A method and system for detecting ionospheric disturbances based on water vapor characteristics

Abstract

This invention provides a method and system for ionospheric disturbance detection based on water vapor characteristics, relating to the field of ionospheric monitoring. The method includes: retrieving total atmospheric water vapor from GNSS observation data and extracting water vapor temporal features; extracting meteorological temporal features from historical meteorological data; constructing a spatial grid based on latitude and longitude, matching the water vapor temporal features and meteorological temporal features to the spatial grid, and fusing the features to obtain a spatiotemporal distribution feature map; constructing a disturbance response feature map based on historical ionospheric disturbance data; training a capsule network using the spatiotemporal distribution feature map as input and the disturbance response feature map as the corresponding output to obtain a detection model; and performing ionospheric disturbance detection based on the detection model. The capsule network of this invention achieves efficient aggregation of basic features into globally evolving features through a routing mechanism. Compared with traditional deep learning models, it is more suitable for prediction scenarios of spatiotemporally continuous fields, effectively improving the targeting and effectiveness of disturbance feature extraction.

Description

Technical Field

[0001] This invention belongs to the field of ionospheric monitoring, specifically relating to a method and system for detecting ionospheric disturbances based on water vapor characteristics. Background Technology

[0002] As a crucial component of Earth's space environment, the ionosphere's disturbances significantly impact human-made technological systems such as radio communication, satellite navigation, and radar detection. Due to the horizontal orientation of the geomagnetic field, the low-latitude ionosphere is the most active and easily disturbed region globally, frequently experiencing phenomena such as the equatorial ionization anomaly (EIA), equatorial electric jet (EEJ), and plasma irregularities. Therefore, accurately monitoring and characterizing the spatiotemporal evolution of ionospheric disturbances is not only a core research topic in space weather science but also has significant application value in enhancing the anti-interference capabilities of technological systems. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for detecting ionospheric disturbances based on water vapor characteristics, thereby improving the aforementioned problems. To achieve this objective, the technical solution adopted by this invention is as follows:

[0004] In a first aspect, this application provides a method for detecting ionospheric disturbances based on water vapor characteristics, comprising:

[0005] Historical GNSS observation data is acquired, and the total atmospheric water vapor is obtained by inversion based on the GNSS observation data, and the temporal characteristics of water vapor are extracted.

[0006] Acquire historical meteorological data and extract meteorological time-series features based on the historical meteorological data;

[0007] A spatial grid is constructed based on latitude and longitude. Water vapor temporal characteristics and meteorological temporal characteristics are matched to the spatial grid and fused to obtain a spatiotemporal distribution feature map.

[0008] Acquire historical perturbation data of the ionosphere and construct a perturbation response feature map based on the historical perturbation data;

[0009] Based on the preset time delay rules, construct sample pairs of spatiotemporal distribution feature maps and disturbance response feature maps;

[0010] The spatiotemporal distribution feature map of the sample pair is used as input, and the perturbation response feature map is used as the corresponding output to train the capsule network and obtain the detection model.

[0011] Ionospheric disturbance detection based on detection model.

[0012] Secondly, this application provides an ionospheric disturbance detection system based on water vapor characteristics, comprising:

[0013] The first module is used to acquire historical GNSS observation data, retrieve the total atmospheric water vapor based on the GNSS observation data, and extract the water vapor temporal characteristics.

[0014] The second module is used to acquire historical meteorological data and extract meteorological time-series features based on the historical meteorological data.

[0015] The third module is used to construct a spatial grid based on latitude and longitude, match water vapor time series features and meteorological time series features to the spatial grid and perform feature fusion to obtain a spatiotemporal distribution feature map;

[0016] The fourth module is used to acquire historical perturbation data of the ionosphere and construct a perturbation response feature map based on the historical perturbation data;

[0017] The fifth module is used to construct sample pairs of spatiotemporal distribution feature maps and disturbance response feature maps according to preset time delay rules;

[0018] The sixth module is used to train the capsule network by taking the spatiotemporal distribution feature map of the sample pair as input and the perturbation response feature map as the corresponding output, and obtaining the detection model.

[0019] The seventh module is used for ionospheric disturbance detection based on the detection model.

[0020] The beneficial effects of this invention are as follows:

[0021] This invention constructs an ionospheric disturbance detection model based on GNSS multi-source data and capsule networks. The capsule network achieves efficient aggregation of basic features into globally evolving features through a routing mechanism. This preserves the spatial location and temporal evolution details of water vapor features while accurately uncovering the occurrence, propagation, and attenuation patterns of ionospheric disturbances driven by water vapor. Compared to traditional deep learning models, it is more suitable for prediction scenarios with spatiotemporal continuous fields, effectively improving the targeting and effectiveness of disturbance feature extraction. Furthermore, this method can output the spatiotemporal evolution results of multiple ionospheric disturbance parameters at once, eliminating the need to build multiple single-parameter models separately, significantly improving modeling efficiency.

[0022] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart of the ionospheric disturbance detection method based on water vapor characteristics according to an embodiment of this application;

[0025] Figure 2 This is a structural diagram of an ionospheric disturbance detection device based on water vapor characteristics, as described in an embodiment of this application.

[0026] Symbol explanation: 800 - Ionospheric disturbance detection method and equipment based on water vapor characteristics; 801 - Processor; 802 - Memory; 803 - Multimedia component; 804 - I / O interface; 805 - Communication component. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0028] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0029] Current ionospheric disturbance detection primarily relies on two main technologies: ground-based GNSS observation networks and space-based satellite platforms. In existing technologies, different disturbance parameters are often detected using different methods: the fragmentation of multi-source information makes it difficult to construct a unified picture of disturbance evolution. Although some studies have attempted multi-source data fusion, the physical relationships between different indices remain unclear, and significant obstacles persist in spatiotemporal registration and scale matching between ground-based and space-based products.

[0030] Example 1:

[0031] See Figure 1 This embodiment provides a method for detecting ionospheric disturbances based on water vapor characteristics, characterized by including steps S100, S200, S300, S400, S500, S600, and S700.

[0032] S100. Acquire historical GNSS observation data, retrieve the total atmospheric water vapor (PWV) based on the GNSS observation data, and extract the water vapor temporal characteristics, as follows:

[0033] S110. Obtain the observation values ​​by eliminating the ionospheric delay based on the dual-frequency GNSS observation values;

[0034] GNSS signals experience frequency-dependent delays when traversing the ionosphere, which is one of the most significant sources of error in tropospheric delay estimation. Ionospheric delay is inversely proportional to the square of the signal frequency. Using dual-frequency observations (such as GPS L1 / L2 or BeiDou B1 / B2a), an ionosphere-free combination can be constructed through linear combination, thereby eliminating the influence of the first-order ionospheric delay. After eliminating the first-order ionospheric term, the remaining parameters are mainly tropospheric delay, geometric distance, clock error, and ambiguity.

[0035] S120. By using precise single-point positioning or double-difference calculation, the total atmospheric delay in the zenith direction is separated from the observations.

[0036] S130, based on the total atmospheric delay in the zenith direction ( ZTD Calculate the zenith wet delay and the water vapor conversion coefficient based on meteorological data;

[0037] Based on air pressure and station location, the zenith dry delay is calculated using the corresponding empirical model. Then, the zenith wet delay is obtained by subtracting the zenith dry delay from the total atmospheric delay in the zenith direction. ZWD );

[0038] The water vapor conversion coefficient is calculated based on meteorological data as follows:

[0039] ;

[0040] in, The water vapor conversion coefficient, The density of liquid water; The constant of water vapor; , It is the atmospheric refractive index constant; This is the atmospheric weighted average temperature.

[0041] S140. Calculate the total atmospheric water vapor based on the water vapor conversion coefficient and zenith wet delay.

[0042] ;

[0043] in, ZWD For zenith moisture delay, The water vapor conversion coefficient; PWV The zenith dry delay.

[0044] S150. Extract the basic, spatial, and dynamic characteristics of water vapor based on the total atmospheric water vapor content;

[0045] Basic characteristics include total atmospheric water vapor. PWV The value also includes the relative humidity equivalent value ( PWV / saturation PWV );

[0046] Spatial features are used to describe the two-dimensional structure of the water vapor field, including:

[0047] A water vapor distribution field is constructed based on the total atmospheric water vapor content, and the water vapor distribution field is binarized to obtain the high water vapor region;

[0048] Using the sliding window method, a water vapor subfield of a preset-sized window (e.g., 50km × 50km) is taken for each grid.

[0049] Calculate the fractal dimension of the high water vapor region in each water vapor subfield, and assign the calculation result to the central grid point to obtain the local fractal dimension of each grid.

[0050] Specifically, the box counting method is used to calculate the fractal dimension. The high water vapor region is covered by a grid with a side length of ε. The number of grids containing the high water vapor region, N(ε), is counted. By changing ε, a series of (ε, N(ε)) are obtained. Then, the log N(ε) ~ log(1 / ε) curve is plotted in a double logarithmic coordinate system. The slope of the linear regression of the curve is the fractal dimension D.

[0051] The optimal feature scale of the grid is determined based on the local fractal dimension, and the water vapor gradient is calculated based on the optimal feature scale to obtain the water vapor gradient characteristics.

[0052] Regions with high fractal dimension indicate complex water vapor structures and dramatic small-scale changes; regions with low fractal dimension show linear water vapor structures (such as fronts), and large-scale gradients better reflect the main characteristics.

[0053] As an alternative approach, the local fractal dimension D can be calculated using windows of different sizes, and then a curve showing the change of fractal dimension with window size can be constructed. The inflection point in the curve can be identified, and the window size corresponding to the inflection point can be taken as the optimal feature scale. This method may find multiple optimal feature scales.

[0054] As an alternative approach, a mapping function between the local fractal dimension D(x,y) and the optimal feature scale can be constructed in advance based on experience, so that the value of the fractal dimension D can be mapped to a suitable scale.

[0055] The water vapor gradient is calculated based on the determined optimal feature scale, and the water vapor gradient characteristics are obtained.

[0056] Within the defined optimal feature scale range, the water vapor gradient at each grid point is calculated, including the gradient magnitude and gradient direction, to obtain the water vapor gradient feature. If there are multiple optimal feature scales, the water vapor gradient within each optimal feature scale range is calculated separately, and then weighted and fused to obtain the comprehensive water vapor gradient feature.

[0057] The dynamic features include PWV Rate of change at multiple time scales (1 hour, 3-6 hours, 6-12 hours).

[0058] S200: Acquire meteorological data and extract meteorological time-series characteristics;

[0059] The acquired meteorological data includes water vapor profiles, temperature profiles, horizontal wind fields, and convection-related indices. These meteorological data need to be time-aligned with the water vapor data.

[0060] Water vapor profiles are used to determine the vertical structure of water vapor and the depth of convection development.

[0061] Temperature profiles are used to determine convective instability and gravity wave propagation conditions.

[0062] Horizontal wind fields are used to reflect water vapor transport, wind shear, and horizontal propagation of gravity waves.

[0063] Convection-related indices such as CAPE (convective available potential energy), CIN (convective inhibition energy), and uplift index can directly reflect the presence of strong convection, which is most likely to excite ionospheric disturbances.

[0064] For the above data, time series data are extracted according to the same time step as the water vapor data to obtain meteorological time series characteristics.

[0065] S300: Construct a spatial grid according to latitude and longitude, match water vapor temporal characteristics and meteorological temporal characteristics to the spatial grid and perform feature fusion to obtain a spatiotemporal distribution feature map;

[0066] In this embodiment, instead of directly using fixed grid parameters, the size and resolution of the spatial grid are determined using the following method when constructing the spatial grid:

[0067] S310. Based on historical ionospheric disturbance data, identify local disturbances and calculate the minimum spatial scale of the local disturbances.

[0068] This step first requires capturing local disturbances such as small- to medium-scale gravity waves and convection-induced disturbances. The spatial scale of these disturbances is approximately 30-100 km, and the final determination is based on the smallest spatial scale obtained from the calculation.

[0069] S320. Determine latitude and longitude resolution based on the minimum spatial scale;

[0070] Converting the smallest spatial scale to latitude and longitude, 30-100km corresponds to a latitude and longitude resolution of 0.25°-0.5° (1°≈111km).

[0071] S330. Obtain the capsule quantity safety threshold, and calculate the grid quantity threshold based on the capsule quantity safety threshold and the data sampling time step.

[0072] As an example, the safe threshold for the number of capsules is 9600, and the data sampling time step is 24; the safe threshold for the number of capsules / the data sampling time step = the grid number threshold, so the number of grids is 400, and the resolution is 20×20;

[0073] S340. Construct a spatial grid based on latitude and longitude resolution and grid number threshold;

[0074] The above method can ensure that the spatial grid is fully adapted to the water vapor-ionospheric coupling scale, while avoiding the waste of computing power.

[0075] Since the data obtained in steps S100 and S200 may come from different observation points and cannot cover the target area, nor can they be directly aligned, it is necessary to resample these data to each grid through bilinear interpolation; then the data features are extracted and fused to obtain the features of the grid points.

[0076] The final result is a spatiotemporal distribution feature map with a resolution of 20×20, where each pixel (grid) has fused multi-dimensional features.

[0077] S400. Obtain historical disturbance data of the ionosphere and construct a disturbance response feature map based on the historical disturbance data;

[0078] Based on historical GNSS observation data, the total electron content of the ionosphere is inverted, and the rate of change of total electron content is calculated. The zenith total electron content VTEC is obtained by using the first-order term elimination and zenith projection method of the ionosphere.

[0079] The standard deviation of the VTEC time series of the same satellite and the same station is calculated by sliding window to obtain the TEC change rate of the satellite. The total electron content change rate index ROTI is obtained by fusing the observation results of multiple satellites, that is, by taking the root mean square of the TEC change rate of all visible satellites.

[0080] Amplitude scintillation index and phase scintillation index of ionospheric history were obtained from the observation station;

[0081] All historical perturbation data of the ionosphere are resampled to the unified spatial grid constructed above through linear interpolation;

[0082] Based on various historical disturbance data in the spatial grid, corresponding disturbance response feature maps are constructed. In this embodiment, a disturbance response feature map is constructed for each type of disturbance feature, which can yield disturbance response feature maps for VTEC, ROTI, etc.

[0083] S500. Construct sample pairs of spatiotemporal distribution feature maps and disturbance response feature maps according to preset time delay rules;

[0084] The influence of atmospheric water vapor on the ionosphere has a time lag, so it is necessary to determine the time lag rule based on experience. The spatiotemporal distribution characteristic map corresponding to time t and the disturbance response characteristic map at time t+n are used as a pair of samples, where n can be 1-6 hours.

[0085] S600. Using the spatiotemporal distribution feature map of the sample pair as input and the perturbation response feature map as the corresponding output, train the capsule network to obtain the detection model.

[0086] Water vapor-ionospheric coupling is characterized by spatiotemporal lag, nonlinearity, and regional differences, making it sensitive to changes in time and space. Traditional CNNs only extract spatial features, while RNNs only process temporal sequences, making it difficult to simultaneously model spatial hierarchy and temporal dependencies. This method uses capsule networks, which can effectively capture the spatial hierarchy and feature pose information of the data, thus overcoming the shortcomings of traditional CNNs / RNNs in spatiotemporal feature modeling.

[0087] By performing sliding convolution on the spatiotemporal distribution feature map in the three directions of time, latitude, and longitude, the basic spatiotemporal blocks are extracted to obtain the convolutional feature map;

[0088] Each input sample = time step T × latitude H × longitude W × feature channel C; simultaneously, sliding convolution is performed in the three directions of time, latitude, and longitude to extract the underlying spatiotemporal features;

[0089] By preserving the feature dimension of the convolutional feature map and merging and flattening the other dimensions, a standard capsule vector is reconstructed. This step is to convert the convolutional feature map into capsule format, which requires merging the other dimensions and retaining only the feature channels as the basic dimensions of the capsule vector. That is, [T, H, W, C] is reconstructed into [N, C], where: N = T×H×W, N represents the total number of spatiotemporal grids (i.e., the number of primary capsules); C represents the vector dimension of each primary capsule (i.e., the length of the capsule).

[0090] Normalize the magnitude of the standard capsule vector to obtain the vector capsule; this step compresses the vector magnitude to between 0 and 1. The magnitude represents the probability of the existence of the feature entity. For example, a magnitude close to 1 indicates that the grid has a strong water vapor gradient, which is a key precursor to ionospheric disturbance. A magnitude close to 0 indicates that there is no significant feature.

[0091] Vector capsules serve as low-level capsules; based on the needs of perturbation evolution, several high-level capsules (corresponding to different perturbation types) are preset, and through dynamic routing iteration, the low-level capsules vote for the high-level capsules, ultimately obtaining a set of core high-level capsules, which are finally decoded into a feature map of future ionospheric perturbation response.

[0092] In this embodiment, the capsule network uses multiple output channels, with each output channel outputting a perturbation response feature map.

[0093] The capsule network described above can predict multiple perturbation parameters of the ionosphere simultaneously using a single model, without the need to build multiple sets of single-parameter computational models separately.

[0094] In another implementation, other influencing features can be constructed based on solar activity data and ionospheric data from the previous 6-12 hours, and used as input to the model, enabling the model to learn the interference caused by other influencing factors.

[0095] S700, Ionospheric disturbance detection based on detection model;

[0096] By inputting the current water vapor and meteorological characteristics into the constructed detection model, future ionospheric disturbance information can be obtained. This information can be directly used for short-term early warning of ionospheric disturbances, providing reliable support for scenarios such as space weather monitoring and satellite communication assurance. At the same time, it provides an efficient technical means for the study of the coupling mechanism between the lower atmosphere and the ionosphere.

[0097] Example 2:

[0098] This embodiment provides an ionospheric disturbance detection system based on water vapor characteristics, including:

[0099] The first module is used to acquire historical GNSS observation data, retrieve the total atmospheric water vapor based on the GNSS observation data, and extract the water vapor temporal characteristics.

[0100] The second module is used to acquire historical meteorological data and extract meteorological time-series features based on the historical meteorological data.

[0101] The third module is used to construct a spatial grid based on latitude and longitude, match water vapor time series features and meteorological time series features to the spatial grid and perform feature fusion to obtain a spatiotemporal distribution feature map;

[0102] The fourth module is used to acquire historical perturbation data of the ionosphere and construct a perturbation response feature map based on the historical perturbation data;

[0103] The fifth module is used to construct sample pairs of spatiotemporal distribution feature maps and disturbance response feature maps according to preset time delay rules;

[0104] The sixth module is used to train the capsule network by taking the spatiotemporal distribution feature map of the sample pair as input and the perturbation response feature map as the corresponding output, and obtaining the detection model.

[0105] The seventh module is used for ionospheric disturbance detection based on the detection model.

[0106] As an optional implementation, the first module includes:

[0107] The first unit is used to eliminate ionospheric delay from dual-frequency GNSS observations to obtain the observation values;

[0108] The second unit is used to separate the total atmospheric delay in the zenith direction from the observations through precise single-point positioning or double-difference calculation;

[0109] The third unit is used to calculate the zenith wet delay based on the total atmospheric delay in the zenith direction and to calculate the water vapor conversion coefficient based on meteorological data.

[0110] The fourth unit is used to calculate the total atmospheric water vapor based on the water vapor conversion coefficient and the zenith wet delay.

[0111] The fifth unit is used to extract the basic, spatial, and dynamic characteristics of water vapor based on the total atmospheric water vapor content.

[0112] As an optional implementation, the system further includes:

[0113] The eighth module is used to perform sliding convolution on the spatiotemporal distribution feature map in the three directions of time, latitude, and longitude to extract the basic spatiotemporal blocks and obtain the convolutional feature map;

[0114] The ninth module is used to reconstruct the standard capsule vector by preserving the feature dimensions in the convolutional feature map, merging and flattening the other dimensions;

[0115] The tenth module is used to normalize the magnitude of the standard capsule vector to obtain the vector capsule.

[0116] As an optional implementation, the fourth module includes:

[0117] The sixth unit is used to invert the total electron content of the ionosphere based on historical GNSS observation data and calculate the total electron content change rate index.

[0118] The seventh unit is used to obtain the amplitude scintillation index and phase scintillation index of the ionospheric history;

[0119] The eighth unit is used to resample all historical ionospheric perturbation data to the spatial grid using linear interpolation;

[0120] Unit 9 is used to construct corresponding disturbance response feature maps based on various historical disturbance data in the spatial grid.

[0121] Example 3:

[0122] Corresponding to the above method embodiments, this embodiment also provides an ionospheric disturbance detection method and device based on water vapor characteristics. The ionospheric disturbance detection method and device based on water vapor characteristics described below can be referred to in correspondence with the ionospheric disturbance detection method based on water vapor characteristics described above.

[0123] Figure 2 This is a block diagram illustrating an ionospheric disturbance detection method and apparatus 800 based on water vapor characteristics, according to an exemplary embodiment. Figure 2 As shown, the ionospheric disturbance detection method device 800 based on water vapor characteristics includes a processor 801 and a memory 802. The ionospheric disturbance detection method device 800 may also include one or more of the following: a multimedia component 803, an input / output (I / O) interface 804, and a communication component 805. The processor 801 controls the overall operation of the ionospheric disturbance detection method device 800 to complete all or part of the steps in the aforementioned ionospheric disturbance detection method based on water vapor characteristics. The memory 802 stores various types of data to support the operation of the ionospheric disturbance detection method device 800. This data may include, for example, commands for any application or method operating on the ionospheric disturbance detection method device 800, and application-related data such as contact data, sent and received messages, images, audio, video, etc. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0124] Multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touchscreen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals.

[0125] The received audio signal can be further stored in memory 802 or transmitted via communication component 805. The audio component also includes at least one speaker for outputting the audio signal. I / O interface 804 provides an interface between processor 801 and other interface modules, such as a keyboard, mouse, buttons, etc. These buttons can be virtual or physical. Communication component 805 is used for wired or wireless communication between the device 800 and other devices based on water vapor characteristics. Wireless communication includes, for example, Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination thereof; therefore, the corresponding communication component 805 may include a Wi-Fi module, a Bluetooth module, or an NFC module.

[0126] Example 4:

[0127] Corresponding to the above embodiment of the ionospheric disturbance detection method based on water vapor characteristics, this embodiment also provides a readable storage medium. The readable storage medium described below can be referred to in correspondence with the ionospheric disturbance detection method based on water vapor characteristics described above.

[0128] A readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described embodiment of the ionospheric disturbance detection method based on water vapor characteristics.

[0129] Specifically, the readable storage medium can be a USB flash drive, external hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, or any other readable storage medium capable of storing program code.

[0130] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0131] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for detecting ionospheric disturbances based on water vapor characteristics, characterized in that, include: Historical GNSS observation data is acquired, and the total atmospheric water vapor is obtained by inversion based on the GNSS observation data, and the temporal characteristics of water vapor are extracted. Acquire historical meteorological data and extract meteorological time-series features based on the historical meteorological data; A spatial grid is constructed based on latitude and longitude. Water vapor temporal characteristics and meteorological temporal characteristics are matched to the spatial grid and fused to obtain a spatiotemporal distribution feature map. Acquire historical perturbation data of the ionosphere and construct a perturbation response feature map based on the historical perturbation data; Based on the preset time delay rules, construct sample pairs of spatiotemporal distribution feature maps and disturbance response feature maps; The spatiotemporal distribution feature map of the sample pair is used as input, and the perturbation response feature map is used as the corresponding output to train the capsule network and obtain the detection model. Ionospheric disturbance detection based on detection model.

2. The ionospheric disturbance detection method based on water vapor characteristics according to claim 1, characterized in that, Historical GNSS observation data is acquired, and the total atmospheric water vapor content is retrieved based on the GNSS observation data. Temporal characteristics of the water vapor are then extracted, including: The observation values ​​are obtained by eliminating the ionospheric delay from the dual-frequency GNSS observation values; The total atmospheric delay in the zenith direction is separated from the observations by precise single-point positioning or double-difference calculation. The zenith wet delay is calculated based on the total atmospheric delay in the zenith direction, and the water vapor conversion coefficient is calculated based on meteorological data. The total amount of atmospheric water vapor is calculated based on the water vapor conversion coefficient and the zenith wet delay. The basic, spatial, and dynamic characteristics of water vapor are extracted based on the total atmospheric water vapor content.

3. The ionospheric disturbance detection method based on water vapor characteristics according to claim 1, characterized in that, The method includes: By performing sliding convolution on the spatiotemporal distribution feature map in the three directions of time, latitude, and longitude, the basic spatiotemporal blocks are extracted to obtain the convolutional feature map; By preserving the feature dimensions in the convolutional feature map and merging and flattening the other dimensions, a standard capsule vector is reconstructed. Normalize the magnitude of the standard capsule vector to obtain the vector capsule.

4. The ionospheric disturbance detection method based on water vapor characteristics according to claim 1, characterized in that, The construction of the disturbance response feature map based on historical disturbance data includes: Based on historical GNSS observation data, the total electron content of the ionosphere is inverted, and the rate of change of total electron content is calculated. Obtain the amplitude scintillation index and phase scintillation index of the ionospheric history; All historical perturbation data of the ionosphere were resampled to the spatial grid using linear interpolation. Construct corresponding disturbance response feature maps based on various historical disturbance data in the spatial grid.

5. The ionospheric disturbance detection method based on water vapor characteristics according to claim 1, characterized in that, The method includes: Based on historical ionospheric disturbance data, local disturbances are identified, and the minimum spatial scale of these local disturbances is calculated. Determine latitude and longitude resolution based on the minimum spatial scale; Obtain the safe threshold for the number of capsules, and calculate the threshold for the number of grids based on the safe threshold for the number of capsules and the data sampling time step; A spatial grid is constructed based on latitude and longitude resolution and a grid number threshold.

6. The ionospheric disturbance detection method based on water vapor characteristics according to claim 2, characterized in that, The extraction of basic, spatial, and dynamic characteristics of water vapor based on total atmospheric water vapor volume includes: A water vapor distribution field is constructed based on the total atmospheric water vapor content, and the water vapor distribution field is binarized to obtain the high water vapor region; The sliding window method is used to capture the water vapor subfield of a window of a preset size around each grid. Calculate the fractal dimension of the high water vapor region in each water vapor subfield, and assign the calculation result to the central grid point to obtain the local fractal dimension of each grid. The optimal feature scale of the grid is determined based on the local fractal dimension, and the water vapor gradient is calculated based on the optimal feature scale to obtain the water vapor gradient characteristics.

7. An ionospheric disturbance detection system based on water vapor characteristics, characterized in that, include: The first module is used to acquire historical GNSS observation data, retrieve the total atmospheric water vapor based on the GNSS observation data, and extract the water vapor temporal characteristics. The second module is used to acquire historical meteorological data and extract meteorological time-series features based on the historical meteorological data. The third module is used to construct a spatial grid based on latitude and longitude, match water vapor time series features and meteorological time series features to the spatial grid and perform feature fusion to obtain a spatiotemporal distribution feature map; The fourth module is used to acquire historical perturbation data of the ionosphere and construct a perturbation response feature map based on the historical perturbation data; The fifth module is used to construct sample pairs of spatiotemporal distribution feature maps and disturbance response feature maps according to preset time delay rules; The sixth module is used to train the capsule network by taking the spatiotemporal distribution feature map of the sample pair as input and the perturbation response feature map as the corresponding output, and obtaining the detection model. The seventh module is used for ionospheric disturbance detection based on the detection model.

8. The ionospheric disturbance detection system based on water vapor characteristics according to claim 7, characterized in that, The first module includes: The first unit is used to eliminate ionospheric delay from dual-frequency GNSS observations to obtain the observation values; The second unit is used to separate the total atmospheric delay in the zenith direction from the observations through precise single-point positioning or double-difference calculation; The third unit is used to calculate the zenith wet delay based on the total atmospheric delay in the zenith direction and to calculate the water vapor conversion coefficient based on meteorological data. The fourth unit is used to calculate the total atmospheric water vapor based on the water vapor conversion coefficient and the zenith wet delay. The fifth unit is used to extract the basic, spatial, and dynamic characteristics of water vapor based on the total atmospheric water vapor content.

9. The ionospheric disturbance detection system based on water vapor characteristics according to claim 7, characterized in that, The system includes: The eighth module is used to perform sliding convolution on the spatiotemporal distribution feature map in the three directions of time, latitude, and longitude to extract the basic spatiotemporal blocks and obtain the convolutional feature map; The ninth module is used to reconstruct the standard capsule vector by preserving the feature dimensions in the convolutional feature map, merging and flattening the other dimensions; The tenth module is used to normalize the magnitude of the standard capsule vector to obtain the vector capsule.

10. The ionospheric disturbance detection system based on water vapor characteristics according to claim 7, characterized in that, The fourth module includes: The sixth unit is used to invert the total electron content of the ionosphere based on historical GNSS observation data and calculate the total electron content change rate index. The seventh unit is used to obtain the amplitude scintillation index and phase scintillation index of the ionospheric history; The eighth unit is used to resample all historical ionospheric perturbation data to the spatial grid using linear interpolation; Unit 9 is used to construct corresponding disturbance response feature maps based on various historical disturbance data in the spatial grid.

Images