Surface rain measurement system based on transparent forwarding payload

Abstract

This application provides an areal rainfall measurement system based on a transparent transponder payload, belonging to the field of rainfall prediction. The system includes: multiple transmitting stations, a satellite, multiple receiving stations, and an areal rainfall monitoring module. The transmitting stations transmit uplink signals to the satellite. The satellite includes a transparent transponder payload module, which receives the uplink signals, converts them into downlink signals, and transmits them to the receiving stations. The areal rainfall detection module constructs an areal rainfall distribution map based on the downlink signals received by the multiple receiving stations. This system simultaneously utilizes the attenuation of both uplink and downlink signals for rainfall prediction, which can improve prediction accuracy.

Description

Technical Field

[0001] This application belongs to the field of rainfall prediction, and in particular relates to a areal rainfall measurement system based on transparent forwarding payload. Background Technology

[0002] Currently, commonly used methods for monitoring rainfall mainly include ground-based rain gauges, weather radar, and satellite remote sensing. While ground-based rain gauges offer high measurement accuracy, their coverage is limited, making it difficult to reflect large-scale, continuous precipitation distribution. Weather radar has good spatial resolution, but suffers from blind spots and terrain obstruction, and its deployment and maintenance costs are high. Satellite remote sensing provides global precipitation coverage, but its resolution and real-time performance are insufficient, and it suffers from indirect estimation errors of cloud precipitation, making it difficult to accurately reflect actual rainfall intensity at the ground.

[0003] To address the aforementioned shortcomings, rain attenuation retrieval methods based on satellite-to-ground microwave links have gradually gained attention in recent years. This method estimates rainfall intensity by measuring the path attenuation of microwave signals under rainfall conditions, and has the advantages of being unaffected by weather or terrain, and enabling real-time continuous observation.

[0004] However, current research and engineering applications mainly rely on the downlink signals of communication satellites, that is, using the power attenuation of the downlink frequency band of satellite communication to invert the average rainfall intensity of the coverage area. Existing rainfall monitoring systems based on satellite-to-ground microwave links generally rely on the unidirectional signals of the downlink of communication satellites, and invert rainfall intensity through empirical models. This only obtains power attenuation information for a single path and a single polarization, lacking a complete characterization of the rainfall field distribution along the entire path. In addition, the frequency band, polarization, and transmit power of commercial communication links are determined by operational requirements, making it impossible to optimize the configuration for rainfall monitoring. This limits the sensitivity to different raindrop spectra and various rainfall processes, and also prevents the flexible adjustment of measurement strategies to cope with complex climate and terrain conditions. Summary of the Invention

[0005] In view of this, this application provides a areal rainfall measurement system based on transparent forwarding payload, aiming to improve the accuracy of rainfall inversion. The system includes: Multiple launch stations, satellites, multiple receiving stations, and areal rainfall monitoring modules; The transmitting station is used to transmit uplink signals to the satellite; The satellite includes a transparent forwarding payload module, which is used to receive uplink signals, convert the uplink signals into downlink signals, and then send them to the receiving station. The areal rainfall detection module constructs an areal rainfall distribution map based on downlink signals received from multiple receiving stations.

[0006] Optionally, the launch station includes: Radio frequency signal source, used to generate microwave signals or continuous waves in the Ka / Ku band as uplink signals; A power amplifier, used to amplify the power of the uplink signal; The transmitting antenna is used to adjust the polarization of the uplink signal and to transmit the uplink signal to the satellite.

[0007] Optionally, the receiving station includes: The receiving antenna is configured to have the same polarization as the transmitting antenna and is used to receive downlink signals; A filter is used to filter downlink signals. Low-noise amplifiers are used to improve the signal-to-noise ratio of downlink signals.

[0008] Optionally, the transparent forwarding payload module is a multi-level superheterodyne link or a direct frequency conversion architecture.

[0009] Optionally, the transparent forwarding payload module includes: An uplink receiving antenna system, wherein the uplink receiving antenna system adopts a multi-beam design and is configured to receive different types of polarized signals and to receive uplink signals from multiple transmitting stations; The uplink RF link is used to filter the uplink signal and improve the signal-to-noise ratio of the uplink signal, and to allocate the uplink signal to different channels according to the polarization type of the uplink signal to avoid crosstalk between polarizations. A frequency conversion unit is used to convert uplink signals into downlink signals with downlink frequency bands; The downlink RF link is used to compensate for the frequency conversion loss of the downlink signal and to dynamically control the power of the downlink signal so that the frequency of the downlink signal is maintained within a preset range. The downlink transmitting antenna system and the downlink receiving antenna system adopt a multi-beam design and are configured to transmit different types of polarized signals and to transmit downlink signals to multiple receiving stations.

[0010] Optionally, the areal rainfall detection module performs areal rainfall monitoring based on downlink signals received from multiple receiving stations, including the following steps: Classify sunny and rainy seasons based on the timing power variation of the downlink signal received by each receiving station; Based on the classification results of sunny and rainy seasons for each receiving station, rain-induced attenuation is calculated for the downlink signal received by each receiving station. Rainfall intensity at each receiving station is inverted based on the rainfall-induced attenuation at each receiving station. Based on the rainfall intensity inversion results from multiple receiving stations, areal rainfall data are constructed.

[0011] Optionally, the step of classifying rainy / sunny periods based on the timing power variation of the downlink signal received by each receiving station includes: Obtain the standard deviation, initial value, and information entropy of the preprocessed downlink signal in historical states to form a multidimensional signal feature vector, and use the multidimensional signal feature vector to form a dataset. Based on the dataset, a rainy season classification model is trained using the rainy season classification results as training labels. Based on the preprocessed downlink signal, the standard deviation, trend and information entropy of the downlink signal are extracted to form a multidimensional signal feature vector. The multidimensional signal feature vector is then input into the pre-trained weather classification model to obtain real-time weather classification results.

[0012] Optionally, based on the rain-induced attenuation classification results for each receiving station, the steps for calculating the rain-induced attenuation of the downlink signal received by each receiving station include: Obtain tropospheric attenuation in historical states This forms a dataset; Based on the dataset, rain-induced attenuation To train the labels, a rain-induced attenuation prediction model is trained. Tropospheric attenuation The input is fed into the pre-trained attenuation baseline prediction model to obtain the corresponding attenuation baseline prediction output. ; Predict output based on the obtained attenuation baseline Calculate rain-induced attenuation If the classification result for sunny / rainy season is rainy season, then in solving for rain-induced attenuation... At that time, the influence of the melted layer is removed.

[0013] Optionally, the step of retrieving the rainfall intensity for each receiving station based on the rainfall-induced attenuation of each receiving station includes: Obtain rain-induced attenuation under historical conditions The dataset is formed by multi-dimensional signal feature vectors; the multi-dimensional signal feature vectors are constructed based on the standard deviation, trend, and information entropy of the downlink signals. Based on the dataset, a rainfall intensity prediction model is trained using rainfall intensity as the training label. Rain-induced attenuation at each receiving station will be acquired in real time. The multidimensional signal feature vector is input into the rainfall intensity prediction model to obtain the rainfall intensity at each receiving station.

[0014] Optionally, the steps for constructing an areal rainfall distribution map based on the rainfall intensity inversion results from multiple receiving stations include: Using the Kriging interpolation method, an areal rainfall distribution map was constructed based on the rainfall intensity at multiple receiving stations.

[0015] The beneficial effects of the technical solution provided in this application include: This application provides a areal rainfall measurement system based on a transparent transponder payload. In this system, the satellite transmits uplink and downlink signals through a transparent transponder payload module. The transparent transponder payload provides a new technical approach for rainfall monitoring based on satellite-to-ground links. A transparent transponder payload is a type of payload that does not digitally demodulate, regenerate, or recode the signal; the satellite only performs analog amplification and frequency conversion on the uplink signal before directly transponding it to the ground. By performing only analog amplification and frequency conversion on the uplink signal, without involving digital demodulation and regeneration, the transparent transponder payload fully preserves the physical attenuation information of the signal in both uplink and downlink paths, avoiding amplitude interference caused by error correction and automatic gain control in traditional downlink monitoring. Compared to methods that only utilize the downlink, transparent transponder can simultaneously capture the rainfall attenuation effects of both uplink and downlink, and also supports flexible configuration of frequency bands and polarization modes, improving sensitivity to different rainfall types and particle size distributions. It possesses higher inversion accuracy and stronger system controllability, making it suitable for constructing dedicated, multi-link areal rainfall measurement networks. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 A schematic diagram of the structure of a surface rainfall measurement system based on a transparent forwarding payload provided in an embodiment of this application; Figure 2 A structural block diagram of a surface rainfall measurement system based on a transparent forwarding payload provided in an embodiment of this application; Figure 3 This is a structural block diagram of a launch station provided in one embodiment of this application; Figure 4 This is a structural block diagram of a transparent forwarding payload module for a multi-level superheterodyne link provided in an embodiment of this application; Figure 5 This is a structural block diagram of a transparent forwarding payload module of a direct frequency conversion architecture provided in an embodiment of this application; Figure 6 This is a structural block diagram of a receiving station provided in an embodiment of this application; Figure 7 A flowchart illustrating a method for constructing an areal rainfall distribution map according to an embodiment of this application.

[0018] The attached figures are labeled as follows: 1: Transmitting station; 11: Radio frequency signal source; 12: Power amplifier; 13: Transmitting antenna; 2: Satellite; 21: Transparent transponder payload module; 211: Uplink receiving antenna system; 212: Uplink RF link; 213: Frequency conversion unit; 214: Downlink RF link; 215: Downlink transmitting antenna system; 216: Wideband low-noise amplifier; 217: Direct conversion mixer; 3: Receiving station; 31: Receiving antenna; 32: Filter; 33: Low-noise amplifier; 4: Surface rainfall detection module. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this 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 this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0020] Figure 1 This is a schematic diagram of a surface rainfall measurement system based on a transparent forwarding payload, provided as an embodiment of this application. See also... Figure 1 The system includes: Multiple transmitting stations 1, satellite 2, multiple receiving stations 3, and areal rainfall monitoring module 4; The transmitting station 1 is used to transmit uplink signals to the satellite 2; The satellite 2 includes a transparent forwarding payload module 21, which is used to receive uplink signals, convert the uplink signals into downlink signals, and send them to the receiving station 3. The areal rainfall detection module 4 constructs an areal rainfall distribution map based on the downlink signals received by multiple receiving stations 3.

[0021] In this embodiment, the ground station network consisting of transmitting station 1 and receiving station 3 is deployed according to the satellite coverage area, including multiple transmitting stations 1 and receiving stations. Transmitting station 1 sends continuous or modulated carrier signals to satellite 2 and records the transmission power. Receiving station 3 receives the signals relayed back from satellite 2 and records changes in the received power. Transmitting station 1 and receiving station 1 need to have high stability and calibrability. Each ground station achieves time-frequency synchronization within the network through GPS timing or satellite two-way link time synchronization, ensuring that observation data from different stations can be compared.

[0022] See Figure 2 To further describe the specific structure of the areal rainfall measurement system based on transparent forwarding payload provided in this application, this application provides a structural block diagram of the areal rainfall measurement system based on transparent forwarding payload.

[0023] See Figure 3 In some examples, launch station 1 includes: Radio frequency signal source 11 is used to generate microwave signals or continuous waves in the Ka / Ku band as uplink signals. Power amplifier 12 is used to amplify the power of the uplink signal; Transmitting antenna 13 is used to adjust the polarization state of the uplink signal and to transmit the uplink signal to the satellite.

[0024] Transmitting station 1 is equipped with a highly stable radio frequency signal source 11, capable of generating continuous wave (CW) or specifically modulated microwave signals in the Ka / Ku uplink frequency band. The generated radio frequency signal is amplified by power amplifier 12 to a level sufficient for uplink transmission to the satellite. Power amplifier 12 supports dynamic power control and calibration mechanisms to ensure precise and controllable transmission power. Transmitting antenna 1 of transmitting station 1 supports dual-polarized or circularly polarized signal transmission and can be configured as horizontal / vertical polarization or left-hand / right-hand circular polarization according to observation requirements, facilitating correspondence with the polarization channel of the satellite's transparent relay. Finally, uplink signal transmission is achieved through transmitting antenna 13.

[0025] In some examples, satellite 2 is a communication satellite. The specific structure of satellite 2 is not related to the inventive concept of this application and belongs to the prior art. The structure of existing communication satellites can be referred to, and this application does not limit it. Compared with the structure of existing communication satellites, the satellite 2 provided by this application additionally has a transparent relay payload module 21.

[0026] See Figure 4 In some examples, the transparent forwarding payload module 21 is a multi-level superheterodyne link or a direct frequency conversion architecture.

[0027] In some examples, the transparent forwarding payload module 21 includes: Uplink receiving antenna system 211, the uplink receiving antenna system adopts a multi-beam design and is configured to receive different types of polarized signals and to receive uplink signals from multiple transmitting stations; Uplink RF link 212 is used to filter uplink signals, improve the signal-to-noise ratio of uplink signals, and allocate uplink signals to different channels according to the polarization type of uplink signals to avoid crosstalk between polarizations. Frequency conversion unit 213 is used to convert uplink signals into downlink signals with downlink frequency bands; The downlink RF link 214 is used to compensate for the frequency conversion loss of the downlink signal and to dynamically control the power of the downlink signal so that the frequency of the downlink signal is maintained within a preset range. The downlink transmitting antenna system 215, which employs a multi-beam design, is configured to transmit different types of polarized signals and to transmit downlink signals to multiple receiving stations.

[0028] In this embodiment, the uplink receiving antenna system 211 is responsible for receiving microwave signals from the ground transmitting station to achieve high sensitivity to rainfall attenuation. The antenna adopts a multi-beam design, which can cover a wide ground area and enable multiple ground stations to access simultaneously. At the same time, the antenna has multi-polarization receiving capability, supporting horizontal polarization, vertical polarization, as well as left-hand and right-hand circular polarization, to adapt to the relay requirements of different polarization signals, facilitating subsequent ground-based inversion of information such as rain attenuation and raindrop size distribution using polarization characteristics.

[0029] In this embodiment, the signal acquired by the receiving antenna of the uplink RF link 212 first enters a high-selectivity bandpass filter to filter out out-of-band interference and clutter, ensuring a clean signal spectrum. Subsequently, the signal passes through a low-noise amplifier (LNA) to boost the amplitude of the received signal while maintaining an extremely low noise figure, ensuring the signal-to-noise ratio meets the requirements of subsequent processing. Based on this, the link is configured with a polarization separation and management module to guide signals of different polarizations to independent channels, preventing inter-polarization crosstalk and ensuring that physical characteristics are not distorted.

[0030] In this embodiment, in the frequency conversion unit 213, the LNA output is mixed with the local oscillator (LO) by the first mixer to be down-converted to an intermediate frequency (IF) signal; the IF signal is then subjected to bandpass filtering and bandwidth limiting before entering the second mixer to be up-converted to the target downlink frequency band. The entire link can be designed with single-stage or multi-stage mixing as required. Both the mixer and the phase-locked loop local oscillator are high-linearity, high-suppression schemes to ensure spectral purity and minimum conversion loss.

[0031] In this embodiment, the downlink RF link 214 first sends the frequency-converted downlink band signal to the pre-amplification module to compensate for conversion losses. It then enters the main power amplification unit, where a traveling wave tube amplifier (TWTA) or a solid-state power amplifier (SSPA) is selected based on power level requirements. The amplification link integrates automatic gain control (AGC) and saturation protection, dynamically adjusting the output power to ensure that the output signal's inversion accuracy is not affected by power overload or underpower throughout the entire link. After power amplification, the signal passes through a downlink bandpass filter to further suppress out-of-band emissions and nonlinear products, ensuring the spectral purity and stability of the downlink signal.

[0032] In this embodiment, in the downlink transmit antenna system 215, the downlink transmit antenna is responsible for radiating the amplified radio frequency signal to the ground receiving station. The antenna design also supports multi-beam and multi-polarization transmission, corresponding to the polarization configuration of the uplink receive antenna, ensuring polarization integrity and undistorted signal characteristics. The antenna features a high-gain, low-sidelobe design, guaranteeing signal coverage for distant ground stations while reducing interference to non-target areas. Furthermore, the antenna system is equipped with a polarization combining and separating module, dynamically selecting and switching between different polarization channels to further improve link stability and signal inversion accuracy.

[0033] See Figure 5 In other embodiments provided in this application, the transparent forwarding payload module can be equivalently replaced by a direct conversion (Zero-IF) architecture instead of a traditional multi-stage superheterodyne link. It utilizes a broadband low-noise amplifier (LNA) and a direct conversion mixer to achieve uplink-to-downlink frequency conversion and amplification. Traveling wave tube amplifiers (TWTA) and solid-state power amplifiers (SSPA) can be interchanged to meet different power and linearity requirements. The antenna can be replaced by a phased array multi-polarized antenna or a doubly fed reflector antenna, which achieves rapid switching between horizontal / vertical and left / right circular polarization through electronic switching.

[0034] See Figure 6 In some examples, receiving station 3 includes: The receiving antenna 31 is configured to have the same polarization as the transmitting antenna for receiving downlink signals; Filter 32 is used to filter the downlink signal; Low-noise amplifier 33 is used to improve the signal-to-noise ratio of the downlink signal.

[0035] Receiving station 3 is equipped with a high-gain receiving antenna 31 in the same frequency band as transmitting station 1, and supports the polarization configuration corresponding to transmitting station 1, ensuring that the received signal retains complete polarization information. After the signal enters the bandpass filter to remove interference, it is then boosted by a low-noise amplifier (LNA) to improve the signal-to-noise ratio, and then the receiving front end analyzes the amplitude, phase, and polarization state of the signal.

[0036] In other embodiments provided in this application, the antennas of the ground transmitting station and the receiving station can be equivalently replaced with high-gain parabolic antennas, planar phased arrays, or adaptive beamforming antennas to flexibly point to different satellites; the LNB and cascaded LNA architecture of the receiving station can be replaced by a distributed front-end array plus a centralized digital downconversion (DDC); in terms of time and frequency synchronization, GPS / GLONASS timing can be replaced with IEEE 1588 PTP network clock synchronization.

[0037] See Figure 7In some examples provided in this application, the areal rainfall detection module 4 performs areal rainfall monitoring based on downlink signals received from multiple receiving stations, including the following steps: S101. Classify the weather based on the timing power changes of the downlink signal received by each receiving station.

[0038] In some examples, step S101 includes: The first step is to obtain the standard deviation, initial value, and information entropy of the preprocessed downlink signal in the historical state, form a multidimensional signal feature vector, and then use the multidimensional signal feature vector to form a dataset.

[0039] The second step is to train a rainy season classification model using the rainy season classification results as training labels based on the dataset.

[0040] The third step is to extract the standard deviation, trend and information entropy of the downlink signal based on the preprocessed downlink signal to form a multidimensional signal feature vector. The multidimensional signal feature vector is then input into the pre-trained weather classification model to obtain the real-time weather classification result.

[0041] In transparent forwarding links, the signal power measured by ground receiving stations includes both uplink and downlink rain-induced attenuation. To accurately identify rainfall periods, it is first necessary to classify the temporal power changes of the received signal into sunny and rainy conditions, providing a reliable basis for subsequent attenuation baseline modeling and rainfall inversion. The principle of the sunny / rainy season differentiation model is mainly based on the attenuation difference of microwave signals during sunny and rainy seasons, automatically distinguishing between sunny and rainy seasons in real time through machine learning. Based on the preprocessed microwave signal, multi-dimensional signal feature vectors such as standard deviation, trend, and information entropy are extracted. These multi-dimensional signal feature vectors are used as input to the machine learning model, outputting real-time sunny / rainy classification results, which serve as input conditions for subsequent attenuation baseline calculation and rainfall inversion modules.

[0042] In other embodiments provided in this application, in the classification of sunny and rainy seasons, the machine learning model can be equivalently replaced by the threshold empirical method or a discrimination scheme that integrates weather radar / rain gauge data.

[0043] S102. Based on the rainy / sunny season classification results of each receiving station, solve for the rain-induced attenuation of the downlink signal received by each receiving station.

[0044] In some examples, step S102 includes: Step 1: Obtain tropospheric attenuation from historical data This forms a dataset.

[0045] The second step is to use the dataset to measure rain-induced attenuation. Using training labels, a rain-induced attenuation prediction model is trained.

[0046] The third step is to attenuate the troposphere. The input is fed into the pre-trained attenuation baseline prediction model to obtain the corresponding attenuation baseline prediction output. .

[0047] Step 4: Predict the output based on the obtained attenuation baseline. Calculate rain-induced attenuation If the classification result for sunny / rainy season is rainy season, then in solving for rain-induced attenuation... At that time, the influence of the melted layer is removed.

[0048] More specifically, step S102 includes: For historical link signals during sunny periods, the system uses a Long Short-Term Memory (LSTM) network to model the normal attenuation characteristics of the link under non-rainy conditions, generating a dynamic attenuation baseline.

[0049] The power relationship of the entire link is as follows:

[0050] In the formula, This indicates the receiving power of the ground station (also the receiving power of the receiving station). This indicates the transmission power of the ground station (also the transmission power of the transmitting station). Indicates the gain of the ground station's transmitting antenna. This indicates the gain of the ground station's receiving antenna. Indicates the satellite receiving antenna gain. Indicates the satellite transmitting antenna gain. Indicates the power amplifier gain; It is free-space propagation attenuation. It refers to tropospheric attenuation, subscript. This represents the uplink (the uplink refers to the connection between the launch station and the satellite). This represents the downlink (the downlink refers to the connection between the satellite and the receiving station). It should be noted that in the above formula... It means Time, like express The ground station's received power at any given time (also known as the receiving station's received power).

[0051] The calculation formula is as follows:

[0052] in, Frequency, in MHz; The distance traveled is measured in km.

[0053] Given the power and gain of the receiving and transmitting antennas at the ground and satellite ends, and free-space propagation attenuation. The tropospheric attenuation can be calculated from the above formula. . This includes attenuation caused by oxygen decay, water vapor decay, liquid water in clouds decay, scintillation decay, precipitation decay, and other factors. The sum of attenuations excluding rainfall attenuation is called the attenuation baseline. Once the attenuation baseline is determined, rainfall attenuation can be calculated. .

[0054]

[0055] In the formula, It is the uplink attenuation baseline. It is the downlink attenuation baseline. This is the total attenuation baseline.

[0056] The attenuation baseline can be assessed using an LSTM (Long Short-Term Memory) model. Modeling and prediction are performed. First, based on the output of the sunny / rainy season classification model, the sunny season is classified... As input to the model, the LSTM network dynamically adjusts its internal parameters based on non-rainfall attenuation and historical state information, thereby achieving accurate prediction of the attenuation baseline at future times. During the rainfall period, due to... Since the data already includes rain decay, the LSTM network relies solely on previously learned time-series patterns and internal state information to continue predicting the decay baseline for the next time step. .

[0057] It is important to note that for rainfall occurring in stratiform clouds, the attenuation caused by the melting layer must be considered; that is, the effect of the melting layer needs to be removed when calculating rain-induced attenuation. The attenuation coefficient in the melting layer... The calculation formula is as follows:

[0058] in, This indicates the distance the object fell within the melt layer. The extinction area of ​​the particle. The equivalent diameter of the melted particles, The equivalent diameter of the raindrop. The maximum equivalent diameter of a raindrop. The smallest equivalent diameter of a raindrop. This represents the volume fraction of snow particles in the melted particles. , The particle densities are those of snow particles and raindrops, respectively. express The particle spectrum distribution.

[0059] In other embodiments provided in this application, the LSTM network used in attenuation baseline modeling can be replaced with ARIMA, GRU, or a causal convolutional network (TCN).

[0060] In other embodiments provided in this application, melting layer correction can be performed in addition to the physical extinction model, using the ITU-RP.840 cloud attenuation empirical formula combined with radar echo top height correction, or linear regression correction based on satellite microwave cloud probe liquid water path (LWP).

[0061] S103. Perform rainfall intensity inversion for each receiving station based on the rainfall-induced attenuation of each receiving station.

[0062] In some examples, step S103 includes: Step 1: Obtain rainfall-induced attenuation under historical conditions The dataset is formed by combining multidimensional signal feature vectors; the multidimensional signal feature vectors are constructed based on the standard deviation, trend, and information entropy of the downlink signal.

[0063] The second step is to train a rainfall intensity prediction model based on the dataset, using rainfall intensity as the training label.

[0064] The third step is to analyze the rain-induced attenuation data acquired in real time at each receiving station. The multidimensional signal feature vector is input into the rainfall intensity prediction model to obtain the rainfall intensity at each receiving station.

[0065] In this embodiment, after identifying the sunny and rainy seasons and determining the attenuation baseline, the rain attenuation component in the microwave link can be accurately extracted. To further convert rainfall attenuation into rainfall intensity, a rainfall intensity inversion model based on a neural network is constructed to achieve a nonlinear mapping from signal features to precipitation.

[0066] The neural network model comprises an input layer, hidden layers, and an output layer. This model takes rainfall attenuation and multi-dimensional signal feature vectors as input. Through a nonlinear activation function, the neural network automatically extracts deep patterns from the input features, learning the mapping relationship between rainfall attenuation and rainfall intensity. The number of hidden layers and neurons is flexibly configured according to the amount of data and prediction accuracy requirements to achieve the model's optimal expressive power. During training, historical rain gauge measurements from the same period are used as supervision labels. The neural network continuously iterates and optimizes the loss function, adjusting internal weights to improve prediction accuracy. The trained model can be directly applied to newly received satellite signal data, outputting rainfall intensity values ​​at the corresponding time and path in real time. Compared to traditional empirical models, this method has stronger nonlinear fitting capabilities and cross-regional adaptability, improving the spatial continuity and accuracy of rainfall retrieval.

[0067] In other embodiments provided in this application, in addition to multi-layer neural networks, rainfall intensity inversion can also employ ensemble learning (XGBoost, LightGBM) or graph neural network (GNN) models.

[0068] S104. Based on the rainfall intensity inversion results from multiple receiving stations, construct an areal rainfall distribution map.

[0069] In some examples, step S104 includes: Using the Kriging interpolation method, an areal rainfall distribution map was constructed based on the rainfall intensity at multiple receiving stations.

[0070] After completing the rainfall intensity inversion for multiple satellite-to-ground microwave links, discrete point rainfall information has been obtained along the paths of each link. To achieve continuous and visualized rainfall distribution over a regional area, it is necessary to extend these non-uniformly distributed point rainfall intensity data into a high spatial resolution areal rainfall field. Therefore, this study employs Kriging interpolation to spatially interpolate the inverted point rainfall intensity data and construct an areal rainfall distribution map.

[0071] Kriging interpolation is an optimal linear unbiased estimation method based on variograms, possessing good data adaptability and statistical optimality, making it particularly suitable for situations where observation points are irregularly distributed in geospatial environments. In the process of constructing areal rainfall data, the spatial covariance structure of rainfall points is first established based on the spatial location of each satellite-to-ground link and its corresponding inverted rainfall intensity, and its variability is modeled. Based on this, the Kriging method utilizes the spatial correlation between points to infer the rainfall intensity at any location, effectively filling in unobserved areas. Compared to traditional distance-weighted averaging or spline interpolation methods, Kriging interpolation not only considers spatial distance but also incorporates the spatial structural characteristics of the data. Therefore, it is more advantageous in handling variables like rainfall, which exhibit spatial heterogeneity and local abrupt changes, especially in scenarios with uneven link distribution or significant localized rainfall, providing more accurate and smooth areal rainfall estimates.

[0072] The entire satellite-to-ground link coverage area is divided into Given a grid, assuming uniform rainfall intensity within each grid and neglecting variations in rainfall intensity along vertical height, then the... Rain attenuation in a satellite-to-ground link is represented as follows:

[0073] in, Indicates the first The number of grids that the link passes through. This represents the horizontal projection length of the link on the ground. This indicates attenuation, expressed in dB / km. Indicates the link is at the 1st Projected length within each grid, This indicates the satellite pitch angle (the angle between the satellite link and the horizontal plane).

[0074] For locations not covered by satellite-to-ground links The rainfall intensity at this location It can be represented as:

[0075] in, The weighting coefficient represents and The degree of correlation can be solved by the following system of linear equations:

[0076] in, It is a Lagrange multiplier. ( , )for and The semivariance function between these two positions, which is the expected value of the difference between the values ​​at these two positions, is defined as:

[0077] in, It is the mathematical expectation, representing the average value over multiple samples or pairs of observations.

[0078] Finally, by summarizing the estimation results of all grid points, the real-time output of two-dimensional surface rainfall can be achieved.

[0079] In other embodiments provided in this application, in the areal rainfall construction stage, Kriging interpolation can be replaced by the inverse distance weighted (IDW) method to accelerate the calculation, or spatial variational interpolation (SVA) can be used in areas with significant topographic relief to enhance the physical continuity of the field; the grid structure can be replaced by a regular square grid to an irregular triangular mesh (TIN) to more flexibly fit the boundary of the study area.

[0080] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A surface rainfall measurement system based on a transparent relay payload, characterized in that, include: Multiple launch stations, satellites, multiple receiving stations, and areal rainfall monitoring modules; The transmitting station is used to transmit uplink signals to the satellite; The satellite includes a transparent forwarding payload module, which is used to receive uplink signals, convert the uplink signals into downlink signals, and then send them to the receiving station. The areal rainfall detection module constructs an areal rainfall distribution map based on downlink signals received from multiple receiving stations.

2. The areal rainfall measurement system based on transparent relay payload according to claim 1, characterized in that, The launch station includes: Radio frequency signal source, used to generate microwave signals or continuous waves in the Ka / Ku band as uplink signals; A power amplifier, used to amplify the power of the uplink signal; The transmitting antenna is used to adjust the polarization of the uplink signal and to transmit the uplink signal to the satellite.

3. The areal rainfall measurement system based on transparent relay payload according to claim 2, characterized in that, The receiving station includes: The receiving antenna is configured to have the same polarization as the transmitting antenna and is used to receive downlink signals; A filter is used to filter downlink signals. Low-noise amplifiers are used to improve the signal-to-noise ratio of downlink signals.

4. The area rainfall measurement system based on transparent relay payload according to claim 1, characterized in that, The transparent forwarding payload module is a multi-level superheterodyne link or a direct frequency conversion architecture.

5. The area rainfall measurement system based on transparent relay payload according to claim 4, characterized in that, The transparent forwarding payload module includes: An uplink receiving antenna system, wherein the uplink receiving antenna system adopts a multi-beam design and is configured to receive different types of polarized signals and to receive uplink signals from multiple transmitting stations; The uplink RF link is used to filter the uplink signal and improve the signal-to-noise ratio of the uplink signal, and to allocate the uplink signal to different channels according to the polarization type of the uplink signal to avoid crosstalk between polarizations. A frequency conversion unit is used to convert uplink signals into downlink signals with downlink frequency bands; The downlink RF link is used to compensate for the frequency conversion loss of the downlink signal and to dynamically control the power of the downlink signal so that the frequency of the downlink signal is maintained within a preset range. The downlink transmitting antenna system and the downlink receiving antenna system adopt a multi-beam design and are configured to transmit different types of polarized signals and to transmit downlink signals to multiple receiving stations.

6. The areal rainfall measurement system based on transparent relay payload according to any one of claims 1 to 5, characterized in that, The areal rainfall detection module performs areal rainfall monitoring based on downlink signals received from multiple receiving stations, including the following steps: Classify sunny and rainy seasons based on the timing power variation of the downlink signal received by each receiving station; Based on the classification results of sunny and rainy seasons for each receiving station, rain-induced attenuation is calculated for the downlink signal received by each receiving station. Rainfall intensity at each receiving station is inverted based on the rainfall-induced attenuation at each receiving station. Based on the rainfall intensity inversion results from multiple receiving stations, areal rainfall data are constructed.

7. The areal rainfall measurement system based on transparent relay payload according to claim 6, characterized in that, The steps for classifying rainy and sunny periods based on the timing power changes of the downlink signals received by each receiving station include: Obtain the standard deviation, initial value, and information entropy of the preprocessed downlink signal in historical states to form a multidimensional signal feature vector, and use the multidimensional signal feature vector to form a dataset. Based on the dataset, a rainy season classification model is trained using the rainy season classification results as training labels. Based on the preprocessed downlink signal, the standard deviation, trend and information entropy of the downlink signal are extracted to form a multidimensional signal feature vector. The multidimensional signal feature vector is then input into the pre-trained weather classification model to obtain real-time weather classification results.

8. The area rainfall measurement system based on transparent relay payload according to claim 6, characterized in that, Based on the rainy / sunny season classification results for each receiving station, the steps for calculating rain-induced attenuation of the downlink signal received by each receiving station include: Obtain tropospheric attenuation in historical states This forms a dataset; Based on the dataset, rain-induced attenuation To train the labels, a rain-induced attenuation prediction model is trained. Tropospheric attenuation The input is fed into the pre-trained attenuation baseline prediction model to obtain the corresponding attenuation baseline prediction output. ; Predict output based on the obtained attenuation baseline Calculate rain-induced attenuation If the classification result for sunny / rainy season is rainy season, then in solving for rain-induced attenuation... At that time, the influence of the melted layer is removed.

9. The area rainfall measurement system based on transparent relay payload according to claim 6, characterized in that, The steps for retrieving rainfall intensity for each receiving station based on rainfall-induced attenuation include: Obtain rain-induced attenuation under historical conditions The dataset is formed by multi-dimensional signal feature vectors; the multi-dimensional signal feature vectors are constructed based on the standard deviation, trend, and information entropy of the downlink signals. Based on the dataset, a rainfall intensity prediction model is trained using rainfall intensity as the training label. Rain-induced attenuation at each receiving station will be acquired in real time. The multidimensional signal feature vector is input into the rainfall intensity prediction model to obtain the rainfall intensity at each receiving station.

10. The area rainfall measurement system based on transparent relay payload according to claim 6, characterized in that, The steps for constructing an areal rainfall distribution map based on the rainfall intensity inversion results from multiple receiving stations include: Using one of the following methods—Kriging interpolation, inverse distance weighting, or spatial variational interpolation—area rainfall distribution maps are constructed based on rainfall intensity at multiple receiving stations.

Images