A regional ionospheric high-precision rapid assimilation method and system

Abstract

The present invention discloses a method and system for high-precision, rapid regional ionospheric assimilation, relating to the field of ionospheric technology. The method comprises: obtaining initial observation data and background data for the region to be assimilated; sequentially screening, removing outliers, and eliminating redundancies from the initial observation data to obtain observation data; grouping the observation data to obtain a plurality of grouped data; and constructing an assimilation model. Based on the assimilation model and in combination with the background data, iteratively assimilate each grouped data to obtain assimilated data. The present invention improves the data assimilation speed and avoids abnormal assimilation results.

Description

Technical Field

[0001] The present invention relates to the field of ionosphere technology, and in particular to a high-precision and rapid assimilation method and system for a regional ionosphere. Background Art

[0002] As humanity enters the information age, radio systems have become widely used. They are currently used in radio radar, satellite navigation, and communications. These systems either rely on the ionosphere's influence on radio waves to operate, or their radio waves must pass through the ionosphere, severely impacting the state of the ionosphere. The ionosphere can also reduce the accuracy of satellite navigation, as well as the resolution and positioning accuracy of synthetic aperture radar (SAR) imaging. Ionospheric scintillation can affect the quality of satellite-to-ground communications and even lead to interruptions. During solar flares, ionospheric absorption can severely interfere with low- and very-low-frequency communications.

[0003] The time-varying characteristics of ionospheric electron density play a crucial role in ionospheric applications. To meet the growing demand for ionospheric monitoring and forecasting, it is necessary to obtain the ionospheric electron density distribution in real time and monitor and study various disturbances or anomalies in the ionosphere.

[0004] Existing ionospheric data assimilation technologies, including Kalman filtering, three-dimensional variational and four-dimensional variational methods, utilize little data, do not perform fast processing, and may lead to abnormal assimilation results due to the unevenness of the acquired data. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and system for high-precision and rapid assimilation of regional ionosphere, which improves the data assimilation speed and avoids abnormal assimilation results.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] A high-precision and rapid assimilation method for regional ionosphere, comprising:

[0008] Obtaining initial observation data and background data of the area to be assimilated; screening, eliminating outliers, and eliminating redundancy on the initial observation data to obtain observation data;

[0009] Grouping the observation data to obtain several groups of grouped data;

[0010] An assimilation model is constructed, and based on the assimilation model and in combination with the background data, each of the grouped data is iteratively assimilated to obtain assimilated data.

[0011] Optionally, the assimilation model is as follows:

[0012] Xa =X b +K(yX b );

[0013] Where: K is the gain matrix, K = B b H T (R+HB b H T ) -1 , X a is the analysis value, X b is the background value, y is the observation value, H is the observation operator, R is the observation error covariance matrix, B b is the background value error covariance matrix, and T is the transpose.

[0014] Optionally, the constructing of the assimilation model is based on the assimilation model and combined with the background data to iteratively assimilate each of the grouped data to obtain assimilated data, specifically:

[0015] For the first group of grouped data, the assimilation model is substituted to obtain a group of grouped data; for the second group of grouped data, based on formula B a =(1-KH)B b Calculate B a , the calculated B a Assign to B b , based on the assigned B b , combined with the assimilation model, a set of grouped data is obtained; and so on, several sets of grouped data are obtained; the assimilated data includes several sets of grouped data.

[0016] Optionally, the assimilated data includes three-dimensional electron density, peak electron density and vertical total electron content.

[0017] Alternatively, for the formula K=B b H T (R+HB b H T ) -1 [R+HBH T ] Perform eigenvalue decomposition and get R+HBH T =UWU T , perform eigenvalue truncation on W and obtain (R+HB b H T ) -1 =UW -1 U T , using UW -1 U T Replace the formula K=B b H T (R+HB b H T )-1 (R+HB b H T ) -1 Calculate, where: U is the eigenvector matrix, W is the eigenvalue matrix.

[0018] The present invention also provides a regional ionospheric high-precision rapid assimilation system, comprising:

[0019] The data acquisition and processing module is used to obtain the initial observation data and background data of the area to be assimilated, and to screen, eliminate outliers and redundancies of the initial observation data in order to obtain the observation data;

[0020] A data grouping module, used for grouping the observation data to obtain a plurality of grouped data;

[0021] The data assimilation module is used to construct an assimilation model, and based on the assimilation model and in combination with the background data, iteratively assimilate each of the grouped data to obtain assimilated data.

[0022] Optionally, the assimilation model is as follows:

[0023] X a =X b +K(yX b );

[0024] Where: K is the gain matrix, K = B b H T (R+HB b H T ) -1 , X a is the analysis value, X b is the background value, y is the observation value, H is the observation operator, R is the observation error covariance matrix, B b is the background value error covariance matrix, and T is the transpose.

[0025] Optionally, the data assimilation module is specifically:

[0026] For the first group of grouped data, the assimilation model is substituted to obtain a group of grouped data; for the second group of grouped data, based on formula B a =(1-KH)B b Calculate B a , the calculated B a Assign to B b , based on the assigned B b , combined with the assimilation model, a set of grouped data is obtained; and so on, several sets of grouped data are obtained; the assimilated data includes several sets of grouped data.

[0027] Optionally, the assimilated data includes three-dimensional electron density, peak electron density and vertical total electron content.

[0028] Alternatively, for the formula K=B b H T (R+HB b H T ) -1 [R+HBH T ] Perform eigenvalue decomposition and get R+HBH T =UWU T , perform eigenvalue truncation on W and obtain (R+HB b H T ) -1 =UW -1 U T , using UW -1 U T Replace the formula K=B b H T (R+BH b H T ) -1 (R+HB b H T ) -1 Calculate, where: U is the eigenvector matrix, W is the eigenvalue matrix.

[0029] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0030] The present invention discloses a method and system for rapid, high-precision regional ionospheric assimilation. The method comprises: obtaining initial observation data and background data for the region to be assimilated; sequentially screening, removing outliers, and eliminating redundancies from the initial observation data to obtain observation data; grouping the observation data to obtain a plurality of groups of grouped data; and constructing an assimilation model. Based on the assimilation model and in combination with the background data, the method iteratively assimilates each group of data to obtain assimilated data. The present invention improves the speed of data assimilation and avoids abnormal assimilation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] 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. 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.

[0032] Figure 1 This is a flow chart of the high-precision and rapid assimilation method of the regional ionosphere of the present invention;

[0033] Figure 2This is a structural diagram of the regional ionospheric high-precision rapid assimilation system of the present invention;

[0034] Figure 3 This is a schematic diagram of the vertical total electron content of the present invention;

[0035] Figure 4 Schematic diagram of the distribution of electron density at 40°N latitude with longitude and altitude;

[0036] Figure 5 Schematic diagram of the distribution of electron density at 120°E meridian with latitude and altitude;

[0037] Figure 6 Schematic diagram of the probability distribution of relative deviation between the peak electron density obtained by the method of the present invention and the peak electron density observed by the vertical meter;

[0038] Figure 7 Schematic diagram of the probability distribution of the absolute deviation between the peak electron density obtained by the method of the present invention and the peak electron density obtained by CODE.

[0039] Explanation of symbols: 1. Data acquisition and processing module; 2. Data grouping module; 3. Data assimilation module. DETAILED DESCRIPTION

[0040] 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.

[0041] The purpose of the present invention is to provide a method and system for high-precision and rapid assimilation of regional ionosphere, which improves the data assimilation speed and avoids abnormal assimilation results.

[0042] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0043] Figure 1 The flowchart of the high-precision and rapid assimilation method of the regional ionosphere of the present invention is as follows: Figure 1 As shown, the present invention provides a high-precision and rapid assimilation method for the regional ionosphere, comprising:

[0044] Step S1: Obtain initial observation data and background data for the area to be assimilated. The initial observation data is sequentially screened, outliers removed, and redundancy removed to obtain observation data. This ensures the quality, validity, and necessity of the observation data being assimilated, thereby improving the accuracy of the assimilation results and the speed of the assimilation operation. In this example, the International Ionospheric Reference Model (IRI2016) is used as the background data.

[0045] When performing regional ionospheric assimilation, in order to avoid data unevenness and the resulting imbalance in assimilation results, when performing ionospheric assimilation, within the horizontal grid of the assimilated area, if there are multiple stations in one horizontal grid, the station closest to the grid center is selected as the observation station, thereby screening the data to ensure that the assimilation results will not be abnormal due to excessive local observation data. Assimilation reduces redundant observation data, improves calculation speed, and reduces calculation time.

[0046] Step S2 groups the observational data to obtain several groups of grouped data. As the amount of observational data increases, the computational effort increases exponentially, as does the demand for computing resources and the time consumed. Therefore, methods are needed to reduce the computational effort and improve timeliness. Based on three-dimensional variational assimilation, grouped statistical optimization is employed for assimilation. The observational data to be assimilated is divided into several groups, and each group of grouped data is assimilated using a three-dimensional variational method.

[0047] Step S3: construct an assimilation model, and iteratively assimilate each of the grouped data based on the assimilation model and the background data to obtain assimilated data. In this embodiment, the assimilated data includes three-dimensional electron density, peak electron density, and vertical total electron content.

[0048] Specifically, starting from the optimal estimation theory, the three-dimensional variational assimilation form can be derived from the Bayesian formula:

[0049] P α (x) = P b (x b )P o (y);

[0050] Where: P o (y) is the probability density distribution value of the observed data, P b (x b ) is the probability density distribution value of the background data, P α (x) is the probability density distribution value under the conditions of observation data and background data.

[0051] When P αWhen the probability density of (x) reaches its maximum value, the obtained x is the optimal value obtained after the observation data and background data are assimilated. Assuming that the probability density distribution obeys the normal distribution, then:

[0052]

[0053] The optimal estimate of the ionospheric electron density state parameter x (called the analytical value) should make the probability density function P α (x) reaches its maximum value, that is, the value function J(x) reaches its minimum value. The value function J(x) is as follows:

[0054]

[0055] Where: x b is the electron density background field output by the ionospheric model, and B is x b The error covariance matrix of y is the observation value of the total electron content of the ionosphere relative to the slant path, H is the observation operator, which maps the three-dimensional ionospheric electron density parameter to the observation value of the total electron content of the slant path, R is the error covariance matrix of y, and T is the transpose.

[0056] The observation operator H is designed based on the principle of least squares, taking advantage of the relationship between observations and grid crossings within the assimilation range. If the spatial location of an observation crosses a spatial grid, the weight parameter is set to the length of the total electron content observation path within the spatial grid. If the observation does not cross the spatial grid, the weight parameter is set to 0. This matrix of weight parameters maps the ionospheric electron density background field provided by the IRI model to the total electron content observations along the slant path. This observation operator setting results in a large number of zero elements in the observation operator H matrix. Therefore, a sparse matrix is ​​used to represent H during data processing to prevent zero elements from participating in large matrix operations. This overcomes the problem of long runtimes in three-dimensional variational assimilation due to large data volumes while ensuring effective assimilation.

[0057] By deriving the value function and performing a series of transformations, the optimal state value of the electron density analysis field x can be solved:

[0058] x=x b +BH T [R+HBH T ] -1 (y-Hx b );

[0059] y-Hx b It is the difference between the true value and the background value expressed in the observation space, that is, the observation growth, BH T [R+HBH T ] -1 is the weight matrix.

[0060] Since the present invention adopts the group assimilation method in order to reduce the amount of calculation, the assimilation model established is as follows:

[0061] X a =X b +K(yX b );

[0062] Where: K is the gain matrix, K = B b H T (R+HB b H T ) -1 , X a is the analysis value, X b is the background value, R is the observation error covariance matrix, B b is the background value error covariance matrix.

[0063] For the first group of grouped data, the assimilation model is substituted to obtain a group of grouped data; for the second group of grouped data, based on formula B a =(1-KH)B b Calculate B a , the calculated B a Assign to B b , based on the assigned B b , combined with the assimilation model, a set of grouped data is obtained; and so on, several sets of grouped data are obtained; the assimilated data includes several sets of grouped data.

[0064] In order to solve the problem of singular values ​​that may appear in the calculation, the formula K=B b H T (R+HB b H T ) -1 [R+HBH T ] Perform eigenvalue decomposition and get R+HBH T =UWU T , perform eigenvalue truncation on W and obtain (R+HB b H T ) -1 =UW -1 U T , using UW -1 U T Replace the formula K=B b H T (R+HB b H T ) -1 (R+HB b H T ) -1Calculations are performed to avoid the influence of singular values; where U is the eigenvector matrix and W is the eigenvalue matrix.

[0065] Specifically, the data of 260 global satellite navigation system GNSS observation stations in the land-based network were used to process the generated ionospheric slant path total electron content data as input. The regional ionospheric assimilation study from January 1 to March 31, 2020 was carried out using the method of the present invention, and the regional ionospheric assimilation results for three months were obtained, including three-dimensional electron density, vertical total electron content and peak electron density.

[0066] When performing regional ionospheric assimilation, in order to avoid the imbalance of assimilation results caused by too many stations in the east and too few stations in the west, when performing ionospheric assimilation, first of all, if there are multiple stations in the horizontal grid of the assimilation area, the station closest to the grid center is selected as the observation station.

[0067] According to such rules, Figure 3 The regional distribution of the vertical total electron content TECu of the assimilation results every 2 hours on January 1, 2020 is given. Figure 4 The distribution of electron density Ne at 40°N latitude with longitude and altitude is given every 2 hours, and the changes of electron density with longitude, altitude and time can be clearly seen. Figure 5 The distribution of electron density Ne at 120°E meridian with latitude and altitude is given every 2 hours, and the changes of electron density with latitude, altitude and time can be clearly seen.

[0068] The accuracy of the ionospheric data generated by the method of the present invention is evaluated using third-party data such as ionospheric plummets and ground-based vertical total electron content. Two methods are used for evaluation and verification: (1) using ionospheric plummet data to compare and verify the ionospheric peak electron density; (2) using the vertical total electron content provided by the global ionospheric grid data CODE to compare and verify the total electron content report obtained by ionospheric assimilation.

[0069] The ionospheric assimilation results are verified using data from an ionospheric plumb-hole instrument. A plumb-hole instrument is a traditional ionospheric sounding tool that transmits radio waves of continuously swept frequencies vertically into the ionosphere. When the transmitted frequency equals the plasma frequency at a certain altitude in the ionosphere, the waves are reflected from that altitude. One frequency corresponds to one altitude, and within a single sounding, a frequency-height map (called a frequency-height map) is generated within the sweeping frequency range. The frequency-height map can be used to very accurately determine the critical frequency of the ionospheric F layer. Appropriate inversion techniques can also be used to determine the electron density profile below the F layer peak altitude. The peak electron density NmF2 (calculated from the critical frequency) and peak altitude hmF2 obtained from plumb-hole soundings are used as a basis for comparison and verification of the ionospheric assimilation results.

[0070] Before conducting inspection and evaluation, the plumb line data needs to be quality controlled. The quality control standards are as follows: if only one of NmF2 and hmF2 has an observation value at a certain moment, the observation value at that moment is discarded; if the difference between a certain NmF2 observation value and the 27-day moving average value at that time point is greater than 40%, the observation value is discarded.

[0071] The matching conditions between the ionospheric assimilation data and the vertical instrument data are as follows: (1) the F2 layer peak value obtained by ionospheric assimilation is interpolated to the geographical location corresponding to the vertical instrument site as the ionospheric assimilation peak electron density to be compared; (2) the assimilation result time is the same as the vertical instrument detection time.

[0072] The test and evaluation mainly involves calculating the variance of the deviation between the ionospheric assimilation results and the vertical meter inversion results. The formula for calculating the variance of the relative deviation is as follows:

[0073]

[0074]

[0075] Where: P represents NmF2, P ASS is the peak electron density of the F2 layer inverted by the vertical meter, P ionosonde is the peak electron density of the F2 layer obtained by ionospheric assimilation, r represents the deviation between the ionospheric assimilation result and the vertical instrument inversion result, N1 is the number of ionospheric assimilation results and vertical instrument inversion results, and RMS is the relative deviation between the ionospheric assimilation result and the vertical instrument inversion result.

[0076] The peak electron density of the regional ionospheric assimilation results from January 1 to March 31, 2020, was compared with the peak electron density observed by the vertical instrument in the region, and the relative deviation between the two was statistically analyzed. The RMS was 19.1%, and the probability distribution of the relative deviation is as follows: Figure 6 shown.

[0077] The vertical total electron content provided by the global ionospheric grid data CODE is used to compare and verify the total electron content reported by ionospheric assimilation.

[0078] The accuracy of the global ionospheric grid data CODE is about 3-5 total electron units TECU, which can be used as a verification basis for the vertical total electron content data obtained by ionospheric assimilation.

[0079] The verification and evaluation mainly calculates the variance of the deviation between the ionospheric assimilation vertical total electron content result and the CODE result. The calculation formula is as follows:

[0080]

[0081] Where: TEC iASSTEC stands for the total vertical electron content of the ionosphere assimilated iCODE Indicates the vertical total electron content provided by the global ionospheric grid data CODE, the number of N2 ionospheric assimilation vertical total electron content results and CODE results, RMS TEC is the relative deviation between the ionospheric assimilation vertical total electron content result and the CODE result.

[0082] When comparing, the data with higher latitude and longitude resolution are interpolated onto a grid with lower latitude and longitude resolution, and then the difference statistics are calculated. That is, when evaluating the regional ionospheric assimilation results, the regional assimilation results are interpolated onto the CODE latitude and longitude grid, and the deviation statistics are calculated compared with the CODE results.

[0083] In this paper, 3 months of ionospheric assimilation data are used for verification. The vertical total electron content of the regional ionospheric assimilation results from January 1 to March 31, 2020 is compared with the CODE-TEC results of the corresponding time and location in the region, and the absolute deviation between the two is statistically analyzed. The RMS is 3.78TECU, and the probability distribution of the absolute deviation is as follows: Figure 7 shown.

[0084] Figure 2 This is the structural diagram of the regional ionosphere high-precision rapid assimilation system of the present invention. Figure 2 As shown, the present invention provides a regional ionosphere high-precision rapid assimilation system, including: a data acquisition and processing module 1, a data grouping module 2 and a data assimilation module 3.

[0085] The data acquisition and processing module 1 is used to obtain initial observation data and background data of the area to be assimilated, and to sequentially screen, eliminate outliers and eliminate redundancy on the initial observation data to obtain observation data.

[0086] The data grouping module 2 is used to group the observation data to obtain several groups of grouped data.

[0087] The data assimilation module 3 is used to construct an assimilation model, and based on the assimilation model and in combination with the background data, iteratively assimilate each of the grouped data to obtain assimilated data.

[0088] Optionally, the assimilation model is as follows:

[0089] X a =X b +K(yX b );

[0090] Where: K is the gain matrix, K = B b H T (R+HB b H T ) -1 , Xa is the analysis value, X b is the background value, y is the observation value, H is the observation operator, R is the observation error covariance matrix, B b is the background value error covariance matrix, and T is the transpose.

[0091] Optionally, the data assimilation module 3 is specifically:

[0092] For the first group of grouped data, the assimilation model is substituted to obtain a group of grouped data; for the second group of grouped data, based on formula B a =(1-KH)B b Calculate B a , the calculated B a Assign to B b , based on the assigned B b , combined with the assimilation model, a set of grouped data is obtained; and so on, several sets of grouped data are obtained; the assimilated data includes several sets of grouped data.

[0093] Optionally, the assimilated data includes three-dimensional electron density, peak electron density and vertical total electron content.

[0094] Alternatively, for the formula K=B b H T (R+HB b H T ) -1 [R+HBH T ] Perform eigenvalue decomposition and get R+HBH T =UWU T , perform eigenvalue truncation on W and obtain (R+HB b H T ) -1 =UW -1 U T , using UW -1 U T Replace the formula K=B b H T (R+HB b H T ) -1 (R+HB b H T ) -1 Calculate, where: U is the eigenvector matrix, W is the eigenvalue matrix.

[0095] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0096] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A high-precision and rapid assimilation method for regional ionosphere, characterized in that: include: Obtain initial observation data and background data for the area to be assimilated; The initial observation data are sequentially screened, outliers are eliminated, and redundancy is eliminated to obtain observation data; Grouping the observation data to obtain several groups of grouped data; Constructing an assimilation model, and iteratively assimilating each of the grouped data based on the assimilation model and in combination with the background data to obtain assimilated data; Constructing an assimilation model, and iteratively assimilating each of the grouped data based on the assimilation model and in combination with the background data to obtain assimilated data, specifically: For the first group of grouped data, the assimilation model is substituted to obtain a group of grouped data; for the second group of grouped data, based on formula B a =(1-KH)B b Calculate B a , where K is the gain matrix, K = B b H T (R+HB b H T ) -1 , H is the observation operator, R is the observation error covariance matrix, B b is the background value error covariance matrix, T is the transpose; the calculated B a Assign to B b , based on the assigned B b , combined with the assimilation model, a set of grouped data is obtained; and so on, several sets of grouped data are obtained; the assimilated data includes several sets of grouped data.

2. The high-precision rapid assimilation method for the regional ionosphere according to claim 1, characterized in that: The assimilation model is as follows: X a =X b +K(y-X b ); Where: K is the gain matrix, K = B b H T (R+HB b H T ) -1 , X a is the analysis value, X b is the background value, y is the observation value, H is the observation operator, R is the observation error covariance matrix, B b is the background value error covariance matrix, and T is the transpose.

3. The high-precision rapid assimilation method for the regional ionosphere according to claim 1, characterized in that: The assimilated data include three-dimensional electron density, peak electron density and vertical total electron content.

4. A regional ionospheric high-precision rapid assimilation system, characterized by: include: The data acquisition and processing module is used to obtain the initial observation data and background data of the area to be assimilated, and to screen, eliminate outliers and redundancies of the initial observation data in order to obtain the observation data; A data grouping module, used for grouping the observation data to obtain a plurality of grouped data; The data assimilation module is used to build an assimilation model and iteratively assimilate each group data based on the assimilation model and the background data to obtain assimilated data. The data assimilation module is specifically as follows: for the first group of group data, substitute the assimilation model to obtain a group of group data; for the second group of group data, based on formula B a =(1-KH)B b Calculate B a , where K is the gain matrix, K = B b H T (R+HB b H T ) -1 , H is the observation operator, R is the observation error covariance matrix, B b is the background value error covariance matrix, T is the transpose; the calculated B a Assign to B b , based on the assigned B b , combined with the assimilation model, a set of grouped data is obtained; and so on, several sets of grouped data are obtained; the assimilated data includes several sets of grouped data.

5. The regional ionospheric high-precision rapid assimilation system according to claim 4, characterized in that: The assimilation model is as follows: X a =X b +K(y-X b ); Where: K is the gain matrix, K = B b H T (R+HB b H T ) -1 , X a is the analysis value, X b is the background value, y is the observation value, H is the observation operator, R is the observation error covariance matrix, B b is the background value error covariance matrix, and T is the transpose.

6. The regional ionospheric high-precision rapid assimilation system according to claim 4, characterized in that: The assimilated data include three-dimensional electron density, peak electron density and vertical total electron content.

Images