[0053] 1. Compared to RTK and traditional static methods, the HCMonitor algorithm for BeiDou + 5G GNSS data processing has the following characteristics: HCMonitor uses raw GNSS observations from the same moment (within 1 microsecond) for differential calculation; while the RTK method does not require the differential corrections to be synchronized with the rover's observation data. The broadcast update rate of differential corrections from a typical reference station receiver is 1Hz. Therefore, the differential corrections are generally delayed by 0.5 to 2 seconds. In special cases, the rover can allow differential corrections from 1 minute ago to participate in the calculation; HCMonitor can use an extended dynamic nonlinear Kalman filter algorithm for differential calculation. The HCMonitor algorithm has higher hardware requirements, usually requiring a high-performance computer, while the RTK algorithm is always provided by the GNSS receiver manufacturer and is embedded within the GNSS receiver; static calculation requires manual intervention and generally uses a double-difference fixed solution method.

[0054] 2. The GNSS data processing method of BeiDou + 5G, HCMonitor directly uses the raw data of the GNSS receiver. The observation data of the reference station and the rover station are strictly synchronized. Therefore, the common error caused by atmospheric delay is canceled to the greatest extent. HCMonitor also uses filtering methods to eliminate various random errors in GNSS dynamic positioning data, making the output positioning results more consistent with the real situation. Therefore, depending on the GNSS receiver and GNSS antenna used, HCMonitor can guarantee millimeter-level positioning accuracy, while the dynamic positioning accuracy of a typical RTK receiver is centimeter-level.

[0055] 3. Regarding the GNSS data processing method using BeiDou + 5G, since HCMonitor only requires receiving raw observation data from the GNSS receiver, in principle, only one-way communication between the software (server) and the GNSS receiver should be required. In contrast, typical RTK methods require communication between the reference station and the rover, and also between the rover and the data center. Attached Figure Description

[0056] Figure 1 This is a data communication diagram under PTK mode;

[0057] Figure 2 This is a data transmission diagram under HCMonitor mode;

[0058] Figure 3 A baseline processing interface diagram for a BeiDou + 5G GNSS data processing method;

[0059] Figure 4 A diagram illustrating the main steps of single-baseline solution;

[0060] Figure 5 This is the coordinate transformation settings interface for a BeiDou + 5G GNSS data processing method. Detailed Implementation

[0061] Please see Figures 1-5 In this embodiment of the invention, a BeiDou + 5G GNSS data processing method is provided. The BeiDou + 5G GNSS data processing method includes the acquisition of raw GNSS data in the monitoring area, and the GNSS monitoring software HCMonitor for system control and data processing. The method is characterized by the use of advanced nonlinear Kalman filtering double-difference and triple-difference algorithms in GNSS data processing, and the addition of an ionospheric correction model, support for multi-reference station processing and real-time independent baseline network adjustment function. The specific accuracy is less than 2mm for plane and 4mm for elevation.

[0062] HCMonitor is a real-time 3D deformation measurement and analysis system software based on the Global Positioning System (GNSS) using a network. HCMonitor can simultaneously perform real-time 3D position calculations on dozens of GNSS receivers placed on target facilities or natural objects, achieving millimeter-level accuracy. The software adopts a C / S architecture, simultaneously realizing real-time differential positioning of monitoring stations, and has functions such as graphic display, receiver settings, monitoring station parameter settings, observation data recording, and alarms. In the deformation monitoring network composed of HCMonitor as the core, each GNSS receiver only needs to output the raw GNSS data and ephemeris. The data is transmitted to the control center through WAN, LAN, serial port, wireless devices, etc. The HCMonitor software in the control center obtains the raw real-time data of each monitoring point according to the IP address and port number corresponding to each GNSS receiver, and then performs real-time differential calculations on these raw data to obtain the coordinates of each monitoring station, which are then stored in the database or sent to the client.

[0063] In a preferred embodiment, HCMonitor supports BDS single-frequency and dual-frequency computation.

[0064] When project costs are low and environmental conditions permit (e.g., baseline <10km), BDS single-frequency solution can be selected; when the accuracy, stability and reliability of the solution results are to be improved, GNSS dual-frequency solution can be selected.

[0065] Automatic optimal search for various frequency combinations, and automatic selection of results for various combinations such as B1 and B2;

[0066] A comprehensive coordinate system management system supports Beijing 54, Xi'an 80, CGCS2000, and self-built systems.

[0067] It has a rich library of GNSS error models and supports high-precision long baseline (500KM) solution;

[0068] Supports various NGS certified antennas and has a complete antenna parameter system;

[0069] A highly efficient GNSS partial ambiguity search algorithm.

[0070] In a preferred embodiment, GNSS ambiguity fixing is the core of the software. Due to the increase in GNSS satellites, the number of ambiguity searches has increased exponentially. If traditional methods are still used, the search efficiency is low and it is greatly affected by "contaminated" satellites, often leading to search failures. The software adopts a proprietary partial ambiguity search algorithm, which not only improves the ambiguity search efficiency but also ensures the reliability of fixing.

[0071] In a preferred embodiment, mean filter and Kalman filter are supported;

[0072] Powerful decoder functionality, supporting data parsing from mainstream motherboards both domestically and internationally;

[0073] Compared to traditional static engine software on the market, this software's solution strategy is more suitable for slow-change monitoring, specifically in the following ways:

[0074] Traditional static engines assume that the monitoring point is stationary for a certain period of time (e.g., 1 hour). They use the traditional static processing mode to calculate the position of the monitoring point through iterative loops. Therefore, this mode cannot fully reflect the displacement changes of the monitoring point.

[0075] The static solution engine assumes that the slowly changing monitoring points are also moving. Based on this, a static solution filtering engine is built. This mode can reflect the displacement changes of each monitoring point in real time, and the accuracy also increases with the increase of observation time.

[0076] To eliminate random errors in GNSS dynamic positioning data using filtering methods, the first thought that comes to mind is to apply the optimal estimation method (i.e., Kalman filter) to estimate the true state (positioning result) from various random disturbances in real time in the best possible way.

[0077] However, to apply the Kalman filter for optimal estimation (filtering), it is necessary to establish a relatively accurate system model and observation model. Therefore, it is necessary not only to establish an accurate and reasonable dynamic model for the moving vehicle, but also to accurately model various random errors.

[0078] The discrete state-space model for GNSS dynamic positioning is as follows:

[0079]

[0080] In the formula for dimensional state vector, For the state dimensional nonlinear function, for Dimensional output quantity, for dimensional system transition matrix, , They are respectively , A sequence of dimensional random vectors, satisfying the following statistical properties:

[0081]

[0082] for A positive semidefinite symmetric matrix. for A positive definite symmetric matrix;

[0083] Further define the initial state A random vector that satisfies the following statistical properties:

[0084] ;

[0085] When a nonlinear system is subjected to a Taylor expansion along its nominal trajectory, approximating its linear portion, and then state-estimated using a standard Kalman filter, the well-known extended Kalman filter is obtained as follows:

[0086] ;

[0087] Compared with RTK and traditional static methods, the static solution algorithm in this application has the following characteristics.

[0088] In a preferred embodiment, the static solution algorithm is used in conjunction with the HCMonitor's instantaneous (within 1 microsecond) raw GNSS observations for differential calculation; while the RTK method does not require the differential corrections to be synchronized with the rover's observation data. The typical reference station receiver's differential correction broadcast update rate is 1Hz. Therefore, the differential corrections are generally delayed by 0.5 to 2 seconds. In special cases, the rover can allow differential corrections from 1 minute ago to participate in the calculation.

[0089] The static solution algorithm can be performed using an extended dynamic nonlinear Kalman filter algorithm for differential solution;

[0090] Static solution algorithms have high hardware requirements and are typically used in high-performance computers, while RTK algorithms are always provided by GNSS receiver manufacturers and are embedded within the GNSS receiver. Static solution requires manual intervention and generally uses a double-difference fixed solution method.

[0091] In a preferred embodiment, the static solution algorithm directly applies the raw data from the GNSS receiver, and the observation data from the reference station and the rover are kept strictly synchronized. Therefore, the common errors caused by atmospheric delay are offset to the greatest extent. The static solution algorithm also uses filtering methods to eliminate various random errors in the GNSS dynamic positioning data, making the output positioning results more consistent with the actual situation. Therefore, depending on the GNSS receiver and GNSS antenna used, the static solution algorithm can guarantee millimeter-level positioning accuracy, while the dynamic positioning accuracy of a typical RTK receiver is centimeter-level.

[0092] In a preferred embodiment, since HCMonitor only requires receiving raw observation data from the GNSS receiver, in principle, it should be required that the software (server) and the GNSS receiver only achieve one-way communication, while the usual RTK method requires communication between the reference station and the rover, and also requires communication between the rover and the data center.

[0093] Figure 1 and 2 This diagram illustrates the data communication links required for HCMonitor and RTK methods in network monitoring using a serial port GNSS receiver. As shown in the diagram, in RTK mode, differential correction values ​​and positioning results need to be transmitted multiple times. Figure 1 To send differential data from a monitoring point to the data center, three serial port data transmissions and two network data transmissions are required.

[0094] Depend on Figure 2 As can be seen, in HCMonitor mode, the raw data from the rover only needs one serial data communication and one network data communication to reach the data center, and the reference station can be directly connected to the data center server.

[0095] RTK is commonly used in surveying, high-precision navigation, etc. For RTK receivers, if the GNSS signal is lost, the receiver needs to be reinitialized and the integer ambiguity needs to be solved, which will result in the inability to output centimeter-level positioning solutions normally within a short time interval.

[0096] Traditional static data processing requires manual intervention for calculation and network adjustment, and also requires long-term observation at each point in different time periods. If the monitoring point is in a moving state, it cannot calculate effective results.

[0097] HCMonitor is designed specifically for deformation monitoring and is suitable for deformation monitoring of bridges, dams, mining areas, landslides, etc. The software can work continuously and reliably for a long time. The shortcomings of RTK, such as the need for frequent re-initialization, the need for manual intervention in static calculation and the inability to monitor moving objects, and poor timeliness, do not exist in HCMonitor.

[0098] HCMonitor software runs on computers in data centers, and its entire computing function can be designed in a redundant mode to increase system reliability. RTK methods cannot achieve similar functionality, and HCMonitor can achieve higher accuracy. When combined with static solution algorithms, HCMonitor can achieve higher reliability than RTK methods and traditional static methods.

[0099] The software moves the calculation work that was originally done in the receiver to a computer with higher performance, faster speed and more stable performance. All the raw GNSS data in the system is transmitted back to the computer in the control center via the network (wired / wireless), and the software performs synchronous and real-time calculation on all the data.

[0100] Meanwhile, all GNSS receiver settings are executed through the computer in the control center, achieving complete unattended operation and remote control;

[0101] According to system parameter settings, Kalman filtering can be applied to the real-time differential results of different monitoring stations to achieve different dynamic and accuracy requirements. It can process data from multiple base stations and multiple monitoring stations simultaneously, and display baseline changes and point movement in real time. The software includes the following views: real-time data view, real-time network diagram, trend chart, satellite view, 3D view, data management, automatic saving function of raw data and solution results, which can be set according to user needs, remote setting function for monitoring stations and base station receivers, and independent monitoring modules for each GNSS receiver. It can send user commands to the GNSS receiver to change parameters (such as sampling interval, altitude cutoff angle, etc.).

[0102] In a preferred embodiment, the GNSS baseline is formed from synchronous observation data from two receivers. The HCMonitor software integrates real-time dynamic processing, quasi-dynamic processing algorithms, and various filtering and ionospheric correction models. Therefore, specific data processing methods and correction models need to be set according to the actual needs of the monitoring project, including limits on satellite elevation angle and PDOP value. The reference station provides reference coordinates and carrier phase observations for relative positioning calculations throughout the system. By differential processing of the reference station data and the data from each monitoring station, the precise three-dimensional coordinates of each monitoring point are calculated.

[0103] Normally, real-time processing of a single base station is sufficient to determine the precise location of a monitoring station. However, to improve the accuracy of the monitoring system and ensure its stability, we typically employ a dual-base station or multi-base station approach. Experiments have demonstrated a significant improvement in accuracy. This is because multiple base stations generate more redundant observations, the calculation results for the same point from different baselines are cross-checked, a weighted average is performed using the least squares method, and real-time independent baseline network adjustment is conducted. Furthermore, multiple base stations can cover the entire monitoring area, allowing for the calculation of an ionospheric correction model that better suits the specific monitoring region.

[0104] In a preferred embodiment, the system employs a single-frequency / dual-frequency hybrid solution method;

[0105] In GNSS positioning, one or more signals modulated on L1 and L2 are often used to calculate the baseline vector between points to be determined, thereby determining the coordinates of the points. The system employs a single-frequency L1 or a hybrid single-frequency / dual-frequency L1 and L2 calculation method.

[0106] signal type L1 L2 Carrier phase observations √ √ C / A code pseudorange √ － P-code pseudorange √ √ Doppler shift √ √

[0107] The main signals of GNSS two carriers

[0108] In addition to using the above observations for data processing, the system also uses some special observations formed by combining the above observations, such as wide-lane observations, narrow-lane observations, and ionosphere-free observations, for data processing.

[0109] In a preferred embodiment, the baseline processing step...

[0110] 1) Baseline solution parameter control

[0111] Baseline optimization can be achieved by setting control parameters. These parameters mainly include the data sampling interval, cutoff angle, reference satellite and its ionosphere, and the solution model.

[0112] 2) Baseline solution self-check

[0113] Before baseline calculation, the software first sets the baseline calculation control parameters, observation data and ephemeris files, starting coordinates, etc.

[0114] 3) Read in ephemeris data

[0115] The ephemeris data portion of the raw data can also be extracted by downloading precise ephemeris data from relevant websites to improve the accuracy of the calculation;

[0116] 4) Read in the observation data

[0117] Read the observation data of the raw data, including the observation data of the starting station and the ending station, the single-point positioning coordinates, the observation time, the C / A code pseudorange, the P code pseudorange, and the carrier phase;

[0118] 5) Three-difference solution

[0119] The double-difference observations are subtracted between epochs to form triple-difference observations. An observation equation is established and solved to obtain a triple-difference solution. However, for short sides, the accuracy of the triple-difference solution is often not high. Usually, the purpose of the triple-difference solution is to obtain a more approximate baseline side, which is convenient for cycle slip repair.

[0120] 6) Cycle slip repair

[0121] The key to baseline resolution is finding the correct integer ambiguity. The prerequisite for solving the integer ambiguity is that the receiver continuously tracks the carrier phase. However, the receiver cannot always track the carrier phase continuously. Obstruction, interference, etc., will cause the tracking of the carrier phase to be interrupted, resulting in the so-called cycle slip of the carrier phase observation between epochs. How to detect and repair cycle slip is often the main problem that baseline processing software needs to solve.

[0122] 7) Perform double-difference floating-point calculations.

[0123] If signals from N satellites are observed, the double-difference observation equation system will have N-1 more unknowns than the triple-difference observation equation system. The double-difference solution yields further unknown point coordinates and integer ambiguities expressed as floating-point numbers. Theoretically, the integer ambiguity should be an integer, but due to the absorption of observation noise and other unmodeled errors during the solution process, it usually only yields a floating-point number. This floating-point number often deviates from the actual integer value, sometimes by several weeks.

[0124] 8) Integer fuzziness decomposition

[0125] Generally speaking, given a sufficiently long synchronous observation time and enough observation data, the correct integer ambiguity can be obtained simply by rounding. However, the Fast Ambiguity Resolution Approach (FARA) and the LAMBDA method can significantly shorten the observation time and improve work efficiency.

[0126] 9) Perform double-difference fixed solution

[0127] After the integer ambiguity is correctly fixed, double-difference fixed solution is performed. Double-difference fixed solution has the highest accuracy, but if the integer ambiguity is incorrect, the accuracy of double-difference fixed solution will also be incorrect.

[0128] GNSS real-time independent baseline network adjustment: The method for network adjustment in this monitoring system is independent baseline network adjustment, that is, when adjusting the same monitoring point, only the baseline related to the reference station is used. The advantage of this design is that it will not move other monitoring points to other monitoring points for adjustment.

[0129] Depending on the monitoring methods and approaches, users can directly connect the system and the electronic observation handbook through the system's interface program to import data automatically or manually, such as inputting internal displacement monitoring results. Gross error inspection is performed according to relevant standards and procedures using appropriate methods for gross error inspection. If gross errors are found, a warning and possible causes will be given to help find the cause and rework and retest. If there are no gross errors, the inspection will be indicated as passed, and the next step of data saving can be performed.

[0130] The data transmission section mainly includes the data interface between the raw monitoring data and the processing module, the storage of raw and processed data in the database, and the data interface between data processing and data analysis. To intuitively express the changing trends and patterns of the monitored area, the coordinates of the GNSS processing results need to be transformed during data analysis. Therefore, coordinate transformation of the monitoring results is required before storing the data in the database.

[0131] Before being transferred to the database module, the results from the raw data processing module need to undergo coordinate transformation. GNSS-acquired data uses WGS84 latitude and longitude coordinates, and even after projection transformation, it can only be a planar coordinate system with true north as the north direction. However, our monitoring point layout is divided into several sections, and the specific monitoring results need to be in coordinates with the direction of the section as the north direction. Therefore, coordinate axis rotation is necessary to meet the requirements. Figure 5 .

[0132] It should be noted that all the above embodiments belong to the same inventive concept, and the descriptions of each embodiment have different focuses. Where the description in a particular embodiment is not detailed, please refer to the description in other embodiments.

[0133] The embodiments described above are merely illustrative of implementation methods of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A GNSS data processing method using BeiDou + 5G, comprising the acquisition and server of raw GNSS data within the monitoring area, and the GNSS monitoring software HCMonitor for system control and data processing, characterized in that... In terms of GNSS data processing, advanced nonlinear Kalman filtering double-difference and triple-difference algorithms are adopted. At the same time, an ionospheric correction model is added, supporting multi-reference station calculation and real-time independent baseline network adjustment functions are added. The specific accuracy is less than 2mm for plane and 4mm for elevation. HCMonitor only requires receiving raw observation data from the GNSS receiver; the communication between the software (server) and the GNSS receiver is one-way. HCMonitor supports BDS single-frequency and dual-frequency decoding; If the environment permits (e.g., baseline <10km), BDS single-frequency solution can be selected; to improve the accuracy, stability and reliability of the solution results, GNSS dual-frequency solution can be selected.

2. The GNSS data processing method of BeiDou + 5G according to claim 1, characterized in that, GNSS ambiguity fixing is the core of the software. With the addition of GNSS satellites, the number of ambiguity searches increases exponentially. A unique partial ambiguity search algorithm is adopted.

3. The GNSS data processing method of BeiDou + 5G according to claim 3, characterized in that, GNSS supports mean filters and Kalman filters; The static solution engine assumes that the slowly changing monitoring points are also moving. Based on this, a static solution filtering engine is built. This mode can reflect the displacement changes of each monitoring point in real time, and the accuracy also increases with the increase of observation time. Filtering methods are used to eliminate random errors in GNSS dynamic positioning data, and the true state (positioning result) is estimated optimally in real time from various random disturbances. The discrete state-space model for GNSS dynamic positioning is as follows: ； In the formula for dimensional state vector, For the state 3D nonlinear function, for Dimensional output quantity, for dimensional system transition matrix, , They are respectively , A sequence of dimensional random vectors, satisfying the following statistical properties: ； for A positive semidefinite symmetric matrix. for A positive definite symmetric matrix; Further define the initial state A random vector that satisfies the following statistical properties: ； When a nonlinear system is subjected to a Taylor expansion along its nominal trajectory, approximating its linear portion, and then state-estimated using a standard Kalman filter, the well-known extended Kalman filter is obtained as follows: 。 4. The GNSS data processing method of BeiDou + 5G according to claim 4, characterized in that, The static solution algorithm is used in conjunction with the GNSS raw observations obtained simultaneously by HCMonitor (within 1 microsecond) for differential calculation. The static solution algorithm uses an extended dynamic nonlinear Kalman filter algorithm for differential calculation.

5. The GNSS data processing method of BeiDou + 5G according to claim 4, characterized in that, The static solution algorithm directly applies the raw data from the GNSS receiver, and the observation data from the reference station and the rover station are kept strictly synchronized. The common errors caused by atmospheric delay are offset to the greatest extent.

6. The GNSS data processing method of BeiDou + 5G according to claim 4, characterized in that, HCMonitor only requires receiving raw observation data from the GNSS receiver. Therefore, the software (server) and the GNSS receiver should only need to achieve one-way communication. In HCMonitor mode, the raw data from the rover only needs one serial data communication and one network data communication to reach the data center. The reference station can be directly connected to the data center server.

7. The GNSS data processing method of BeiDou + 5G according to claim 1, characterized in that, The GNSS baseline is formed by synchronous observation data from two receivers. The HCMonitor software integrates real-time dynamic processing, quasi-dynamic processing algorithms, and different filtering and ionospheric correction models. Therefore, it is necessary to set specific data processing methods, correction models, etc., according to the actual needs of the monitoring project, and even include limits on satellite elevation angle and PDOP value. The reference station provides reference coordinates and carrier phase observations for relative positioning calculation in the whole system. By differential processing of the reference station data and the data of each monitoring station, the precise three-dimensional coordinates of each monitoring point are calculated.

8. The GNSS data processing method of BeiDou + 5G according to claim 1, characterized in that, The HCMonitor system employs a hybrid single-frequency / dual-frequency solution method. In GNSS positioning, one or more signals modulated on L1 and L2 are often used to calculate the baseline vector between the points to be determined and to determine the coordinates of the points. The system uses a single-frequency L1 or a single-frequency / dual-frequency L1 and L2 hybrid solution method.

9. The GNSS data processing method of BeiDou + 5G according to claim 1, characterized in that, Baseline processing steps, Baseline solution parameter control Baseline optimization can be achieved by setting control parameters. These parameters mainly include the data sampling interval, cutoff angle, reference satellite and its ionosphere, and the solution model. b Baseline solution self-check Before baseline calculation, the software first sets the baseline calculation control parameters, observation data and ephemeris files, and starting coordinates; c import ephemeris data The ephemeris data portion of the raw data can also be extracted by downloading precise ephemeris data from relevant websites to improve the accuracy of the calculation; d Read in observation data Read the observation data of the raw data, including the observation data of the starting station and the ending station, the single-point positioning coordinates, the observation time, the C / A code pseudorange, the P code pseudorange, and the carrier phase; e Tri-difference solution Subtract the double-difference observations between epochs to form triple-difference observations, establish the observation equation, solve it, and obtain the triple-difference solution; f-cycle slip repair The key to baseline resolution is finding the correct integer ambiguity. The prerequisite for solving the integer ambiguity is that the receiver continuously tracks the carrier phase. The main problem that baseline processing software needs to solve is that the receiver continuously tracks the carrier phase. Obstruction, interference, etc. can cause the tracking of the carrier phase to be interrupted. g performs double-difference floating-point calculations If signals from N satellites are observed, the double-difference observation equation system will have N-1 more unknowns than the triple-difference observation equation system. The double-difference solution yields further unknown point coordinates and integer ambiguities expressed in floating-point numbers. The integer ambiguities should be integers. h Integer fuzziness decomposition The Fast Ambiguity Resolution Approach (FARA) and the LAMBDA method are employed. i performs double-difference fixed solution After the integer ambiguity is correctly fixed, double difference fixed solution is performed.

10. The GNSS data processing method of BeiDou + 5G according to claim 1, characterized in that, GNSS real-time independent baseline network adjustment: The method for adjusting the network in this monitoring system is independent baseline network adjustment, which means that when adjusting the same monitoring point, only the baselines related to the reference station are used.

Images