Monte carlo based non-coherent scatter ionospheric parameter unambiguous estimation method

Abstract

The application belongs to the technical field of signal processing, and particularly relates to a non-coherent scattering ionospheric parameter non-ambiguous estimation method based on Monte Carlo, aiming to solve the problems of inaccurate parameter estimation and poor robustness in ionospheric parameter inversion in the prior art. The method comprises the following steps: receiving measured echo data, estimating the signal noise level and determining the prior information combination mode; obtaining an expected index based on a signal noise threshold mapping library, and judging whether a preset decision threshold is met; if the preset decision threshold is met, starting an optimization selection strategy based on Monte Carlo, and obtaining a candidate solution set; performing cluster analysis on the candidate solution set, selecting a solution with the optimal fitting degree from the solution cluster containing the largest number of solutions, and taking the solution as the final non-ambiguous estimation value. Through systematic Monte Carlo simulation, the application can quantitatively evaluate the inversion reliability under the current data condition, effectively avoid invalid calculation under adverse conditions such as excessive noise or insufficient prior information, and improve the robustness and success rate of estimation.

Description

Technical Field

[0001] This application belongs to the field of signal processing technology, and specifically relates to a Monte Carlo-based method for ambiguity-free estimation of incoherent scattering ionospheric parameters. Background Technology

[0002] Incoherent scattering radar (ISR) is one of the most powerful ground-based techniques for detecting ionospheric parameters such as electron density, electron temperature, ion temperature, ion composition, and drift velocity. However, within a specific altitude range (e.g., approximately 130 to 300 kilometers) such as the F1 region of the ionosphere, specific combinations of different ion compositions (e.g., molecular ions and atomic oxygen ions) with plasma temperatures can produce physically very similar backscattering spectra. This leads to an inherent "temperature-ion composition ambiguity" (TICA) problem when estimating ion temperature and ion composition using traditional inversion algorithms.

[0003] Existing technologies have significant shortcomings in attempting to solve this ambiguity problem. First, traditional nonlinear least squares optimization algorithms heavily rely on the selection of initial parameters. When the measured data has high signal noise, the algorithm is prone to converging to an incorrect local optimum, leading to significant deviations in the estimation results. Second, some methods that rely on auxiliary measurement information such as plasma lines to eliminate ambiguity place stringent requirements on the quality of the radar signal (i.e., signal-to-noise ratio), which is difficult to meet under conventional observation modes. Furthermore, existing technologies lack a clear quantitative criterion to pre-determine under what signal conditions the ambiguity problem can be reliably resolved.

[0004] Therefore, when faced with actual observational data with poor noise levels or uncertain prior information, existing technologies generally lack a processing procedure that can adaptively assess the feasibility of inversion and systematically ensure the robustness of the inversion results. A new method is urgently needed to address this challenge. Summary of the Invention

[0005] To address the aforementioned problems in the prior art, namely the inaccurate parameter estimation and poor robustness caused by the "temperature-ion composition ambiguity" issue during ionospheric parameter inversion by incoherent scattering radar (ISR), this application provides a Monte Carlo-based unambiguous estimation method for incoherent scattering ionospheric parameters. The method includes the following steps:

[0006] Receive measured echo data, estimate the actual signal-to-noise level of the measured echo data, and determine the currently available combination of prior information.

[0007] Based on the actual signal-noise level and the combination pattern of prior information, the signal-noise threshold mapping library is queried to obtain the expected index characterizing the reliability of parameter estimation under the current conditions; the signal-noise threshold mapping library stores the mapping relationship between different combination patterns of prior information and different signal-noise levels and one or more expected indices.

[0008] Determine whether the expected index meets the preset decision threshold. If it does, then based on the Monte Carlo optimization selection strategy, use multiple sets of different initial parameters to independently perform multiple optimization fittings on the measured echo data to obtain a candidate solution set composed of multiple fitting results.

[0009] Statistical analysis is performed on the candidate solution set to identify a candidate subset representing the global optimal solution, and the solution with the best fit is determined from the candidate subset as the final estimate of the ionospheric parameter.

[0010] In some preferred embodiments, the signal-noise threshold mapping library is established through the following steps:

[0011] For each combination of prior information, sampling is performed within a preset ionospheric parameter space to generate multiple sets of true values ​​of ionospheric parameters.

[0012] For each set of true values ​​of ionospheric parameters, Monte Carlo simulations were performed multiple times under multiple signal and noise levels, combined with the theoretical model of incoherent scattering radar, to obtain a database containing a large number of simulation spectra.

[0013] For each simulated spectrum in the database, parameter inversion is performed, and based on the relationship between the inversion results and the true values ​​of ionospheric parameters, the expected indicators corresponding to different signal noise levels under each prior information combination mode are statistically calculated. This determines the mapping relationship between different prior information combination modes and different signal noise levels and one or more expected indicators, thereby forming the signal noise threshold mapping library.

[0014] In some preferred embodiments, during the Monte Carlo simulation, the result of each parameter inversion is judged. If the normalized chi-square value of a certain inversion is less than the chi-square threshold determined based on a preset confidence level, it is judged as a valid convergence.

[0015] In some preferred embodiments, the statistical analysis of the candidate solution set includes:

[0016] Cluster analysis is performed on the ion composition values ​​of each solution in the candidate solution set to divide the candidate solution set into at least one solution cluster, and the target solution cluster containing the maximum number of solutions is determined.

[0017] From the target solution cluster, a solution with the best fit is selected as the final estimate of the ionospheric parameters.

[0018] In some preferred embodiments, the solution with the best fit is the solution in the target solution family that has the smallest fitting residual among all solutions that meet the physical constraints.

[0019] In some preferred embodiments, a comprehensive confidence index corresponding to the final estimate is generated and output by combining the expected index obtained from the signal-noise threshold mapping library with the statistical characteristics of the target solution cluster.

[0020] In some preferred embodiments, the expected indicators of the reliability of the characterization parameter estimation are the probability of correct estimation and / or the probability of effective convergence.

[0021] In some preferred embodiments, the prior information combination mode is defined as: a specific combination of applying prior constraints to one or more parameters in a set of ionospheric parameters to be estimated, wherein the prior constraints are derived from ionospheric theoretical models or plasmaline measurements.

[0022] In some preferred embodiments, estimating the actual signal-to-noise level of the measured echo data includes:

[0023] The measured echo data is preprocessed to obtain the autocorrelation function or power spectrum at the target altitude; based on the autocorrelation function or power spectrum, and combined with the noise power measured in the no-signal region, the signal-to-noise ratio at the target altitude is calculated; the actual signal-to-noise level is determined according to the signal-to-noise ratio and the number of pulse accumulations.

[0024] In some preferred embodiments, when it is determined that the expected index does not meet the preset decision threshold, a warning is triggered, indicating that the current data conditions cannot obtain a reliable solution, or the solution process is terminated.

[0025] The beneficial effects of this application are:

[0026] (1) The method of this application can quantitatively evaluate the reliability of inversion under the current data conditions by querying the preset mapping library based on the actual signal noise level and prior information before formal solution, thereby making intelligent decisions on whether to start the complex solution process, realizing the pre-assessment and avoidance of risks, effectively avoiding invalid calculations under adverse conditions such as excessive noise or insufficient prior information, saving computing resources and issuing timely warnings to users.

[0027] (2) The Monte Carlo-based optimization selection strategy is adopted. By starting from a large number of random initial points and performing multiple independent fittings, the entire parameter space is systematically explored. This effectively overcomes the defect of traditional single-fitting methods that are prone to getting trapped in local optima due to improper initial value selection, and greatly enhances the global convergence ability and robustness of the algorithm.

[0028] (3) By performing statistical analysis such as clustering on the candidate solution set generated by multiple fittings, the "majority solution" is used to approximate the global optimal solution, and on this basis, the physically optimal solution is selected as the final estimate. Compared with the single minimum residual criterion, the decision-making mechanism based on statistical laws is more resistant to noise interference and significantly improves the accuracy of parameter estimation and the certainty of the final result. Attached Figure Description

[0029] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0030] Figure 1 This is a schematic diagram of the overall process of an embodiment of the Monte Carlo-based incoherent scattering ionospheric parameter ambiguity estimation method provided in this application;

[0031] Figure 2 This is a flowchart illustrating a Monte Carlo optimization selection strategy provided in one embodiment of this application;

[0032] Figure 3 This is a schematic diagram of the framework of a Monte Carlo-based incoherent scattering ionospheric parameter ambiguity estimation system provided in one embodiment of this application;

[0033] Figure 4 This is a schematic diagram of the structure of a computer system used to implement the methods, systems, and electronic devices of this application. Detailed Implementation

[0034] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0035] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0036] To address the technical challenges of inaccurate parameter estimation and poor robustness caused by the "temperature-ion composition ambiguity" problem in existing technologies, this application provides a Monte Carlo-based unambiguous estimation method for incoherent scattering ionospheric parameters. After receiving measured echo data from an incoherent scattering radar, the method estimates the actual signal-noise level and determines the currently available prior information combination pattern. Then, based on this condition, it queries a pre-established signal-noise threshold mapping library to obtain an expected index characterizing the reliability of the parameter estimation. Next, it determines whether this expected index meets a preset decision threshold. If it does, a Monte Carlo-based optimization selection strategy is initiated, which independently performs multiple optimization fittings on the measured data using multiple sets of different random initial parameters to obtain a candidate solution set. Through statistical analysis such as clustering of this candidate solution set, the solution with the best fit is selected from the cluster containing the most solutions as the final unambiguous estimate of the ionospheric parameters. This application can pre-assess inversion risks, avoid invalid calculations under poor data conditions, and approximate the global optimum by statistically determining the majority solution, significantly improving the success rate and reliability of parameter estimation.

[0037] It should be noted that the implementation process of this invention involves two types of data: one type is simulated data generated by the Monte Carlo method for offline establishment of the signal-noise threshold mapping library; the other type is measured echo data that needs to be processed online in practical applications. The processes involving these two types of data will be described below.

[0038] To more clearly explain the ambiguity-free estimation method for incoherent scattering ionospheric parameters based on Monte Carlo in this application, the following will combine... Figure 1 The steps in the embodiments of this application are described in detail.

[0039] The first embodiment of this application uses a Monte Carlo-based method for unambiguous estimation of incoherent scattering ionospheric parameters, steps S100-S600, each described in detail below:

[0040] S100. Measured echo data from ISR is used to obtain the ISR spectrum or autocorrelation function at the target altitude, estimate the actual signal-noise level of the measured echo data, and determine the currently available combination mode of prior information.

[0041] Preferably, the system receives measured echo data acquired by the radar system, calculates the corresponding autocorrelation function (ACF), and estimates the signal-to-noise ratio (SNR). Simultaneously, based on auxiliary information such as the altitude, local time, and season of the detection mission, one or more possible combinations of prior information are determined. For example, in the top region of layer F, the dominant ion component might be... and Therefore, the prior information combination mode can be set as a combination of... Inversion of mixing ratios.

[0042] Preferably, estimating the actual signal-to-noise level of the measured echo data includes:

[0043] The measured echo data is preprocessed to obtain the autocorrelation function or power spectrum at the target height; and the signal-to-noise ratio is calculated based on the autocorrelation function or power spectrum to determine the actual signal noise level.

[0044] More preferably, in a specific embodiment, receiving and preprocessing the measured raw echo data from the ISR specifically includes:

[0045] 1) Decoding and deblurring of raw IQ data;

[0046] 2) Calculate the autocorrelation function or its Fourier transform (power spectrum) for each distance gate (corresponding to a specific height);

[0047] 3) Background noise and interference removal.

[0048] More preferably, in a specific embodiment, calculating the signal-to-noise ratio based on the autocorrelation function or power spectrum, and then determining the actual signal noise level, specifically includes:

[0049] For preprocessed ISR spectrum or ACF data, the actual signal-noise level of the current data is estimated based on the root mean square error of the plasma autocorrelation estimate. (For specific calculation formulas, please refer to the following preset signal noise levels) Calculation method); this The value will serve as a key input for comparison with the mapping library.

[0050] Preferably, all available auxiliary information sources (such as plasma line measurement results, theoretical model predictions, etc.) are collected and organized to determine the currently available prior information combination patterns and to assess the uncertainty of each prior information.

[0051] More preferably, in a specific embodiment, the currently available prior information combination patterns are determined, and the following operations are performed in parallel:

[0052] Extracting plasma line information: Analyzing the plasma line resonance frequencies from the same radar observations to accurately invert the electron density. By utilizing the asymmetry of the upper and lower sidebands of the plasma line, or by combining this with the total power received by the antenna, the electron temperature or electron-ion temperature ratio can be estimated. Record these estimates and their uncertainties (such as standard errors).

[0053] Accessing external model data: Calling the international reference ionospheric model to obtain data at the current time and altitude. , , Predicted values ​​of iso-ionospheric parameters and their uncertainty assessed based on model performance.

[0054] Determine the combination pattern and uncertainty: Integrate all available information to determine the combination pattern of currently available prior information, such as known prior information. (From plasma beam) and known (From the model). Simultaneously, an uncertainty measure is assigned to each known parameter. (In percentage form), for example, from plasma line measurements. Uncertainty is From model predictions Uncertainty is .

[0055] More preferably, the prior information combination mode is defined as: a specific combination of applying prior constraints to one or more parameters in a set of ionospheric parameters to be estimated, wherein the prior constraints are derived from ionospheric theoretical models or plasmaline measurements.

[0056] More preferably, in one embodiment, the prior information combination pattern includes:

[0057] Mode A: No prior information mode, four unknown parameters ( and Mode B: Known electron density mode, three unknown parameters ( and ), prior known Mode C: Known electron density and electron-ion temperature ratio mode, with two unknown parameters ( and ), known and Mode D: Known electron density and electron temperature mode, two unknown parameters ( and ), known and ).

[0058] S200. Based on the actual signal noise level and the combination pattern of prior information, query the signal noise threshold mapping library to obtain the expected index characterizing the reliability of parameter estimation under the current conditions; the signal noise threshold mapping library stores the mapping relationship between different combination patterns of prior information and different signal noise levels and one or more expected indices.

[0059] Preferably, obtaining the expected index characterizing the reliability of parameter estimation under current conditions specifically includes:

[0060] Based on the determined prior information combination pattern, the complete probability curve dataset under that pattern is retrieved from the pre-established signal-noise threshold mapping library.

[0061] Preferably, the expected indicators of the reliability of the characterization parameter estimation include the probability of correct estimation and / or the probability of effective convergence.

[0062] More preferably, in one specific embodiment, the actual signal-to-noise level of the current measured data is used. and the uncertainty of prior parameters As input, the nearest sampling point is located in the mapping library, and the expected effective convergence probability under the current conditions is calculated through bilinear interpolation. And the expected correct estimate of probability .

[0063] This application describes the signal-noise threshold mapping library: The library is constructed through extensive theoretical simulations or historical data analysis. Optimization algorithms (such as Levenberg-Marquardt) are used to independently fit each set of simulated data multiple times, and the expected indices (effective convergence probability and correct estimation probability) are statistically analyzed. The library provides the highest permissible noise level threshold for ensuring the correctness of parameter estimation under different prior combinations and specific confidence requirements, and records the expected indices at different noise levels. For example, the expected indices can be the inverted parameter (such as the ratio of electron temperature to ion temperature). The expected standard deviation of the data is calculated. Therefore, querying the mapping library yields an expected indicator that matches the current measured data.

[0064] The preset signal-noise threshold mapping library is established through the following steps:

[0065] For each combination of prior information, sampling is performed within a preset ionospheric parameter space to generate multiple sets of true values ​​of ionospheric parameters.

[0066] For each set of true values ​​of ionospheric parameters, Monte Carlo simulations were performed multiple times under multiple signal and noise levels, combined with the theoretical model of incoherent scattering radar, to obtain a database containing a large number of simulation spectra.

[0067] For each simulated spectrum in the database, parameter inversion is performed, and based on the relationship between the inversion results and the true values ​​of ionospheric parameters, the expected indicators corresponding to different signal noise levels under each prior information combination mode are statistically calculated. This determines the mapping relationship between different prior information combination modes and different signal noise levels and one or more expected indicators, thereby forming the signal noise threshold mapping library.

[0068] More preferably, during the Monte Carlo simulation, the result of each parameter inversion is judged. If the normalized chi-square value of a certain inversion is less than the chi-square threshold determined based on a preset confidence level, it is judged as a valid convergence.

[0069] More preferably, in a specific embodiment, establishing a signal-noise threshold mapping library specifically includes:

[0070] (1) Based on the theory of incoherent scattering, a forward model of the incoherent scattering spectrum is established:

[0071] The ionospheric parameter vectors involved in the model include electron density. Electronic temperature ion temperature Plasma line-of-sight drift velocity and ionic components In this embodiment, the plasma line-of-sight drift velocity is taken into account. The inversion constraint only affects the Doppler shift of the spectrum without changing other parameters, therefore it can be inverted independently of other parameters. In actual inversion, As known prior information is input into the model, a more stable joint inversion of the remaining parameters is achieved; the types of ions considered include molecular ions ( and ) and oxygen ions ionic components Defined as ;in, express density.

[0072] The selection range of the true values ​​of each parameter is based on satellite observations and model results, covering the typical ionospheric parameter range of each latitude zone at an altitude of 130–300 km under different solar and geomagnetic activity conditions, and then uniformly selected within the following range: , , , Ionic composition parameters exist Interval.

[0073] (2) For each set of preset true parameters Generate Monte Carlo simulation data according to the following procedure:

[0074] Calculating the noise-free theoretical spectrum using an incoherent scattering spectrum theoretical spectral model ;

[0075] Based on preset signal noise level (Expressed as a percentage, such as) , , , , (etc.), calculate the standard deviation of additive white Gaussian noise. Among them, the preset signal noise level The root mean square error of the plasma autocorrelation estimate is defined as follows:

[0076] ;

[0077] Wherein, SNR is the signal-to-noise ratio of the radar backscatter echo signal in the corresponding altitude range, and N is the number of pulses in the accumulation time range;

[0078] Generate compliance Distributed random noise vector This yields simulated observation data that includes noise. ;

[0079] For different combinations of prior information, the known parameters are fixed to their true values ​​(or uncertainties are added around them) during the fitting process, and only the unknown parameters are inverted.

[0080] (3) For each set of simulated data, perform optimization fitting and probability statistics:

[0081] The Levenberg-Marquardt (LM) nonlinear least squares optimization algorithm is used to minimize the theoretical autocorrelation and noisy autocorrelation data. The cost function is fitted to the objective.

[0082] To ensure that the simulation results are independent of the selected initial parameters, different initial parameters are used in each repeated estimation of the Monte Carlo simulation; these initial parameters are randomly selected from a uniform distribution centered on the actual input ionospheric parameters. The range of the initial parameters is as follows:

[0083] ;

[0084] in, This represents the size of the initial value search range, i.e., the uncertainty of the initial parameters of the theoretical model. Percentage of the initial parameter search range ( ), Select the minimum value within the range for the parameter. Select the maximum value within the range of parameters.

[0085] The result determination includes:

[0086] Effective convergence: If the fitting result satisfies Less than the preset threshold If the convergence is successful, it is considered a valid convergence.

[0087] Correct estimation: In the results of effective convergence, the estimation error of ion composition is assessed using the Expectation-Maximization (EM) algorithm. Perform cluster analysis, Input the true value for the ionic composition. This represents the estimated ionic composition. Specifically, assuming the error distribution follows a bi-Gaussian mixture model, the EM algorithm iteratively optimizes and estimates the mean, variance, and mixture weights of the two Gaussian components. After clustering, components with a mean close to 0 are labeled as "correctly classified," and components with a mean deviating from 0 are labeled as "incorrectly classified." Each sample is assigned based on its posterior probability of belonging to each category, and the fitting result belonging to the "correctly classified" category is determined to be a correct estimate.

[0088] More preferably, in one specific embodiment, due to Following a chi-square distribution, this embodiment selects the corresponding normal distribution. probability level ( The chi-square critical value is used as a threshold to ensure that fitting results exceeding this threshold can be judged as invalid convergence.

[0089] Statistical calculations are performed on the judgment results:

[0090] The probability of effective convergence of the fit is defined as the probability of finding the minimum value of the sum of squared residuals using the Levenberg-Marquardt algorithm. :

[0091] ;

[0092] in, The number of valid convergent results. This represents the total number of simulation parameters. The number of different input parameters for the Monte Carlo simulation. The number of repetitions simulated for each input parameter.

[0093] The probability of a correct estimate is defined as the probability that the sum of squared residuals reaches the global minimum and converges to a valid solution. for:

[0094] ;

[0095] in, This represents the number of correct solutions calculated by the EM clustering algorithm in a single simulation.

[0096] (4) Based on the fitted results, establish a signal-noise threshold mapping library:

[0097] Repeat the above steps Iterate through all preset combinations of real parameters, all prior modes, and all signal-noise levels. For each prior mode, the analysis... and The relationship curve; determining the threshold: setting a target confidence level (e.g., ,Right now Find the highest value that satisfies this condition. The value is denoted as the signal-to-noise threshold for this mode. Store the following mapping relationships in the database in a structured manner:

[0098] (Prior information combination mode, signal-noise level) ) (effective convergence probability) Correctly estimate probability );

[0099] (Prior information combination pattern, target confidence level) (Signal noise threshold) ).

[0100] Preferably, in this embodiment, the parameter range used to construct the mapping library is selected to fully cover the various physical states that may occur during periods of ionospheric geomagnetic calm and geomagnetic storms, so as to ensure that the generated mapping library has good universality, rather than being used to limit the range of measured data that the method of the present invention can process.

[0101] S300. Determine whether the expected indicator meets the preset decision threshold. If it does, process the measured echo data based on the Monte Carlo optimization selection strategy:

[0102] Multiple sets of different initial parameters are used to independently perform multiple optimization fittings on the measured echo data to obtain a candidate solution set composed of multiple fitting results.

[0103] Preferably, the obtained expected indicators are compared with a preset decision threshold, which is a standard for measuring whether the inversion is likely to be successful.

[0104] If the expected indicators are not met (i.e., the expected indicators are worse than the decision threshold), it is determined that under the current data conditions (excessive noise or poor prior information), the optimization algorithm is highly likely to fail to converge effectively or converge to an incorrect solution. The current inversion can be abandoned, or the data can be accumulated to improve the signal-to-noise ratio, thereby avoiding ineffective computation. Simultaneously, an early warning is triggered; the system generates and issues a warning message indicating "High risk of failure in the current height interval parameter inversion. Recommendation: a) Increase data integration time to reduce..." b) Attempt to obtain more precise prior information (e.g., initiate plasma line observations).

[0105] If the expected indicators are met (i.e., the expected indicators are better than the decision threshold), then the basic conditions for reliable inversion are considered to be met, and the Monte Carlo-based optimization selection strategy is initiated to make robust estimates of the ionospheric parameters.

[0106] For example, data under these observation conditions can be obtained through segmented statistical analysis of long-term satellite / radar observation data (based on local time, season, geomagnetic activity, latitude zone, etc.). A preset standard deviation threshold is used, and this implementation preferably sets the preset standard deviation threshold to 0.2. If the expected index (expected standard deviation) obtained from the query is 0.5, it means that the current data quality is poor, and the risk of failure in direct inversion is very high. Conversely, if the expected index is 0.1, which is less than the decision threshold, it indicates that the data quality is sufficient to support a reliable inversion.

[0107] More preferably, such as Figure 2 As shown, the measured echo data is processed using an optimization selection strategy based on Monte Carlo methods, specifically as follows:

[0108] Multiple sets of randomly generated initial parameters are used to independently perform multiple optimization fittings on the same measured echo data (ACF) to obtain a candidate solution set composed of multiple fitting results.

[0109] Specifically, obtaining a candidate solution set consisting of multiple fitting results includes the following steps:

[0110] Set an independent fitting number K, and generate K different initial parameter values ​​independently, uniformly and randomly within the entire physical reasonable range of the parameters;

[0111] Using the LM algorithm, K independent nonlinear least squares fitting operations were performed in parallel with the same measured data but different random initial values; all K sets of fitting results were collected, including the estimated values ​​of each parameter. , , , ) and its corresponding value;

[0112] Extract the estimated ionic composition values ​​from all K results. This yields a candidate solution set. Since the TICA problem primarily manifests itself in… In the bimodal ambiguity, this parameter is key to identifying the correct solution.

[0113] Regarding the number of independent fits, K, generally, the larger the K value, the more thoroughly the solution space is explored, and the more reliable the statistical results. In some embodiments, a suitable K value can be selected by testing the convergence of the final estimate under different K values. For example, initially setting K=500, it can be adjusted during cluster analysis. For instance, if it is found that 85% of the results are concentrated around one set of solutions, and the rest are scattered into two minor clusters, K can be increased to 600 and run again. The cluster structure stabilizes until the proportion of principal solutions is high. Meanwhile, the maximum number of independent fitting attempts is set to 1000.

[0114] For example, 100 different combinations of initial parameters (including electron density, electron temperature, ion temperature, ion composition ratio, etc.) can be randomly generated. Each set of initial parameters is iteratively optimized using a nonlinear least squares fitting algorithm (such as the Levenberg-Marquardt algorithm) to finally obtain a solution for the ionospheric parameters. Due to the existence of "fuzziness," different initial parameters may lead the fitting process to converge to different local optima. In this way, after 100 independent fittings, a candidate solution set containing 100 possible solutions will be obtained.

[0115] S400. Perform statistical analysis on the candidate solution set to determine the final estimated value of the ionospheric parameter from the candidate solution set.

[0116] Preferably, the statistical analysis of the candidate solution set includes:

[0117] Cluster analysis is performed on the ion composition values ​​of each solution in the candidate solution set to divide the candidate solution set into at least one solution cluster, and the target solution cluster containing the maximum number of solutions is determined.

[0118] From the target solution cluster, a solution with the best fit is selected as the final estimate of the ionospheric parameters.

[0119] More preferably, determining the final estimated value as an ionospheric parameter specifically includes:

[0120] For the K candidate solutions obtained Cluster analysis was performed on the values; in the unambiguous case, all The solutions cluster around the true value. In the case of fuzziness, two main clusters are formed, corresponding to the correct solution and the fuzzy solution, respectively. According to the evaluation criteria, the solution with the highest frequency of ion component values, the smallest residual, and the physical constraints is selected as the final estimate of the ionospheric parameters, and a comprehensive confidence index based on the expected correct probability and the convergence probability is added.

[0121] The correct true solution should be stable, meaning that no matter how the initial parameters are selected within a reasonable range, the fitting result should converge to its vicinity with a high probability. Therefore, in the multidimensional parameter space, a dense "solution cluster" will form near the true solution, while the erroneous solutions caused by "fuzziness" will be distributed more discretely.

[0122] In this embodiment, key parameters (such as...) in the candidate solution set can be... Cluster analysis can be performed using the proportion of ion components, for example, by using simple histogram statistics or more complex clustering algorithms (such as K-means) to identify the cluster containing the most solutions.

[0123] Specifically, in one particular embodiment, the evaluation criteria are:

[0124] A. High-frequency optimization (main criterion): Select the cluster with the largest number of samples, assuming that this cluster corresponds to the correct solution. The statistical basis for this is that in a large number of repeated estimations, the correct solution, as the global optimal solution, should have a higher convergence probability.

[0125] B. Residual Verification (Secondary Criterion): Within high-frequency clusters, select... The set of results with the smallest value is considered the best fit in physics.

[0126] C. Physical constraint check: Verify the selected results. , , Does it meet the physical constraints? In this embodiment, the physical constraints are preferably... Is it greater than 1? Is it greater than , Is it in between.

[0127] From the solution cluster containing the largest number of solutions, the optimal solution is selected as the final unambiguous estimate. Here, "optimal" is usually measured by goodness of fit, with the most commonly used metric being the chi-square value (CQS). That is, among all the solutions contained in the largest solution cluster (the target solution cluster), the solution that minimizes the chi-square value between the theoretical ACF and the measured ACF is selected. This solution satisfies both the statistical "majority principle" and the physical "best fit," and is therefore considered the most reliable and least ambiguous estimate in this inversion.

[0128] More preferably, in a specific embodiment, an additional comprehensive confidence index based on the expected correctness probability and the convergence probability is provided, specifically including:

[0129] Generation of comprehensive confidence index: Combining the expected effective convergence probability obtained in step S200 And the expected correct estimate of probability The confidence score, ranging from 0 to 1, is generated by weighting the proportion of correctly clustered groups selected in this Monte Carlo selection or by establishing scoring rules. For example: Confidence = Clustering ratio. This confidence index is output along with the final parameter estimates, providing a quantitative assessment of the reliability of the inversion results.

[0130] Preferably, when it is determined that the expected indicator does not meet the preset decision threshold, a warning is triggered, indicating that the current data conditions cannot obtain a reliable solution, or the solution process is terminated.

[0131] Although the steps in the above embodiments are described in the above order, those skilled in the art will understand that in order to achieve the effect of this embodiment, different steps do not need to be executed in such order. They can be executed simultaneously (in parallel) or in reverse order. These simple changes are all within the protection scope of this application.

[0132] A second embodiment of this application provides a Monte Carlo-based unambiguous estimation system for incoherent scattering ionospheric parameters, used to implement the aforementioned Monte Carlo-based unambiguous estimation method for incoherent scattering ionospheric parameters. The system includes:

[0133] The threshold mapping library module is configured to store a pre-established signal-noise threshold mapping library, which stores the mapping relationship between different prior information combination modes and different signal-noise levels and expected indices of parameter estimation.

[0134] The noise assessment module is configured to estimate the actual signal-noise level of the measured echo data after receiving the measured echo data from the incoherent scattering radar, and determine the currently available combination mode of prior information.

[0135] The risk prediction module is configured to query the signal noise threshold mapping library based on the actual signal noise level and the combination pattern of prior information to obtain the expected index characterizing the reliability of parameter estimation under the current conditions, and compare it with the preset decision threshold.

[0136] The optimization selection module is configured to decide whether to start and execute the Monte Carlo optimization selection strategy based on the comparison result between the expected index and the preset decision threshold. The strategy includes using multiple sets of different initial parameters to independently perform multiple optimization fittings on the measured echo data to obtain a candidate solution set.

[0137] The optimal solution selection module is configured to perform statistical analysis on the candidate solution set to determine the final estimated value of the ionospheric parameter from the candidate solution set.

[0138] Preferably, the system also includes a confidence output module connected to the optimal solution selection module, used to receive the final estimate. This module can also combine the expected indicators provided by the risk prediction module and the solution cluster distribution characteristics provided by the statistical analysis submodule to generate and output a comprehensive confidence index corresponding to the final estimate.

[0139] Preferably, the risk prediction module is connected to the noise assessment module. Based on the signal-noise level and prior information combination pattern provided by the noise assessment module, it queries an internally stored or externally connected signal-noise threshold mapping library to obtain an expected index characterizing the reliability of the parameter estimation. This module compares this expected index with a preset decision threshold and, based on the comparison result, decides whether to transfer control and related data to the optimization selection module.

[0140] Preferably, the optimization selection module can be further divided into:

[0141] Multiple Fitting Submodule: Configured to use multiple sets of randomly generated initial parameters to independently perform multiple nonlinear optimization fittings on the echo data provided by the data receiving module, thereby generating a candidate solution set containing multiple fitting results;

[0142] Statistical Analysis Submodule: Configured to perform statistical analysis (e.g., through cluster analysis) on the candidate solution set generated by the multifit submodule to identify the solution cluster containing the largest number of solutions.

[0143] Preferably, the threshold mapping library module can be implemented as a local database associated with the prediction and decision module or a remote database accessed through a network interface.

[0144] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process and related descriptions of the system described above can be found in the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0145] It should be noted that the Monte Carlo-based incoherent scattering ionospheric parameter ambiguity estimation system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the modules or steps in the embodiments of this application can be further decomposed or combined. For example, the modules in the above embodiments can be merged into one module, or further divided into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of this application are only for distinguishing the various modules or steps and are not considered as an improper limitation of this application.

[0146] A device according to a third embodiment of this application includes:

[0147] At least one processor;

[0148] and a memory communicatively connected to at least one of the processors;

[0149] The memory stores instructions that can be executed by the processor to implement the aforementioned Monte Carlo-based method for unambiguous estimation of incoherent scattering ionospheric parameters.

[0150] A computer-readable storage medium according to a fourth embodiment of this application stores computer instructions that are executed by the computer to implement the above-described Monte Carlo-based method for unambiguous estimation of incoherent scattering ionospheric parameters.

[0151] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process and related descriptions of the storage device and processing device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0152] The following is for reference. Figure 4 It shows a schematic diagram of the structure of a computer system for implementing embodiments of the systems, methods, and electronic devices of this application. Figure 4 The server shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0153] like Figure 4As shown, the computer system 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 section 408 into Random Access Memory (RAM) 403. RAM 403 also stores various programs and data required for system operation. The CPU 401, ROM 402, and RAM 403 are interconnected via bus 404. Input / output (I / O) interface 405 is also connected to bus 404.

[0154] 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.

[0155] Specifically, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure 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 the functions defined in the methods of this application. It should be noted that the computer-readable medium described above in 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 computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk 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 connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A 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 a computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0156] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0157] 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. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing 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 the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can 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.

[0158] The terms “first”, “second”, etc., are used to distinguish similar objects, not to describe or indicate a specific order or sequence.

[0159] The term "comprising" or any other similar term is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus / device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent in such process, method, article, or apparatus / device.

[0160] The technical solutions of this application have been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of this application is obviously not limited to these specific embodiments. Without departing from the principles of this application, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of this application.

Claims

1. A fuzz-free estimation method for incoherent scattering ionospheric parameters based on Monte Carlo methods, characterized in that, Includes the following steps: Receive measured echo data, estimate the actual signal-to-noise level of the measured echo data, and determine the currently available combination of prior information. Based on the actual signal-noise level and the combination pattern of prior information, the signal-noise threshold mapping library is queried to obtain the expected index characterizing the reliability of parameter estimation under the current conditions; the signal-noise threshold mapping library stores the mapping relationship between different combination patterns of prior information and different signal-noise levels and one or more expected indices. Determine whether the expected index meets the preset decision threshold. If it does, then based on the Monte Carlo optimization selection strategy, use multiple sets of different initial parameters to independently perform multiple optimization fittings on the measured echo data to obtain a candidate solution set composed of multiple fitting results. Statistical analysis is performed on the candidate solution set to identify a candidate subset representing the global optimal solution, and the solution with the best fit is determined from the candidate subset as the final estimate of the ionospheric parameter. The establishment of the signal-noise threshold mapping library is achieved through the following steps: For each combination of prior information, sampling is performed within a preset ionospheric parameter space to generate multiple sets of true values ​​of ionospheric parameters. For each set of true values ​​of ionospheric parameters, Monte Carlo simulations were performed multiple times under multiple signal and noise levels, combined with the theoretical model of incoherent scattering radar, to obtain a database containing a large number of simulation spectra. For each simulated spectrum in the database, parameter inversion is performed, and based on the relationship between the inversion results and the true values ​​of ionospheric parameters, the expected indicators corresponding to different signal noise levels under each prior information combination mode are statistically calculated. This determines the mapping relationship between different prior information combination modes and different signal noise levels and one or more expected indicators, thereby forming the signal noise threshold mapping library.

2. The method for ambiguity-free estimation of ionospheric parameters based on Monte Carlo incoherent scattering according to claim 1, characterized in that, In the Monte Carlo simulation process, the result of each parameter inversion is judged. If the normalized chi-square value of a certain inversion is less than the chi-square threshold determined based on the preset confidence level, it is judged as a valid convergence.

3. The ambiguity-free estimation method for incoherent scattering ionospheric parameters based on Monte Carlo as described in claim 1, characterized in that, Statistical analysis is performed on the candidate solution set, including: Cluster analysis is performed on the ion composition values ​​of each solution in the candidate solution set to divide the candidate solution set into at least one solution cluster, and the target solution cluster containing the maximum number of solutions is determined. From the target solution cluster, a solution with the best fit is selected as the final estimate of the ionospheric parameters.

4. The method for ambiguity-free estimation of ionospheric parameters based on Monte Carlo incoherent scattering according to claim 3, characterized in that, The solution with the best fit is the solution with the smallest fitting residual among all solutions that meet the physical constraints in the target solution family.

5. The ambiguity-free estimation method for incoherent scattering ionospheric parameters based on Monte Carlo as described in claim 4, characterized in that, By combining the expected index obtained from the signal-noise threshold mapping library with the statistical characteristics of the target solution cluster, a comprehensive confidence index corresponding to the final estimated value is generated and output.

6. The ambiguity-free estimation method for incoherent scattering ionospheric parameters based on Monte Carlo as described in claim 1, characterized in that, The expected indicators for the reliability of the characterization parameter estimation are the probability of correct estimation and / or the probability of effective convergence.

7. The method for ambiguity-free estimation of ionospheric parameters based on Monte Carlo incoherent scattering according to claim 1, characterized in that, The prior information combination mode is defined as: a specific combination of applying prior constraints to one or more parameters in a set of ionospheric parameters to be estimated, wherein the prior constraints are derived from ionospheric theoretical models or plasmaline measurements.

8. The ambiguity-free estimation method for incoherent scattering ionospheric parameters based on Monte Carlo as described in claim 1, characterized in that, The estimation of the actual signal-to-noise level of the measured echo data includes: The measured echo data is preprocessed to obtain the autocorrelation function or power spectrum at the target altitude; based on the autocorrelation function or power spectrum, and combined with the noise power measured in the no-signal region, the signal-to-noise ratio at the target altitude is calculated; the actual signal-to-noise level is determined according to the signal-to-noise ratio and the number of pulse accumulations.

9. The ambiguity-free estimation method for incoherent scattering ionospheric parameters based on Monte Carlo as described in claim 1, characterized in that, When it is determined that the expected indicator does not meet the preset decision threshold, a warning is triggered, indicating that the current data conditions cannot obtain a reliable solution, or the solution process is terminated.

Images