Intelligent detection method for binding information of motor vehicle inspection and detection equipment software

Abstract

The invention relates to the field of intelligent detection, and provides a method for intelligently detecting software binding information of motor vehicle inspection and detection equipment, which comprises the following steps of: acquiring a time queue of behavior data of equipment to be detected, and performing embedded coding on the behavior data of the equipment to be detected by adopting an AI-based data analysis algorithm; and on the basis of each behavior data embedding coding feature, time sequence feature propagation aggregation representation based on the principal component is carried out to automatically judge whether the to-be-detected equipment has abnormal activity or not. Thus, abnormal activities of the equipment can be found in time by monitoring the behavior mode of the equipment in real time, the detection efficiency is improved, dependence on human resources is reduced, the reliability and accuracy of a detection result are improved, and therefore automatic detection of the behavior mode of the equipment is achieved.

Description

Technical Field

[0001] The present application relates to the field of intelligent detection, and more specifically, to a method for intelligent detection of software binding information of motor vehicle inspection and testing equipment. Background Art

[0002] With the rapid development of the automotive industry and the advancement of technology, modern motor vehicles are equipped with more and more electronic control systems and sensors. These devices are responsible for monitoring various parameters of the vehicle, such as engine status, braking system, tire pressure, etc., to ensure the safety and reliability of the vehicle. With the development of connected vehicles and autonomous driving technology, the number of sensors and control units inside the vehicle is increasing, which has led to a growing demand for efficient and accurate detection of these devices. During the vehicle detection process, it is very important to ensure the correct binding of all sensors and LAN devices. Incorrect binding will not only lead to inaccurate detection results, but may also bring safety hazards.

[0003] However, traditional methods often lack effective security mechanisms when detecting device software binding information, and usually rely on manual records and document verification, which is not only inefficient but also prone to human errors. In addition, due to the lack of a centrally managed standard database, device authentication becomes complicated and unreliable, increasing the risk of malicious attacks or device impersonation. In other words, there is no unified database to store and manage device registration information, making the verification process vulnerable to forged data or unauthorized access, which in turn leads to system security vulnerabilities and potential safety hazards.

[0004] Therefore, an optimized solution for detecting software binding information of motor vehicle inspection and testing equipment is desired. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present application provides a method for intelligently detecting software binding information of motor vehicle inspection and testing equipment.

[0006] According to one aspect of the present application, a method for intelligently detecting software binding information of a motor vehicle inspection and testing device is provided, which includes: Building a standard database, the standard database contains standard information of various sensors and local area network devices that have been registered and bound, the standard information includes device ID, model, binding status and binding time; Using a network scanning tool to detect the device to be detected in the current local area network, and identifying the information of the device to be detected; Verify whether the device to be detected is consistent with the registration information based on the comparison between the information of the device to be detected and the standard information in the standard database; In response to the device to be detected being consistent with the registration information, the behavior pattern of the device to be detected is analyzed, and it is confirmed whether the device to be detected has abnormal activities, including: obtaining a time queue of the behavior data of the device to be detected; embedding and encoding each behavior data in the time queue of the behavior data to obtain a time queue of behavior data embedded coding vectors; performing temporal principal component feature propagation aggregation on the time queue of the behavior data embedded coding vectors to obtain a semantic temporal principal component propagation aggregation representation of device behavior data; based on the semantic temporal principal component propagation aggregation representation of the device behavior data, generating a recognition result, and the recognition result is used to indicate whether the device to be detected has abnormal activities.

[0007] This application has significant technical effects due to the adoption of the above technical solutions: The intelligent detection method for software binding information of motor vehicle inspection and testing equipment provided by the present application obtains the time queue of the behavior data of the device to be detected, and adopts an AI-based data analysis algorithm to embed and encode the behavior data of the device to be detected, thereby automatically judging whether the device to be detected has abnormal activities based on the main component-based temporal feature propagation aggregation representation between the embedded coding features of each behavior data. In this way, by real-time monitoring of the behavior pattern of the device, abnormal activities of the device can be discovered in time, which not only improves the detection efficiency, but also reduces the dependence on human resources, improves the reliability and accuracy of the detection results, and thus realizes the automatic detection of the device behavior pattern. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0009] Figure 1 It is a flowchart of a method for intelligently detecting software binding information of motor vehicle inspection and testing equipment according to an embodiment of the present application.

[0010] Figure 2 A flowchart of a method for intelligently detecting software binding information of motor vehicle inspection and testing equipment according to an embodiment of the present application, in response to the device to be detected being consistent with the registration information, analyzing the behavior pattern of the device to be detected, and confirming whether there is abnormal activity in the device to be detected.

[0011] Figure 3In the intelligent detection method for software binding information of motor vehicle inspection and testing equipment according to the embodiment of the present application, in response to the device to be detected being consistent with the registration information, the behavior pattern of the device to be detected is analyzed, and a data flow diagram is provided to confirm whether there is abnormal activity in the device to be detected.

[0012] Figure 4 A flowchart of a method for intelligently detecting software binding information of motor vehicle inspection and testing equipment according to an embodiment of the present application, in which temporal principal component feature propagation aggregation is performed on the time queue of the behavior data embedded in the coding vector to obtain a semantic temporal principal component propagation aggregation representation of the device behavior data.

[0013] Figure 5 In the method for intelligent detection of software binding information of motor vehicle inspection and testing equipment according to an embodiment of the present application, a flowchart is provided for extracting a sequence of current behavior data embedded coding vectors and historical behavior data embedded coding vectors from a time queue of the behavior data embedded coding vectors, and then performing principal component extraction to obtain a sequence of principal component representation of current behavior data embedded coding features and principal component representation of historical behavior data embedded coding.

[0014] Figure 6 In the method for intelligently detecting software binding information of motor vehicle inspection and testing equipment according to an embodiment of the present application, a flowchart is provided for performing time-degradation dynamic aggregation on the sequence of the principal component representation of the embedded coded features of the current behavior data and the principal component representation of the embedded coded features of the historical behavior data to obtain the semantic time-series principal component propagation aggregation representation of the device behavior data. DETAILED DESCRIPTION

[0015] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0016] With the rapid development of the automotive industry and the continuous advancement of technology, modern motor vehicles are equipped with a large number of electronic control systems and sensors. These devices are responsible for monitoring key parameters of the vehicle, including but not limited to engine performance, brake system status, and tire pressure, to ensure the safety and performance reliability of the vehicle. With the rise of vehicle networking and autonomous driving technology, the number of sensors and control units inside vehicles has increased exponentially, and the demand for efficient and accurate detection of these devices has also increased.

[0017] When conducting vehicle inspections, it is critical to ensure accurate binding of all sensors and local area network devices. Improper binding will not only lead to distorted test results, but may also cause security issues. However, current traditional detection methods often lack effective security measures in software binding information detection, and mainly rely on manual records and document verification, which is not only inefficient but also prone to errors. In addition, the lack of a centralized, standardized database to manage the device's identity authentication information makes the authentication process complicated and unreliable, increasing the risk of malicious attacks or device misuse. In short, the lack of a unified database to store and manage device registration information makes the verification process vulnerable to forged data or unauthorized access, resulting in system security vulnerabilities and potential security risks.

[0018] Based on this, the present application proposes an intelligent detection method for software binding information of motor vehicle inspection and testing equipment. Figure 1 Flowchart of the intelligent detection method of software binding information of motor vehicle inspection and testing equipment according to the embodiment of the present application. Figure 1 As shown, according to the embodiment of the present application, the intelligent detection method for software binding information of motor vehicle inspection and testing equipment includes: S110, building a standard database, the standard database contains standard information of various types of registered and bound sensors and local area network devices, the standard information includes device ID, model, binding status and binding time; S120, using a network scanning tool to detect the device to be detected in the current local area network, and identify the information of the device to be detected; S130, based on the comparison between the information of the device to be detected and the standard information in the standard database, verifying whether the device to be detected is consistent with the registration information; S140, in response to the device to be detected being consistent with the registration information, analyzing the behavior pattern of the device to be detected, and confirming whether the device to be detected has abnormal activity.

[0019] In step S110, a standard database is constructed, and the standard database contains standard information of various registered and bound sensors and local area network devices, and the standard information includes device ID, model, binding status and binding time. It should be understood that by establishing a centrally managed standard database to store the standard information of all registered and bound sensors and local area network devices, benchmark data can be provided for subsequent device information comparison, ensuring the accuracy and consistency of device information. And by constructing a unified database, the security risks caused by data dispersion can be reduced.

[0020] In step S120, a network scanning tool is used to detect the devices to be detected in the current local area network, and the information of the devices to be detected is identified. It should be understood that by using a network scanning tool to scan the devices to be detected in the current local area network, some basic information of the devices to be detected can be identified and collected in real time. Compared with the traditional manual recording method, the automatic collection and recording of the data of the devices to be detected by the scanning tool can improve the collection efficiency of the device data and ensure the timeliness and accuracy of the information of the devices to be detected.

[0021] In step S130, based on the comparison between the information of the device to be detected and the standard information in the standard database, it is verified whether the device to be detected is consistent with the registration information. It should be understood that by comparing the information of the device to be detected with the information of the standard database one by one, it is verified whether the device to be detected is consistent with the registration information, thereby determining whether the device has been correctly bound.

[0022] In step S140, in response to the device to be detected being consistent with the registration information, the behavior pattern of the device to be detected is analyzed, and it is confirmed whether the device to be detected has abnormal activities. After confirming that the device to be detected is consistent with the registration information, its behavior pattern is further analyzed. That is, by carefully analyzing the behavior data of the device to be detected and identifying whether there are abnormal activities, potential safety hazards can be discovered early to ensure the safety and reliability of motor vehicle operation.

[0023] In general, the intelligent detection method for software binding information of motor vehicle inspection and testing equipment provided by this application can solve the problem of lack of a unified database in traditional methods, ensure the consistency and reliability of equipment information, and reduce the security risks caused by data dispersion. At the same time, it replaces traditional manual records and document verification, improves the efficiency and accuracy of verification, and reduces the possibility of human errors.

[0024] It should be understood that the safe operation of motor vehicles is directly related to the life safety of passengers. When the information of the equipment is consistent with the records in the standard database, it is necessary to further analyze the status and abnormal activities of the equipment. In other words, by detecting whether there is abnormal activity in the equipment, equipment failure or abnormal behavior can be discovered in time, thereby preventing traffic accidents caused by equipment problems. However, traditional equipment abnormal activity detection usually requires regular manual inspections, which is not only time-consuming and labor-intensive, but may also result in a long detection cycle and failure to detect problems in time. In addition, regular manual inspections are difficult to adapt to such dynamic changes, which may cause equipment problems to gradually accumulate during the detection cycle, ultimately leading to serious failures.

[0025] Accordingly, in response to the device to be detected being consistent with the registration information, analyzing the behavior pattern of the device to be detected, and confirming whether the device to be detected is in abnormal activity, the technical concept of the present application is to obtain the time queue of the behavior data of the device to be detected, and use AI-based data processing and analysis algorithms to embed and encode the behavior data of the device to be detected, so as to automatically determine whether the device to be detected has abnormal activity based on the principal component-based temporal feature propagation aggregation representation between the embedded coding features of each behavior data. In this way, by real-time monitoring of the behavior pattern of the device, abnormal activities of the device can be discovered in time, which not only improves the detection efficiency, but also reduces the dependence on human resources, improves the reliability and accuracy of the detection results, and thus realizes the automated detection of the device behavior pattern.

[0026] Figure 2 A flowchart of a method for intelligently detecting software binding information of motor vehicle inspection and testing equipment according to an embodiment of the present application, in response to the device to be detected being consistent with the registration information, analyzing the behavior pattern of the device to be detected, and confirming whether there is abnormal activity in the device to be detected. Figure 3 A data flow diagram of an intelligent detection method for software binding information of motor vehicle inspection and testing equipment according to an embodiment of the present application, in response to the device to be detected being consistent with the registration information, analyzing the behavior pattern of the device to be detected, and confirming whether the device to be detected has abnormal activities. Figure 2 and Figure 3 As shown, in response to the device to be detected being consistent with the registration information, the behavior pattern of the device to be detected is analyzed, and it is confirmed whether the device to be detected has abnormal activities, including: S141, obtaining the time queue of the behavior data of the device to be detected; S142, embedding and encoding each behavior data in the time queue of the behavior data to obtain a time queue of behavior data embedded coding vectors; S143, performing temporal principal component feature propagation aggregation on the time queue of the behavior data embedded coding vectors to obtain a semantic temporal principal component propagation aggregation representation of device behavior data; S144, generating a recognition result based on the semantic temporal principal component propagation aggregation representation of the device behavior data, and the recognition result is used to indicate whether the device to be detected has abnormal activities.

[0027] In step S141, the time queue of the behavior data of the device to be detected is obtained. It should be understood that the behavior data of the device to be detected generally includes but is not limited to the behavior type data (start, stop, accelerate, decelerate) of the device, the behavior parameter data (speed, temperature of each key part), etc. The time queue of the behavior data of the device to be detected can capture the dynamic changes of the device. Through the time queue data, the trend of the device behavior can be analyzed, and the difference between the normal state and the abnormal state of the device can be identified, so as to realize the detection of the abnormal state of the device.

[0028] In step S142, each behavior data in the time queue of the behavior data is embedded and encoded to obtain a time queue of behavior data embedded coding vectors. Specifically, in an embodiment of the present application, each behavior data in the time queue of the behavior data is embedded and encoded to obtain a time queue of behavior data embedded coding vectors, including: using a behavior data embedding coding matrix to embed coding each behavior data in the time queue of the behavior data to obtain a time queue of behavior data embedded coding vectors. It should be understood that each behavior data in the time queue of the behavior data contains key feature information of the behavior data at different time points. Based on this, in order to better capture and mine the complex relationships and patterns within the data, thereby extracting more meaningful features, in the technical solution of the present application, a behavior data embedding coding matrix is ​​used to embed coding each behavior data in the time queue of the behavior data to obtain a time queue of behavior data embedded coding vectors. In this way, the behavior patterns of the device at different time points can be better reflected, which helps to improve the accuracy of subsequent anomaly detection analysis.

[0029] In step S143, the time queue of the behavior data embedded coding vector is subjected to time series principal component feature propagation aggregation to obtain the device behavior data semantic time series principal component propagation aggregation representation. Accordingly, considering that each behavior data embedded coding vector in the time queue of the behavior data embedded coding vector represents the behavior feature information of a time period or a time point, and the behavior data will change with time. In order to capture these behavior data time series features, so as to better understand the dynamic behavior change trend of the device over time and capture more complex behavior data interaction patterns, in the technical solution of the present application, the time queue of the behavior data embedded coding vector is subjected to time series principal component feature propagation aggregation to obtain the device behavior data semantic time series principal component propagation aggregation representation vector as the device behavior data semantic time series principal component propagation aggregation representation. That is, the time series principal component feature propagation aggregation extracts the most representative principal component by performing principal component analysis (PCA) on the node feature vector at each time node. Then, it calculates the rate of change of the principal component between different time nodes and determines the decay entropy factor associated therewith. And through the modulation and gating mechanism of time span, key information is filtered to capture the dynamic evolution characteristics of node features in time series.

[0030] In detail, first, the current behavior data embedding coding vector (current node feature vector) and the sequence of historical behavior data embedding coding vectors (sequence of historical node feature vectors) are extracted from the time queue of the behavior data embedding coding vector, and principal component analysis is performed to obtain the current behavior data embedding coding feature principal component representation vector and the sequence of historical behavior data embedding coding principal component representation vector. In particular, principal component analysis (PCA) is a common dimensionality reduction technique that can help identify which features are most important for distinguishing normal behavior from abnormal behavior, that is, PCA can identify the direction that best represents the overall change trend in the data set, that is, the so-called principal components. These principal components usually contain most of the variance information in the data, so the information in the original data set can be approximated with a smaller number, and the high-dimensional data can be projected into a lower-dimensional space, providing a basis for subsequent feature propagation. Afterwards, in order to evaluate and quantify the trend of device behavior changes over time and its stability, the absolute factor of the temporal propagation attenuation entropy between each historical behavior data embedding coding principal component representation vector and the current behavior data embedding coding feature principal component representation vector is calculated to measure the degree of change of device behavior over time to obtain the sequence of the absolute factor of the temporal propagation attenuation entropy of the behavior data. Then, the sequence of absolute factors of the temporal propagation attenuation entropy of the behavior data is modulated in the time dimension based on the time span between each historical principal component representation and the current principal component representation. It should be understood that the introduction of the time span can help identify the long-term trend of device behavior. For example, if the behavior data of the device shows a trend of gradually deviating from the normal range over a period of time, even if the deviation is not large each time, this trend can be identified through the modulation of the time dimension and early warning can be given. Furthermore, in order to identify the most influential propagation path and assign corresponding weights accordingly, the modulated factors are input into the information transmission screening module based on the gating function to obtain the sequence of the behavior data temporal span modulation propagation attenuation weights. In particular, this module is similar to the gate mechanism in GRU or LSTM, which can help screen out the really important information. Through the gating mechanism, the weights can be dynamically adjusted according to the behavior data of the current device to determine which information should be retained and which should be ignored, so as to ensure that only those factors closely related to the current device state are used to calculate the weights. Then, the filtered weights are used to weighted aggregate the sequence of the historical behavior data embedded encoding principal component representation vectors to integrate the overall trend of short-term and long-term behavior data. Finally, the aggregated historical behavior data embedding encoding principal component significant transfer aggregation representation vector and the current behavior data embedding encoding feature principal component representation vector are added by position to fully utilize the long-term trends in historical data and the immediate changes in current data, provide a more complete deep relationship of device status, and obtain the semantic temporal principal component propagation aggregation representation vector of device behavior data.

[0031] Specifically, Figure 4 This is a flowchart of performing time series principal component feature propagation aggregation on the time queue of the behavior data embedded in the coding vector to obtain the semantic time series principal component propagation aggregation representation of the device behavior data in the intelligent detection method of the software binding information of the motor vehicle inspection and testing equipment according to the embodiment of the present application. Figure 4 As shown, the time queue of the behavior data embedded coding vector is subjected to temporal principal component feature propagation aggregation to obtain the device behavior data semantic temporal principal component propagation aggregation representation, including: S1431, extracting the sequence of the current behavior data embedded coding vector and the historical behavior data embedded coding vector from the time queue of the behavior data embedded coding vector, and then performing principal component extraction to obtain the sequence of the current behavior data embedded coding feature principal component representation and the historical behavior data embedded coding principal component representation; S1432, performing temporal attenuation dynamic aggregation on the sequence of the current behavior data embedded coding feature principal component representation and the historical behavior data embedded coding principal component representation to obtain the device behavior data semantic temporal principal component propagation aggregation representation.

[0032] More specifically, Figure 5 The present invention is a flowchart of the method for intelligently detecting software binding information of motor vehicle inspection and testing equipment according to the embodiment of the present application, in which the sequence of current behavior data embedding coding vector and historical behavior data embedding coding vector is extracted from the time queue of the behavior data embedding coding vector, and then the principal component is extracted to obtain the sequence of the principal component representation of the current behavior data embedding coding feature and the principal component representation of the historical behavior data embedding coding. Figure 5 As shown, after extracting the sequence of the current behavior data embedded coding vector and the historical behavior data embedded coding vector from the time queue of the behavior data embedded coding vector, principal component extraction is performed to obtain a sequence of the current behavior data embedded coding feature principal component representation and the historical behavior data embedded coding principal component representation, including: S14311, extracting the current behavior data embedded coding vector from the time queue of the behavior data embedded coding vector, and performing eigenvalue-based principal component extraction on the current behavior data embedded coding vector to obtain the current behavior data embedded coding feature principal component representation vector as the current behavior data embedded coding feature principal component representation; S14312, defining other behavior data embedded coding vectors in the time queue of the behavior data embedded coding vector as historical behavior data embedded coding vectors to obtain a sequence of historical behavior data embedded coding vectors, and performing eigenvalue-based principal component extraction on each historical behavior data embedded coding vector in the sequence of historical behavior data embedded coding vectors to obtain a sequence of historical behavior data embedded coding principal component representation vectors as the sequence of historical behavior data embedded coding principal component representation.

[0033] More specifically, Figure 6The flowchart of the method for intelligently detecting software binding information of motor vehicle inspection and testing equipment according to the embodiment of the present application is to perform time-series decay dynamic aggregation on the sequence of the current behavior data embedded coding feature principal component representation and the historical behavior data embedded coding principal component representation to obtain the device behavior data semantic time series principal component propagation aggregation representation. Figure 6 As shown, the sequence of the current behavior data embedded coding feature principal component representation and the historical behavior data embedded coding feature principal component representation is dynamically aggregated with time decay to obtain the semantic time series principal component propagation aggregate representation of the device behavior data, including: S14321, calculating the time series propagation attenuation entropy absolute factor of each historical behavior data embedded coding principal component representation vector in the sequence of the historical behavior data embedded coding principal component representation vector relative to the current behavior data embedded coding feature principal component representation vector to obtain a sequence of behavior data time series propagation attenuation entropy absolute factors; S14322, calculating the time series propagation attenuation entropy absolute factor of each behavior data in the sequence of the behavior data embedded coding principal component representation vector based on the time span between each historical behavior data embedded coding principal component representation vector in the sequence of the historical behavior data embedded coding principal component representation vector and the current behavior data embedded coding feature principal component representation vector. The absolute factor of the propagation attenuation entropy is modulated in the time dimension to obtain a sequence of behavior data timing span modulation propagation attenuation entropy factors; S14323, the sequence of behavior data timing span modulation propagation attenuation entropy factors is input into the information transmission screening module based on the gating function to obtain a sequence of behavior data timing span modulation propagation attenuation weights; S14324, based on the sequence of behavior data timing span modulation propagation attenuation weights, the weighted sum of the sequence of historical behavior data embedded coding principal component representation vectors is calculated to obtain the historical behavior data embedded coding principal component significant transfer aggregation representation vector; S14325, the positional sum of the historical behavior data embedded coding principal component significant transfer aggregation representation vector and the current behavior data embedded coding feature principal component representation vector is calculated to obtain the device behavior data semantic timing principal component propagation aggregation representation vector as the device behavior data semantic timing principal component propagation aggregation representation.

[0034] More specifically, in an embodiment of the present application, the absolute factors of the time-series propagation attenuation entropy of each historical behavior data embedded coding principal component representation vector in the sequence of the historical behavior data embedded coding principal component representation vector relative to the current behavior data embedded coding feature principal component representation vector are calculated to obtain a sequence of absolute factors of the time-series propagation attenuation entropy of the behavior data, including: calculating the positional division of the corresponding positions of the historical behavior data embedded coding principal component representation vector and the current behavior data embedded coding feature principal component representation vector to obtain the behavior data embedded coding feature principal component attenuation vector; calculating the logarithmic function value with base two of the absolute value of each eigenvalue of the behavior data embedded coding feature principal component attenuation vector to obtain the behavior data embedded coding feature principal component attenuation logarithmic vector; calculating the positional dot product of the historical behavior data embedded coding principal component representation vector and the behavior data embedded coding feature principal component attenuation logarithmic vector, and then performing positional multiplication on the obtained dot product vector to obtain the behavior data time-series propagation attenuation value; calculating an exponential function with the natural constant e as the base, which is an exponential of the behavior data time-series propagation attenuation value, to obtain the behavior data time-series propagation attenuation entropy absolute factors.

[0035] More specifically, in an embodiment of the present application, based on the time span between each historical behavior data embedded coding principal component representation vector in the sequence of the historical behavior data embedded coding principal component representation vector and the current behavior data embedded coding feature principal component representation vector, each behavior data temporal propagation attenuation entropy absolute factor in the sequence of the behavior data temporal propagation attenuation entropy absolute factors is modulated in the time dimension to obtain a sequence of behavior data temporal span modulated propagation attenuation entropy factors, including: subtracting the timestamp of the current behavior data embedded coding feature principal component representation vector from the timestamp of the historical behavior data embedded coding principal component representation vector and rounding down to obtain the behavior data time span value; calculating an exponential function with the natural constant e as the base and the behavior data time span value as the exponent to obtain the behavior data time span modulation value; dividing the behavior data temporal propagation attenuation entropy absolute factor corresponding to the historical behavior data embedded coding principal component representation vector by the behavior data time span modulation value to obtain the behavior data temporal span modulated propagation attenuation entropy factor.

[0036] More specifically, in an embodiment of the present application, the sequence of the behavior data timing span modulation propagation attenuation entropy factors is input into an information transmission screening module based on a gating function to obtain a sequence of the behavior data timing span modulation propagation attenuation weights, including: comparing each behavior data timing span modulation propagation attenuation entropy factor in the sequence of the behavior data timing span modulation propagation attenuation entropy factors with a predetermined threshold to obtain the sequence of the behavior data timing span modulation propagation attenuation weights; wherein, in response to the behavior data timing span modulation propagation attenuation entropy factor being greater than the predetermined threshold, the behavior data timing span modulation propagation attenuation entropy factor greater than the predetermined threshold is input into a sigmoid function; in response to the behavior data timing span modulation propagation attenuation entropy factor being less than or equal to the predetermined threshold, the behavior data timing span modulation propagation attenuation entropy factor less than or equal to the predetermined threshold is set to zero.

[0037] In the embodiment of the present application, specifically, the time queue of the behavior data embedded in the coding vector is subjected to time series principal component feature propagation aggregation, which can be expressed by the formula:

[0038] in, a temporal sequence for embedding encoding vectors for the behavioral data, are the first, second, ..., first in the time queue of the behavior data embedded in the encoding vector. ,..., Behavior data embedding encoding vector, Yes Perform principal component extraction based on eigenvalues. for The corresponding behavior data embedding encoding principal component representation vector, for The transposed vector of is the number of eigenvalues ​​in the behavior data embedding coding vector, yes The corresponding behavioral data covariance matrix, yes The corresponding principal component orthogonal matrix of behavioral data is, for The transposed matrix of is each behavioral data principal component vector in the sequence of behavioral data principal component vectors, for The corresponding diagonal matrix of behavioral data, The diagonal elements of the matrix are The behavioral data diagonal matrix, are the weight values ​​of the principal component vectors of each behavioral data, To return the maximum value value, is the maximum approximate matching value, embedding the behavioral data into a sequence of principal component representation vectors, are the first, second, ..., first, second, third, fourth, fifth, fifth, sixth, sixth, sixth, seventh ... ,..., The behavior data is embedded in the encoding principal component representation vector, for The eigenvalues ​​at each position in , for The eigenvalues ​​at each position in for The number of eigenvalues ​​in , represents the logarithmic function value with base 2, Indicated by natural constant The exponential function with base , for and The absolute factor of the decay entropy of the behavioral data time series propagation between and Respectively represent and The behavior data is embedded in the timestamp of the principal component representation vector, To round down, yes and The behavior data temporal span modulates the propagation attenuation entropy factor, For masking, yes function, is a predetermined threshold, is the timing span modulation propagation attenuation weight of each behavior data in the sequence of the timing span modulation propagation attenuation weight of the behavior data, is the number of vectors in the sequence of principal component representation vectors of the behavioral data embedding encoding, It is the semantic temporal principal component propagation aggregation representation vector of the device behavior data.

[0039] In step S144, a recognition result is generated based on the semantic temporal principal component propagation aggregation representation of the device behavior data, and the recognition result is used to indicate whether the device to be detected has abnormal activity. Specifically, in an embodiment of the present application, a recognition result is generated based on the semantic temporal principal component propagation aggregation representation of the device behavior data, including: inputting the semantic temporal principal component propagation aggregation representation vector of the device behavior data into an abnormal behavior identifier based on a classifier to obtain the recognition result. That is, the semantic temporal principal component propagation aggregation representation of the device behavior data obtained by performing temporal principal component aggregation using the time queue of the behavior data embedded in the coding vector is used for classification processing, so as to automatically determine whether the device to be detected has abnormal activity. In this way, by real-time monitoring of the behavior pattern of the device, abnormal activity of the device can be discovered in a timely manner, which not only improves the detection efficiency, but also reduces the dependence on human resources, improves the reliability and accuracy of the detection results, and realizes the automatic detection of device behavior patterns.

[0040] In the technical solution of the present application, each behavior data embedding coding vector in the time queue of the behavior data embedding coding vector respectively represents a low-dimensional semantic embedding coding vector of the behavior data of the device to be detected. When it is input into the feature temporal dynamic propagation network based on the node feature principal component gradient attenuation, the low-dimensional embedding representation of the behavior data embedding coding vector becomes more sparse after the feature principal component extraction, resulting in the feature manifold of the device behavior data semantic temporal principal component propagation aggregation representation vector obtained after the dynamic propagation of temporal gradient attenuation. There is a lack of propagation aggregation feature instance determination in the high-dimensional feature space, which affects the accuracy of the recognition result obtained by the classifier-based abnormal behavior identifier.

[0041] Based on this, in a preferred embodiment, the semantic temporal principal component propagation aggregation representation vector of the device behavior data is input into a classifier-based abnormal behavior identifier to obtain an identification result, including: Calculate the distance between each pair of eigenvalues ​​of the device behavior data semantic time series principal component propagation aggregation representation vector, such as the L2 distance, and take the square root of the distance to obtain the device behavior data semantic time series principal component propagation aggregation distance representation matrix, that is,

[0042] in, The semantic time series principal component propagation aggregation representation vector of the device behavior data is represented. and The first and Eigenvalues, The first and The distance between eigenvalues, Represents the semantic time series of device behavior data, principal component propagation, and aggregation distance representation matrix The eigenvalue of the position; Obtain the device behavior data semantic time series principal component propagation aggregation autocorrelation matrix of the device behavior data semantic time series principal component propagation aggregation representation vector as a row vector, that is, ,in, The semantic time series principal component propagation aggregation representation vector of the device behavior data is represented. represents the transpose of a vector, represents the matrix multiplication operation, Represents the semantic temporal principal component propagation aggregated autocorrelation matrix of device behavior data; The device behavior data semantic time series principal component propagation aggregation representation vector is matrix-multiplied with the device behavior data semantic time series principal component propagation aggregation distance representation matrix to obtain the device behavior data semantic time series principal component propagation aggregation first-level mapping vector, that is, ,in, The semantic time series principal component propagation aggregation representation vector of the device behavior data is represented. Represents the device behavior data semantic time series principal component propagation aggregation distance representation matrix, represents the matrix multiplication operation, Represents the semantic temporal principal component propagation aggregation first-level mapping vector of device behavior data; The device behavior data semantic time series principal component propagation aggregation first-level mapping vector is matrix-multiplied with the device behavior data semantic time series principal component propagation aggregation distance representation matrix and the device behavior data semantic time series principal component propagation aggregation self-correlation matrix to obtain the device behavior data semantic time series principal component propagation aggregation multi-level mapping vector ,in, Represents the semantic time series of device behavior data, principal component propagation aggregation, and first-level mapping vector, Represents the device behavior data semantic time series principal component propagation aggregation distance representation matrix, represents the matrix multiplication operation, Represents the semantic temporal principal component propagation aggregation autocorrelation matrix of device behavior data, Represents the semantic temporal principal component propagation aggregation multi-level mapping vector of device behavior data; The device behavior data semantic time series principal component propagation aggregation multi-level mapping vector is interpolated with the device behavior data semantic time series principal component propagation aggregation associated eigenvector composed of the eigenvalues ​​of the device behavior data semantic time series principal component propagation aggregation auto-association matrix to obtain an optimized device behavior data semantic time series principal component propagation aggregation representation vector, wherein interpolation or zero padding is performed when the eigenvalue is insufficient; The optimized device behavior data semantic temporal principal component propagation aggregation representation vector is input into a classifier-based abnormal behavior identifier to obtain an identification result.

[0043] Therefore, through the linear target mapping representation based on the similarity distance representation matrix of the semantic temporal principal component propagation aggregation representation vector of the device behavior data, the self-correlation complete similarity instantiation of the semantic temporal principal component propagation aggregation representation vector of the device behavior data is performed with a secondary target mapping representation based on a multi-level distribution hierarchy, and the association mismatch negative influence factor is compensated by the association fusion kernel bias to improve the degree of instance determination of the eigenvalue of the semantic temporal principal component propagation aggregation representation vector of the device behavior data under the similarity constraint, that is, the degree of significance of the eigenvalue as an instance for classification regression determination, so as to improve the accuracy of the recognition result obtained by the classifier-based abnormal behavior identifier of the semantic temporal principal component propagation aggregation representation vector of the device behavior data.

[0044] In summary, the feature manifold optimization process of the semantic temporal principal component propagation aggregation representation vector of the device behavior data is explained, which extracts the best prompt sample based on the fine-grained target retrieval of random perturbations, and then guides adversarial robustness learning based on the best prompt sample to complete the structure of the high-dimensional manifold of the feature vector in the feature space to improve the detail and structure of the feature expression. In this way, by monitoring the behavior pattern of the device in real time, abnormal activities of the device can be discovered in time, which not only improves the detection efficiency, but also reduces the dependence on human resources, improves the reliability and accuracy of the detection results, and realizes the automatic detection of the device behavior pattern.

[0045] In summary, the intelligent detection method for software binding information of motor vehicle inspection and testing equipment based on the embodiment of the present application is explained, which obtains the time queue of the behavior data of the device to be detected, and adopts an AI-based data analysis algorithm to embed and encode the behavior data of the device to be detected, so as to automatically judge whether there is abnormal activity in the device to be detected based on the main component-based temporal feature propagation aggregation representation between the embedded coding features of each behavior data. In this way, by real-time monitoring of the behavior pattern of the device, abnormal activities of the device can be discovered in time, which not only improves the detection efficiency, but also reduces the dependence on human resources, improves the reliability and accuracy of the detection results, and thus realizes the automatic detection of the device behavior pattern.

[0046] The basic principles of the present application are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present application are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present application. In addition, the specific details of the above application are only for the purpose of illustration and ease of understanding, rather than limitation, and the above details do not limit the present application to the use of the above specific details to be implemented.

[0047] In this application, words such as "including", "comprising", "having", etc. are open words, meaning "including but not limited to", and can be used interchangeably therewith. The words "or" and "and" used herein mean the words "and / or" and can be used interchangeably therewith, unless the context clearly indicates otherwise. The word "such as" used herein means the phrase "such as but not limited to", and can be used interchangeably therewith.

[0048] It should also be noted that in the method of the present application, each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent solutions of the present application.

Claims

1. A method for intelligently detecting software binding information of motor vehicle inspection and testing equipment, characterized in that: include: Building a standard database, the standard database contains standard information of various sensors and local area network devices that have been registered and bound, the standard information includes device ID, model, binding status and binding time; Using a network scanning tool to detect the device to be detected in the current local area network, and identifying the information of the device to be detected; Verify whether the device to be detected is consistent with the registration information based on the comparison between the information of the device to be detected and the standard information in the standard database; In response to the device to be detected being consistent with the registration information, analyzing the behavior pattern of the device to be detected, and confirming whether the device to be detected has abnormal activities, including: obtaining a time queue of the behavior data of the device to be detected; Each behavior data in the time queue of the behavior data is embedded and encoded to obtain a time queue of behavior data embedded coding vectors; the time queue of the behavior data embedded coding vectors is subjected to time series principal component feature propagation aggregation to obtain a semantic time series principal component propagation aggregation representation of device behavior data; based on the semantic time series principal component propagation aggregation representation of device behavior data, an identification result is generated, and the identification result is used to indicate whether the device to be detected has abnormal activity.

2. The intelligent detection method for software binding information of motor vehicle inspection and testing equipment according to claim 1 is characterized in that: Embedding and encoding each behavior data in the time queue of the behavior data to obtain a time queue of the behavior data embedded coding vector includes: using a behavior data embedding coding matrix to embed and encode each behavior data in the time queue of the behavior data to obtain a time queue of the behavior data embedded coding vector.

3. The intelligent detection method for software binding information of motor vehicle inspection and testing equipment according to claim 2 is characterized in that: Performing time series principal component feature propagation aggregation on the time queue of the behavior data embedding encoding vector to obtain a semantic time series principal component propagation aggregation representation of the device behavior data, including: Extracting the sequence of current behavior data embedding coding vector and historical behavior data embedding coding vector from the time queue of the behavior data embedding coding vector, and then performing principal component extraction to obtain a sequence of current behavior data embedding coding feature principal component representation and historical behavior data embedding coding principal component representation; The sequences of the embedded coding feature principal component representation of the current behavior data and the embedded coding principal component representation of the historical behavior data are dynamically aggregated with time decay to obtain the semantic time series principal component propagation aggregation representation of the device behavior data.

4. The intelligent detection method for software binding information of motor vehicle inspection and testing equipment according to claim 3 is characterized in that: Extracting the sequence of the current behavior data embedding coding vector and the historical behavior data embedding coding vector from the time queue of the behavior data embedding coding vector and then performing principal component extraction to obtain a sequence of the current behavior data embedding coding feature principal component representation and the historical behavior data embedding coding principal component representation, including: Extracting a current behavior data embedding coding vector from the time queue of the behavior data embedding coding vector, and performing principal component extraction based on eigenvalues ​​on the current behavior data embedding coding vector to obtain a current behavior data embedding coding feature principal component representation vector as the current behavior data embedding coding feature principal component representation; Define other behavior data embedding coding vectors in the time queue of the behavior data embedding coding vector as historical behavior data embedding coding vectors to obtain a sequence of historical behavior data embedding coding vectors, and perform eigenvalue-based principal component extraction on each historical behavior data embedding coding vector in the sequence of historical behavior data embedding coding vectors to obtain a sequence of historical behavior data embedding coding principal component representation vectors as the sequence of historical behavior data embedding coding principal component representations.

5. The intelligent detection method for software binding information of motor vehicle inspection and testing equipment according to claim 4 is characterized in that: Performing time-decay dynamic aggregation on the sequence of the current behavior data embedded coding feature principal component representation and the historical behavior data embedded coding principal component representation to obtain the device behavior data semantic time-series principal component propagation aggregation representation, including: Calculate the absolute factor of the temporal propagation attenuation entropy of each historical behavior data embedded coding principal component representation vector in the sequence of the historical behavior data embedded coding principal component representation vector relative to the current behavior data embedded coding feature principal component representation vector to obtain a sequence of the absolute factors of the temporal propagation attenuation entropy of the behavior data; Based on the time span between each historical behavior data embedding coding principal component representation vector in the sequence of historical behavior data embedding coding principal component representation vectors and the current behavior data embedding coding feature principal component representation vector, each behavior data temporal propagation attenuation entropy absolute factor in the sequence of behavior data temporal propagation attenuation entropy absolute factors is modulated in time dimension to obtain a sequence of behavior data temporal span modulated propagation attenuation entropy factors; Inputting the sequence of behavior data time series span modulation propagation attenuation entropy factors into the information transfer screening module based on the gating function to obtain a sequence of behavior data time series span modulation propagation attenuation weights; Based on the sequence of the behavior data temporal span modulation propagation attenuation weights, calculating the weighted sum of the sequence of the historical behavior data embedded coding principal component representation vectors to obtain the historical behavior data embedded coding principal component significant transfer aggregation representation vector; Calculate the positional sum of the historical behavior data embedded coding principal component significant transfer aggregation representation vector and the current behavior data embedded coding feature principal component representation vector to obtain the device behavior data semantic temporal principal component propagation aggregation representation vector as the device behavior data semantic temporal principal component propagation aggregation representation.

6. The intelligent detection method for software binding information of motor vehicle inspection and testing equipment according to claim 5 is characterized in that: Calculating the absolute factor of the temporal propagation attenuation entropy of each historical behavior data embedding encoding principal component representation vector in the sequence of the historical behavior data embedding encoding principal component representation vector relative to the current behavior data embedding encoding feature principal component representation vector to obtain a sequence of absolute factors of the temporal propagation attenuation entropy of the behavior data, including: Calculate the positional division of the historical behavior data embedded coding principal component representation vector and the current behavior data embedded coding feature principal component representation vector at corresponding positions to obtain the behavior data embedded coding feature principal component attenuation vector; Calculate the logarithmic function value with base 2 of the absolute value of each eigenvalue of the principal component attenuation vector of the behavior data embedding coding feature to obtain the principal component attenuation logarithmic vector of the behavior data embedding coding feature; After calculating the positional point product of the historical behavior data embedded coding principal component representation vector and the behavior data embedded coding feature principal component attenuation logarithm vector, the obtained point product vector is multiplied by position point to obtain the behavior data time series propagation attenuation value; An exponential function of the behavioral data time series propagation attenuation value is calculated with the natural constant e as the base to obtain the behavioral data time series propagation attenuation entropy absolute factor.

7. The method for intelligently detecting software binding information of motor vehicle inspection and testing equipment according to claim 6 is characterized in that: Based on the time span between each historical behavior data embedding coding principal component representation vector in the sequence of historical behavior data embedding coding principal component representation vectors and the current behavior data embedding coding feature principal component representation vector, each behavior data temporal propagation attenuation entropy absolute factor in the sequence of behavior data temporal propagation attenuation entropy absolute factors is modulated in time dimension to obtain a sequence of behavior data temporal span modulated propagation attenuation entropy factors, including: Subtract the timestamp of the principal component representation vector of the current behavior data embedded in the coding feature from the timestamp of the principal component representation vector of the historical behavior data embedded in the coding feature and round down to obtain a behavior data time span value; Calculating an exponential function with the natural constant e as the base and the behavior data time span value as the exponent to obtain the behavior data time span modulation value; The behavior data temporal propagation attenuation entropy absolute factor corresponding to the historical behavior data embedding encoding principal component representation vector is divided by the behavior data time span modulation value to obtain the behavior data temporal span modulation propagation attenuation entropy factor.

8. The intelligent detection method for software binding information of motor vehicle inspection and testing equipment according to claim 7 is characterized in that: Inputting the sequence of behavior data time series span modulation propagation attenuation entropy factors into the information transfer screening module based on the gating function to obtain a sequence of behavior data time series span modulation propagation attenuation weights, including: Comparing each behavior data timing span modulation propagation attenuation entropy factor in the sequence of behavior data timing span modulation propagation attenuation entropy factors with a predetermined threshold value to obtain a sequence of behavior data timing span modulation propagation attenuation weights; Among them, in response to the behavior data timing span modulation propagation attenuation entropy factor being greater than the predetermined threshold, the behavior data timing span modulation propagation attenuation entropy factor greater than the predetermined threshold is input into the sigmoid function; in response to the behavior data timing span modulation propagation attenuation entropy factor being less than or equal to the predetermined threshold, the behavior data timing span modulation propagation attenuation entropy factor less than or equal to the predetermined threshold is set to zero.

9. The method for intelligently detecting software binding information of motor vehicle inspection and testing equipment according to claim 8 is characterized in that: Based on the semantic temporal principal component propagation aggregation representation of the device behavior data, generating a recognition result includes: inputting the semantic temporal principal component propagation aggregation representation vector of the device behavior data into a classifier-based abnormal behavior identifier to obtain the recognition result.

Images