Ionized layer three-dimensional electron density reconstruction method during magnetic storm based on diffusion model

Abstract

The invention discloses a diffusion model-based ionosphere three-dimensional electron density reconstruction method in a magnetic storm period, and the method is based on a diffusion model and an ionosphere three-dimensional electron density reconstruction technology principle. The problem that three-dimensional inversion of the ionosphere is not constant due to the fact that the number of foundation observation stations and geometric sight lines between the observation stations and satellites are limited can be relieved; according to the ionosphere three-dimensional ionosphere electron density reconstruction method based on the diffusion model, the three-dimensional ionosphere inversion precision and resolution during the magnetic storm period are remarkably improved; the three-dimensional reconstruction efficiency of the ionosphere can be improved, and the spatial dynamic change process of the ionosphere during the magnetic storm period can be revealed.

Description

Technical Field

[0001] The invention relates to the technical field of ionosphere research and data processing thereof, and in particular to a method for reconstructing three-dimensional electron density of the ionosphere during a magnetic storm based on a diffusion model. Background Art

[0002] The ionosphere is an important part of the Earth's near-Earth space. Affected by solar activity, the Earth's magnetic field and atmospheric fluctuations, the temporal and spatial motion characteristics of the ionosphere are extremely complex. On the one hand, it has a significant impact on radio communications and navigation positioning. On the other hand, the ionosphere has a complex coupling mechanism with the crust. The reconstruction of the ionosphere's three-dimensional electron density is an advanced detection technology that reveals the spatial structure of the ionosphere. High-precision ionosphere three-dimensional electron density can construct an accurate ionospheric delay model, thereby improving the accuracy of navigation positioning, especially in low- and mid-latitude regions where the ionosphere is very active. In addition, by studying the relationship between crustal activities such as earthquakes, tsunamis and volcanic eruptions and ionospheric disturbances, a new technology can be provided for the field of geological disaster warning.

[0003] Magnetic storms are strong disturbances of the Earth's magnetic field caused by solar activities such as the corona and solar flares. This disturbance can cause drastic changes in the ionospheric electron density. In addition, magnetic storms may also cause abnormal phenomena such as ionospheric scintillation and plasma irregularities, which will affect the stability and reliability of satellite signals. Existing methods for reconstructing 3D ionospheric electron density mainly rely on traditional algorithms (iterative reconstruction algorithms, singular value decomposition and function models, etc.) and limited observation data, especially the inversion ill-posed problem caused by limited ground observation stations and limited observation angles, resulting in insufficient reconstruction accuracy and resolution. The method based on the diffusion model can effectively deal with the 3D inversion ill-posed problem under extreme magnetic storm conditions and improve the reconstruction accuracy and resolution, providing a new technical means for improving the real-time monitoring and forecasting of 3D ionospheric electron density.

[0004] The diffusion model is a mathematical model used to describe the diffusion process of particles, energy or matter in space. In recent years, the diffusion model has shown great potential and superiority in image reconstruction, text processing and 3D reconstruction. However, this technology has not been used to describe the spatial structure of the ionosphere. In ionospheric research, the diffusion model is mainly used to simulate the movement and distribution of electrons in the ionosphere. The ionospheric three-dimensional electron density reconstruction method based on the diffusion model can effectively and accurately reconstruct the spatial structure of the ionosphere. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides a method for reconstructing the three-dimensional electron density of the ionosphere during a magnetic storm based on a diffusion model; this method can effectively alleviate the ill-posed problem of three-dimensional inversion and improve the reconstruction accuracy of the three-dimensional electron density of the ionosphere during a magnetic storm.

[0006] In order to achieve the above technical effects, the present invention is implemented by the following technical solutions: a method for reconstructing the three-dimensional electron density of the ionosphere during a magnetic storm based on a diffusion model, characterized in that it includes the following steps:

[0007] S1. Obtain the original observation data and satellite navigation messages of the ground-based observation stations in the reconstruction area, and use the carrier phase smoothed pseudorange algorithm to calculate the slant total electron content (STEC) on the line of sight from the observation station to the satellite;

[0008] S2, discretizing the ionosphere in the reconstruction area into a three-dimensional space voxel grid;

[0009] S3, eliminating the geometric sight lines with low elevation angles from the ground-based observation station to the satellite, and constructing the observation intercept matrix between the sight lines and the voxel grid;

[0010] S4, using the International Ionospheric Reference Model (IRI) to obtain the ionospheric three-dimensional electron density in the inversion area and at the inversion time; further matching the electron density at the center of each voxel grid as the initial value of iterative reconstruction;

[0011] S5. Use the efficient multiplication iterative reconstruction algorithm (MART) to preliminarily reconstruct the 3D electron density of the ionosphere in the inversion area.

[0012] S6. The electron density data obtained by the IRI2020 model is used as the benchmark for the forward training of the diffusion model. The voxel grid that is not penetrated by the rays during the MART reconstruction process will be regarded as measurement noise and denoised in the inverse processing to improve the overall reconstruction accuracy and quality.

[0013] S7. Use the ionospheric sounding instrument data as the true value to measure the accuracy of the ionospheric three-dimensional electron density reconstruction results based on the diffusion model; if the expected effect is not achieved, adjust the diffusion model parameters in S6 to ultimately achieve high-quality ionospheric three-dimensional electron density reconstruction.

[0014] Furthermore, the formula for calculating STEC from the station to the satellite line of sight using carrier phase smoothed pseudorange in S1 is as follows:

[0015]

[0016] Where f1 and f2 are the carrier frequencies of L1 and L2 bands, usually 1.57542 GHz and 1.2276 GHz respectively; P sm It is a smoothed pseudo-range combined observation value, which reduces the influence of noise and multipath; DCB i and DCB j are the hardware delay deviations of the station and the satellite respectively; the unit of STEC is TECU, 1TECU = 10 16Ne / m 2 ;

[0017] STEC refers to the total amount of electrons from the satellite to the receiver in the satellite signal propagation path; its integral expression is:

[0018]

[0019] Where Ne is the electron density, which indicates the The number of electrons in the ionosphere at a certain location on Earth;

[0020] When doing 3D inversion, the above formula is usually discretized, and its discretized expression is:

[0021]

[0022] In the formula, A ij is the intercept matrix between the ray and the voxel, indicating the length of the path i passing through the jth voxel; ∈ i is the measurement error or noise; x j is the electron density of the jth voxel, which is the unknown quantity to be solved.

[0023] Furthermore, the intercept matrix in S3 is calculated using the following formula:

[0024]

[0025] Where d is the intercept value of the ray passing through the voxel grid, X1, Y1, Z1 are the spatial positions of the ray entering one side of the voxel; X2, Y2, Z2 are the spatial positions of the ray passing through the other side of the voxel, and the unit of d is generally m.

[0026] Furthermore, the formula of the iterative reconstruction algorithm MART in S5 is as follows:

[0027]

[0028] In the formula, x j k+1 is the electron density of the jth voxel after the k+1th iteration, x j k is the estimated value at the kth iteration; i T ,x k > is the TEC calculated based on the estimated value at the kth iteration; λ is the relaxation factor, which ranges from 0<λ<1; a ij is the intercept value of the i-th ray and the j-th voxel; ||a|| is the sum of the intercepts of all voxels passed by the i-th ray.

[0029] ​Furthermore, in S6, the forward process of the diffusion model converts the original electron density data into noise close to a normal distribution by gradually adding noise to the empirical model electron density data; then in the reverse process, the voxel electron density with low reconstruction quality and accuracy of some voxels caused by the ill-posed problem of three-dimensional inversion in the MART reconstruction process is treated as noise, and the global voxel electron density with high accuracy and quality is gradually reconstructed.

[0030] Furthermore, in S7, the root mean square error (RMSE) is used as the accuracy measurement indicator; the specific indicator size can be adjusted according to actual needs; the root mean square error formula is as follows:

[0031]

[0032] In the formula, x i is the true electron density extracted by ionospheric sounding instrument or incoherent scatter radar, x^ i is the electron density value after denoising by the diffusion model.

[0033] The beneficial effects of the present invention are:

[0034] The present invention alleviates the ill-posed problem of ionospheric three-dimensional inversion caused by the limited number of ground-based observation stations and the limited geometric line of sight between the observation stations and the satellites;

[0035] The three-dimensional ionospheric electron density reconstruction method based on the diffusion model of the present invention significantly improves the three-dimensional ionospheric inversion accuracy and resolution during magnetic storms;

[0036] The high efficiency of the ionosphere three-dimensional reconstruction of the present invention can reveal the dynamic change process of the ionosphere space during magnetic storms. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0038] Figure 1 It is a flow chart of the high-precision ionospheric electron density reconstruction technology during magnetic storms of the present invention;

[0039] Figure 2 This is the Kp and Dst index chart on March 17, 2015;

[0040] Figure 3 This is a diagram of the distribution of electron density at different altitudes over Europe under magnetic storm conditions at 08:00, 10:00 and 12:00 UT on March 17, 2015;

[0041] Figure 4 This is a comparison of the vertical contours of the electron density at the same longitude and latitude derived by IRI2020, MART and the diffusion model under the European regional magnetic storm environment on March 17, 2015. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] Example 1

[0044] like Figure 1 As shown, the present invention provides a method for reconstructing the three-dimensional electron density of the ionosphere during a magnetic storm based on a diffusion model, which mainly includes the following steps.

[0045] Step 1: To calculate the geometric line of sight and STEC, the present invention example obtains the observation data and navigation messages of the European region with longitudes of -10° to 30° and latitudes of 35° to 70° on March 17, 2015. The carrier phase smoothed pseudorange algorithm is used to solve the STEC on the line of sight from the station to the satellite. The main calculation formula is:

[0046]

[0047] In formula (1), f1 and f2 are the carrier frequencies of L1 band and L2 band, which are usually 1.57542 GHz and 1.2276 GHz respectively; P sm It is a smoothed pseudo-range combined observation value, which reduces the influence of noise and multipath; DCB i and DCB j They are the hardware delay deviations of the station and the satellite respectively. The unit of STEC is TECU, 1TECU = 10 16 Ne / m 2 .

[0048] STEC usually refers to the total amount of electrons from the satellite to the receiver on the satellite signal propagation path. Its integral expression is:

[0049]

[0050] In formula (2), Ne is the electron density, which means the The number of electrons in the ionosphere at a certain position on the surface. When doing 3D inversion, the above formula is usually discretized, so its discretized expression is:

[0051]

[0052] In formula (3), A ij is the intercept matrix between the ray and the voxel, indicating the length of the path i passing through the jth voxel; ∈ i is the measurement error or noise; x j is the electron density of the jth voxel, which is the unknown quantity to be solved.

[0053] Step 2: The reconstruction range is -10° to 30° in longitude, 35° to 70° in latitude and 90 to 1000 km in altitude. The discretized spatial resolution is 1°*1°*30 km voxels, with a total of 42,000 voxels.

[0054] Step 3: Eliminate the geometric sight lines with an elevation angle less than 15° between the ground-based observation station and the satellite, and construct the observation intercept matrix A between the sight line and the voxel model. The calculation formula of the intercept is as follows:

[0055]

[0056] In formula (4), d is the intercept value of the ray passing through the voxel, X1, Y1, Z1 are the spatial positions of the ray entering a certain side of the voxel; X2, Y2, Z2 are the spatial positions of the ray passing through the other side of the voxel, and the unit of d is generally m.

[0057] Step 4: Use the International Ionospheric Reference Model IRI2020 to obtain the three-dimensional electron density in the inversion area and the corresponding time, and further match the electron density at the center of each voxel as the initial value for iterative reconstruction.

[0058] Step 5: Use the efficient multiplication iterative reconstruction algorithm (MART) to preliminarily reconstruct the ionospheric three-dimensional electron density with a certain accuracy. The expression of MART is as follows:

[0059]

[0060] In formula (5), x j k+1 is the electron density of the jth voxel after the k+1th iteration, x j k is the estimated value at the kth iteration; i T ,x k > is the TEC calculated based on the estimated value at the kth iteration; λ is the relaxation factor, the value range is 0<λ<1, and in the example of the present invention, λ=0.2; a​ij is the intercept value of the i-th ray and the j-th voxel; ||a|| is the sum of the intercepts of all voxels passed by the i-th ray; the maximum number of iterations using MART is set to 30.

[0061] Step 6: Use the electron density data obtained by the IRI2020 model as the benchmark for forward training of the diffusion model. Voxels that are not penetrated by rays during the MART reconstruction process will be regarded as measurement noise and denoised in the reverse process to improve the overall reconstruction accuracy and quality. The main algorithm process of the example of the present invention is as follows:

[0062] Table 1 Ionospheric 3D electron density reconstruction algorithm based on diffusion model

[0063]

[0064]

[0065] Step 7: Use the ionospheric sounding instrument data as the true value to measure the accuracy of the ionospheric 3D electron density reconstruction results based on the diffusion model. If the expected effect is not achieved, adjust the diffusion model parameters in step 6 to finally achieve high-quality ionospheric 3D electron density reconstruction. This example uses the root mean square error (RMSE) as the accuracy measurement indicator. The specific indicator conditions can be adjusted according to actual needs.

[0066]

[0067] In formula (6), x i is the true electron density extracted by ionospheric sounding instrument or incoherent scatter radar, x^ i is the electron density value after denoising by the diffusion model.

[0068] like Figure 2 As shown in the figure, the Earth's magnetic field experienced a severe disturbance on March 17, 2015, with the geomagnetic Kp index exceeding 7 and the Dst index dropping below -200nT. This strong geomagnetic disturbance was triggered by the arrival of a coronal mass ejection on March 15, which caused strong geomagnetic disturbances in the ionosphere over Europe.

[0069] like Figure 3 As shown, left Figure 3 (a), (c) and (e) are the reconstruction results of the traditional MART iterative reconstruction method; Figure 3 (b), (d) and (f) are the reconstruction results based on the diffusion model method. Comparing the images on both sides, it is obvious that the image reconstructed by the diffusion model method has higher quality. Figure 3The data clearly shows the distribution of ionospheric electron density in Europe at 08:00, 10:00 and 12:00 UTC on March 17, 2015. The data shows that the electron density is mainly concentrated at an ionospheric altitude of about 380 kilometers, especially in low-latitude areas. In contrast, the electron density in the ionospheric regions of 100-200 kilometers and above 700 kilometers is relatively low.

[0070] like Figure 4 As shown in the figure, this example verifies the vertical profiles of electron density derived from IRI2020, traditional MART, diffusion model and ionospheric radiosonde at the same longitude and latitude at 08:00 and 10:00 UT on March 17, and takes the radiosonde profile as the true value for comparison. Figure 4 From the cross-section comparison diagram, it can be seen that the reconstruction method of the present invention is closer to the real cross-section of the ionospheric electron density. Compared with the traditional MART iterative reconstruction method, the reconstruction result of the method of the present invention is smoother and can effectively reflect the distribution of the three-dimensional electron density of the ionosphere under a strong magnetic storm environment.

Claims

1. A method for reconstructing the three-dimensional electron density of the ionosphere during a magnetic storm based on a diffusion model, characterized in that: The following steps are involved: S1. Obtain the original observation data and satellite navigation messages of the ground-based observation stations in the reconstruction area, and use the carrier phase smoothed pseudorange algorithm to calculate the slant total electron content (STEC) on the line of sight from the observation station to the satellite; S2, discretizing the ionospheric space of the reconstruction area to form a three-dimensional voxel grid; S3, eliminating the geometric sight lines with low elevation angles from the ground-based observation station to the satellite, and constructing the observation intercept matrix between the sight lines and the voxel model; S4, using the International Ionospheric Reference Model (IRI) to obtain the three-dimensional electron density of the ionosphere in the inversion area and at the inversion time; further matching the electron density at the center of each voxel as the initial value of iterative reconstruction; S5. Use the efficient multiplication iterative reconstruction algorithm (MART) to preliminarily reconstruct the 3D electron density of the ionosphere in the inversion area. S6. The electron density data obtained by the IRI2020 model is used as the benchmark for the forward training of the diffusion model. The voxels that are not penetrated by the rays during the MART reconstruction process will be regarded as measurement noise and denoised in the inverse processing to improve the overall reconstruction accuracy and quality. S7. Use the ionospheric sounding instrument data as the true value to measure the accuracy of the ionospheric three-dimensional electron density reconstruction results based on the diffusion model; if the expected effect is not achieved, adjust the diffusion model parameters in S6 to ultimately achieve high-quality ionospheric three-dimensional electron density reconstruction.

2. The method for reconstructing the three-dimensional electron density of the ionosphere during a magnetic storm based on a diffusion model according to claim 1, characterized in that: The formula for calculating STEC on the line of sight from the station to the satellite using carrier phase smoothed pseudorange in S1 is as follows: Where f1 and f2 are the carrier frequencies of L1 and L2 bands, usually 1.57542 GHz and 1.2276 GHz respectively; P sm It is a smoothed pseudo-range combined observation value, which reduces the influence of noise and multipath; DCB i and DCB j are the hardware delay deviations of the station and the satellite respectively; the unit of STEC is TECU, 1TECU = 10 16 Ne / m 2 ; STEC refers to the total amount of electrons from the satellite to the receiver in the satellite signal propagation path; Its integral expression is: Where Ne is the electron density, which indicates the The number of electrons in the ionosphere at a certain location on Earth; When doing 3D inversion, the above formula is usually discretized, and its discretized expression is: In the formula, A ij is the intercept matrix between the ray and the voxel, indicating the length of the path i passing through the jth voxel; ∈ i is the measurement error or noise; x j is the electron density of the jth voxel, which is the unknown quantity to be solved.

3. The method for reconstructing the three-dimensional electron density of the ionosphere during a magnetic storm based on a diffusion model according to claim 1, characterized in that: The intercept matrix in S3 is calculated using the following formula: In the formula, d is the intercept value of the ray passing through the voxel, X1, Y1, Z1 are the spatial positions of the ray entering one side of the voxel; X2, Y2, Z2 are the spatial positions of the ray passing through the other side of the voxel, and the unit of d is generally m.

4. The method for reconstructing the three-dimensional electron density of the ionosphere during a magnetic storm based on a diffusion model according to claim 1, characterized in that: The formula of the iterative reconstruction algorithm MART in S5 is as follows: In the formula, x j k+1 is the electron density of the jth voxel after the k+1th iteration, x j k is the estimated value at the kth iteration; i T ,x k > is the TEC calculated based on the estimated value at the kth iteration; λ is the relaxation factor, which ranges from 0<λ<1; a ij is the intercept value of the i-th ray and the j-th voxel; ||a|| is the sum of the intercepts of all voxels passed by the i-th ray.​ 5. The method for reconstructing the three-dimensional electron density of the ionosphere during a magnetic storm based on a diffusion model according to claim 1, characterized in that: In S6, the forward process of the diffusion model gradually adds noise to the empirical model electron density data, converting the original electron density data into noise close to a normal distribution; then in the reverse process, the voxel electron density with low reconstruction quality and accuracy caused by the three-dimensional inversion ill-posed problem in the MART reconstruction process is treated as noise, and the global voxel electron density with high accuracy and quality is gradually reconstructed.

6. The method for reconstructing the three-dimensional electron density of the ionosphere during a magnetic storm based on a diffusion model according to claim 1, characterized in that: In S7, the root mean square error (RMSE) is used as the accuracy measurement indicator; the specific indicator size can be adjusted according to actual needs; the root mean square error formula is as follows: In the formula, x i is the true electron density extracted by ionospheric sounding instrument or incoherent scatter radar, x^ i is the electron density value after denoising by the diffusion model.

Images