A cluster GNSS antenna calibration method

Abstract

The present invention discloses a cluster GNSS antenna calibration method, which relates to the field of satellite navigation and positioning technology. Since a large number of GNSS antennas are used and their use range is widely distributed, the mutual influence of the collaborative operation between the antennas is difficult to accurately evaluate and calibrate, and there may be differences in the signals received by different antennas. However, it is difficult for traditional methods to effectively coordinate and unify these differences, resulting in unstable overall system performance. The present invention can compare the influence of actual signal parameters received by each antenna on the positioning accuracy, and perform targeted calibration on the antenna to provide antenna correction parameters. The correction parameters can assist in improving the performance during use, thereby improving the stability of the cluster GNSS system, utilizing a phase winding model in combination with the azimuth of antenna rotation to eliminate phase winding errors, and at the same time weakening the influence of multipath effects through the similarity of signals in a short period of time, thereby effectively improving the calibration accuracy of the antenna phase center.

Description

Technical Field

[0001] The present invention relates to the field of satellite navigation and positioning technology, and in particular to a clustered GNSS antenna calibration method. Background Art

[0002] Clustered GNSS antenna refers to a cluster of multiple global navigation satellite system antennas in one system. Compared with a single antenna, a clustered GNSS antenna contains multiple antenna units, which are usually distributed in a certain area. This enables the system to receive the same satellite signals at the same time, improving the consistency and accuracy of antenna phase center detection.

[0003] In this regard, the patent CN106443744B discloses a method for calibrating and calibrating the attitude of a GNSS dual antenna, including: step 1, obtaining observation data; step 2, calibration of the attitude deflection angle of the GNSS dual antenna, this step further includes: constructing a DCM matrix using the yaw angle, pitch angle and roll angle provided by the inertial navigation; constructing a DCM matrix using the heading angle and pitch angle provided by the GNSS dual antenna to obtain the angle relative relationship matrix between the GNSS dual antenna and the inertial navigation; inversely calculating the difference in heading angle and pitch angle between the GNSS dual antenna and the inertial navigation according to the angle relative relationship matrix to obtain a difference sequence; averaging the difference sequences of the heading angle and pitch angle to obtain the calibration values ​​of the heading deflection angle and the pitch deflection angle. The strapdown inertial navigation system and the GNSS dual antenna are installed on the same platform, and the error angle is calibrated and the data is corrected, which greatly simplifies the calibration process and reduces the difficulty of operation.

[0004] Combining the above patent documents and prior art, the current antenna calibration method is usually only performed on a single antenna, and the calibration process is cumbersome and inefficient. In a cluster system with a large number of antennas, calibrating the antennas one by one is time-consuming and labor-intensive, and it is difficult to meet the needs of efficient calibration in practical applications. Due to the large number of antennas and their wide distribution, the mutual influence between antennas is difficult to accurately evaluate and calibrate, and the signals received by different antennas may be different. Traditional methods are difficult to effectively coordinate and unify these differences, resulting in unstable overall system performance.

[0005] In view of the above problems, a cluster GNSS antenna calibration method is proposed. Summary of the invention

[0006] The purpose of the present invention is to provide a cluster GNSS antenna calibration method, which solves the problem of calibration differences and difficulty in calibration in the background technology.

[0007] To achieve the above object, the present invention provides the following technical solution: a cluster GNSS antenna calibration method, comprising:

[0008] A cluster GNSS antenna calibration method, comprising:

[0009] Step 1: Establish a position reference. Set up a continuous tracking station-type GNSS receiver at the reference point in the GNSS ultra-short baseline field, conduct 24-hour observation, obtain the precise station position through static PPP solution, select the longest baseline side in the GNSS ultra-short baseline field as the directional baseline, and use a 0.5″ class total station to accurately determine the station center coordinates of each station with a coordinate accuracy of better than 1mm. The position coordinates can be used to constrain and verify the baseline solution;

[0010] Step 2: Select the calibration time and select one or more specific time periods;

[0011] Step 3: Synchronously receive the actual signal. In the first stage, one antenna is fixed, and the other antennas are rotated by a precision electric-controlled pan-tilt. The rotation step is 10°, and the antennas rotate within the range of 0-350°. The antennas stay for 10 seconds every 10° for data collection.

[0012] In the second stage, the other antennas were fixed and the antenna that was fixed in the first stage was rotated with a rotation step of 10° within the range of 0-350°, staying at each 10° for 10 seconds for data collection;

[0013] Step 4: Receiving and recording. After each antenna receives the actual signal, it records the signal's arrival time, signal strength, and key parameters of phase observation. The parameters are transmitted to the control computer, which selects the receiver used to record antenna data for time synchronization.

[0014] Step 5: Data analysis and calibration. The central processing unit processes and analyzes the received data, compares the differences between the actual signal parameters received by each antenna, and calibrates the antenna according to the analysis results. The antenna calibration includes adjusting the direction, position, and time matching of the antenna, and eliminating the phase winding error through the phase winding model combined with the azimuth of the antenna rotation. The central processing unit processes and analyzes the received data.

[0015] The antenna phase center is calculated in the ultra-short baseline field. In the short baseline data processing, the following formula can represent any system and frequency point:

[0016] Undifferenced observation equation:

[0017]

[0018] In the formula, is the carrier phase observation value; λ is the wavelength of the carrier phase at the corresponding frequency point; is the geometric distance between the receiver r and the satellite s; c is the speed of light in a vacuum; dt s is the satellite s clock error; dtr is the receiver r clock error; is the ionospheric refraction delay; is the refraction delay in the troposphere; is the integer ambiguity of the carrier phase; is the antenna phase center deviation; The effects of receiver measurement noise and other minor errors in carrier phase observations;

[0019] The single-difference observation equation between stations compares the differences between the actual signal parameters received by each antenna;

[0020] Station single difference observation equation:

[0021]

[0022] Where Δ represents the difference in the phase observations of the same satellite carrier between stations at the same time;

[0023] Using the formula of the station-satellite double difference observation equation, calculate the single difference of the difference between the carrier phase observation values ​​of different satellites at the same time between stations, and then further calculate the difference of the single difference to obtain the station-satellite double difference observation value;

[0024] The station-satellite double difference observation equation is:

[0025]

[0026] In the formula Δ represents the difference between the single differences of the carrier phase observations;

[0027] By linearizing and arranging the double difference observation equation using known approximate coordinates, we can obtain:

[0028]

[0029] Where, (dx dy dz) T is the position parameter to be determined, (lmn) is (dx dy dz) T The partial derivative of is the integer ambiguity parameter to be determined;

[0030] The solution uses an ultra-short baseline field that is precisely known for data collection, and the ionospheric delay can be ignored in the solution. Tropospheric delay The error effect of the phase center delay is λ During the solution, its value can be considered as zero, because it will be absorbed into the coordinates (x, y, z) of the short baseline solution.

[0031] The solution results of each device are unified into the framework of the ultra-short baseline field, eliminating the translation and rotation effects in the results of the single baseline solution;

[0032] Station center coordinate conversion, after the short baseline solution, the coordinates of the measured point need to be marked as Q point conversion value in the station center coordinate system with reference point marked as P as the starting point, recorded as P-NEU, the coordinates of point Q in this coordinate system are according to the formula:

[0033]

[0034] Where B and L are the latitude and longitude of point P; X P , Y P , Z P is the Earth-centered rectangular coordinate of point P, mm; X Q , Y Q , Z Q is the geocentric and earth-fixed rectangular coordinate of point Q, mm; N, E, U are the station center coordinates of point Q, mm;

[0035] The calculated station center coordinates (N, E, U) are used to evaluate the stability and deviation of PCO based on the difference between the same antenna at different time periods and the position relationship of known points;

[0036] Antenna phase center evaluation index, the calculation is expressed as a+b×D, and its standard deviation σ is calculated according to the following formula:

[0037] π=a+b×D

[0038] Where: a is the nominal fixed error, mm; b is the nominal proportional error, mm / km; D ​​is the distance between the measured points, km. When D<0.5km, take D=0.5km;

[0039] Step 6: Iterative optimization. After completing the initial calibration, repeat steps 2, 3, and 4 to further improve the calibration accuracy through multiple iterative optimizations. Adjust the calibration frequency and calibration accuracy according to the actual situation of the clustered GNSS system.

[0040] Preferably, the stability and strength of the GNSS satellite signal are evaluated, and the appropriate number of rotations and interval time are determined by monitoring the signal strength index and the multipath effect, and the range and level of the signal strength index are defined. The signal strength is divided into three levels: strong, medium, and weak, corresponding to different value ranges, and the evaluation method and index of the multipath effect are determined;

[0041] Before calibration, the cluster GNSS antenna is in a stable state. It receives GNSS signals for a period of time, records the average, maximum, minimum, and RMS values ​​of the signal strength during the period, and presets the rotation step and initial interval time. It rotates the antenna, records the data after each rotation, and compares it with the previous data to analyze the changing trends of the signal strength and multipath effect.

[0042] Based on the monitoring data, the changes in signal strength and multipath effect are judged, and the automated algorithm dynamically adjusts the rotation parameters. During calibration, the changes in signal strength and multipath effect are continuously monitored to optimize the number of rotations and interval time.

[0043] Preferably, the signal strength and multipath effect data are statistically analyzed, a curve of signal strength versus time and a curve of RMS value of multipath effect versus number of rotations are plotted, the optimal number of rotations and interval time are determined, the calibration effect is regularly evaluated, and it is checked whether the signal quality meets the requirements.

[0044] Preferably, observation stations are arranged in the ultra-short baseline field to ensure that the distance between the observation stations meets the requirements of the ultra-short baseline. The receivers of each observation station are installed and started to collect data continuously. After the collection is completed, the observation data is processed by static PPP solution software to obtain the precise station location. The longest baseline side is determined from the ultra-short baseline field as the directional baseline, and the station center coordinates of each station are accurately measured using a 0.5″ class total station, and the measurement data is recorded to ensure that the coordinate accuracy is better than 1 mm.

[0045] Preferably, the cluster GNSS system data acquisition denoises the received signal through a digital filter, removes high-frequency noise and random interference, identifies and processes outliers in the signal, and uses a machine learning-based outlier detection algorithm to extract statistical characteristics of the signal strength, including mean, variance, and peak value, and analyze the characteristics of the multipath effect, such as the amplitude and periodic changes of the multipath error.

[0046] Preferably, historical data is analyzed, a mathematical model between signal strength and number of rotations is established, and a root mean square error evaluation index is selected.

[0047] Preferably, a time period with good signal quality is selected as the calibration time, and the arrival time, signal strength, and phase key parameters of the signal are recorded and transmitted to the central processing unit. The central processing unit processes and analyzes the received data, and compares the differences between the actual signal parameters received by each antenna. According to the analysis results, the antenna is calibrated.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] 1. A cluster GNSS antenna calibration method provided by the present invention reduces manual intervention and operation steps in the calibration process by synchronously receiving actual signals and rotating antennas in stages to collect data, thereby greatly improving the calibration efficiency. Secondly, it is able to compare the differences between the actual signal parameters received by each antenna and perform targeted calibration on the antenna to ensure that the signals received by each antenna are consistent and stable, thereby improving the stability of the cluster GNSS system.

[0050] 2. The cluster GNSS antenna calibration method provided by the present invention eliminates the phase winding error by introducing the phase winding model and combining the azimuth of the antenna rotation. In multi-period data collection, it avoids the operation caused by the power off and restart of the equipment, reduces the operation time, and improves the detection efficiency. At the same time, the influence of the multipath effect is weakened by the similarity of the signal in a short time, and the calibration accuracy of the antenna phase center is effectively improved.

[0051] 3. The present invention provides a cluster GNSS antenna calibration method, which introduces a constraint method for the antenna array, establishes a local station center coordinate frame and calibrates the antenna installation position, introduces constraints in the antenna phase center solution, detects solution errors in real time, and ensures the accuracy of the solution process. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a schematic diagram of the ultra-short baseline of the present invention;

[0053] Figure 2 This is a schematic diagram of the station center coordinates of the present invention;

[0054] Figure 3 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0055] 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.

[0056] In order to further understand the content of the present invention, the present invention is described in detail in conjunction with the accompanying drawings.

[0057] Combination Figure 1-Figure 3 A cluster GNSS antenna calibration method of the present invention comprises:

[0058] Step 1: Establish a position reference. Set up a continuous tracking station-type GNSS receiver at the reference point in the GNSS ultra-short baseline field, conduct 24-hour observation, obtain the precise station position through static PPP solution, select the longest baseline side in the GNSS ultra-short baseline field as the directional baseline, and use a 0.5″ class total station to accurately determine the station center coordinates of each station with a coordinate accuracy of better than 1mm. The position coordinates can be used to constrain and verify the baseline solution;

[0059] Step 2: Select the calibration time and select one or more specific time periods;

[0060] Step 3: Synchronously receive the actual signal. In the first stage, one antenna is fixed, and the other antennas are rotated by a precision electric-controlled pan-tilt. The rotation step is 10°, and the antennas rotate within the range of 0-350°. The antennas stay for 10 seconds every 10° for data collection.

[0061] In the second stage, the other antennas were fixed and the antenna that was fixed in the first stage was rotated with a rotation step of 10° within the range of 0-350°, staying at each 10° for 10 seconds for data collection.

[0062] Step 4: Receiving and recording. After each antenna receives the actual signal, it records the signal's arrival time, signal strength, and phase key parameters. The parameters are transmitted to the control computer, and the control computer selects the receiver used to record antenna data for time synchronization.

[0063] Step 5: Data analysis and calibration. The central processing unit processes and analyzes the received data, compares the differences between the actual signal parameters received by each antenna, and calibrates the antenna according to the analysis results. Antenna calibration includes adjusting the direction, position, and time matching of the antenna, and eliminating the phase winding error through the phase winding model combined with the azimuth of the antenna rotation. The central processing unit processes and analyzes the received data;

[0064] The antenna phase center is calculated in the ultra-short baseline field. In the short baseline data processing, the following formula can represent any system and frequency point:

[0065] Undifferenced observation equation:

[0066]

[0067] In the formula, is the carrier phase observation value; λ is the wavelength of the carrier phase at the corresponding frequency point; is the geometric distance between the receiver r and the satellite s; c is the speed of light in a vacuum; dt s is the satellite s clock error; dt r is the receiver r clock error; is the ionospheric refraction delay; is the refraction delay in the troposphere; is the integer ambiguity of the carrier phase; is the antenna phase center deviation; The effects of receiver measurement noise and other minor errors in carrier phase observations;

[0068] Inter-station difference observation equation:

[0069]

[0070] Where Δ represents the difference in the phase observations of the same satellite carrier between stations at the same time;

[0071] Using the formula of the station-satellite double difference observation equation, calculate the single difference of the difference between the carrier phase observation values ​​of different satellites at the same time between stations, and then further calculate the difference of the single difference to obtain the station-satellite double difference observation value;

[0072] The station-satellite double difference observation equation is:

[0073]

[0074] In the formula Δ represents the difference between the single differences of the carrier phase observations;

[0075] By linearizing and arranging the double difference observation equation using known approximate coordinates, we can obtain:

[0076]

[0077] Where, (dx dy dz) T is the position parameter to be determined, (lmn) is (dx dy dz) T The partial derivative of is the integer ambiguity parameter to be determined;

[0078] The solution uses an ultra-short baseline field that is precisely known for data collection, and the ionospheric delay can be ignored in the solution. Tropospheric delay The error effect of the phase center delay is λ During the solution, its value can be considered as zero, because it will be absorbed into the coordinates (x, y, z) of the short baseline solution.

[0079] The data is solved and the solution results of each device are unified into the framework of the ultra-short baseline field to eliminate the translation and rotation effects in the results of single baseline solution;

[0080] Station center coordinate conversion, after the short baseline solution, the coordinates of the measured point need to be marked as Q point conversion value in the station center coordinate system with reference point marked as P as the starting point, recorded as P-NEU, the coordinates of point Q in this coordinate system are according to the formula:

[0081]

[0082] Where B and L are the latitude and longitude of point P; X P , Y P , Z P is the Earth-centered rectangular coordinate of point P, mm; X Q , Y Q , Z Q is the geocentric and earth-fixed rectangular coordinate of point Q, mm; N, E, U are the station center coordinates of point Q, mm;

[0083] The calculated station center coordinates (N, E, U) are used to evaluate the stability and deviation of PCO based on the difference between the same antenna at different time periods and the position relationship of known points;

[0084] Antenna phase center evaluation index, the calculation is expressed as a+b×D, and its standard deviation σ is calculated according to the following formula:

[0085] π=a+b×D

[0086] Where: a is the nominal fixed error, mm; b is the nominal proportional error, mm / km; D ​​is the distance between the measured points, km. When D<0.5km, take D=0.5km;

[0087] Evaluation and calculation of the antenna phase center; the coordinate deviation of each point from the reference station center in the reference frame coordinates is known to be (N0, E0, U0); the reference station center (N0, E0, U0) obtained by baseline solution and adjustment calculation in four periods i ,E i ,U i )The average coordinate is (N ave ,E ave ,U ave ), the final phase center deviation PCO can be obtained, the formula is as follows:

[0088]

[0089] Calculate the total deviation S of the Euclidean distance PCO as follows:

[0090]

[0091] The maximum value and the difference between the maximum values ​​in the four time periods are taken as the detection result of the consistency of the antenna phase center (N t ,E t ,U t ), calculated as follows:

[0092]

[0093] For effective testing, an ultra-short baseline field should be used, and each point should be free of obstruction and electromagnetic interference at a viewing angle of more than 15 degrees. Select six points A, B, C, D, E, and F. Obtain the precise point coordinates of these points in the global frame and the accurate relative position of each point to the rest of the points (N, E, U);

[0094] The calibration time is selected when the signal quality is good. The signal arrival time, signal strength, and phase key parameters are recorded and transmitted to the central processing unit. The central processing unit processes and analyzes the received data and compares the differences between the actual signal parameters received by each antenna. Based on the analysis results, the antenna is calibrated.

[0095] Step 6: Iterative optimization. After completing the initial calibration, repeat steps 2, 3, and 4 to further improve the calibration accuracy through multiple iterative optimizations. Adjust the calibration frequency and calibration accuracy according to the actual situation of the clustered GNSS system.

[0096] In a cluster GNSS system with 10 antennas, first select a time period with good signal quality as the calibration time. Then, each antenna in the cluster synchronously receives the actual signal from the GNSS satellite during the selected calibration time period. After each antenna receives the actual signal, it records the signal's arrival time, signal strength, and phase parameters, and transmits these parameters to the central processing unit. The central processing unit processes and analyzes the received data and compares the differences between the actual signal parameters received by each antenna. Based on the analysis results, the antenna is calibrated. After completing the preliminary calibration, the above steps can be repeated for iterative optimization until the preset calibration accuracy requirements are met;

[0097] In the process of handling GNSS system antenna calibration, the key steps include signal reception and recording, data processing and analysis, and iterative optimization of the antenna. First, the signal parameters received by each antenna, such as arrival time, signal strength and phase, are recorded and synchronously transmitted to the control computer. Subsequently, the central processing unit conducts in-depth analysis of the data through the station single difference observation equation and the station-satellite double difference observation equation, which involves calculating the signal difference obtained by different satellites and ignoring the ionospheric delay and tropospheric delay errors, because these have little effect on the precise data acquisition of the ultra-short baseline field.

[0098] During the solution process, the Earth-centered Earth-fixed coordinate system and the station-centered coordinate system are converted to ensure positioning accuracy. In addition, the antenna calibration is further optimized by calculating the deviation and stability of the antenna phase center. Finally, through multiple iterative optimizations, the direction, position, and time matching of the antenna are adjusted to improve the calibration accuracy and reliability of the entire system. This process requires not only precise mathematical models and algorithms, but also relies on professional equipment and strict operating procedures to ensure that the final positioning results are both accurate and stable.

[0099] In addition, by recording the GNSS signal strength value received after each rotation, the average value is used as a statistic, and by comparing it with the known standard position, the calibrated position error is calculated to analyze the amplitude and phase changes of the multipath error;

[0100] The average signal strength, calibration accuracy RMSE and multipath effect RMS of each group are shown in the following table:

[0101]

[0102]

[0103] Evaluate the stability and strength of GNSS satellite signals, determine the appropriate number of rotations and intervals by monitoring signal strength indicators and multipath effects, define the range and level of signal strength indicators, divide signal strength into three levels: strong, medium, and weak, corresponding to different value ranges, and determine the evaluation method and indicators of multipath effects;

[0104] Before calibration, the cluster GNSS antenna is in a stable state. It receives GNSS signals for a period of time, records the average, maximum, minimum, and RMS values ​​of the signal strength during the period, and presets the rotation step and initial interval time. It rotates the antenna, records the data after each rotation, and compares it with the previous data to analyze the changing trends of the signal strength and multipath effect.

[0105] According to the monitoring data, the changes in signal strength and multipath effect are judged, and the automatic algorithm dynamically adjusts the rotation parameters. During calibration, the changes in signal strength and multipath effect are continuously monitored to optimize the number of rotations and interval time;

[0106] Based on the monitored data, determine the changes in signal strength and multipath effect. If the signal strength drops significantly or the multipath effect increases seriously, you can appropriately reduce the number of rotations or increase the interval time. If the signal quality is good, consider increasing the number of rotations or shortening the interval time. During the entire calibration process, continuously monitor the changes in signal strength and multipath effect, and continuously optimize the number of rotations and interval time.

[0107] Perform statistical analysis on signal strength and multipath effect data, draw curves of signal strength versus time and multipath effect RMS value versus rotation times, determine the best rotation times and interval time, evaluate calibration results regularly, and check whether signal quality meets requirements;

[0108] Arrange observation stations in the ultra-short baseline field, ensure that the distance between observation stations meets the requirements of ultra-short baseline, install and start the receivers of each observation station, and conduct continuous data collection. After the collection is completed, the observation data is processed by static PPP solution software to obtain the precise station location, determine the longest baseline side from the ultra-short baseline field as the directional baseline, and use a 0.5″ class total station to accurately measure the station center coordinates of each station, record the measurement data and ensure that the coordinate accuracy is better than 1mm;

[0109] The data acquisition of the cluster GNSS system uses digital filters to denoise the received signals, remove high-frequency noise and random interference, identify and process outliers in the signals, and use machine learning-based outlier detection algorithms to extract statistical features of signal strength, including mean, variance, and peak values, and analyze the features of multipath effects, such as the amplitude and phase changes of multipath errors.

[0110] Analyze historical data, establish a mathematical model between signal strength and number of rotations, and select the root mean square error evaluation index. It is used to evaluate the calibration accuracy and multipath effect. The formula is:

[0111]

[0112] where y i is the observed value, y true is the true value, and n is the number of observations.

[0113] For example:

[0114] According to the collected signals, the signal strength is divided into three levels: strong, medium and weak. Signal strength greater than -125dBm is defined as strong, -125dBm to -135dBm is defined as medium, and less than -135dBm is defined as weak. The evaluation method and index for determining the multipath effect is to use the root mean square value (RMS) of the multipath error as the evaluation index. An RMS value less than 0.2 meters is a slight multipath effect, 0.2 meters to 0.4 meters is a medium multipath effect, and greater than 0.4 meters is a severe multipath effect;

[0115] Before calibration, let the cluster GNSS antenna be in a stable state and receive GNSS signals for 30 minutes. During this time, record the average, maximum, minimum and RMS values ​​of the signal strength and multipath effect. These data will serve as an initial reference for subsequent comparison with the data after rotating the antenna;

[0116] The preset rotation step is 10°, and the initial interval is 10 seconds. After starting to rotate the antenna, pause for a while after each rotation to allow the system to record the data on signal strength and multipath effect. Compare the data after each rotation with the previous data to analyze the changing trends of signal strength and multipath effect. If the signal strength drops from a strong level to a medium level, or the RMS value of the multipath effect more than doubles, appropriately reduce the number of rotations or increase the interval time. Conversely, if the signal quality is good, consider increasing the number of rotations or shortening the interval time. During the entire calibration process, the signal is fully monitored and analyzed every 5 minutes. Based on the monitoring results, the number of rotations and interval time are continuously optimized to ensure that mechanical wear on the antenna is minimized while ensuring calibration accuracy;

[0117] Perform statistical analysis on the monitored signal strength and multipath effect data, and draw a curve of signal strength changing over time. This will allow you to intuitively see the fluctuations in signal strength at different time points. At the same time, draw a curve of the RMS value of the multipath effect changing with the number of rotations to understand the relationship between the multipath effect and antenna rotation. The selection of optimal parameters should take into account factors such as signal quality, calibration accuracy, and efficiency. If, under a certain combination of rotation times and interval time, the signal strength is always maintained at a relatively strong level, the RMS value of the multipath effect is also small, and the calibration accuracy meets the requirements, then this combination can be considered optimal. Regularly evaluate the calibration effect to check whether the signal quality meets the requirements. Set the evaluation indicators to be the stability of the signal strength, the degree of control of the multipath effect, and the maintenance of the calibration accuracy. If it is found that the signal quality has deteriorated or the calibration accuracy does not meet the requirements, it is necessary to re-monitor the signal and optimize the parameters;

[0118] When arranging observation stations in an ultra-short baseline field, ensure that the distance between the observation stations is within 30 meters to meet the requirements of the ultra-short baseline. According to the actual site conditions, the location of the observation station should be reasonably selected so that the distance between each observation station can not only meet the definition of the ultra-short baseline, but also cover the range of the GNSS antenna that needs to be calibrated. Ensure that the location of the observation station is selected in a place with a panoramic elevation angle of more than 15 degrees without obstruction and electromagnetic interference, install and start the receiver of each observation station for continuous data collection. The receiver should have high precision, high stability and real-time data output functions so that it can accurately record the changes in GNSS signals. The data collection time is 24 hours to obtain sufficient data for subsequent processing and analysis. During the collection process, ensure that the working state of the receiver is stable to avoid equipment failure or data loss;

[0119] After the acquisition is completed, the observation data is processed by static PPP solution software. Static PPP solution is a high-precision processing method that can use the observation data of the global navigation satellite system to achieve high-precision solution of the observation point position through accurate estimation of satellite orbits and clock errors. Through static PPP solution, the precise site location can be obtained. Determine the longest baseline side from the ultra-short baseline field as the directional baseline, which will be used as a reference for subsequent coordinate measurement and calibration. Use a 0.5″ class total station to accurately measure the station center coordinates of each site. During the measurement process, strictly follow the measurement specifications to ensure the accuracy and reliability of the measurement data. Record the measurement data and ensure that the coordinate accuracy is better than 1mm;

[0120] During the data collection process of the cluster GNSS system, the received signal is denoised through a median filter. An outlier detection algorithm based on machine learning is used to identify and process outliers in the signal. Through the machine learning algorithm, outliers can be automatically identified, and corresponding processing measures can be taken to delete outliers or perform interpolation processing; statistical characteristics of signal strength can be extracted, including mean, variance, and peak value. The characteristics of the multipath effect are analyzed, such as the amplitude and phase change of the multipath error. Correlation analysis and spectrum analysis are used to extract the characteristics of the multipath effect, and historical data are analyzed. A linear regression model is used to establish a mathematical model between signal strength and the number of rotations.

[0121] Linear regression or polynomial regression is used to establish the relationship between signal strength and the number of rotations. The linear regression model can be expressed as:

[0122] y=β0+β1x+∈

[0123] Where y is the signal strength, x is the number of rotations, β0 and β1 are model parameters, and ∈ is the error value.

[0124] The signal strength is used as the dependent variable and the number of rotations as the independent variable to establish a linear relationship between the two. The root mean square error is selected as the evaluation indicator to measure the prediction accuracy of the model. By calculating the root mean square error, the accuracy and reliability of the model can be evaluated, providing a basis for subsequent parameter optimization.

[0125] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0126] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A cluster GNSS antenna calibration method, characterized in that: include: Step 1: Set up a continuously operating reference station type GNSS receiver at the GNSS ultra-short baseline field benchmark point, observe for 24 hours, calculate the station position by static PPP, select the longest baseline side as the directional baseline, and use a total station to measure the station center coordinates with an accuracy better than 1mm; Step 2: Select the calibration time and select one or more specific time periods; Step 3: Rotate the antenna to collect data in two stages and receive the actual signal synchronously. In the first stage, fix one antenna and use the precision electric-controlled pan-tilt to rotate the other antennas. The rotation step is 10° and the antenna rotates within the range of 0-350°. The antenna stays for 10 seconds every 10° to collect data. In the second stage, the other antennas were fixed and the antenna that was fixed in the first stage was rotated with a rotation step of 10° within the range of 0-350°, staying at each 10° for 10 seconds for data collection; Step 4: Receiving and recording. After each antenna receives the actual signal, it records the signal's arrival time, signal strength, and key parameters of phase observation. The parameters are transmitted to the control computer, which selects the receiver used to record antenna data for time synchronization. Step 5: Data analysis and calibration. The central processing unit processes and analyzes the received data, establishes the satellite data time series and triangulation network, eliminates the phase winding error through the phase winding model combined with the azimuth of the antenna rotation, solves each ultra-short baseline, compares the differences between the actual signal parameters received by each antenna in each period, and gives the antenna calibration parameters according to the analysis results. The antenna calibration includes adjusting the direction, position, and time matching of the antenna. The central processing unit processes and analyzes the received data. Step 6: After the initial calibration, perform iterative optimization to adjust the calibration frequency and accuracy. After completing the initial calibration, repeat steps 2, 3, and 4. Further improve the calibration accuracy through multiple iterative optimizations. Adjust the calibration frequency and calibration accuracy according to the actual situation of the clustered GNSS system.

2. A cluster GNSS antenna calibration method according to claim 1, characterized in that: The antenna phase center is carried out in the ultra-short baseline field. In the short baseline data processing, the following formula can represent any system and frequency point; Undifferenced observation equation: In the formula, is the carrier phase observation value; λ is the wavelength of the carrier phase at the corresponding frequency point; is the geometric distance between the receiver r and the satellite s; c is the speed of light in a vacuum; dt s is the satellite s clock error; dt r is the receiver r clock error; is the ionospheric refraction delay; is the refraction delay in the troposphere; is the integer ambiguity of the carrier phase; is the antenna phase center deviation; The effects of receiver measurement noise and other minor errors in carrier phase observations; The single-difference observation equation between stations compares the differences between the actual signal parameters received by each antenna; Station single difference observation equation: Where Δ represents the difference in the phase observations of the same satellite carrier between stations at the same time; Using the formula of the station-satellite double difference observation equation, calculate the single difference of the difference between the carrier phase observation values ​​of different satellites at the same time between stations, and then further calculate the difference of the single difference to obtain the station-satellite double difference observation value; The station-satellite double difference observation equation is: In the formula represents the difference between single differences of carrier phase observations; By linearizing and arranging the double difference observation equation using known approximate coordinates, we can obtain: Where, (dx dy dz) T is the position parameter to be determined, (lmn) is (dx dy dz) T The partial derivative of is the integer ambiguity parameter to be determined; The solution uses an ultra-short baseline field that is precisely known for data collection, and the ionospheric delay can be ignored in the solution. Tropospheric delay The error effect of the phase center delay is λ During the solution, its value can be considered to be zero, because it will be absorbed into the coordinates (x, y, z) of the short baseline solution. The solution results of each device are unified into the framework of the ultra-short baseline field, eliminating the translation and rotation effects in the results of the single baseline solution; Station center coordinate conversion, after the short baseline solution, the coordinates of the measured point need to be marked as Q point conversion value in the station center coordinate system with reference point marked as P as the starting point, recorded as P-NEU, the coordinates of point Q in this coordinate system are according to the formula: Where B and L are the latitude and longitude of point P; X P , Y P , Z P is the Earth-centered rectangular coordinate of point P in earth-fixed space, mm; X Q , Y Q , Z Q is the geocentric and earth-fixed rectangular coordinate of point Q, mm; N, E, U are the station center coordinates of point Q, mm; The calculated station center coordinates (N, E, U) are used to evaluate the stability and deviation of PCO based on the difference between the same antenna at different time periods and the position relationship of known points; Antenna phase center evaluation index, the calculation is expressed as a+b×D, and its standard deviation σ is calculated according to the following formula: σ=a+b×D Where: a is the nominal fixed error, mm; b is the nominal proportional error, mm / km; D ​​is the distance between the measured points, km. When D<0.5km, take D=0.5km.

3. The cluster GNSS antenna calibration method according to claim 1, characterized in that: Evaluate the stability and strength of GNSS satellite signals, determine the appropriate number of rotations and intervals by monitoring signal strength indicators and multipath effects, define the range and level of signal strength indicators, divide signal strength into three levels: strong, medium, and weak, corresponding to different value ranges, and determine the evaluation method and indicators of multipath effects; Before calibration, the cluster GNSS antenna is in a stable state. It receives GNSS signals for a period of time, records the average, maximum, minimum, and RMS values ​​of the signal strength during the period, and presets the rotation step and initial interval time. It rotates the antenna, records the data after each rotation, and compares it with the previous data to analyze the changing trends of the signal strength and multipath effect. Based on the monitoring data, the changes in signal strength and multipath effect are judged, and the automated algorithm dynamically adjusts the rotation parameters. During calibration, the changes in signal strength and multipath effect are continuously monitored to optimize the number of rotations and interval time.

4. The cluster GNSS antenna calibration method according to claim 3, characterized in that: Perform statistical analysis on signal strength and multipath effect data, draw curves showing changes in signal strength over time and changes in the RMS value of the multipath effect over the number of rotations, determine the optimal number of rotations and interval time, evaluate the calibration effect regularly, and check whether the signal quality meets the requirements.

5. A cluster GNSS antenna calibration method according to claim 4, characterized in that: Arrange observation stations in the ultra-short baseline field to ensure that the distance between observation stations meets the requirements of ultra-short baselines. Install and start the receivers of each observation station to conduct continuous data collection. After the collection is completed, the observation data is processed by static PPP solution software to obtain the precise station location. The longest baseline side in the ultra-short baseline field is determined as the directional baseline, and the station center coordinates of each station are accurately measured using a 0.5″ class total station. The measurement data is recorded and the coordinate accuracy is ensured to be better than 1mm. The calibrated coordinates are used to establish the ultra-short baseline network constraint framework in the station center coordinate system to constrain the GNSS baseline solution.

6. A cluster GNSS antenna calibration method according to claim 5, characterized in that: The cluster GNSS system data acquisition uses digital filters to denoise the received signals, remove high-frequency noise and random interference, identify and process outliers in the signals, and use machine learning-based outlier detection algorithms to extract statistical features of signal strength, including mean, variance, and peak values, and analyze the features of multipath effects, such as the amplitude and periodic changes of multipath errors.

7. A cluster GNSS antenna calibration method according to claim 6, characterized in that: The historical data were analyzed, a mathematical model between signal strength and number of rotations was established, and the root mean square error evaluation indicator was selected.

8. The cluster GNSS antenna calibration method according to claim 7, characterized in that: A time period with good signal quality is selected as the calibration time, and the signal arrival time, signal strength, and phase key parameters are recorded. These parameters are transmitted to the central processing unit, which processes and analyzes the received data, compares the differences between the actual signal parameters received by each antenna, and calibrates the antenna based on the analysis results.

9. A cluster GNSS antenna calibration method according to claim 1, characterized in that: A GNSS antenna phase winding correction algorithm has been introduced. In conventional detection, when the GNSS antenna changes, it is necessary to cut off the power and restart data collection. By using the GNSS antenna phase winding correction algorithm, the signal can be tracked continuously, reducing the time cost of restarting the power supply and improving the verification efficiency.

10. The cluster GNSS antenna calibration method according to claim 1, characterized in that: Multiple antennas installed in the ultra-short baseline field use pre-calibrated precise coordinates to solve constraints, improve the accuracy of antenna phase center solution, use programmable control motors to achieve coordinated rotation of antennas, improve rotation efficiency, and ensure that the antenna phase center constraint model always corresponds to the antenna attitude.

Images