A method and device for constructing a high-resolution ionospheric model

Abstract

The present invention discloses a method and device for constructing a high-resolution ionospheric model in the technical field of ionospheric modeling. The method comprises: obtaining GNSS receiver observation data, precision products and satellite DCB products of each time period within a target time interval; constructing a training set according to the GNSS receiver observation data, precision products and satellite DCB products of each time period, wherein the training samples in the training set use the ionospheric observation value of the puncture point as label data, and use the spatiotemporal information of the puncture point and the ionospheric physical parameters as input data; within the target time interval, segmented modeling is performed according to a predetermined time interval to generate an ionospheric model for each time interval; for each ionospheric model, the corresponding training sample is taken from the training set, and a machine learning method is used for training to obtain the final ionospheric model for each time interval; the present invention can construct an ionospheric model with high resolution and accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of ionospheric modeling, and in particular to a method and device for constructing a high-resolution ionospheric model. Background Art

[0002] The Global Navigation Satellite System (GNSS) sends navigation signals to provide real-time, continuous, all-weather positioning, navigation, and timing services to users around the world. After the GNSS signal enters the atmosphere, it is affected by the refraction of the ionosphere, the propagation speed of the navigation signal changes, and the ionospheric delay effect occurs. Using the GNSS multi-frequency observations, the total electron volume (TEC) of the signal propagation path from the satellite to the station can be extracted. Through the projection function, the propagation direction TEC is projected to the zenith direction to obtain the total electron volume (VTEC) in the zenith direction. Using the TEC observations in the region, regional ionosphere modeling can be completed. According to the regional ionosphere model, the TEC at any position in the region can be calculated to realize the monitoring of ionosphere changes, support users to correct ionosphere delay, thereby expanding the GNSS service field and improving the GNSS positioning accuracy. Therefore, building a high-precision regional ionosphere model is of great significance in the fields of space weather monitoring and forecasting, GNSS local enhancement, CORS services, etc.

[0003] At present, all major navigation systems generally use empirical ionospheric models to correct ionospheric errors. For example, the Klobuchar model used by the GPS system, the NeQuick model used by the Galileo system, and the BDGIM model used by the BeiDou-3 system. These models are based on simple mathematical functions to describe the horizontal distribution of ionospheric TEC, which can effectively capture the large-scale structure of the ionosphere (thousands of kilometers). However, these models show obvious deficiencies in dealing with small and medium-scale irregular phenomena in the ionosphere (such as traveling disturbances, plasma bubbles, tongue-shaped ionospheric anomalies, etc.).

[0004] Its main defects include:

[0005] (1) Oversmoothing: Due to the limitations of simple mathematical functions, the model is difficult to reflect the ionospheric structure at small and medium scales (hundreds of kilometers and below);

[0006] (2) Complex models are difficult to generalize: Although higher-order mathematical models can improve accuracy, they increase model complexity and computational burden and may lead to over-parameterization;

[0007] (3) Lack of physical explanation: Traditional models usually rely on statistical or empirical formulas and cannot directly explain the physical phenomena of the ionosphere; Summary of the invention

[0008] The purpose of the present invention is to overcome the deficiencies in the prior art and to provide a method and device for constructing a high-resolution ionospheric model, which has significant advantages in processing complex nonlinear relationships and large-scale data sets through machine learning technology, can capture deep patterns and associations in the data, and provide more accurate predictions than traditional methods.

[0009] To achieve the above object, the present invention is implemented by adopting the following technical solutions:

[0010] In a first aspect, the present invention provides a method for constructing a high-resolution ionospheric model, comprising:

[0011] Obtain GNSS receiver observation data, precision products and satellite DCB products for each time period within the target time interval;

[0012] A training set is constructed based on GNSS receiver observation data, precision products and satellite DCB products in each time period, wherein the training samples in the training set use the ionospheric observation values ​​of the puncture points as label data, and use the spatiotemporal information of the puncture points and the ionospheric physical parameters as input data;

[0013] In the target time interval, segmented modeling is performed according to predetermined time intervals to generate an ionospheric model for each time interval;

[0014] For the ionosphere model of each time interval, corresponding training samples are taken from the training set, and training is performed using a machine learning method to obtain the final ionosphere model of each time interval.

[0015] Optionally, constructing a training set based on GNSS receiver observation data, precision products, and satellite DCB products in each time period includes:

[0016] Let t=2, and repeat the following steps until t>T, where t is the time period index and T is the total number of time periods:

[0017] The GNSS receiver observation data of the t-1th time period is solved based on the GNSS receiver DCB data of the t-1th time period, the precision product of the tth time period and the satellite DCB product to obtain the GNSS receiver DCB data of the tth time period and the ionospheric observation value and spatiotemporal information of the puncture point;

[0018] Integrate the ionospheric observation value and the spatiotemporal information of the puncture point in the t-th time period to generate a STEC observation value, and transform the STEC observation value using a projection function to generate a VTEC observation value;

[0019] The VTEC observations of the tth time period are augmented with ionospheric physical parameters using the International Ionospheric Reference Model;

[0020] The spatiotemporal information and ionospheric physical parameters of the puncture point in the tth time period are used as input data, and the ionospheric observation values ​​of the puncture point are used as label data to generate training samples and added to the training set;

[0021] Let t=t+1.

[0022] Optionally, the spatiotemporal information of the puncture point includes: ionospheric observation time, altitude angle and longitude and latitude of ionospheric observation values.

[0023] Optionally, the step of constructing a training set based on GNSS receiver observation data, precision products, and satellite DCB products in each time period further includes:

[0024] When t=1, t is the time period index, and the non-difference non-combined PPP method is used to solve the GNSS receiver observation data, precision products and satellite DCB products of the first time period to obtain the GNSS receiver DCB data of the first time period.

[0025] In a second aspect, the present invention provides a device for constructing a high-resolution ionospheric model, comprising:

[0026] A data acquisition module is configured to acquire GNSS receiver observation data, precision products, and satellite DCB products for each time period within a target time interval;

[0027] A training set construction module is configured to construct a training set according to GNSS receiver observation data, precision products and satellite DCB products in each time period, wherein the training samples in the training set use the ionospheric observation values ​​of the puncture points as label data, and use the spatiotemporal information of the puncture points and the ionospheric physical parameters as input data;

[0028] The model building module is configured to perform segmented modeling at predetermined time intervals within a target time interval to generate an ionospheric model for each time interval;

[0029] The model training module is configured to extract corresponding training samples from the training set for the ionospheric model of each time interval, and use a machine learning method to perform training to obtain the final ionospheric model of each time interval.

[0030] In a third aspect, the present invention provides an electronic device, including a processor and a storage medium;

[0031] The storage medium is used to store instructions;

[0032] The processor is used to operate according to the instructions to execute the steps according to the above method.

[0033] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above method when executed by a processor.

[0034] In a fifth aspect, the present invention provides a computer program product, comprising a computer program / instruction, which implements the steps of the above method when executed by a processor.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] The present invention provides a method and device for constructing a high-resolution ionospheric model, which uses the spatiotemporal information of the puncture point and the physical parameters of the ionosphere as input data, fully utilizing the spatiotemporal information of each original puncture point, and avoiding the fitting error caused by the image input of the traditional neural network model; the characteristics of the physical model are added to the data of the puncture point, which improves the interpretability of the ionosphere model and makes the model physically meaningful. Through segmented modeling, the amount of training data is small at a time, and compared with the traditional neural network model, it does not require long-term input data, the calculation speed is fast and the accuracy is higher, and the established model is more in line with the actual ionosphere changes. The high-resolution ionosphere model constructed by the present invention can generate ionosphere products with specified spatiotemporal resolution, avoiding excessive smoothing caused by using mathematical functions to express ionosphere TEC. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a flow chart of a method for constructing a high-resolution ionospheric model provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0038] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0039] Embodiment 1:

[0040] like Figure 1 As shown, an embodiment of the present invention provides a method for constructing a high-resolution ionosphere model, comprising the following steps:

[0041] Step S1, obtaining GNSS receiver observation data, precision products and satellite DCB products for each time period within a target time interval.

[0042] A GNSS receiver is a device used to receive, track, process and measure Global Navigation Satellite System (GNSS) signals. It can receive signals from different satellites, process and analyze these signals through built-in antennas, receiving units and processing units, and then calculate the receiver's position, speed, time and other information.

[0043] Precision products are products used for PPP navigation and positioning. Their specific function is to provide accurate satellite positions and satellite clock information. Precision products include .clk files (precision satellite clock information) and sp3 (precision satellite orbit information).

[0044] Satellite DCB (Differential Code Bias) products are an important data product in the Global Navigation Satellite System (GNSS). DCB is caused by the influence of satellite and receiver hardware delays, and there is a systematic deviation between the observations obtained through different signals. This deviation must be compensated in precise positioning to improve positioning accuracy.

[0045] Step S2: construct a training set based on the GNSS receiver observation data, precision products and satellite DCB products of each time period. The training samples in the training set use the ionospheric observation values ​​of the puncture point as label data, and the spatiotemporal information of the puncture point and the ionospheric physical parameters as input data.

[0046] Specifically in this embodiment, constructing a training set based on GNSS receiver observation data, precision products, and satellite DCB products in each time period includes:

[0047] When t=1, t is the time period index, and the non-difference non-combined PPP method is used to solve the GNSS receiver observation data, precision products and satellite DCB products of the first time period to obtain the GNSS receiver DCB data of the first time period. The main purpose of this step is to obtain the GNSS receiver DCB data of the first time period through initialization solution, so as to provide initial data for subsequent cyclic solution.

[0048] Let t=2, and repeat steps S10 to S50 until t>T, where t is the time period index and T is the total number of time periods:

[0049] Step S10, based on the GNSS receiver DCB data of the t-1th time period, the precision product of the tth time period and the satellite DCB product, the GNSS receiver observation data of the tth time period is solved to obtain the GNSS receiver DCB data of the tth time period and the ionospheric observation value and spatiotemporal information of the puncture point.

[0050] Specifically in this embodiment, the spatiotemporal information of the puncture point includes the ionospheric observation time, the altitude angle and the longitude and latitude of the ionospheric observation value. In other optional embodiments, other information can also be solved to prepare for subsequent model construction.

[0051] Step S20: Integrate the ionospheric observation value and the spatiotemporal information of the puncture point in the t-th time period to generate the STEC observation value, and use the projection function to transform the STEC observation value to generate the VTEC observation value.

[0052] In GNSS (Global Navigation Satellite System) positioning and ionosphere research, the STEC (Slant Total Electron Content) observation value and VTEC (Vertical Total Electron Content) observation value at the puncture point are two important parameters.

[0053] STEC describes the total electron content along the signal propagation path when the GNSS signal passes through the ionosphere. It is an important parameter of the impact of the ionosphere on the GNSS signal, and its value can be calculated from GNSS observation data. VTEC describes the total number of electrons in a column per unit area above the puncture point in the ionosphere, and is a key indicator for evaluating the impact of the ionosphere on the GNSS signal.

[0054] Step S30: using the International Ionospheric Reference Model to add ionospheric physical parameters to the VTEC observation values ​​of the t-th time period.

[0055] The physical parameters of the ionosphere use the IRI model, namely the International Reference Ionosphere (IRI), which is an internationally agreed standard model used to describe various parameters in the Earth's ionosphere. The physical parameters of the ionosphere include: AE index at the observation time, Kp index at the observation time, and Dst index at the observation time. Among them, the Kp index at the observation time and the Dst index at the observation time can be obtained from the US NOAA Space Weather Forecast Center, GFZ German Research Centre for Geosciences, etc. The acquisition method is resampling: Taking the Kp index at the observation time as an example, because the time resolution of the physical parameters is 3h, it is necessary to resample and interpolate to the observation time point of the observation value as the observation time Kp at that time point.

[0056] Step S40: Use the spatiotemporal information and ionospheric physical parameters of the puncture point in the t-th time period as input data, and the ionospheric observation value of the puncture point as label data to generate training samples, and add them to the training set.

[0057] Step S50, let t=t+1.

[0058] Step S3: within the target time interval, perform segmented modeling according to predetermined time intervals to generate an ionospheric model for each time interval.

[0059] By segmenting the model, the accuracy of the ionospheric model at each time interval can be improved.

[0060] Step S4: For the ionospheric model of each time interval, corresponding training samples are taken from the training set, and training is performed using a machine learning method to obtain the final ionospheric model of each time interval.

[0061] Taking the construction of a high-resolution ionospheric model for 2023 as an example, the specific steps are as follows:

[0062] 1. Prepare global GNSS receiver observation data for the whole year of 2023, and prepare corresponding precision products (clk, sp3) and satellite DCB products for each day.

[0063] 2. The non-differenced and non-combined PPP method is used to bring in the GNSS receiver observation data, satellite DCB and precision products of the first day to calculate the DCB data of all GNSS receivers on the first day.

[0064] 3. Starting from the second day, the DCB data of the receiver from the previous day and the satellite DCB products of the day are brought in to solve the GNSS receiver observation data of the day. The receiver DCB data of the day and the ionospheric observation value of the puncture point, the longitude, latitude and altitude angle of the puncture point and other information are solved from the GNSS receiver observation data of the day.

[0065] 4. Repeat step 3, and from the second day onwards, use the GNSS receiver DCB data calculated on the previous day and the satellite DCB product of the day to perform the solution.

[0066] 5. Finally, the ionospheric observation values ​​of each day's puncture point and the spatiotemporal information data of the puncture point are integrated together, and the STEC observation values ​​are converted into VTEC observation values ​​through the projection function, in preparation for the subsequent ionospheric model construction and product generation.

[0067] 6. Add the ionospheric physical parameters (AE index, Kp index and Dst index) of IRI to the puncture point data according to the observation time of the puncture point, and resample each physical parameter to the observation of the puncture point.

[0068] 7. Segment modeling is performed according to the time interval of 12 hours, and the data of each time interval is trained by machine learning methods. The content of each training data includes: ionospheric observation time, latitude and longitude of ionospheric observation value, altitude angle of ionospheric observation value, AE index at observation time, Kp index at observation time, and Dst index at observation time. The ionospheric observation value is used as a label, and all training data within 12 hours are input into the machine learning algorithm to learn multi-dimensional features. The effect that the model can achieve after training is: input the specified time, longitude and latitude, altitude angle of observation value, AE index at observation time, Kp index at observation time, and Dst index at observation time, so as to obtain the predicted value of the ionosphere.

[0069] 8. Production of high-resolution ionospheric product maps: The trained model is used to generate ionospheric VTEC MAP images with a time resolution of 5 minutes and a spatial resolution of 0.1°*0.1° within the interval. It is only necessary to specify the interval time and the longitude and latitude of the interval, traverse each feature one by one and input them into the model, and then obtain the ionospheric VTEC MAP of the specified resolution in the area.

[0070] 9. Finally, repeat step 7 to generate an ionospheric model with a temporal resolution of 5 minutes and a spatial resolution of 0.1°*0.1° for the whole year of 2023.

[0071] Embodiment 2:

[0072] The present invention provides a device for constructing a high-resolution ionosphere model, comprising:

[0073] A data acquisition module is configured to acquire GNSS receiver observation data, precision products, and satellite DCB products for each time period within a target time interval;

[0074] The training set construction module is configured to construct a training set based on the GNSS receiver observation data, precision products and satellite DCB products of each time period, wherein the training samples in the training set use the ionospheric observation values ​​of the puncture point as label data, and use the spatiotemporal information of the puncture point and the ionospheric physical parameters as input data;

[0075] The model building module is configured to perform segmented modeling at predetermined time intervals within a target time interval to generate an ionospheric model for each time interval;

[0076] The model training module is configured to extract corresponding training samples from the training set for the ionospheric model of each time interval, and use a machine learning method to perform training to obtain the final ionospheric model of each time interval.

[0077] Embodiment three:

[0078] Based on the method for constructing a high-resolution ionosphere model provided in Embodiment 1, an embodiment of the present invention provides an electronic device, including a processor and a storage medium;

[0079] The storage medium is used to store instructions;

[0080] The processor is used to operate according to the instructions to execute the steps according to the above method.

[0081] Embodiment 4:

[0082] Based on the method for constructing a high-resolution ionosphere model provided in Example 1, an embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of the above method are implemented.

[0083] Embodiment five:

[0084] Based on the method for constructing a high-resolution ionospheric model provided in Example 1, an embodiment of the present invention provides a computer program product, including a computer program / instructions, which implement the steps of the above method when executed by a processor.

[0085] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0086] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0087] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0088] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0089] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for constructing a high-resolution ionospheric model, characterized in that: include: Obtain GNSS receiver observation data, precision products and satellite DCB products for each time period within the target time interval; A training set is constructed based on GNSS receiver observation data, precision products and satellite DCB products in each time period, wherein the training samples in the training set use the ionospheric observation values ​​of the puncture points as label data, and use the spatiotemporal information of the puncture points and the ionospheric physical parameters as input data; In the target time interval, segmented modeling is performed according to predetermined time intervals to generate an ionospheric model for each time interval; For the ionosphere model of each time interval, corresponding training samples are taken from the training set, and training is performed using a machine learning method to obtain a final ionosphere model of each time interval; The step of constructing a training set based on GNSS receiver observation data, precision products, and satellite DCB products in each time period includes: Let t=2, and repeat the following steps until t>T, where t is the time period index and T is the total number of time periods: The GNSS receiver observation data of the t-1th time period is solved based on the GNSS receiver DCB data of the t-1th time period, the precision product of the tth time period and the satellite DCB product to obtain the GNSS receiver DCB data of the tth time period and the ionospheric observation value and spatiotemporal information of the puncture point; Integrate the ionospheric observation value and the spatiotemporal information of the puncture point in the t-th time period to generate a STEC observation value, and transform the STEC observation value using a projection function to generate a VTEC observation value; The VTEC observations of the tth time period are augmented with ionospheric physical parameters using the International Ionospheric Reference Model; The spatiotemporal information and ionospheric physical parameters of the puncture point in the tth time period are used as input data, and the ionospheric observation values ​​of the puncture point are used as label data to generate training samples and added to the training set; Let t=t+1.

2. The method for constructing a high-resolution ionospheric model according to claim 1, characterized in that: The spatiotemporal information of the puncture point includes: ionospheric observation time, altitude angle and longitude and latitude of ionospheric observation value.

3. The method for constructing a high-resolution ionospheric model according to claim 1, characterized in that: The construction of the training set based on the GNSS receiver observation data, precision products and satellite DCB products in each time period also includes: When t=1, t is the time period index, and the non-difference non-combined PPP method is used to solve the GNSS receiver observation data, precision products and satellite DCB products of the first time period to obtain the GNSS receiver DCB data of the first time period.

4. A device for constructing a high-resolution ionospheric model, characterized in that: include: A data acquisition module is configured to acquire GNSS receiver observation data, precision products, and satellite DCB products for each time period within a target time interval; A training set construction module is configured to construct a training set according to GNSS receiver observation data, precision products and satellite DCB products in each time period, wherein the training samples in the training set use the ionospheric observation values ​​of the puncture points as label data, and use the spatiotemporal information of the puncture points and the ionospheric physical parameters as input data; The model building module is configured to perform segmented modeling at predetermined time intervals within a target time interval to generate an ionospheric model for each time interval; The model training module is configured to extract corresponding training samples from the training set for the ionosphere model of each time interval, and use a machine learning method to perform training to obtain a final ionosphere model of each time interval; The step of constructing a training set based on GNSS receiver observation data, precision products, and satellite DCB products in each time period includes: Let t=2, and repeat the following steps until t>T, where t is the time period index and T is the total number of time periods: The GNSS receiver observation data of the t-1th time period is solved based on the GNSS receiver DCB data of the t-1th time period, the precision product of the tth time period and the satellite DCB product to obtain the GNSS receiver DCB data of the tth time period and the ionospheric observation value and spatiotemporal information of the puncture point; Integrate the ionospheric observation value and the spatiotemporal information of the puncture point in the t-th time period to generate a STEC observation value, and transform the STEC observation value using a projection function to generate a VTEC observation value; The VTEC observations of the tth time period are augmented with ionospheric physical parameters using the International Ionospheric Reference Model; The spatiotemporal information and ionospheric physical parameters of the puncture point in the tth time period are used as input data, and the ionospheric observation values ​​of the puncture point are used as label data to generate training samples and added to the training set; Let t=t+1.

5. An electronic device, characterized in that: including processor and storage medium; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1-3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 3 are implemented.

7. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 3 are implemented.

Images