Consistency checking method and system applied to multi-source communication data

Abstract

The present application relates to the technical field of data check processing, and particularly relates to a consistency check method and system applied to multi-source communication data. The technical scheme of the present application obtains historical data of various communication parameters of multiple data sources and target data for consistency check of the target parameters by the historical data; obtains influence strength indexes of various communication parameters on the deviation of the target parameters based on fluctuation correlation characteristic analysis of the historical data and the target data; performs influence weight analysis according to the differences between the influence strength indexes and the operation requirement values of the various communication parameters to obtain weighted difference scores of inconsistent performances between the data sources; combines the weighted difference scores and the untrustworthiness of the data sources in the historical period to correct the deviation caused by the historical stability difference; and marks real-time data according to the inconsistency scores corresponding to the data sources when the target parameters perform the check task, so that accurate identification of abnormal data is realized.

Description

Consistency Verification Method and System for Multi-Source Communication Data Technical Field

[0001] This invention relates to the technical field of data verification and processing, and specifically to a consistency verification method and system for multi-source interconnected data. Background Technology

[0002] Multi-source communication data consistency verification is a process that processes the same target / data item from multiple data sources (such as different sensors and communication detection devices) to filter out unreliable data. Cross-validation significantly improves the accuracy and reliability of situational awareness. This processing method can effectively identify and locate sensor errors, data conflicts, and potential deception, filtering out unreliable information. Based on this, more accurate verification results can be generated, providing a high-quality data foundation for subsequent decision-making, target allocation, and automated response, ultimately improving decision-making quality. Traditional methods only verify data consistency based on the differences in parameters across multiple data sources, which is not accurate enough in judging the consistency of acquired data and is prone to identifying abnormal errors. Summary of the Invention

[0003] To address the technical problem of inaccurate data consistency verification based on various communication parameters from multiple data sources, this invention aims to provide a consistency verification method and system for multi-source communication data. This method considers the impact of different communication parameters and subsequent application operations on data consistency verification, thereby improving the accuracy of data consistency verification across multiple data sources. The specific technical solution adopted is as follows:

[0004] In a first aspect, embodiments of the present invention provide a consistency verification method for multi-source communication data, the method comprising:

[0005] Acquire historical data of various communication parameters from multiple data sources, as well as target data for consistency verification of target parameters against historical data;

[0006] Analyze the fluctuation correlation characteristics between historical data and target data to obtain the influence intensity index of each communication parameter of each data source on the target parameter;

[0007] Impact weight analysis is performed based on the differences between the various impact intensity indices and the operational requirements of each communication parameter to obtain a weighted difference score for the inconsistent performance between the various data sources.

[0008] Based on the weighted difference score and the unreliability of the target parameter of each data source in the historical period, the inconsistency score between the current data source and other data sources is obtained.

[0009] When performing the verification task for the target parameters, the real-time data currently collected from the corresponding data source is marked according to each inconsistency score.

[0010] In an optional embodiment, before obtaining historical data of various communication parameters from multiple data sources, the method further includes:

[0011] Configure spatial precision, timestamp precision, time synchronization error, data update rate, and location consistency score calculated based on historical data as various communication parameters of the data source in the spatiotemporal dimension;

[0012] Configure the model identification confidence, parameter measurement error, identification accuracy calculated based on historical data, and confusion matrix as various communication parameters of the data source in the attribute dimension;

[0013] Configure the reporting metrics for communication relationship detection capability, false alarm rate, false negative rate, and relationship persistence as various communication parameters of the data source in the data relationship dimension.

[0014] In one optional embodiment, the fluctuation correlation characteristics between historical data and target data are analyzed to obtain an influence strength index of each communication parameter from each data source on the target parameter, including:

[0015] Based on the historical data differences of various communication parameters from multiple data sources in the same data acquisition task, the sensitivity of the differences between each communication parameter from each data source to the target parameter is obtained.

[0016] Based on the frequency of fluctuations exceeding the difference threshold in historical data from each data source and the average value of all sensitivities, the influence intensity index of each communication parameter from each data source on the target parameter is obtained.

[0017] In one optional embodiment, based on the historical data differences of various communication parameters from multiple data sources in the same data acquisition task, the sensitivity of the differences between each communication parameter from each data source to the target parameter is obtained, including:

[0018] In the same data acquisition task, based on the absolute value of the difference between the current data source and other data sources with the same communication parameters in historical data, the cumulative summation of the absolute differences between the current data source and other data sources with the corresponding communication parameters is obtained;

[0019] Based on the ratio of the cumulative summation result of the current data source to the maximum summation result of all data sources, the difference results of each communication parameter of the current data source in the corresponding data acquisition task are obtained;

[0020] The sensitivity of each data source is obtained based on the changes in target data and the differences in various communication parameters of the current data source in different data acquisition tasks.

[0021] In one optional embodiment, an influence weighting analysis is performed based on the differences between the various influence intensity indices and the operational requirements of each connectivity parameter to obtain a weighted difference score for inconsistencies in the performance of each data source, including:

[0022] Based on the operational requirement values ​​of each communication parameter and the maximum requirement values ​​of all corresponding operations, the first weighting coefficient of each communication parameter is obtained;

[0023] Based on the summation of the differences in the influence intensity indices of the same communication parameters from various data sources, the second weighting coefficients of each communication parameter are obtained.

[0024] Based on the first and second weight coefficients of the same communication parameter, the influence weight of the corresponding communication parameter on the target parameter is obtained;

[0025] Based on the summation of the differences between the influence weights of each communication parameter and the influence intensity indices, a weighted difference score is obtained for the inconsistency performance of the target parameters of each data source.

[0026] In an optional embodiment, before obtaining the inconsistency score between the current data source and other data sources based on the weighted difference score and the unreliability of each data source in verifying the target parameter over a historical period, the method further includes:

[0027] Based on the frequency of inconsistencies in the performance of the target parameters when the current data source is verified in historical periods, the first unreliability coefficient of the current data source in the dimension of data anomalies is obtained between the current data source and other data sources.

[0028] Based on the degree of deviation of the current data source from the inconsistent performance of the target parameters when verifying them in historical periods, a second unreliability coefficient between the current data source and other data sources is obtained.

[0029] Based on the first and second unreliability coefficients of each data source, the unreliability of the target parameter for verification during the historical period is obtained.

[0030] In one optional embodiment, a first unreliability coefficient is obtained between the current data source and other data sources in the dimension of data anomalies, based on the frequency of inconsistencies in the current data source's behavior when verifying target parameters over historical periods. This includes:

[0031] Based on the differences in various communication parameters of the current data source during historical periods in the data acquisition task, obtain the abnormal indicators of the corresponding communication parameters;

[0032] Statistical analysis is performed on abnormal indicators where various communication parameters exceed the indicator threshold to obtain the degree of abnormality of each data source;

[0033] The first unreliability coefficient of each data source is obtained by the ratio of the degree of anomaly of each data source to the maximum value of all anomalies.

[0034] In one optional embodiment, a second unreliability coefficient between the current data source and other data sources is obtained based on the degree of deviation of the current data source from inconsistent performance when verifying target parameters in historical periods, including:

[0035] The consistency difference of the current data source in the historical period is obtained by comparing the initial consistency level with the final consistency level of the current data source in the historical period.

[0036] The second unreliability coefficient of each data source is obtained by comparing the difference in consistency between the data sources with the average consistency of the corresponding data sources over a historical period.

[0037] In one optional embodiment, an inconsistency score between the current data source and other data sources is obtained based on a weighted difference score and the unreliability of each data source in verifying the target parameter over a historical period, including:

[0038] Based on the unreliability of the target parameters verified by each data source in historical time periods and the preset coefficients, the weighted correction amount of each data source is obtained.

[0039] Based on the weighted correction amount and corresponding weighted difference score of each data source, the inconsistency score between the current data source and other data sources is obtained.

[0040] Secondly, embodiments of the present invention also provide a consistency verification system for multi-source communication data. The verification system is a system corresponding to any method in the first aspect, and the system includes:

[0041] The acquisition module is used to acquire historical data of various communication parameters from multiple data sources, as well as target data for consistency verification of the target parameters against the historical data.

[0042] The first acquisition module is used to analyze the fluctuation correlation characteristics between historical data and target data in order to obtain the influence intensity index of each communication parameter of each data source on the target parameter;

[0043] The second acquisition module is used to perform influence weight analysis based on the differences between the influence intensity indices and the operational requirements of each communication parameter, so as to obtain a weighted difference score of inconsistent performance between the data sources.

[0044] The third acquisition module is used to obtain the inconsistency score between the current data source and other data sources based on the weighted difference score and the unreliability of the target parameter of each data source in the historical period.

[0045] The verification marking module is used to mark the real-time data currently collected by the corresponding data source according to each inconsistency score when performing verification tasks on the target parameters.

[0046] The present invention has the following beneficial effects:

[0047] The technical solution of this invention acquires historical data of various communication parameters from multiple data sources, as well as target data for consistency verification of target parameters based on historical data. Based on the fluctuation correlation characteristics analysis between historical data and target data, it obtains the influence intensity index of each communication parameter on the deviation of the target parameter, thereby deriving differentiated results of the consistency performance of different communication parameters on the target parameter. Further, it performs influence weight analysis based on the differences between each influence intensity index and the operational requirements of each communication parameter to obtain a weighted difference score for inconsistencies between data sources. Then, it combines the weighted difference score with the unreliability correction of each data source over historical periods to address biases caused by historical stability differences. This allows the inconsistency score obtained from the data sources to simultaneously reflect the magnitude of the influence of different communication parameters and the magnitude of the influence of subsequent application operations on data verification. When the target parameter performs a verification task, real-time data is marked according to the inconsistency score corresponding to each data source, achieving accurate identification of abnormal data. This technical solution is based on a verification mechanism constructed from a multi-level inference chain. It can simultaneously consider the sensitivity of different communication parameters, the operational requirements under the task scenario, and the impact of the long-term stability of the data source on the verification. This makes the consistency verification results of multiple data sources closer to the actual deviation, thereby improving the accuracy of consistency verification of multi-source communication data. It solves the technical problem of inaccurate consistency verification caused by inconsistent contributions of different parameters and differences in the stability of different data sources in the existing technology. Attached Figure Description

[0048] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 is a flowchart of a consistency verification method for multi-source communication data provided in an embodiment of the present invention;

[0050] Figure 2 is a flowchart of calculating the influence intensity index provided by an embodiment of the present invention;

[0051] Figure 3 is a flowchart of calculating the weighted difference score provided in an embodiment of the present invention;

[0052] Figure 4 is a schematic diagram of the structure of a consistency verification system for multi-source communication data provided in an embodiment of the present invention. Detailed Implementation

[0053] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a consistency verification method and system for multi-source communication data proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0054] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0055] Multi-source communication data consistency verification is a cross-validation data processing method that can significantly improve the accuracy and reliability of data situational awareness. It effectively identifies and locates sensor errors, data conflicts, and potential deception, filtering out unreliable information. Based on this, it can generate a more accurate and unified situational map, providing a high-quality data foundation for subsequent command decisions, target allocation, and automated response, ultimately improving efficiency and decision-making quality. Traditional methods only verify consistency based on the differences in parameters across multiple data sources. However, different parameters have varying impacts on data verification, and subsequent data applications also have related influences. Therefore, a new technical solution is needed to address these technical problems.

[0056] The following describes in detail, with reference to the accompanying drawings, a specific scheme for a consistency verification method and system for multi-source communication data provided by the present invention. Please refer to Figure 1, which shows a flowchart of a consistency verification method for multi-source communication data according to an embodiment of the present invention. This method can be applied to a data processing terminal to perform consistency verification of target parameters from multiple data sources. The processing terminal can be a computer device or a server, as long as it can run the method; no specific limitations are imposed here. The verification method includes:

[0057] S11: Obtain historical data of various communication parameters from multiple data sources, as well as target data for consistency verification of target parameters against historical data.

[0058] Specifically, communication parameters are used to verify the consistency of the target parameters. There are i data sources, and each data source has k communication parameters, where i and k are natural numbers greater than 1. Let i and k be the indices of the data source and the communication parameter, respectively. The size of the historical data corresponding to communication parameter k for data source i is denoted as... The target parameter is denoted as j, and the size of the target data is denoted as . .

[0059] For example, in a scenario where three communication terminals coordinate for positioning, the first data source is historical data of various communication parameters collected by the UAV terminal, the second data source is historical data of various communication parameters collected by the ground mobile platform, and the third data source is historical data of various communication parameters collected by the fixed base station. The target parameter is the location to be located, and the target data can be spatial coordinates. Communication parameters can include timestamp accuracy, time synchronization error, etc. The consistency of the three data sources is checked to evaluate the inconsistency of each data source in the current task. Of course, the technical solution of this invention can also be applied to scenarios where multiple data sources are terminals of the same type, such as determining the location of a mobile terminal through multiple base stations, where the communication parameters of the multiple base stations are the same.

[0060] In the process of consistency verification of multi-source communication data, different data sources may exhibit differences in different dimensions. In order to effectively identify the data deviations between data sources and improve the accuracy of verification results, this invention configures the communication parameters of each data source according to the data attribute characteristics they reflect.

[0061] For example, spatial accuracy, timestamp accuracy, time synchronization error, data update rate, and location consistency score calculated based on historical data can be configured as various communication parameters of the data source in the spatiotemporal dimension. Spatial accuracy and timestamp accuracy are used to describe the basic situation of the data source in terms of spatial positioning and acquisition time control, time synchronization error is used to characterize the time deviation between different data sources, data update rate reflects the real-time performance of its uploaded data, and historical location consistency score reflects the long-term performance of the data source in terms of spatial data reliability.

[0062] The model identification confidence score, parameter measurement error, identification accuracy calculated based on historical data, and confusion matrix are configured as the communication parameters of the data source in the attribute dimension. The model identification confidence score and identification accuracy are used to evaluate the data source's ability to judge the target attribute, the confusion matrix is ​​used to characterize the misclassification relationship between different categories, and the parameter measurement error reflects the stability of the data source in the attribute data acquisition process.

[0063] The reporting metrics for communication relationship detection capability, false alarm rate, false negative rate, and relationship persistence are configured as various communication parameters for the data source in the data relationship dimension. Detection capability measures the effectiveness of the data source in identifying communication relationships; false alarm rate and false negative rate reflect the probability of misjudgment and missed judgment, respectively; and the relationship persistence metric measures the consistency of its tracking of relationship links. By systematically configuring the communication parameters in these three dimensions, the sources of deviation from different data sources can be comprehensively evaluated from three levels: spatiotemporal consistency, attribute judgment consistency, and communication relationship consistency. This provides a structured and quantifiable data input foundation for subsequent consistency verification.

[0064] Furthermore, the system can be configured to collect inherent performance parameters and dynamic performance parameters from various data sources. Inherent performance parameters are primarily obtained through laboratory calibration, equipment factory specifications, and technical manuals; dynamic performance parameters rely on an online learning and feedback loop. Dynamic performance parameters are calculated by continuously comparing multi-source data and using statistical tests (such as the chi-square test) to determine spatiotemporal consistency scores. The system updates attribute recognition accuracy and confusion matrix by recording the differences between identification results and final verification conclusions. Furthermore, it dynamically corrects false alarm rates and false negative rates by analyzing the differences between relationship detection records and the actual situation. This method ensures that the evaluation parameters incorporate prior knowledge and continuously self-optimize as the system operates.

[0065] At this point, historical data of various communication parameters from multiple data sources and target data of target parameters have been obtained, and we proceed to step S12.

[0066] S12: Analyze the fluctuation correlation characteristics between historical data and target data to obtain the influence intensity index of each communication parameter of each data source on the target parameter.

[0067] Specifically, inconsistencies in communication parameters acquired from different data sources do not necessarily have a significant impact on the consistency of the final collected data. To assess the consistency of data acquired from different data sources, the impact of different communication parameters on the collected data should be prioritized. For example, frequent inconsistencies in a particular communication parameter indicate that its corresponding measurement stage is inherently vulnerable, susceptible to interference, or has design flaws. In complex electromagnetic environments, the attribute parameters of communication signals (such as modulation identification) are more prone to general deviations than basic spatiotemporal parameters. Therefore, inconsistencies in this particular communication parameter often have a relatively small impact on the final data consistency assessment. Furthermore, if a parameter exhibits significant inconsistencies, but the historical data acquired from this data source shows little difference from historical data acquired from other data sources, this also indicates a relatively small impact of that parameter. Therefore, based on the prevalence of inconsistencies in a single communication parameter in historical data and their impact on the final results, the influence of inconsistencies in different communication parameters can be analyzed, leading to an influence strength index.

[0068] For example, please refer to Figure 2, which is a flowchart of the calculation of the influence intensity index. Step S12 includes sub-steps S12-1 to S12-2, which are described in detail below:

[0069] S12-1: Based on the historical data differences of various communication parameters from multiple data sources in the same data acquisition task, obtain the sensitivity of the differences between each communication parameter of each data source to the target parameter. The number of data acquisition tasks is denoted as p. The differences of similar communication parameters from multiple data sources in the same acquisition task can be calculated, and then summed to obtain the historical data difference performance. Analyzing the historical data difference performance yields the sensitivity of each communication parameter to the target parameter. For example, the following three steps can be used for analysis.

[0070] The first step, within the same data acquisition task, is to obtain the cumulative sum of the absolute differences between the current data source and other data sources at the same communication parameters, based on the absolute values ​​of the differences in historical data with the same communication parameters. This is achieved using the formula: Calculate the current data source in data acquisition task p. The absolute value of the difference in the communication parameter k ; The number of multiple data sources; For the current data source The historical data size of the communication parameter k; Other data sources The historical data size of the communication parameter k.

[0071] The second step is to obtain the differences in various communication parameters of the current data source in the corresponding data acquisition task based on the ratio of the cumulative sum of the current data source to the maximum sum of all data sources. Continue using the formula: Calculate the current data source in data acquisition task p. Difference results of communication parameter k , This refers to all data sources retrieved during all historical retrievals of data j. The maximum value in the range, i.e., the maximum summation result, should be noted that... The value is not 0, in When the value is 0, it means that the cumulative summation result of the entire data source is 0. At this time, it has no analytical significance and is not within the scope of analysis of this application.

[0072] The third step is to obtain the sensitivity of each data source based on the changes in target data and the differences in various communication parameters of the current data source in different data acquisition tasks. This is further demonstrated using the formula: Calculate the sensitivity s; The change in the target data. Calculation. At that time, Replace with the value of the target parameter j This yields the difference in the number of times the target parameter j is acquired, p. Smaller When the value is large, it indicates that a small change in the communication parameter k is more likely to have a significant impact on the target parameter j. The function represents the normalization function. In one embodiment of the invention, the normalization process can specifically be, for example, maximum-minimum value normalization. Furthermore, the normalization in subsequent steps can all adopt maximum-minimum value normalization. In other embodiments of the invention, other normalization methods can be selected according to the specific range of the values, which will not be elaborated further. Adding 0.1 to the denominator is to prevent calculation problems caused by a value of 0.

[0073] S12-2: Based on the frequency of fluctuations exceeding the difference threshold in historical data from each data source and the average of all sensitivities, obtain the influence strength index of each communication parameter from each data source on the target parameter. The difference threshold can be set, and the fluctuation frequency of historical data can be statistically analyzed. For example, setting the difference threshold to 0.7, the influence strength index of each communication parameter on the target parameter j when it is acquired can be calculated. The number of occurrences of >0.7 And the total number of times the target parameter j was obtained from all different historical data sources. The ratio of the two is .when The larger the value of parameter k, the greater its impact on the numerical difference of data j. The smaller the value of the communication parameter k, the more frequently data differences occur when the target parameter j is obtained from multiple data sources. Historically, the data differences of the communication parameter k have a relatively small impact on the data differences of the target parameter j. Even a large numerical difference in the communication parameter k will not cause a significant change in the numerical value of the target parameter j. Therefore, the reference weight of the communication parameter k should be smaller when analyzing data consistency.

[0074] The specific values ​​(e.g., 0.7) given in the embodiments of this invention are empirical values ​​obtained under typical hardware configurations and test scenarios, intended to facilitate understanding of this invention. In practical applications, those skilled in the art can adjust, calibrate, or optimize these parameters according to specific hardware performance, scenario complexity, and data characteristics, which does not constitute a limitation of this invention.

[0075] When calculating the influence intensity index, it is necessary to first calculate the impact of the differences in the communication parameter k obtained from all data sources throughout history on the value of the target parameter j. mean This is used to represent the impact of differences in the communication parameter k on the numerical differences in the target parameter j; the smaller the value, the higher the consistency. Based on the formula: The influence of communication parameter k on the consistency judgment of target parameter j is calculated, that is, the influence strength index of communication parameter k on target parameter j. This method is used to obtain the impact of different communication parameters on the consistency judgment of the target parameter j, so as to ensure the accurate removal of inconsistent data in the final data set.

[0076] At this point, the influence intensity index of each communication parameter of each data source on the target parameter has been obtained based on the above method, and we proceed to step S13.

[0077] S13: Perform influence weight analysis based on the differences between the various influence intensity indices and the operational requirements of each communication parameter to obtain a weighted difference score for the inconsistent performance between the various data sources.

[0078] Specifically, when conducting influence weight analysis, it is necessary to consider the degree of correlation between the next operation after obtaining different communication parameters. The greater the correlation, the greater the influence weight needs to be; conversely, the influence weight should be reduced. By considering the differences in inconsistent parameter performance during historical data transmission of the current data source as a reference for judging the reliability of the data source, the accuracy of target parameter verification can be effectively improved.

[0079] Because the standards for acquiring communication parameters vary depending on the usage scenario—for example, when acquiring location information—the accuracy requirements for that location information may differ depending on the subsequent operation. When high accuracy is required, the consistency of accuracy-related parameters is often crucial; otherwise, the acquired data may not meet the needs of the next operation. Therefore, based on the varying requirements of the data in the next operation, inconsistency tolerance limits are imposed on the relevant parameters to obtain a corresponding weighted difference score. The degree of requirement for the data corresponding to the communication parameters can be characterized based on operational requirement values, such as the magnitude of synchronization error.

[0080] For example, please refer to Figure 3, which is a flowchart of the weighted difference score calculation. Step S13 includes sub-steps S13-1 to S13-4, which are described in detail below:

[0081] S13-1: Based on the operational requirement values ​​of each communication parameter and the maximum requirement values ​​of all corresponding operations, obtain the first weighting coefficient for each communication parameter. The operational requirement values ​​are denoted as... The maximum required value for all operations is denoted as Taking the communication parameter k as an example, there are several related requirements b. Corresponding The mean of the absolute values ​​of the differences is The first weighting coefficient can be derived from this value, using the following formula: .

[0082] The operational requirement value ( This refers to the performance requirements of downstream application tasks for various communication parameters. This value can be preset by the task planning system according to the task type. For example, when performing a "precision strike" task, the operational requirement value for the "spatial accuracy" parameter will be set very high, such as 0.8; while when performing a "wide-area surveillance" task, the requirement value for the "spatial accuracy" parameter will be relatively low, such as 0.5, and preset according to the actual situation.

[0083] S13-2: Based on the summation of the differences in the influence intensity indices of the same communication parameters from various data sources, the second weighting coefficients for each communication parameter are obtained. The influence of communication parameter k on the consistency judgment of target parameter j is represented by the influence intensity index. Other parameters The impact of the consistency judgment on the target parameter j is the influence intensity index. Through the formula: The second weighting coefficient is calculated. This represents the number of influence intensity indices associated with the communication parameter k.

[0084] S13-3: Based on the first and second weighting coefficients of the same communication parameter, obtain the influence weight of the corresponding communication parameter on the target parameter. The influence weight of communication parameter k on target parameter j is denoted as... It can be done through the formula: Calculations are performed using the max-min normalization method. After normalization, we get Its range is [0,1]. This can be understood as... The larger, The smaller the value, the greater the relationship between the next operation after obtaining the target parameter j and the communication parameter k, the smaller the tolerance for its inconsistency, and the greater its influence on the final consistency judgment of the obtained data.

[0085] S13-4: Based on the summation of the differences between the influence weights of each communication parameter and the influence intensity indices, obtain the weighted difference score for the inconsistency performance of the target parameters for each data source. This can be achieved using the formula: This yields the weighted difference score between the target parameter j obtained from data source i in the current iteration and other data sources. , Given the number of parameters, this formula considers the differences between the communication parameters k of data source i and the corresponding communication parameters of other data sources in the current detection of target parameter j by multiple data sources (p). Multiply by the weight of the effect of parameter k on the objective parameter j And sum them up. Use weighted difference scoring. As a preliminary assessment, it serves as one of the references for the final consistency assessment, ensuring the accuracy of the data consistency assessment.

[0086] Due to issues with data measurement equipment or interference from the local environment, data sources may exhibit significant inconsistencies over a period of time. In such cases, the reliability of the acquired data can be dynamically adjusted based on its consistency over a given time period, serving as one of the references for judging consistency. Therefore, the verification method of this invention further includes:

[0087] S13-5: Based on the frequency of inconsistencies in the target parameter's behavior during historical data verification, obtain the first unreliability coefficient between the current data source and other data sources in the dimension of data anomalies. A single data source may exhibit similarities in different data acquisitions of the same communication parameter. When a parameter shows a significant difference compared to previous acquisitions, and this difference is greater than the difference from other data sources, the degree of difference between this acquired data and previous data is greater, resulting in lower data reliability and greater data inconsistency. The first unreliability coefficient can be calculated based on this data difference characteristic. Specifically, the first unreliability coefficient is calculated using the following method.

[0088] The first step is to obtain the abnormal indicators of the corresponding communication parameters based on the differences in the historical data acquisition results of various communication parameters from the current data source. For example, when acquiring the target parameter j in p historical data acquisitions, the differences in the corresponding communication parameter k from data source i are as follows: The differences in communication parameter k are calculated when the target parameter j is obtained multiple times from data source i during historical time periods. mean According to the formula: Calculate the abnormal indicators of the communication parameter k corresponding to the current data source i p. Adding 0.1 to the denominator is to prevent calculation problems caused by a value of 0.

[0089] The second step involves statistically analyzing abnormal indicators where each communication parameter exceeds a threshold value to determine the degree of anomaly for each data source. This includes analyzing the abnormal indicators corresponding to different communication parameters k. Normalization is performed using the maximum and minimum values ​​to obtain the normalized result. Its range is [0,1]. Statistics Number of communication parameters k Normalized results of all communication parameters k The sum is: , The number of parameters. For historical periods The number of parameters, the maximum value is The ratio of the two is . The larger the cumulative sum, the higher the ratio. The larger the value, the greater the difference between the inconsistency in the acquisition of target parameter j by the current data source i and previous instances, indicating a higher likelihood of anomalies or deception, resulting in lower reliability and consistency. Therefore, the degree of anomaly in the inconsistency of communication parameters in the acquisition of target parameter j by the current data source i can be determined. This serves as one of the criteria for judging the reliability of data j obtained from data source i, making the consistency judgment of the target parameter j more accurate.

[0090] The third step is to obtain the first unreliability coefficient for each data source based on the ratio of the anomaly degree of each data source to the maximum value of all anomaly degrees. The anomaly degree of inconsistency in the acquisition of target parameter j by data source i in the current p-th iteration is: The maximum value for all abnormalities is The first unreliability coefficient is .

[0091] It should be noted that, , The value is not 0, in When the value is 0, it means that none of the conditions are met. In this case, it has no analytical significance and is not within the scope of analysis of this application. If such a situation occurs, the system will directly report an error.

[0092] S13-6: Based on the degree of deviation of the current data source from the inconsistencies in the target parameter verification during historical periods, obtain the second unreliability coefficient between the current data source and other data sources. Using the target parameter j acquired in the current instance as a reference, calculate N times backwards to obtain the historical data acquisition period g. For each instance of target parameter j acquisition p within historical period g, the consistency of the target parameter j verified by data source i is denoted as... The average consistency of data source i across multiple data collection tasks during historical time periods g can be calculated based on a computational model or formula. .

[0093] Based on the difference between the initial consistency level and the final consistency level of the current data source in the historical time period, obtain the consistency level difference of the current data source in the historical time period g. Obtain the initial consistency level at the start time of the historical time period g. Consistency with termination time The difference between the two is: This yields the consistency difference, which can be normalized to ensure its value falls within a preset range. Based on the consistency difference of each data source and the average consistency of that data source over a historical period, a second unreliability coefficient for each data source can be obtained using the ratio of the two.

[0094] S13-7: Based on the first and second unreliability coefficients of each data source, obtain the unreliability of the target parameter for each data source during the historical period. The unreliability of the target parameter j for the current verification of data source i is denoted as... Through the formula: The unbelievability level is calculated, where norm is the normalization function. Similarly, the max-min normalization method can be used to evaluate the calculation results. After normalization, we get Its range is [0,1]. The closer the unreliability score is to 1, the lower the reliability of the data source; conversely, the closer the unreliability score is to 0, the higher the reliability of the data source. From the above formula, it can be seen that the mean... The smaller the value, the greater the difference in consistency. The larger the ratio The larger the value, the more likely the data source is to have measurement anomalies within the time period, and the lower the reliability of the obtained data. Adding 0.1 to the denominator is a safety factor to prevent the denominator from being zero; its dimensions are... Consistent with.

[0095] At this point, the weighted difference scores and unreliability scores for the inconsistent performance among the various data sources have been obtained, and we proceed to step S14.

[0096] S14: Based on the weighted difference score and the unreliability of the target parameter of each data source in the historical period, obtain the inconsistency score between the current data source and other data sources.

[0097] Specifically, the weighted difference score is derived from the correlation analysis of the next step after obtaining different communication parameters. It is rather limited to use it as a result of inconsistency between data sources. Therefore, it can be further analyzed by combining unreliability and weighted difference score to obtain the inconsistency score of data sources.

[0098] It's understandable, when The larger the value, the lower the reliability of the target parameter j obtained by data source i, and the lower the weighted difference score of the target parameter j obtained by data source i in the current p-th generation with other data sources. The more you increase the weighted difference score, the better. This characterizes the inconsistency between data source i and other data sources. Conversely, the greater the inconsistency, the more it should be reduced. Therefore, the inconsistency score can be expressed by the formula: The calculation shows that the natural number 1 in the formula is a preset coefficient. This preset coefficient can be set based on the range of unreliability values, ensuring accurate adjustment of the weighted difference score. (Inconsistency scoring) This characterizes the inconsistency between the target parameter j of the current data source i and other data sources.

[0099] It should be noted that after obtaining the inconsistency score based on the above method, the min-max normalization method can also be used to further refine the score. After normalization, we get Its range is [0,1]. When When the value is greater than 0.7, it is determined that the target parameter j obtained by data source i in the current time p is inconsistent with the target parameter j obtained by other data sources, indicating an anomaly or deception. The reliability of the data can be determined based on the inconsistency score of each data source.

[0100] At this point, an inconsistency score has been obtained based on the weighted difference score and the unreliability score, and we proceed to step S15.

[0101] S15: When performing the verification task for the target parameters, mark the real-time data currently collected from the corresponding data source according to each inconsistency score.

[0102] Specifically, the above methods are used to identify inconsistencies in target data acquired from different data sources. These inconsistencies are then characterized by an inconsistency score. During the validation process for the target parameters, an inconsistency score for each data source is derived based on this method and clearly marked. For example, problematic inconsistencies from different data acquisition sessions can be selected. After acquiring inconsistent and abnormal data, each piece of data can be tagged with a timestamp, data source identifier, and conflict type label, and then stored in a dedicated database in association with the original data. This structured record not only preserves complete audit trails but also provides crucial information for subsequent analysis of data source reliability and optimization of validation algorithms.

[0103] Furthermore, at the visualization level, abnormal data is marked with prominent visual elements. For example, on the situation map, conflicting trajectories are displayed as red dashed lines, warning icons are added next to untrustworthy attributes, and contradictory relationships are identified using flashing warning symbols. This instant highlighting ensures that operators can identify data inconsistencies immediately, providing intuitive support for human-machine collaborative decision-making.

[0104] Based on the same technical concept as the verification method, this embodiment of the invention also provides a consistency verification system for multi-source communication data. Please refer to Figure 4, which is a schematic diagram of the structure of the verification system. The verification system is the system corresponding to any of the above methods. The verification system includes an acquisition module 401, a first acquisition module 402, a second acquisition module 403, a third acquisition module 404, and a verification mark module 405.

[0105] The acquisition module 401 is used to acquire historical data of various communication parameters from multiple data sources, as well as target data for consistency verification of the target parameters against the historical data.

[0106] The first acquisition module 402 is used to analyze the fluctuation correlation characteristics between historical data and target data to obtain the influence intensity index of each communication parameter of each data source on the target parameter.

[0107] The second acquisition module 403 is used to perform influence weight analysis based on the differences between each influence intensity index and the operational requirements of each communication parameter, so as to obtain a weighted difference score of inconsistent performance between each data source.

[0108] The third acquisition module 404 is used to obtain the inconsistency score between the current data source and other data sources based on the weighted difference score and the unreliability of the target parameter of each data source in the historical period.

[0109] The verification marking module 405 is used to mark the real-time data currently collected by the corresponding data source according to each inconsistency score when performing a verification task on the target parameters.

[0110] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0111] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A consistency verification method for multi-source communication data, characterized in that, The method includes: acquiring historical data of various communication parameters from multiple data sources, and target data for consistency verification of the target parameter using the historical data; analyzing the fluctuation correlation characteristics between the historical data and the target data to obtain an influence strength index of each communication parameter from each data source on the target parameter; performing influence weight analysis based on the differences between the influence strength indices and the operational requirement values ​​of each communication parameter to obtain a weighted difference score of inconsistencies between data sources; obtaining an inconsistency score between the current data source and other data sources based on the weighted difference score and the unreliability of each data source in verifying the target parameter during historical periods; marking the real-time data currently collected by the corresponding data source based on each inconsistency score when the target parameter is being verified; and determining the influence strength index by: obtaining the sensitivity of the differences between each communication parameter from each data source to the target parameter based on the historical data differences of the various communication parameters from multiple data sources in the same data collection task; and obtaining the influence strength index of each communication parameter from each data source to the target parameter based on the frequency of fluctuations exceeding the difference threshold in the historical data of each data source and the average of all sensitivities.

2. The consistency verification method for multi-source communication data according to claim 1, characterized in that, Before acquiring historical data of various communication parameters from multiple data sources, the method further includes: configuring spatial precision, timestamp precision, time synchronization error, data update rate, and location consistency score calculated based on historical data as various communication parameters of the data source in the spatiotemporal dimension; configuring model identification confidence, parameter measurement error, identification accuracy calculated based on historical data, and confusion matrix as various communication parameters of the data source in the attribute dimension; and configuring communication relationship detection capability, false alarm rate, false negative rate, and relationship persistence reporting indicators as various communication parameters of the data source in the data relationship dimension.

3. The consistency verification method for multi-source communication data according to claim 1, characterized in that, The step of obtaining the sensitivity of the differences between each communication parameter of each data source to the target parameter based on the historical data differences of the multiple data sources in the same data acquisition task includes: obtaining the cumulative summation of the absolute differences between the current data source and other data sources in the corresponding communication parameter in the same data acquisition task, based on the absolute value of the differences between the historical data of the current data source and other data sources in the same communication parameter; obtaining the difference results of each communication parameter of the current data source in the corresponding data acquisition task based on the ratio of the cumulative summation result of the current data source to the maximum summation result of all data sources; and obtaining the sensitivity of each data source based on the change of the target data in different data acquisition tasks and the difference results of each communication parameter of the current data source.

4. The consistency verification method for multi-source communication data according to claim 1, characterized in that, The step of performing influence weight analysis based on the differences between various influence intensity indices and the operational requirement values ​​of various communication parameters to obtain a weighted difference score for inconsistencies between data sources includes: obtaining a first weight coefficient for each communication parameter based on the operational requirement values ​​of each communication parameter and the maximum requirement value of all corresponding operations; obtaining a second weight coefficient for each communication parameter based on the summation of the differences in influence intensity indices of the same communication parameter across data sources; obtaining the influence weight of the corresponding communication parameter on the target parameter based on the first and second weight coefficients of the same communication parameter; and obtaining a weighted difference score for verifying the inconsistencies in the target parameter across data sources based on the summation of the influence weights of each communication parameter and the differences between various influence intensity indices.

5. The consistency verification method for multi-source communication data according to claim 1, characterized in that, Before obtaining the inconsistency score between the current data source and other data sources based on the weighted difference score and the unreliability of each data source in verifying the target parameter during historical periods, the method further includes: obtaining a first unreliability coefficient between the current data source and other data sources in the data anomaly dimension based on the abnormal frequency of the inconsistency performance of the current data source in verifying the target parameter during historical periods; obtaining a second unreliability coefficient between the current data source and other data sources based on the deviation degree of the inconsistency performance of the current data source in verifying the target parameter during historical periods; and obtaining the unreliability of each data source in verifying the target parameter during historical periods based on the first and second unreliability coefficients of each data source.

6. The consistency verification method for multi-source communication data according to claim 5, characterized in that, The step of obtaining the first unreliability coefficient of the current data source in the dimension of data anomaly between the current data source and other data sources based on the frequency of inconsistencies in the target parameters when the current data source verifies them in historical periods includes: obtaining anomaly indicators for corresponding communication parameters based on the differences in the data acquisition results of various communication parameters of the current data source in historical periods; statistically analyzing the anomaly indicators of various communication parameters that are greater than the indicator threshold to obtain the degree of anomaly of each data source; and obtaining the first unreliability coefficient of each data source based on the ratio of the degree of anomaly of each data source to the maximum value of all anomaly degrees.

7. The consistency verification method for multi-source communication data according to claim 5, characterized in that, The step of obtaining a second unreliability coefficient between the current data source and other data sources based on the degree of deviation of the current data source from inconsistent performance when verifying the target parameter in historical periods includes: obtaining a consistency difference of the current data source in historical periods based on the difference between the initial consistency degree and the final consistency degree of the current data source in historical periods; and obtaining a second unreliability coefficient of each data source based on the consistency difference of each data source and the average consistency degree of the corresponding data source in historical periods.

8. The consistency verification method for multi-source communication data according to claim 1, characterized in that, The step of obtaining the inconsistency score between the current data source and other data sources based on the weighted difference score and the unreliability of the target parameter verified by each data source in historical time periods includes: obtaining the weighted correction amount of each data source based on the unreliability of the target parameter verified by each data source in historical time periods and a preset coefficient; and obtaining the inconsistency score between the current data source and other data sources based on the weighted correction amount of each data source and the corresponding weighted difference score.

9. A consistency verification system for multi-source communication data, characterized in that, The verification system is the system corresponding to any one of the methods described in claims 1-8. The system includes: an acquisition module, used to acquire historical data of various communication parameters from multiple data sources, and target data for consistency verification of the target parameter against the historical data; a first acquisition module, used to analyze the fluctuation correlation characteristics between the historical data and the target data to obtain the influence intensity index of each communication parameter from each data source on the target parameter; a second acquisition module, used to perform influence weight analysis based on the differences between the influence intensity indices and the operational requirement values ​​of each communication parameter to obtain a weighted difference score of inconsistencies between data sources; a third acquisition module, used to obtain an inconsistency score between the current data source and other data sources based on the weighted difference score and the unreliability of each data source in verifying the target parameter during historical periods; and a verification marking module, used to mark the real-time data currently collected by the corresponding data source based on each inconsistency score when the target parameter performs a verification task.

Images