A high-precision GNSS data processing method that takes into account shade from dense forests

Abstract

The present invention discloses a GNSS high-precision data processing method that takes into account the obstruction of dense forest shade, which relates to the technical field of GNSS data processing. The specific method comprises: step 1, dividing time periods, and determining the satellite signal data observed by the GNSS receiver in each time period; step 2, judging whether the GNSS observation area in different time periods belongs to the tree shade obstruction area; step 3, preliminarily processing the GNSS data, and screening and eliminating the sampling points in the tree shade obstruction area; step 4, completing the data at the vacant sampling points based on conventional data, and then performing high-precision data processing; based on the present invention, the demand for high-precision measurement can be met even in complex environments such as dense forest shade, so that the GNSS high-precision positioning technology can be applied in more complex environments, which not only improves the positioning accuracy and data reliability, but also expands the application scope of the GNSS technology.

Description

Technical Field

[0001] The present invention belongs to the technical field of GNSS data processing, and in particular relates to a GNSS high-precision data processing method that takes into account shade from dense forests. Background Art

[0002] The Global Navigation Satellite System (GNSS) covers a variety of global and regional satellite navigation systems, such as the United States' Global Positioning System (GPS), China's Beidou Navigation Satellite System (BDS), Russia's Global Navigation Satellite System (GLONASS), Europe's Galileo Navigation Satellite System (Galileo), and Japan's Quasi-Zenith Satellite System (QZSS). Together, they weave a positioning, navigation and timing service network covering the world or specific regions, and are widely used in various industries.

[0003] GNSS applications often face challenges in complex environments, particularly those shrouded by dense forests and tree cover. In these areas, dense trees not only block some satellite signals but also introduce multipath effects through reflection and refraction, leading to signal interference, interruptions, and a reduction in the number of visible satellites. This, in turn, affects the geometric distribution of the satellite constellation, introduces significant gross errors into GNSS positioning, and severely impacts the accuracy and reliability of parameter estimation.

[0004] As users' expectations for the real-time, accuracy, and reliability of GNSS positioning services continue to grow, how to efficiently and accurately handle the impact of the environment on the accuracy of observation data has become a key technical challenge that needs to be solved urgently, especially in complex scenarios such as multi-frequency and multi-system fusion applications and shade from dense forests. Summary of the Invention

[0005] The purpose of the present invention is to provide a GNSS high-precision data processing method that takes into account the obstruction of dense forest shade, which is used to solve the technical problem in the existing technology that it is difficult to achieve GNSS high-precision data processing when taking into account the obstruction of dense forest shade.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] Step 1: Divide the time periods and determine the satellite signal data observed by the GNSS receiver in each time period; divide the day into different time periods of equal length, determine the length of each time period as t, and classify the time periods into morning, noon, and dusk according to the changes in tree shade over time. At the same time, set the GNSS to obtain observation data in a dense forest environment, record the satellite signal data in each time period, and record it as dense forest data. The satellite signal data includes: the number of satellites, satellite signal strength, and observed satellite numbers. The satellite number set of the open data in n time periods includes the satellite number set of the dense forest data.

[0008] Step 2: Determine whether the GNSS observation area at different time periods is shaded by trees. Obtain GNSS data from n time periods in a real forest environment. Simulate the real forest environment to establish a simulation environment. Use the simulation experiment to set the shaded area to have no effect on the accuracy of the GNSS observation data. Simulate each time period to obtain open data. Determine the true values ​​of the forest data and open data in a certain time period, and obtain the observation factors of the forest data and open data in the time period. The calculation formula of the observation factor G is as follows:

[0009]

[0010] Among them, K m Indicates the size of the time period type factor, m indicates the type of time period, which includes morning, noon and dusk, i indicates that the number of satellites observed is i, N indicates that a total of N satellites are observed in this time period, S i Indicates the signal strength of satellite number i within time t.

[0011] Based on the calculation formula of the observation factor, the observation factor of the dense forest data in time period a is obtained, recorded as the dense forest factor, and the observation factor of the open data in time period a is recorded as the open factor. The ratio g of the dense forest factor to the open factor is calculated, and the shade shielding threshold L is set. When g is less than or equal to L, the GNSS observation area in time period a is judged to belong to the shade shielding area. Otherwise, the GNSS observation area in time period a is judged to belong to the open area. L is set according to actual needs and specific conditions, and all GNSS data belonging to the shade shielding area in different time periods are collected.

[0012] Step 3: Preliminary processing of GNSS data, screening and eliminating sampling points in the area blocked by tree shade; filtering the received GNSS data, using digital filtering technology to remove noise interference, such as: Kalman filtering, wavelet transform or median filtering, etc., through digital filtering technology to suppress random noise, retain useful information in the signal, and improve the signal-to-noise ratio of the data. After filtering, the smoothness and stability of GNSS data are significantly improved, and the GNSS data is verified in real time to verify the integrity of the data and eliminate erroneous data, including: checking the integrity of the data packet, the continuity of the timestamp and the standardization of the data format, using redundant observations and internal consistency check methods to identify and eliminate potential erroneous data or abnormal data, and correcting problems that may cause data deviations, such as clock error, atmospheric environment and ionospheric delay, using the satellite clock error information provided by the precise ephemeris, through The influence of clock error is eliminated by over-differentiation technology. The atmospheric environment deviation is eliminated by constructing an atmospheric delay model and estimating it using meteorological parameters and observation data. The deviation of ionospheric delay is further weakened by using multi-frequency observation data and linear combination technology. The changing trend of GNSS data in each time period is obtained through simulation experiments. The sampling frequency is determined based on the changing trend of GNSS data in each time period. The signal quality of GNSS data in different time periods in the tree-shaded area is evaluated according to the sampling point. The signal-to-noise ratio and cycle slip indicators of the signal are evaluated, and the signal-to-noise ratio threshold and cycle slip threshold are set. When the signal-to-noise ratio or cycle slip of the signal is greater than or equal to the corresponding threshold, the data of the sampling point is eliminated to obtain a vacant sampling point, and the data that is not eliminated is recorded as regular data. The signal-to-noise ratio threshold is obtained by integrating historical GNSS Maolin data, and the cycle slip threshold is obtained by screening outliers or mutation points in the data sequence in the GNSS data.

[0013] Step 4: Complete the data at the missing sampling points based on conventional data, and then perform high-precision data processing; collect the data of the first N sampling points of each missing sampling point to form a prediction data set, complete each missing sampling point in sequence according to the time series, extract the data feature values ​​of the sampling points, such as pseudorange observation values ​​and carrier phase observation values, use trend extrapolation to predict the data feature values ​​of the missing sampling points, and predict the data of the N+1th sampling point from the data of the first N sampling points by establishing a polynomial model. The polynomial model is specifically shown as follows:

[0014]

[0015] Where M(N+1) represents the characteristic data value of the N+1th sampling point, {a, a′,…, a (x)} represents the model coefficient, x represents the highest order of the polynomial model, and the range of the model coefficient is determined by comprehensively considering the data change trends of the N sampling points before the vacant sampling point. By predicting the data of the non-vacant sampling points, the true value is compared with the predicted value to determine the prediction accuracy of the polynomial model at different orders, thereby determining the highest order of the polynomial model.

[0016] The data in the missing sampling points are completed by a polynomial model, and complete GNSS data is obtained after all the missing sampling points are completed. For the complete GNSS data, time correlation analysis is used to assist in the detection of gross errors in the data, and an interpolation method is used to process the gross errors. The interpolation method includes: a classical spatiotemporal interpolation algorithm and an interpolation method based on a neural network. The GNSS data is baseline solved, and the obtained baseline solution results are adjusted. The accuracy of the baseline solution is adjusted through error analysis. The data processing process is monitored in real time, and the GNSS data quality after each data processing step is evaluated, for example, the baseline vector correction number, the constrained adjustment baseline vector correction number and the unconstrained baseline vector correction number of the same name are checked, and the correctness of the data processing steps is monitored in real time.

[0017] Furthermore, precise ephemeris is used to reduce the impact of GNSS satellite orbit errors on the accuracy of baseline solutions. Multi-band observation data is used for integer solutions to improve the redundancy of observation data, separate and eliminate errors such as ionospheric delay, and fix the ambiguity parameters to integers through integer solutions to improve the accuracy of baseline solutions. Multi-band observation data is used to constrain the solution of ambiguities and improve the stability of integer solutions. During the GNSS network adjustment process, an iterative algorithm is used to gradually approach the optimal solution, eliminating the probability that the optimal solution cannot be obtained when solving the adjustment equation. The iterative algorithm continuously adjusts the parameter estimates and recalculates the residual and covariance matrices until certain convergence conditions are met.

[0018] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0019] 1. Through detailed time period division, the present invention can accurately identify which areas are most affected by tree shade and which areas are least affected by tree shade. For time periods with less tree shade, the present invention can fully utilize these high-quality observation data to improve the accuracy and reliability of overall data processing. By identifying and eliminating data from time periods with greater tree shade effects, the data processing workload is reduced and processing efficiency is improved.

[0020] 2. The present invention adjusts the sampling frequency to ensure that sufficient data can be obtained even in obscured areas. Based on high-quality prediction and analysis of non-obstructed data, the present invention fills in the data segments that are eliminated due to tree shade, thereby maintaining the continuity and integrity of the data and reducing the impact of data mutations or outliers on GNSS data accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0022] Figure 1 A flow chart of a method for processing high-precision GNSS data taking into account shade from dense forests is shown;

[0023] Figure 2 A step diagram of a data processing method for improving the accuracy of complete data is shown. DETAILED DESCRIPTION

[0024] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0025] like Figure 1 、 Figure 2 As shown, this embodiment proposes a method for processing GNSS high-precision data that takes into account shade from dense forests. The method specifically includes:

[0026] Step 1: Divide the time periods and determine the satellite signal data observed by the GNSS receiver in each time period; divide the day into different time periods of equal length, determine the length of each time period as t, and classify the time periods into morning, noon, and dusk according to the changes in tree shade over time. At the same time, set the GNSS to obtain observation data in a dense forest environment, record the satellite signal data in each time period, and record it as dense forest data. The satellite signal data includes: the number of satellites, satellite signal strength, and observed satellite numbers. The satellite number set of the open data in n time periods includes the satellite number set of the dense forest data.

[0027] Step 2: Determine whether the GNSS observation area at different time periods is shaded by trees. Obtain GNSS data from n time periods in a real forest environment. Simulate the real forest environment to establish a simulation environment. Use the simulation experiment to set the shaded area to have no effect on the accuracy of the GNSS observation data. Simulate each time period to obtain open data. Determine the true values ​​of the forest data and open data in a certain time period, and obtain the observation factors of the forest data and open data in the time period. The calculation formula of the observation factor G is as follows:

[0028]

[0029] Among them, K m Indicates the size of the time period type factor, m indicates the type of the time period, and the time period types include: morning, noon and evening. In this embodiment, the time period type factors are: when the time period belongs to morning, K m =1.1, time period belongs to noon K m =0.8, the time period belongs to dusk K m =1, i means the number of the observed satellite is i, N means a total of N satellites were observed in this time period, S i Indicates the signal strength of satellite number i within time t.

[0030] Based on the calculation formula of the observation factor, the observation factor of the dense forest data in time period a is obtained, recorded as the dense forest factor, and the observation factor of the open data in time period a is recorded as the open factor. The ratio g of the dense forest factor to the open factor is calculated, and the shade shielding threshold L is set. When g is less than or equal to L, the GNSS observation area in time period a is judged to belong to the shade shielding area. Otherwise, the GNSS observation area in time period a is judged to belong to the open area. L is set according to actual needs and specific conditions, and all GNSS data belonging to the shade shielding area in different time periods are collected.

[0031] Step 3: Preliminary processing of GNSS data, screening and eliminating sampling points in the area blocked by tree shade; filtering the received GNSS data, using digital filtering technology to remove noise interference, such as: Kalman filtering, wavelet transform or median filtering, etc., through digital filtering technology to suppress random noise, retain useful information in the signal, and improve the signal-to-noise ratio of the data. After filtering, the smoothness and stability of the GNSS data are significantly improved. Real-time verification of GNSS data is performed to verify the integrity of the data and eliminate erroneous data, including: checking the integrity of the data packet, the continuity of the timestamp and the standardization of the data format, using redundant observations and internal consistency checks, etc., to identify and eliminate potential erroneous data or outliers, and to correct problems that may cause data deviations, such as: clock error, atmospheric environment and ionospheric delay. Clock error correction uses the satellite clock error information provided by the precise ephemeris. The influence of clock error is eliminated through differential technology. The atmospheric environment error is eliminated by constructing an atmospheric delay model and estimating the error using meteorological parameters and observation data. The influence of ionospheric delay is further weakened by using multi-frequency observation data and linear combination technology. The changing trend of GNSS data in each time period is obtained through simulation experiments. The sampling frequency is determined based on the changing trend of GNSS data in each time period. The signal quality of GNSS data in the tree-shaded area in different time periods is evaluated according to the sampling point. The signal-to-noise ratio and cycle slip indicators of the signal are evaluated, and the signal-to-noise ratio threshold and cycle slip threshold are set. When the signal-to-noise ratio or cycle slip of the signal is greater than or equal to the corresponding threshold, the data of the sampling point is eliminated to obtain a vacant sampling point, and the data that is not eliminated is recorded as regular data. The signal-to-noise ratio threshold is obtained by integrating historical GNSS Maolin data, and the cycle slip threshold is obtained by screening outliers or mutation points in the data sequence in the GNSS data.

[0032] Step 4: Based on conventional data, fill in the data at the missing sampling points and then perform high-precision data processing; collect the data of the first N sampling points of each missing sampling point to form a prediction data set, fill in each missing sampling point in sequence according to the time series, extract the data features of the sampling points, such as pseudorange observations and carrier phase observations, use trend extrapolation to predict the data feature values ​​of the missing sampling points, and establish a polynomial model to predict the data of the N+1th sampling point from the data of the first N sampling points. The polynomial model is specifically shown as follows:

[0033]

[0034] Where M(N+1) represents the characteristic data value of the N+1th sampling point, {a, a′,…, a (x)} represents the model coefficient, x represents the highest order of the polynomial model, and the range of the model coefficient is determined by comprehensively considering the data change trends of the N sampling points before the vacant sampling point. By predicting the data of the non-vacant sampling points, the true value is compared with the predicted value to determine the prediction accuracy of the polynomial model at different orders, thereby determining the highest order of the polynomial model.

[0035] The data in the missing sampling points are completed by a polynomial model, and complete GNSS data is obtained after all the missing sampling points are completed. For the complete GNSS data, time correlation analysis is used to assist in the detection of gross errors in the data, and an interpolation method is used to process the gross errors. The interpolation method includes: a classical spatiotemporal interpolation algorithm and an interpolation method based on a neural network. The GNSS data is baseline solved, and the obtained baseline solution results are adjusted. The accuracy of the baseline solution is adjusted through error analysis. The data processing process is monitored in real time, and the GNSS data quality after each data processing step is evaluated, for example, the baseline vector correction number, the constrained adjustment baseline vector correction number and the unconstrained baseline vector correction number of the same name are checked, and the correctness of the data processing steps is monitored in real time.

[0036] Furthermore, precise ephemeris is used to reduce the impact of GNSS satellite orbit errors on the accuracy of baseline solutions. Multi-band observation data is used for integer solutions to improve the redundancy of observation data, separate and eliminate errors such as ionospheric delay, and fix the ambiguity parameters to integers through integer solutions to improve the accuracy of baseline solutions. Multi-band observation data is used to constrain the solution of ambiguities and improve the stability of integer solutions. During the GNSS network adjustment process, an iterative algorithm is used to gradually approach the optimal solution, eliminating the probability that the optimal solution cannot be obtained when solving the adjustment equation. The iterative algorithm continuously adjusts the parameter estimates and recalculates the residual and covariance matrices until certain convergence conditions are met.

[0037] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

[0038] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A high-precision GNSS data processing method that takes into account shade from dense forests, characterized in that: include: Step 1: Divide the time period and determine the satellite signal data observed by the GNSS receiver in each time period; Step 2: Determine whether the GNSS observation area at different time periods belongs to the tree shade area; Obtain GNSS Maolin data for n time periods in a real Maolin environment, restore the real Maolin environment to establish a simulation environment, determine the true value of Maolin data and open data in a certain time period, and use the formula The observation factors of Maolin data and open data are obtained, where K m Indicates the size of the type factor for time period m. Time period types include: morning, noon and dusk. Each time period type corresponds to a type factor size. i indicates that the number of the observed satellite is i, and N indicates that a total of N satellites are observed in this time period. S i The signal strength of satellite number i within time t is used to determine the time period that falls within the tree shade area based on the calculation formula of the observation factor. This gives the time period when GNSS is affected by tree shade in a real dense forest environment. Step 3: Preliminary processing of GNSS data to filter out sampling points within tree shade areas. This preliminary processing eliminates the effects of noise, abnormal data, and data deviations. After preliminary processing, the signal quality of each GNSS sampling point is evaluated to filter out data from sampling points that are significantly affected by tree shade. Step 4: Based on the conventional data, the data at the vacant sampling points are completed, and then high-precision data processing is performed; the data of the vacant sampling points are predicted by the trend extrapolation method, and the data parts that were eliminated at the vacant sampling points are completed to obtain complete data, and high-precision data processing is performed on the complete data.

2. The method for processing high-precision GNSS data taking into account shade from dense forests according to claim 1, characterized in that: The specific method of step one is: A day is divided into different time periods of equal length, with the length determined as t. Based on the changes in tree shade over time, the time periods are classified into morning, noon, and dusk. A GNSS is set up to obtain observation data in a dense forest environment, and the satellite signal data in each time period is recorded, which is recorded as the dense forest data. The satellite signal data includes: the number of satellites, satellite signal strength, and the observed satellite number.

3. The method for processing high-precision GNSS data taking into account shade from dense forests according to claim 1, characterized in that: The calculation formula based on the observation factor is used to determine the time period belonging to the tree shade area. The specific method is as follows: For time period a, based on the calculation formula of the observation factor, the observation factor of the dense forest data in time period a is obtained, recorded as the dense forest factor, and the observation factor of the open data in time period a is recorded as the open factor. The ratio g of the dense forest factor to the open factor is calculated, and the shade blocking threshold L is set. When g is less than or equal to L, the GNSS observation area in time period a is judged to belong to the shade blocking area. Otherwise, the GNSS observation area in time period a is judged to belong to the open area. L is set according to actual needs and specific conditions, and all GNSS data belonging to the shade blocking area in different time periods are collected.

4. The method for processing GNSS high-precision data taking into account shade from dense forests according to claim 1, characterized in that: Eliminate the effects of noise, abnormal data, and data deviations by: Eliminate noise influences; filter the received GNSS data using digital filtering techniques including Kalman filtering, wavelet transform or median filtering to suppress random noise and remove noise interference; Eliminate the impact of abnormal data; perform real-time verification of GNSS data, including checking the integrity of data packets, the continuity of timestamps, and the standardization of data formats. Utilize redundant observations and internal consistency checking methods to identify and eliminate potential erroneous or abnormal data. Eliminate the impact of data bias; correct possible data bias problems, including: using the satellite clock error information provided by precise ephemeris to eliminate the clock error through differential technology; by building an atmospheric delay model, using meteorological parameters and observation data to estimate and eliminate atmospheric environment bias; using multi-frequency observation data, and further weakening the ionospheric delay bias through linear combination technology.

5. The method for processing GNSS high-precision data taking into account shade from dense forests according to claim 1, characterized in that: Filter out the data at sampling points that are greatly affected by tree shade. The specific method is as follows: Through simulation experiments, the changing trend of GNSS data in each time period is obtained. Based on the changing trend of GNSS data in each time period, the sampling frequency is determined. The signal quality of GNSS data in different time periods in the tree-shaded area is evaluated according to the sampling points. The signal-to-noise ratio and cycle slip indicators of the signal are evaluated, and the signal-to-noise ratio threshold and cycle slip threshold are set. When the signal-to-noise ratio or cycle slip of the signal is greater than or equal to the corresponding threshold, the data of the sampling point is eliminated to obtain a vacant sampling point. The data that is not eliminated is recorded as regular data. The signal-to-noise ratio threshold is obtained by integrating historical GNSS Maolin data, and the cycle slip threshold is obtained by screening outliers or mutation points in the data sequence in the GNSS data.

6. The method for processing high-precision GNSS data taking into account shade from dense forests according to claim 1, characterized in that: The data of the missing sampling points are predicted by trend extrapolation method to fill in the data that were eliminated at the missing sampling points. The specific method is as follows: Collect the data of the first N sampling points of each vacant sampling point to form a prediction data set. Fill in each vacant sampling point in turn according to the time series, extract the data characteristic values ​​of the sampling points, such as pseudo-range observation values ​​and carrier phase observation values, and use the trend extrapolation method to predict the data characteristic values ​​of the vacant sampling points. By establishing a polynomial model, the data of the N+1 sampling point is predicted from the data of the first N sampling points, and the formula is used. Represents a polynomial model, where M(N+1) represents the characteristic data value of the N+1th sampling point, {a, a′,…, a (x) } represents the model coefficient, x represents the highest order of the polynomial model, and the data change trend of the N sampling points before the vacant sampling point is integrated to determine the range of the model coefficient. By predicting the data of the non-vacant sampling points, the true value is compared with the predicted value to determine the prediction accuracy of the polynomial model at different orders, thereby determining the highest order of the polynomial model, and the polynomial model is used to complete the data in the vacant sampling points.

7. The method for processing GNSS high-precision data taking into account shade from dense forests according to claim 1, characterized in that: Perform high-precision data processing on the complete data. The specific method is as follows: For complete GNSS data, time correlation analysis is first used to assist in the detection of gross errors in the data, and the interpolation method is used to handle the gross errors. Then, the GNSS data is baseline solved, and the baseline solution results are adjusted. The accuracy of the baseline solution is adjusted through error analysis. The data processing process is monitored in real time, and the GNSS data quality after each data processing step is evaluated. Precise ephemeris is used to eliminate the impact of the orbit error of the GNSS satellite on the accuracy of the baseline solution. Integer solution is performed using multi-band observation data. Through integer solution, the ambiguity parameters are fixed to integers. Multi-band observation data is used to constrain the resolution of ambiguities. During the GNSS network adjustment process, an iterative algorithm is used to gradually approach the optimal solution. The iterative algorithm continuously adjusts the parameter estimates and recalculates the residuals and covariance matrices until the optimal solution is obtained.

Images