Soil humidity inversion method, device and equipment and storage medium

Abstract

The invention provides a soil humidity inversion method, device and equipment and a storage medium, and the method comprises the steps: receiving a navigation signal of a low-orbit satellite, and determining a corresponding SNR observation value sequence; determining a fitting result of a direct component and a reflection component in the SNR observation value sequence, and deducting the fitting result from the SNR observation value sequence to obtain an SNR multi-path signal; performing normalization processing on the SNR multi-path signal to obtain a normalized SNR multi-path signal; calculating an intensity value of the normalized SNR multi-path signal; and converting the intensity value of the normalized SNR multi-path signal into the soil humidity by using the trained soil humidity inversion model. According to the method, soil humidity inversion is realized through the low-orbit satellite navigation signal, the limitation of the prior art on time resolution, low-reflection scene precision and short data processing capability is effectively solved, and reliable technical support is provided for remote sensing measurement of the low-orbit satellite navigation signal.

Description

Technical Field

[0001] This invention relates to the field of remote sensing measurement of low-orbit satellite navigation signals, and in particular to a method, apparatus, equipment and storage medium for soil moisture inversion. Background Technology

[0002] After decades of development, GNSS (Global Navigation Satellite System) has become increasingly mature in research and application in navigation, positioning, and related fields. In recent years, with the research and application of L-band radar signals in remote sensing, GNSS remote sensing, with its advantages of abundant reflected signals, wide coverage, high spatiotemporal resolution, and all-weather capability, is gradually forming a research system for Earth remote sensing based on GNSS reflected signals. GNSS interferometry-reflectometry (GNSS-IR) uses ordinary geodetic receivers to simultaneously receive and process both direct and reflected GNSS signals. Based on the amplitude, phase, and frequency characteristics of the reflected signals, it achieves the inversion of surface physical parameters. It features low cost, all-weather capability, and multiple signal sources, and is widely used in sea breeze inversion, snow cover parameter inversion, soil moisture monitoring, and water level measurement. However, existing GNSS-IR technology has limitations such as insufficient temporal resolution, limited inversion accuracy in low-reflection scenarios, and weak processing capabilities for equal- or short-term non-equidistant data. Summary of the Invention

[0003] This application provides a soil moisture inversion method, apparatus, device, and storage medium for using low-orbit satellite navigation signals to realize the soil moisture inversion method, thereby solving the technical problems in the background art.

[0004] According to a first aspect of the embodiments of this application, a soil moisture inversion method is provided, comprising: receiving navigation signals from low-orbit satellites and determining a sequence of SNR (signal-noise ratio) observations corresponding to the navigation signals; determining the fitting results of the direct component and the reflected component in the SNR observation sequence, subtracting the fitting results from the SNR observation sequence to obtain an SNR multipath signal; normalizing the SNR multipath signal to obtain a normalized SNR multipath signal; calculating the intensity value of the normalized SNR multipath signal; and converting the intensity value of the normalized SNR multipath signal into soil moisture using a trained soil moisture inversion model.

[0005] This application effectively addresses the limitations of existing technologies in terms of temporal resolution, accuracy in low-reflection scenarios, and short data processing capabilities by using signal processing methods adapted to the characteristics of low-orbit satellites and a soil moisture inversion model, thus providing more reliable technical support for remote sensing measurement of low-orbit satellite navigation signals.

[0006] In one embodiment of this application, the method further includes characterizing the normalized SNR multipath signal using physical parameters, specifically including: characterizing the amplitude of the direct signal using antenna gain, and characterizing the amplitude of the reflected signal using antenna gain and reflection coefficient; the reflection coefficient is determined by the satellite elevation angle and the complex permittivity of the reflecting surface; determining the phase of the SNR multipath signal caused by reflection; and characterizing the normalized SNR multipath signal using the intensity of the direct signal, the intensity of the reflected signal, and the phase of the SNR multipath signal caused by reflection to obtain the normalized SNR multipath signal characterized by physical parameters.

[0007] In one embodiment of this application, the calculation of the intensity value of the normalized SNR multipath signal specifically includes: constructing a piecewise cubic spline function of the normalized SNR multipath signal; and using the Simpson integral method to iteratively calculate the intensity value of the normalized SNR multipath signal for the piecewise cubic spline function.

[0008] In one embodiment of this application, the construction of the piecewise cubic spline function of the normalized multipath signal specifically includes: determining the sampling interval based on the sampling distance between adjacent sampling points in the normalized SNR multipath signal; constructing a cubic spline interpolation function in each sampling interval by interpolation; establishing a recursive formula and calculating the coefficients of each cubic spline interpolation function according to the recursive formula; and determining the piecewise cubic spline function of the normalized SNR multipath signal.

[0009] In one embodiment of this application, the step of iteratively calculating the intensity value of the normalized SNR multipath signal using the Simpson integral method specifically includes: determining the integration interval, integration limit, and initial integration step size of the normalized SNR multipath signal, wherein the integration interval is determined based on a preset elevation angle range; dividing the integration interval according to the integration step size to determine sub-intervals and sub-interval endpoint values; determining the SNR value of each sub-interval endpoint value according to a piecewise cubic spline function; determining the SNR amplitude integral value of each sub-interval and the SNR amplitude integral value of the entire integration interval at the current integration step size according to the Simpson integral formula; adjusting the integration step size so that the SNR amplitude integral value of the entire integration interval at the current integration step size satisfies the relationship with the integration limit, and using the SNR amplitude integral value of the entire integration interval at the current integration step size as the intensity value of the normalized SNR multipath signal.

[0010] In one embodiment of this application, adjusting the integration step size to ensure that the SNR amplitude integral value of the entire integration interval under the current integration step size satisfies the relationship with the integration limit difference specifically includes: determining whether the interpolation of the SNR amplitude integral value of the entire integration interval under the current integration step size and the previous integration step size is less than the integration limit difference; if it is less, it means that the current SNR amplitude integral value satisfies the relationship with the integration limit difference; if it is greater, it means that the current SNR amplitude integral value does not satisfy the relationship with the integration limit difference, and the integration step size needs to be adjusted and the interpolation of the SNR amplitude integral value of the entire integration interval under the new integration step size needs to be re-determined to see if it is less than the integration limit difference.

[0011] In one embodiment of this application, when the current SNR amplitude integral value does not satisfy the relationship with the integration limit difference, the current integration step size is halved, and the integration interval is divided with the new integration step size. The SNR amplitude integral value of each sub-interval and the SNR amplitude integral value of the entire integration interval are recalculated according to the Simpson integral formula.

[0012] In one embodiment of this application, the training process of the soil moisture inversion model includes: constructing a GA-BP neural network model; obtaining the intensity value of the normalized SNR multipath signal used for training based on the navigation signal, and completing the training of the GA-BP neural network model to obtain the soil moisture inversion model.

[0013] In one embodiment of this application, the construction of the GA-BP neural network model specifically includes: constructing a BP neural network, including determining the number of nodes in the input layer, hidden layer, and output layer of the BP neural network, wherein the input layer is used to receive signal strength values, and the output layer is used to generate predicted values; initializing the weights and biases of the BP neural network using a genetic algorithm, and evaluating the performance of the BP neural network using a fitness function; generating a new population through selection, crossover, and mutation using a genetic algorithm, updating the weights and biases of the BP neural network, and obtaining the GA-BP neural network model.

[0014] In one embodiment of this application, the step of acquiring the intensity value of the normalized SNR multipath signal used for training based on the navigation signal and completing the training of the GA-BP neural network model to obtain the soil moisture inversion model specifically includes: acquiring the intensity value of the normalized SNR multipath signal used for training based on the navigation signal; dividing the intensity value of the normalized SNR multipath signal into a training set and a test set; training the GA-BP neural network model using the training set and the measured values ​​of the soil moisture meter; completing the test using the test set to form the soil moisture inversion model.

[0015] According to a second aspect of the embodiments of this application, an apparatus is provided, comprising: a receiving module for receiving navigation signals from low-Earth orbit satellites and determining a sequence of SNR observations corresponding to the navigation signals; a signal separation module for separating SNR multipath signals from the SNR observation sequence; a signal processing module for normalizing the SNR multipath signals and characterizing the normalized SNR multipath signals using physical parameters; the physical parameters including antenna gain, reflection coefficient, and complex permittivity of the reflecting surface; a signal strength calculation module for calculating the intensity value of the normalized SNR multipath signals; a model building and training module for building a GA-BP neural network model and training the GA-BP neural network model using the intensity value of the normalized SNR multipath signals to form a soil moisture inversion model; and a soil moisture inversion module for calling the soil moisture inversion model to convert the intensity value of the actually acquired SNR multipath signals into soil moisture.

[0016] According to a third aspect of the present application, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as described in the first aspect for the soil moisture inversion method.

[0017] According to a fourth aspect of the present application, a computer-readable storage medium is provided, on which computer program instructions are stored, which, when executed by a processor, are used to implement the process corresponding to the soil moisture inversion method described in the first aspect. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0019] Figure 1 This is a flowchart of the soil moisture inversion method proposed in the embodiments of this application.

[0020] Figure 2 This is a schematic diagram of the SNR sequence and gift box results of a low-orbit satellite in an embodiment of this application.

[0021] Figure 3 This is a schematic diagram of the SNR multipath signal in an embodiment of this application.

[0022] Figure 4 This is a flowchart illustrating the normalized SNR multipath signal strength acquisition process in an embodiment of this application.

[0023] Figure 5This is a schematic diagram of the GA-BP neural network model in an embodiment of this application.

[0024] Figure 6 This is a schematic diagram illustrating the implementation of the soil moisture inversion method in this application.

[0025] Figure 7 This is a schematic diagram of an apparatus according to an embodiment of this application.

[0026] Figure 8 This is a schematic diagram of an electronic device according to an embodiment of this application.

[0027] Figure 9 This is a schematic diagram of the structure of a computer system suitable for implementing the embodiments of this application. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application. Unless otherwise specified, the embodiments and features in the embodiments of this application can be arbitrarily combined with each other. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than that shown here.

[0029] The term "comprising" and any variations thereof in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive protection. For example, a process, method, system, product, or apparatus that comprises a series of steps or units is not limited to the steps or units listed, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus.

[0030] Currently, most traditional methods are based on GNSS satellites to develop GNSS-IR soil moisture inversion methods. This method has the following drawbacks: (1) Insufficient time resolution: Due to the high orbital altitude and slow speed of GNSS satellites, the transit time of a single satellite can be as long as several hours, resulting in low monitoring frequency and difficulty in capturing rapidly changing surface parameters (such as sudden changes in soil moisture after heavy rain). (2) Limited inversion accuracy in low reflectivity scenarios: GNSS signals have long propagation paths and high loss, resulting in weak reflected signal energy. In low reflectivity scenarios with low soil moisture, the amplitude and phase characteristics of SNR multipath signals are not obvious, resulting in insufficient inversion accuracy. (3) Weak ability to process short-term non-equal interval data: The measured discrete signal-to-noise ratio multipath signals are very complex. Traditional methods rely on long-term SNR data (such as continuous observation for several hours) and do not perform targeted processing on discrete SNR signals, resulting in large errors in the extraction of SNR multipath signal features.

[0031] Low-Earth orbit (LEO) internet systems are a crucial component of 5G, 6G, and future networks, and a vital space information infrastructure. Besides communication functions, LEO satellites can also broadcast navigation signals as a signal source. Using LEO satellite L-band navigation signals can compensate for the shortcomings of existing GNSS-IR technologies: 1. LEO satellites have lower orbital altitudes, resulting in less signal loss during propagation in space and greater landing power; 2. LEO satellites move at higher speeds and have shorter transit times, effectively improving the temporal resolution of traditional GNSS-IR technologies; 3. The large number of LEO satellites provides abundant signal sources. Therefore, conducting research on remote sensing measurements integrating LEO internet satellite navigation signals, improving the accuracy and spatiotemporal resolution of existing GNSS-IR technologies, and overcoming the limitations of current methods has significant research value and practical implications for the field of navigation signal remote sensing measurement and the integration of communication, navigation, and remote sensing technologies using LEO satellites. Currently, multipath effect modeling and standardized signal extraction methods for LEO satellites are not yet mature, and there are no technical solutions for soil moisture inversion using LEO satellite navigation signals. Therefore, establishing a multipath signal model that adapts to the characteristics of low-Earth orbit satellites, proposing standardized signal extraction methods, and constructing a high-precision and high-stability soil moisture inversion model are the foundation for soil moisture inversion of low-Earth orbit satellite navigation signals.

[0032] Based on this, this application proposes a soil moisture inversion method based on low-orbit satellite navigation signals, applied to a ground receiver, mainly solving the following problems: (1) Improving time resolution: Low-orbit satellites have low orbital altitude, fast movement speed, and a large number of satellites. The transit time of a single satellite is only a few minutes, and multiple satellites can be networked to achieve high-frequency monitoring, effectively capturing rapidly changing surface parameters and providing high-precision, high-time-resolution soil moisture inversion products. (2) Enhancing inversion accuracy in low-reflection scenarios: Low-orbit satellite signals have short propagation paths, low loss, and greater landing power, resulting in stronger reflected signal energy. Even in dry soil, the amplitude and phase characteristics of the SNR multipath signal are still clearly distinguishable, significantly improving inversion accuracy. (3) Optimize the processing of short-term non-equal interval data: A special algorithm was designed for short-term, non-equal interval SNR data: a signal strength calculation method of piecewise cubic spline interpolation + Simpson integral, which transforms the discrete non-equal interval "elevation angle sine - SNR" points into a smooth curve, solves the problem of smoothing short sequences, and then performs high-precision integration on the smoothed curve to accurately calculate the signal strength.

[0033] Please refer to Figure 1 , Figure 6 The soil moisture inversion method proposed in this application specifically includes the following steps: S101. Receive navigation signals from low-orbit satellites and determine the SNR observation sequence corresponding to the navigation signals.

[0034] Signal-to-noise ratio (SNR) is one of the output observations of modern navigation and positioning receivers, and a quantitative indicator characterizing the strength of navigation signals. Its physical meaning is the ratio of carrier power to white noise power spectral density, measured in dBHz. According to this definition, SNR can be expressed as: (1) In the formula N 0 represents the power spectral density of white noise; P represents the power of the navigation signal received by the receiver, and we have: (2) Please refer to Figure 2 , Figure 3 The navigation signal received by the receiver is an interference signal formed by the superposition of direct and reflected signals, and the mathematical model for the signal-to-noise ratio observation is as follows: (3) In equation (3), the first term A d For the direct component, the second term A m The first term is the reflection component, and the third term is the multipath component, which is formed by the interference of the reflected signal and the direct signal.

[0035] In practical applications, the elevation and azimuth angles of each satellite at any time during the observation period are calculated using observation and navigation files. Then, based on the files output by the receiver, the SNR observation values ​​of a specified constellation and a specified frequency are extracted one by one by limiting the elevation and azimuth angles, thus determining the SNR observation value sequence.

[0036] It should be noted that this embodiment is mainly designed for navigation signals from low-Earth orbit (LEO) internet satellites. In other embodiments, it can also be modified to fuse multiple LEO satellite systems, improving the stability of the inversion results through the redundancy of multi-source signals; or further fuse signals from high-Earth orbit (HEO) GNSS satellites and LEO satellites, combining long-term data from HEO satellites with high-frequency data from LEO satellites to achieve "high-low orbit collaborative" soil moisture inversion.

[0037] S102. Determine the fitting results of the direct component and the reflected component in the SNR observation sequence, and subtract the fitting results from the SNR observation sequence to obtain the SNR multipath signal.

[0038] Since the first and second terms in equation (3) are low-frequency components that vary with the satellite elevation angle, they can be estimated well using a low-order polynomial fitting method. The fitting result is as follows: Figure 2 As shown: (4) Subtracting the fitted result from the SNR observation sequence yields the SNR multipath signal, such as... Figure 3 As shown: (5) In the formula, The phase delay of the reflected signal relative to the direct signal is mainly related to the antenna height, while The initial value of the phase is an important observation for soil moisture inversion.

[0039] Compared to traditional methods where the long propagation path and high loss of high-orbit satellite signals result in weak reflected signal energy and blurred amplitude and phase characteristics of SNR multipath signals in low-reflectivity scenarios (such as dry soil), this embodiment utilizes the advantages of short path and low loss of low-orbit satellite signals, resulting in stronger reflected signal energy. Even in dry soil, the amplitude and phase characteristics of the SNR multipath signal remain clearly discernible.

[0040] S103. Normalize the SNR multipath signal to obtain a normalized SNR multipath signal.

[0041] Because direct sunlight signals are affected by factors such as the atmosphere, ionosphere, and satellite transmission power, even for the same satellite, frequency band, and location, the SNR observation values ​​will differ at different times. To eliminate this problem caused by the intensity of the incident signal, this embodiment also includes normalization of the multipath signal: (6) in, SNR N For normalized SNR multipath signals, SNR M For SNR multipath signals, SNR F The direct signal is obtained by fitting a quadratic polynomial. φ ( t () represents the carrier phase observation value of the direct signal, in cycles; φ For SNR multipath signal phase; A d , A m These are the amplitudes of the direct signal and the reflected signal, respectively, reflecting the intensity of the direct signal and the reflected signal.

[0042] To establish the correlation between SNR multipath signals and soil moisture, the physical model of the normalized SNR multipath signal is derived and its parameters are analyzed in depth based on the actual relay characteristics of navigation signals. The specific process is as follows: Assume the amplitude of the satellite incident signal is μ 0. Considering antenna gain, the amplitudes of the direct and reflected signals received by the receiver can be expressed as follows: (7) In the formula, G d , G RR , G RL These represent the antenna gains for the right-hand circular polarization component in the direct signal, the reflected signal, and the left-hand circular polarization component in the reflected signal, respectively. R RR , R RL Let be the right-hand circular polarization reflection coefficient and the left-hand circular polarization reflection coefficient of the antenna, respectively, and we have: (8) In the formula, R H , R V Let be the reflection coefficients of the horizontal and vertical polarization components of the electromagnetic wave, respectively, and we have: (9) in, The complex permittivity of soil, Assuming the satellite elevation angle is denoted by , it can be seen that the complex permittivity of the reflecting surface and the satellite elevation angle are the main factors determining the reflection coefficient of the reflecting surface. Substituting the physical model of the direct and reflected signal amplitudes into the normalized SNR multipath mathematical model yields: (10) Furthermore, the phase of the SNR multipath signal caused by reflection can be expressed as: (11) From the antenna gain and reflection coefficient models in equations (7) to (11), it can be seen that the amplitude and phase of the normalized SNR multipath signal are parameters related to the antenna gain and the complex permittivity of the reflector, while the frequency of the normalized SNR multipath signal is a parameter related to the antenna height. For a certain type of antenna, the amplitude and phase of the normalized SNR multipath signal are only affected by the complex permittivity of the reflector. Furthermore, substituting the phase expression (11) into the normalized SNR multipath signal expression yields: (12) According to equation (12), the normalized SNR multipath signal is a quasi-cosine function that oscillates periodically with the change of the elevation angle sine. Its oscillation frequency (the main frequency of the normalized SNR multipath signal) is related to the antenna height and the wavelength of the direct signal.

[0043] By incorporating key physical parameters such as antenna gain, reflection coefficient, and complex permittivity of the reflecting surface, this study clarifies the direct correlation between the amplitude and phase of the normalized SNR signal, the dielectric constant of the reflecting surface, and soil moisture. Unlike traditional schemes that merely simplify the ratio of direct to reflected signals, this embodiment reveals that the amplitude of the normalized SNR not only reflects the energy ratio of the reflected to direct signals but is also directly determined by the complex permittivity of the reflecting surface. This in-depth analysis of the physical meaning of the normalized SNR multipath signal provides a more physically meaningful and explicit theoretical support for its amplitude parameters in soil moisture inversion.

[0044] S104. Calculate the intensity value of the normalized SNR multipath signal.

[0045] Considering that the amplitude of the SNR multipath signal is a "time" series that varies with the satellite elevation angle, it must be quantized in order to use the amplitude parameter for remote sensing inversion. The intensity of the SNR multipath signal can also be expressed as the integral value of the amplitude per unit "time", that is: (13) In the formula, sin θ 1. sin θ 2. Observe the lower and upper limits of the sine wave of the satellite elevation angle within the observation period. For ordinary receivers, due to the influence of antenna gain and reflectivity of the reflector, the strength of the SNR multipath signal is weak when the satellite elevation angle is large, which is not conducive to the application of low-orbit satellite navigation signal remote sensing measurement. Therefore, in this embodiment, the normalized SNR multipath signal within the elevation angle range of 5°~25° is used to calculate the signal strength.

[0046] In this embodiment, for the normalized SNR multipath signal, a piecewise cubic spline function with a continuous second derivative is first constructed; then, the piecewise cubic spline function is used as the integration function, and the intensity value of the normalized SNR multipath signal is obtained by iterative calculation using the Simpson integral method.

[0047] Specifically, this embodiment provides a detailed process for constructing a piecewise cubic spline function: The process of constructing a piecewise cubic spline function is essentially to transform the normalized SNR multipath signal into a smooth curve, which can be mathematically represented as: Given sin θ 0, sin θ 1. sin θ 2…sin θ n The function values ​​SNR at these points N (sin θ i )=SNR Ni Construct a cubic spline interpolation function S (sin θ To make it satisfy the following conditions: ① SNR N (sin θ i ) = SNR Ni , I = 0, 1, 2, … n; ② In each small interval [sin θ i , sin θ i+1 Above, S (sin θ ) is a cubic polynomial; ③ S(sin θ ) in [sin θ 0, sin θ n It has a second continuous derivative, that is: (14) in,k =0,1,2; i =1,2,3,…, n -1, where the normalized SNR multipath signal constitutes a "sine value of elevation angle (sin)". θ The sequence of “normalized SNR” is sin θ As the independent variable. Specifically, sin θ i Represents the satellite elevation angle at the i-th observation. θ i The sine value, and sin θ i ∈[0,1].

[0048] Based on the above conditions, the process of constructing the piecewise cubic spline function of the normalized SNR multipath signal is as follows: (1) In each interval [sin θ i , sin θ i+1 The cubic spline interpolation function is obtained using the Heimite interpolation formula on (i = 0, 1, 2, …, n-1). S (sin θ The formula for calculating ) is: (15) (2) Using condition ③, and adding boundary conditions: (16) The system of equations can be obtained as follows: (17) in: (18) The coefficient matrix of the equation system above is a strong triangular matrix with a non-zero determinant, therefore the solution to the equation system exists and is unique. Solving the equation system yields the recurrence relation: (19) in: (20) (3) Use the above formula to find a i , b i ( i = 1,2,…, n ),make m i+1 =0, find the answer. m n , mn-1 ,…, m 0, the parameter to be obtained m i and observed scatter values ​​[sin θ i SNR N (sin θ i Substituting into the recursive formula (19), the piecewise cubic spline function of the discrete normalized SNR multipath signal can be obtained. S (sin θ After that, the Simpson quadrature formula can be constructed to calculate the signal strength.

[0049] Please refer to Figure 4 In practical applications, the sampling interval is first determined based on the sampling distance between adjacent sampling points in the normalized SNR multipath signal; then, the coefficients of the equation system are calculated according to formula (17). , Calculate according to formula (18) , The coefficients of the spline functions are calculated one by one according to the recursive formula (19). The coefficients of the piecewise cubic spline function are obtained; finally, the piecewise cubic spline function of the normalized SNR multipath signal can be determined.

[0050] In some embodiments, if the data noise is high, B-spline interpolation or adaptive interpolation can be introduced to improve noise resistance.

[0051] Similarly, this embodiment also provides the iterative calculation process of Simpson's integral method, as follows: When the upper and lower limits of integration are symmetric with respect to the center point of the expansion, the Taylor integral expansion contains f ( x The terms of odd-order derivatives are all zero. Using this property, by performing a Taylor series expansion over the area of ​​two adjacent subintervals, we can obtain the Simpson integral formula for a certain interval: (twenty one) In the formula, h Let be the length of the integration interval. The accuracy of this integration formula is 1 / N 4 Using the central difference, f ( x The second derivative of is approximately expressed as: (twenty two) Substituting Simpson's integral formula for a certain interval into the second-order approximation expression yields the subinterval [ x i-1 , xi+1 The approximate formula for integrals within the brackets is: (twenty three) Then, the entire integration interval [ a , b The integral result of ] can be expressed as: (twenty four) Because it is used in the integral approximation formula f ( x The second derivative is used to approximate the result. In Simpson's integral method, the influence of odd-numbered lattice points on the result is twice that of even-numbered lattice points. To achieve the required accuracy for integration, iterative calculations are necessary. Let the required accuracy for integration be... ε The influence of the function values ​​at the endpoints of each integration interval on the integral can be divided into three parts: the endpoint regions, odd-numbered points, and even-numbered points, and we have: (25) in: (26) Therefore, the integral value within the interval can be used S d , S o , S e Represented as: (27) Therefore, the integral value for each iteration can be obtained by using the value calculated in the previous iteration. S d , S o and S e Perform the calculation. If the difference between the current integral value and the previous integral value is less than... ε If the result is converged, then the calculation result can be considered converged; otherwise, increase the number of subintervals, shorten the length of the subintervals, and perform the integration calculation again until the result converges.

[0052] In some embodiments, if the integration accuracy requirement is reduced, it can be simplified to trapezoidal integration or rectangular integration; if higher accuracy is required, Gaussian integration or adaptive Simpson integration can be used.

[0053] Please continue to refer to this. Figure 4In practical applications, the integration interval, integration limit, and initial integration step size of the normalized SNR multipath signal are first determined. In this embodiment, the integration interval is determined based on a preset elevation angle range; generally, the integration interval is the sine value corresponding to an elevation angle of 5° to 25°. Then, the integration interval is divided according to the integration step size, determining sub-intervals and their endpoint values. Next, the SNR value of each sub-interval endpoint is determined using a piecewise cubic spline function. Then, the SNR amplitude integral value of each sub-interval at the current integration step size is determined using Simpson's integral formula, and the SNR amplitude integral value of the entire integration interval is calculated. Finally, the integration step size is adjusted so that the SNR amplitude integral value of the entire integration interval at the current integration step size satisfies the relationship with the integration limit, and the SNR amplitude integral value of the entire integration interval at the current integration step size is used as the intensity value of the normalized SNR multipath signal.

[0054] In this embodiment, the relationship is determined by judging whether the interpolation of the SNR amplitude integral value of the entire integration interval under the current integration step size and the previous integration step size is less than the integration limit difference. If it is less, it means that the current SNR amplitude integral value satisfies the relationship with the integration limit difference; if it is greater, it means that the current SNR amplitude integral value does not satisfy the relationship with the integration limit difference, and the integration step size needs to be adjusted and the interpolation of the SNR amplitude integral value of the entire integration interval under the new integration step size needs to be re-judged to see if it is less than the integration limit difference.

[0055] Furthermore, when the current SNR amplitude integral value does not satisfy the relationship with the integration limit, the current integration step size is halved, and the integration interval is divided with the new integration step size. The SNR amplitude integral value of each sub-interval and the SNR amplitude integral value of the entire integration interval are recalculated according to the Simpson integral formula, and their relationship with the integration limit is redefined.

[0056] For short-term, non-uniformly spaced data from low-Earth orbit (LEO) satellites, this embodiment addresses the smoothing problem of short-sequence data by constructing a smooth curve of the second-order continuous derivative through piecewise cubic spline interpolation. Based on the constructed smooth curve, the signal strength (the integral value of amplitude per unit time) is iteratively calculated using the Simpson integral method, which effectively improves the integration accuracy of short-term data and reduces the impact of random errors at discrete points. This data processing method effectively solves the problems of smoothing and signal extraction accuracy for short-term, non-uniformly spaced data, avoids feature extraction errors caused by global fitting failure in traditional schemes, and ensures the usability of short-sequence LEO satellite data.

[0057] S105. Use the trained soil moisture inversion model to convert the intensity values ​​of the normalized SNR multipath signal into soil moisture.

[0058] Current research on soil moisture detection using GNSS reflected signals interferometry still lacks a precise analytical model. Typically, studies only examine the strong correlation between the physical parameters of the interferometric signal (phase shift, amplitude) and soil moisture. Many research findings remain focused on inverting and monitoring soil moisture trends. Aside from the limited application of linear formulas for soil moisture detection, specific inversion equations and models for soil moisture exist. In this embodiment, by inputting the intensity values ​​of the normalized SNR multipath signal into a pre-trained soil moisture inversion model, the corresponding soil moisture can be obtained.

[0059] Furthermore, this embodiment presents a method for training a soil moisture inversion model based on a GA-BP neural network model. Specifically, by constructing a GA-BP neural network model, signal intensity is used as the input, and the effects of surface roughness and vegetation cover are treated as noise. The GA-BP neural network model is then used to obtain soil moisture. The GA-BP network model utilizes a genetic algorithm to refine the initial weights and thresholds, addressing the issue of random initial weights and thresholds in BP neural networks. This allows the model to combine the advantages of nonlinear mapping in BP neural networks with the global search capabilities of genetic algorithms, further improving the model's prediction accuracy and ultimately obtaining the globally optimal solution.

[0060] Specifically, the construction process of this GA-BP neural network model includes: constructing a BP neural network, which involves determining the number of nodes in the input layer, hidden layer, and output layer, wherein the input layer is used to receive signal strength values, and the output layer is used to generate predicted values; then, initializing the weights and biases of the BP neural network using a genetic algorithm, and evaluating the performance of the BP neural network using a fitness function; finally, generating a new population through selection, crossover, and mutation using a genetic algorithm, updating the weights and biases of the BP neural network, and obtaining the GA-BP neural network model. This GA-BP neural network model is as follows: Figure 5 As shown.

[0061] After constructing the GA-BP neural network model, in this embodiment, the intensity values ​​of the normalized SNR multipath signal used for training are obtained based on the navigation signal. This process can be implemented using steps S101 to S104. Then, the intensity values ​​of the normalized SNR multipath signal are divided into a training set and a test set. Finally, the GA-BP neural network model is trained using the training set and measured values ​​from a soil moisture meter, and tested using the test set, thus forming a soil moisture inversion model.

[0062] The soil moisture inversion method proposed in this application is adapted to the characteristics of low-Earth orbit satellites, such as high speed, short data sequences, and high signal energy, overcoming the limitations of traditional GNSS-IR methods in terms of temporal resolution, accuracy in low-reflection scenarios, and short data processing capabilities. The minute-level monitoring capability of low-Earth orbit satellites can capture rapidly changing surface parameters, such as those resulting from sudden changes in soil moisture after heavy rain, providing data support for agricultural and ecological research, as well as disaster emergency response.

[0063] Furthermore, the soil moisture inversion method based on low-orbit satellite navigation signals in this application improves accuracy and temporal resolution, broadens the application scope of GNSS-IR soil moisture inversion technology, and the fusion characteristics of low-orbit satellite communication, navigation and remote sensing (such as combination with 5G / 6G networks) promote the upgrading of remote sensing technology towards high precision and multi-source.

[0064] It should be further noted that the method proposed in this embodiment focuses on soil moisture inversion. In some embodiments, it can be adapted for multi-parameter remote sensing applications, such as the liquid water content of snow: utilizing the reflection characteristics of low-orbit satellite signals on the snow surface (snow dielectric constant and snow liquid water content), the liquid water content of snow is inverted through the amplitude and phase characteristics of the normalized SNR signal. Vegetation water content inversion: combining the attenuation characteristics of vegetation on L-band signals, a vegetation water content inversion model is established through the empirical relationship between signal intensity attenuation and vegetation water content.

[0065] Since the method proposed in this embodiment mainly focuses on establishing an inversion model for a single station, each time a period of calibration is required, and only signal strength and soil moisture are input, in practical applications, it can collect measured data (soil moisture sensor values ​​+ low-orbit satellite inversion values, thermometer observation values) from multiple stations and multiple vegetation types (such as herbaceous plants, shrubs, and trees), automatically learn the influence of vegetation type, height, water content, air temperature and other factors on the inversion results, and improve the accuracy of correction.

[0066] Furthermore, to achieve hardware adaptation of the method proposed in this embodiment, a geodetic receiver can be used, which can be transformed into multiple hardware adaptation schemes. For example, for low-cost receivers, the antenna gain parameters in the normalization model can be adjusted to adapt to their signal reception characteristics; for antennas with different polarizations (such as left-hand circular polarization or linear polarization antennas), the polarization component weights in the reflection coefficient model can be modified to ensure the accuracy of the normalized SNR signal.

[0067] Please refer to Figure 7This application also provides an apparatus 200, comprising: a receiving module 201, configured to receive navigation signals from low-orbit satellites and determine the SNR observation sequence corresponding to the navigation signals; a signal separation module 202, configured to separate SNR multipath signals from the SNR observation sequence; a signal processing module 203, configured to normalize the SNR multipath signals and characterize the normalized SNR multipath signals using physical parameters; the physical parameters include antenna gain, reflection coefficient, and complex permittivity of the reflecting surface; a signal strength calculation module 204, configured to calculate the intensity value of the normalized SNR multipath signals; a model building and training module 205, configured to build a GA-BP neural network model and use the intensity value of the normalized SNR multipath signals to complete the training of the GA-BP neural network model, forming a soil moisture inversion model; and a soil moisture inversion module 206, configured to call the soil moisture inversion model to convert the intensity value of the actually acquired SNR multipath signals into soil moisture.

[0068] Please refer to Figure 8 An electronic device 300 according to an embodiment of this application includes a memory 301 and a processor 302. The memory 301 stores a computer program that can be loaded by the processor 302 and executed to correspond to the aforementioned soil moisture inversion method. It should be noted that the electronic device also has a display screen for displaying a user interface (UI). The UI may include graphics, text, icons, videos, and any combination thereof. When the display screen is a touch screen, it also has the ability to collect touch signals on or above the surface of the display screen. The touch signals can be input to the processor as control signals for processing. In this case, the display screen can also be used to provide virtual buttons and / or a virtual keyboard, also known as soft buttons and / or a soft keyboard. In some embodiments, the display screen can be a single screen, the front panel of the electronic device; in other embodiments, there can be at least two screens, respectively disposed on different surfaces of the electronic device or in a folded design; in still other embodiments, the display screen can be a flexible screen, disposed on a curved surface or a folded surface of the electronic device. Furthermore, the display screen can also be configured as a non-rectangular irregular shape, i.e., a non-rectangular screen. The display screen can be made of materials such as LCD (Liquid Crystal Display) and OLED (Organic Light-Emitting Diode).

[0069] Figure 9 A schematic diagram of the structure of a computer system suitable for implementing the electronic device of the present application is shown.

[0070] It should be noted that, Figure 9The computer system 400 of the electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0071] like Figure 9 As shown, the computer system 400 includes a Central Processing Unit (CPU) 401, which can perform various appropriate actions and processes based on programs stored in Read-Only Memory (ROM) 402 or programs loaded from storage portion 408 into Random Access Memory (RAM) 403, such as performing the methods described in the above embodiments. Various programs and data required for system operation are also stored in RAM 403. The CPU 401, ROM 402, and RAM 403 are interconnected via bus 404. An Input / Output (I / O) interface 405 is also connected to bus 404.

[0072] The following components are connected to I / O interface 405: an input section 406 including a keyboard, mouse, etc.; an output section 407 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to I / O interface 405 as needed. A removable medium 411, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 410 as needed so that computer programs read from it can be installed into storage section 408 as needed.

[0073] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 409, and / or installed from removable medium 411. When the computer program is executed by central processing unit (CPU) 401, it performs various functions defined in the system of this application.

[0074] It should be noted that the computer-readable medium shown in the embodiments of this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such transmitted data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0075] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0076] The units described in the embodiments of this application can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the specific unit itself.

[0077] In another aspect, this application also provides a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the soil moisture inversion method described in the above embodiments.

[0078] In another aspect, this application also provides a computer-readable medium, which may be included in the electronic device described in the above embodiments; or it may exist independently and not assembled into the electronic device. The computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to implement the soil moisture inversion method described in the above embodiments.

[0079] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to the embodiments of this application, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0080] Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, touch terminal, or network device, etc.) to execute the method according to the embodiments of this application.

[0081] For those skilled in the art, the specific meanings of the above terms in this invention can be understood according to the specific circumstances; the accompanying drawings in the embodiments are used to clearly and completely describe the technical solutions in the embodiments of this invention. Obviously, the described embodiments are some embodiments of this invention, but not all embodiments. Generally, the components of the embodiments of this invention described and shown in the accompanying drawings can be arranged and designed in various different configurations.

[0082] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for soil moisture inversion, characterized in that, include: Receive navigation signals from low-Earth orbit satellites and determine the SNR observation sequence corresponding to the navigation signals; Determine the fitting results of the direct component and the reflected component in the SNR observation sequence, and subtract the fitting results from the SNR observation sequence to obtain the SNR multipath signal; The multi-SNR path signal is normalized to obtain a normalized SNR multi-path signal; Calculate the intensity value of the normalized SNR multipath signal; The intensity values ​​of the normalized SNR multipath signal are converted into soil moisture using a trained soil moisture inversion model.

2. The soil moisture inversion method according to claim 1, characterized in that, It also includes characterizing the normalized SNR multipath signal using physical parameters, specifically including: The amplitude of the direct signal is characterized by antenna gain, and the amplitude of the reflected signal is characterized by antenna gain and reflection coefficient; the reflection coefficient is determined by the satellite elevation angle and the complex permittivity of the reflecting surface. Determine the phase of the SNR multipath signal caused by reflection; The normalized SNR multipath signal is characterized by the amplitude of the direct signal, the amplitude of the reflected signal, and the phase of the SNR multipath signal caused by reflection, to obtain the normalized SNR multipath signal characterized by physical parameters.

3. The soil moisture inversion method according to claim 1, characterized in that, The calculation of the intensity value of the normalized SNR multipath signal specifically includes: Construct a piecewise cubic spline function for the normalized SNR multipath signal; For the piecewise cubic spline function, the intensity value of the normalized SNR multipath signal is obtained by iterative calculation using the Simpson integral method.

4. The soil moisture inversion method according to claim 3, characterized in that, The construction of the piecewise cubic spline function for normalized multipath specifically includes: The sampling interval is determined based on the sampling distance between adjacent sampling points in the normalized SNR multipath signal; A cubic spline interpolation function is constructed for each sampling interval using interpolation. Establish a recursive formula and calculate the coefficients of each cubic spline interpolation function based on the recursive formula; Determine the piecewise cubic spline function for the normalized SNR multipath signal.

5. The soil moisture inversion method according to claim 3 or 4, characterized in that, The method of iteratively calculating the intensity value of the normalized SNR multipath signal using Simpson's integral method specifically includes: The integration interval, integration limit, and initial integration step size of the normalized SNR multipath signal are determined; the integration interval is determined based on a preset elevation angle range. The integration interval is divided according to the integration step size, and the sub-intervals and their endpoint values ​​are determined. Determine the SNR value of each sub-interval endpoint based on the piecewise cubic spline function; Based on Simpson's integral formula, determine the SNR amplitude integral value of each sub-interval and the SNR amplitude integral value of the entire integration interval at the current integration step size. By adjusting the integration step size so that the integral value of the SNR amplitude of the entire integration interval under the current integration step size satisfies the relationship with the integration limit difference, the integral value of the SNR amplitude of the entire integration interval under the current integration step size is used as the intensity value of the normalized SNR multipath signal.

6. The soil moisture inversion method according to claim 5, characterized in that, The step of adjusting the integration step size so that the integral value of the SNR amplitude over the entire integration interval at the current integration step size satisfies the relationship with the integration limit error specifically includes: Determine whether the interpolated SNR amplitude integral value of the entire integration interval under the current integration step size and the previous integration step size is less than the integration limit. If it is less, it means that the current SNR amplitude integral value satisfies the relationship with the integration limit. If it is greater, it means that the current SNR amplitude integral value does not satisfy the relationship with the integration limit. The integration step size needs to be adjusted and the interpolated SNR amplitude integral value of the entire integration interval under the new integration step size needs to be re-determined to see if it is less than the integration limit.

7. The soil moisture inversion method according to claim 6, characterized in that, When the current SNR amplitude integral value does not satisfy the relationship with the integration limit difference, the current integration step size is halved, and the integration interval is divided with the new integration step size. The SNR amplitude integral value of each sub-interval and the SNR amplitude integral value of the entire integration interval are recalculated according to the Simpson integral formula.

8. The soil moisture inversion method according to claim 1, characterized in that, The training process of the soil moisture inversion model includes: Construct a GA-BP neural network model; The intensity values ​​of the normalized SNR multipath signal used for training are obtained based on the navigation signal, and the training of the GA-BP neural network model is completed to obtain the soil moisture inversion model.

9. The soil moisture inversion method according to claim 8, characterized in that, The construction of the GA-BP neural network model specifically includes: Constructing a BP neural network includes determining the number of nodes in the input layer, hidden layer, and output layer of the BP neural network, wherein the input layer is used to receive the intensity value of the signal, and the output layer is used to generate the predicted value; The weights and biases of the BP neural network are initialized using a genetic algorithm, and the performance of the BP neural network is evaluated using a fitness function. A new population is generated through selection, crossover, and mutation using a genetic algorithm. The weights and biases of the updated BP neural network are then adjusted to obtain the GA-BP neural network model.

10. The soil moisture inversion method according to claim 9, characterized in that, The process of acquiring the intensity values ​​of the normalized SNR multipath signal used for training based on navigation signals and completing the training of the GA-BP neural network model to obtain the soil moisture inversion model specifically includes: The intensity values ​​of the normalized SNR multipath signal used for training are obtained based on the navigation signal; The intensity values ​​of the normalized SNR multipath signal are divided into a training set and a test set; The GA-BP neural network model was trained using the training set and measured values ​​from a soil moisture meter, and tested using a test set to form a soil moisture inversion model.

11. An apparatus, characterized in that, include: The receiving module is used to receive navigation signals from low-orbit satellites and determine the SNR observation sequence corresponding to the navigation signals; A signal separation module is used to separate the SNR multipath signal from the SNR observation sequence; The signal processing module is used to normalize the SNR multipath signal and characterize the normalized SNR multipath signal using physical parameters; the physical parameters include antenna gain, reflection coefficient, and complex permittivity of the reflecting surface. The signal strength calculation module is used to calculate the strength value of the normalized SNR multipath signal; The model building and training module is used to build a GA-BP neural network model and use the intensity value of the normalized SNR multipath signal to complete the training of the GA-BP neural network model, forming a soil moisture inversion model. The soil moisture inversion module is used to call the soil moisture inversion model to convert the intensity value of the actually acquired normalized SNR multipath signal into soil moisture.

12. An electronic device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed corresponding to the soil moisture inversion method as described in any one of claims 1 to 10.

13. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the program instructions are executed by the processor, they are used to implement the process corresponding to the soil moisture inversion method according to any one of claims 1 to 10.

Images