Vehicle network threat identification method and device, electronic equipment and vehicle

By obtaining vehicle network operation data and preset feature tables, and combining pre-constructed models to identify vehicle network threats, the problem of inefficient manual identification methods is solved, the identification accuracy and efficiency are improved, and users and driving safety are ensured.

CN119052798AInactive Publication Date: 2024-11-29WUHAN LOTUS CARS CO LTD
View PDF 9 Cites 0 Cited by

Patent Information

Application Number
CN202411251072.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-06
Publication Date
2024-11-29
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

In the prior art, the identification of vehicle network threats relies on manual methods, resulting in inaccuracy and inefficiency, inability to effectively identify and respond to network threats, affecting user information security and driving safety.

Method used

By obtaining the vehicle's network operation data and preset data feature tables, feature extraction is performed, and combined with the pre-built network threat identification model, network threats are identified to improve the accuracy and efficiency of identification.

Benefits of technology

It improves the accuracy and efficiency of identifying vehicle cyber threats, can more effectively identify and respond to cyber threats, and ensure user information and driving safety.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119052798A_ABST
    Figure CN119052798A_ABST
Patent Text Reader

Abstract

The embodiment of the invention provides a vehicle network threat identification method and device, electronic equipment and a vehicle. The method comprises the following steps: acquiring network operation data of a vehicle and a preset data feature table; the preset data feature table comprises a plurality of vehicle data features, and the vehicle data features are data features determined based on the network function of the vehicle; performing feature extraction processing on the network operation data according to a preset data feature table to obtain network data features of the vehicle; wherein the network data features correspond to vehicle data features included in a preset data feature table; according to the network data features and a pre-constructed network threat identification model set, determining an identification result of a network threat; wherein the pre-constructed network threat identification model set comprises at least one pre-constructed network threat identification model. The method is used for achieving the effect of improving the recognition accuracy and efficiency of the vehicle network threat.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and in particular to a method, device, electronic device and vehicle for identifying vehicle network threats. Background Art

[0002] The Internet of Vehicles mainly carries core businesses such as sensitive data interaction between the vehicle and the cloud, and vehicle operation control, and is vulnerable to network threats. Network threats can not only cause user information leakage, driving failures, and other problems, but can even affect user driving safety.

[0003] In the existing technology, basic network threat screening and analysis are mainly performed manually.

[0004] However, the manual approach will lead to problems of low accuracy and low efficiency. Summary of the invention

[0005] The embodiments of the present application provide a method, device, electronic device and vehicle for identifying vehicle network threats, so as to achieve the effect of improving the accuracy and efficiency of identifying vehicle network threats.

[0006] In a first aspect, an embodiment of the present application provides a method for identifying vehicle network threats, including:

[0007] Acquire network operation data of the vehicle and a preset data feature table; wherein the network operation data is network data generated during the operation of the vehicle; the preset data feature table includes a plurality of vehicle data features, and the vehicle data features are data features determined based on the network functions of the vehicle;

[0008] Performing feature extraction processing on the network operation data according to the preset data feature table to obtain network data features of the vehicle; wherein the network data features correspond to the vehicle data features included in the preset data feature table;

[0009] The network threat identification result is determined according to the network data characteristics and a pre-constructed network threat identification model set; wherein the pre-constructed network threat identification model set includes at least one pre-constructed network threat identification model.

[0010] In one possible implementation, the process of obtaining the preset data feature table includes: determining multiple network functions of the vehicle; obtaining network operation data corresponding to each of the multiple network functions of the vehicle; performing feature extraction processing on the network operation data corresponding to each network function to obtain vehicle data features corresponding to each network function; and summarizing the vehicle data features corresponding to each network function to obtain the preset data feature table.

[0011] In a possible implementation, feature extraction processing is performed on the network operation data according to the preset data feature table to obtain network data features of the vehicle, including: extracting network operation sub-data corresponding to each vehicle data feature included in the preset data feature table from the network operation data; determining a feature extraction method set corresponding to each vehicle data feature according to the network operation sub-data corresponding to each vehicle data feature; wherein the feature extraction method set includes at least one feature extraction method; and performing feature extraction processing on the corresponding network operation sub-data according to the determined feature extraction method set for each vehicle data feature to obtain the network data features.

[0012] In a possible implementation, if the pre-constructed network threat identification model set includes multiple pre-constructed network threat identification models, then the network threat identification result is determined according to the network data features and the pre-constructed network threat identification model set, including: inputting the network data features into each of the pre-constructed network threat identification models included in the pre-constructed network threat identification model set, and obtaining a sub-identification result of the network threat corresponding to each of the pre-constructed network threat identification models; obtaining a model weight set of the pre-constructed network threat identification model set; wherein the model weight set includes the weight value of each pre-constructed network threat identification model included in the pre-constructed network threat identification model set; according to the weight value of each pre-constructed network threat identification model included in the pre-constructed network threat identification model set included in the model weight set, the sub-identification result of the network threat corresponding to each of the pre-constructed network threat identification models is fused to obtain the network threat identification result.

[0013] In one possible implementation, the process of constructing the pre-constructed network threat identification model includes: obtaining historical network operation data, and constructing a model training set and a model test set based on the preset data feature table and the historical network operation data; based on the model training set and the model test set, performing model training processing and model testing processing on the initial network threat identification model to obtain the pre-constructed network threat identification model.

[0014] In a possible implementation, the historical network operation data includes first historical network operation sub-data and second historical network operation sub-data; wherein the first historical network operation sub-data indicates a first network threat set; the second historical network operation sub-data indicates a second network threat set; the first network threat set includes M network threat data; the second network threat set includes N network threat data among the M network threat data included in the first network threat set; M and N are positive integers, and M is greater than N; constructing a model training set and a model test set according to the preset data feature table and the historical network operation data, including: constructing the model training set according to the preset data feature table and the second historical network operation sub-data; constructing the model test set according to the preset data feature table and the first historical network operation sub-data.

[0015] In a possible implementation, before determining the identification result of the network threat based on the network data features and the pre-built network threat identification model set, the method further includes: performing dimensionality reduction processing on the network data features according to a preset feature screening method.

[0016] In a second aspect, an embodiment of the present application provides a vehicle network threat identification device, including:

[0017] An acquisition module, used to acquire network operation data of the vehicle and a preset data feature table; wherein the network operation data is network data generated during the operation of the vehicle; the preset data feature table includes a plurality of vehicle data features, and the vehicle data features are data features determined based on the network functions of the vehicle;

[0018] A feature extraction module, used to perform feature extraction processing on the network operation data according to the preset data feature table to obtain network data features of the vehicle; wherein the network data features correspond to the vehicle data features included in the preset data feature table;

[0019] An identification module is used to determine the identification result of the network threat according to the network data characteristics and a pre-constructed network threat identification model set; wherein the pre-constructed network threat identification model set includes at least one pre-constructed network threat identification model.

[0020] In a third aspect, an embodiment of the present application provides an electronic device, including: a memory, a processor;

[0021] The memory stores computer-executable instructions;

[0022] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementations of the first aspect.

[0023] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the first aspect above and / or various possible implementations of the first aspect.

[0024] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above first aspect and / or various possible implementation methods of the first aspect.

[0025] In a sixth aspect, an embodiment of the present application provides a vehicle, comprising an electronic device that implements the first aspect above and / or various possible implementations of the first aspect.

[0026] The vehicle network threat identification method, device, electronic device and vehicle provided in the embodiments of the present application obtain network operation data of the vehicle and a preset data feature table, perform feature extraction processing on the network operation data according to the preset data feature table to obtain network data features of the vehicle, and determine the identification result of the network threat according to the network data features and a pre-constructed network threat identification model set. Specifically, feature extraction processing on the network operation data according to the preset data feature table can improve the accuracy of the obtained network data features of the vehicle. On this basis, the obtained network data features are analyzed according to the pre-constructed network threat identification model set to improve the accuracy and efficiency of network threat identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0028] Figure 1 Schematic diagram of the process of identifying vehicle network threats provided by this application Figure 1 ;

[0029] Figure 2 Schematic diagram of the process of identifying vehicle network threats provided by this application Figure 2 ;

[0030] Figure 3 Schematic diagram of the process of identifying vehicle network threats provided by this application Figure 3 ;

[0031] Figure 4 Schematic diagram of the process of identifying vehicle network threats provided by this application Figure 4 ;

[0032] Figure 5 A schematic diagram of a process flow of a method for identifying vehicle network threats provided in this application;

[0033] Figure 6 A schematic diagram of the structure of a vehicle network threat identification device provided by the present application;

[0034] Figure 7 A schematic diagram of the structure of the electronic device provided in this application.

[0035] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0036] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0037] In the prior art, basic network threat screening and analysis are mainly performed manually, but the manual method will lead to problems of low accuracy and low efficiency.

[0038] The vehicle network threat identification method provided by the present application obtains the vehicle's network operation data and a preset data feature table, performs feature extraction processing on the network operation data according to the preset data feature table to obtain the vehicle's network data features, and determines the network threat identification result according to the network data features and a pre-constructed network threat identification model set, wherein the feature extraction processing on the network operation data according to the preset data feature table can improve the accuracy of the obtained vehicle network data features, and on this basis, the obtained network data features are analyzed according to the pre-constructed network threat identification model set to improve the accuracy and efficiency of network threat identification.

[0039] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0040] Figure 1 Schematic diagram of the process of identifying vehicle network threats provided by this application Figure 1 ,like Figure 1 As shown, the method includes:

[0041] Step S101: Acquire the network operation data of the vehicle and a preset data feature table.

[0042] Specifically, the network operation data of the vehicle and the preset data feature table can be obtained. Among them, the network operation data is the network data generated during the operation of the vehicle. Specifically, the network data generated during the operation of the vehicle is usually the data transmitted in the Internet of Vehicles when the vehicle is running, wherein the Internet of Vehicles usually includes an Internet of Vehicles terminal, a background system and a control terminal. The background system is used to connect the control terminal and the Internet of Vehicles terminal to realize the data and command transmission between the two. Optionally, the control terminal is usually a mobile phone with a specific app installed. By operating the app, control instructions can be sent to the Internet of Vehicles terminal through the background system. The Internet of Vehicles terminal is installed in the car and connected to the car control bus. It can obtain and control the state and behavior of the car and upload data or obtain instructions to the control terminal app through the background system. Among them, this application does not limit the process of obtaining the network operation data of the vehicle. Any process that can obtain the network operation data of the vehicle can be used as the process of obtaining the network operation data of the vehicle provided by this application. For example, based on the above description of the Internet of Vehicles, network operation data, that is, data transmitted in the Internet of Vehicles, includes: instructions sent by the mobile app to the background system, instructions sent by the background system to the gateway, instructions sent by the gateway to the Internet of Vehicles terminal, instructions sent by the Internet of Vehicles terminal to the vehicle control bus, data generated by the vehicle bus controlling the vehicle to execute instructions, and other data.

[0043] Optionally, after obtaining the network operation data of the vehicle, the obtained network operation data of the vehicle can be preprocessed, wherein the present application does not limit the preprocessing process. Optionally, the preprocessing process may include: one or more of: missing value processing process, outlier processing process, noise processing process, format unification processing process and the like. For example: the network operation data includes variable-rate CAN messages (CAN with Flexible Data rate, abbreviated as CANFD) and CAN messages (Controller Area Network, abbreviated as CAN), wherein the data formats of the variable-rate CAN messages and the CAN messages are different, for example: parameters such as data length, frame format, ID length, etc. are all different, but the same as the CAN message is that the variable-rate CAN FD message also has seven parts: frame start SOF, arbitration segment, control segment, data field, check field CRC, confirmation field ACK, and frame end. Therefore, after obtaining the variable-rate CAN message and the CAN message, format unification processing can be performed. Among them, the present application does not limit the method for performing format unification processing between variable-rate CAN messages and CAN messages. Any method that can generate variable-rate CAN messages and CAN messages with a unified format can be used as the method for performing format unification processing between variable-rate CAN messages and CAN messages provided in the present application.

[0044] Among them, the preset data feature table is a preset table for extracting features from the network operation data of the acquired vehicle. Among them, the preset data feature table includes multiple vehicle data features, and the vehicle data features are data features determined based on the network functions of the vehicle. Specifically, during the operation of the vehicle, the harm of network threats is mainly reflected in the impact on the realization of each network function. Therefore, in the process of identifying the network threats of the vehicle, data features related to the network functions of the vehicle can be extracted, thereby improving the accuracy of identifying network threats. Among them, when the configuration of the vehicle is different, the network functions that can be realized are also different, so the corresponding preset data feature tables are also different. Among them, the process of obtaining the preset data feature table itself is not limited, and any process that can obtain the preset data feature table can be used as the process of obtaining the preset data feature table provided in this application. Optionally, multiple network functions of the vehicle can be determined first; then the network operation data corresponding to each of the multiple network functions of the vehicle is obtained; then, the network operation data corresponding to each network function is subjected to feature extraction processing to obtain the vehicle data features corresponding to each network function; finally, the vehicle data features corresponding to each network function are summarized and processed to obtain the preset data feature table.

[0045] Step S102: performing feature extraction processing on the network operation data according to a preset data feature table to obtain network data features of the vehicle.

[0046] Specifically, according to the preset data feature table, the network operation data can be subjected to feature extraction processing to obtain the network data features of the vehicle. The network data features are data features extracted from the network operation data, and the network data features correspond to the vehicle data features included in the preset data feature table. Specifically, according to the above description of the preset data feature table, if the acquired network operation data is acquired when all network functions of the vehicle are turned on, the network data features extracted from the network operation data are in a one-to-one correspondence with the vehicle data features included in the preset data feature table; if the acquired network operation data is acquired when some network functions of the vehicle are turned on, the network data features extracted from the network operation data are in a one-to-one correspondence with some vehicle data features included in the preset data feature table.

[0047] Among them, the present application does not limit the process of performing feature extraction processing on network operation data according to a preset data feature table to obtain network data features of the vehicle. Any process that can perform feature extraction processing on network operation data according to a preset data feature table to obtain network data features of the vehicle can be used as the process provided by the present application of performing feature extraction processing on network operation data according to a preset data feature table to obtain network data features of the vehicle. Optionally, a set of feature extraction methods for each vehicle data feature included in a preset data feature table can be determined; wherein the feature extraction method set includes at least one feature extraction method; extracting network operation sub-data corresponding to each vehicle data feature included in the preset data feature table from the network operation data; and then performing feature extraction processing on the corresponding network operation sub-data according to the determined feature extraction method set for each vehicle data feature to obtain network data features.

[0048] Step S103: Determine the identification result of the network threat according to the network data characteristics and the pre-built network threat identification model set.

[0049] Specifically, according to the network data features obtained in step S102 and the set of pre-constructed network threat identification models, the identification result of the network threat can be determined. Among them, the set of pre-constructed network threat identification models includes at least one pre-constructed network threat identification model. The pre-constructed network threat identification model is a pre-constructed model for identifying network threats to vehicles. Among them, the present application does not limit the pre-constructed network threat identification model, and any pre-constructed model that can identify network threats to vehicles can be used as the pre-constructed network threat identification model provided by the present application. Optionally, the pre-constructed network threat identification model can be a pre-constructed network threat identification model based on Bayes, a pre-constructed network threat identification model based on a decision tree, a pre-constructed network threat identification model based on a neural network, a pre-constructed network threat identification model based on a support vector machine, a pre-constructed network threat identification model based on a random forest, and a pre-constructed network threat identification model based on a machine learning algorithm. Among them, the identification result of the network threat characterizes the probability of the occurrence of the network threat.

[0050] Among them, the present application does not limit the process of determining the identification result of the network threat based on the network data characteristics and the pre-constructed network threat identification model set, wherein the specific process of determining the identification result of the network threat based on the network data characteristics and the pre-constructed network threat identification model set is related to the number of constructed network threat identification models included in the pre-constructed network threat identification model set. Optionally, if the pre-constructed network threat identification model set includes a pre-constructed network threat identification model, then determining the identification result of the network threat based on the network data characteristics and the pre-constructed network threat identification model set may include: inputting the network data characteristics into a pre-constructed network threat identification model included in the pre-constructed network threat identification model set, and outputting the identification result of the network threat. Optionally, if the pre-constructed network threat identification model set includes multiple pre-constructed network threat identification models, determining the network threat identification result based on the network data features and the pre-constructed network threat identification model set may include: inputting the network data features into each pre-constructed network threat identification model included in the pre-constructed network threat identification model set, and obtaining the sub-identification result of the network threat corresponding to each pre-constructed network threat identification model; obtaining a model weight set of the pre-constructed network threat identification model set; wherein the model weight set includes the weight value of each pre-constructed network threat identification model included in the pre-constructed network threat identification model set; and according to the weight value of each pre-constructed network threat identification model included in the pre-constructed network threat identification model set included in the model weight set, fusing the sub-identification result of the network threat corresponding to each pre-constructed network threat identification model to obtain the network threat identification result.

[0051] The vehicle network threat identification method provided in the embodiment of the present application obtains the vehicle's network operation data and a preset data feature table, performs feature extraction processing on the network operation data according to the preset data feature table to obtain the vehicle's network data features, and determines the network threat identification result according to the network data features and a pre-constructed network threat identification model set. Specifically, feature extraction processing on the network operation data according to the preset data feature table can improve the accuracy of the obtained vehicle network data features. On this basis, the obtained network data features are analyzed according to the pre-constructed network threat identification model set to improve the accuracy and efficiency of network threat identification.

[0052] Figure 2 Schematic diagram of the process of identifying vehicle network threats provided by this application Figure 2 ,like Figure 2 As shown, in this embodiment Figure 1 Based on the illustrated embodiment, the process of obtaining a preset data feature table is described in detail. The method includes:

[0053] Step S201: Determine multiple network functions of the vehicle.

[0054] Specifically, multiple network functions of the vehicle can be determined. According to the description in step S101, when the configuration of the vehicle is different, the network functions that can be realized are also different. Therefore, the configuration information of the vehicle can be obtained first, and then the multiple network functions of the vehicle can be determined according to the configuration information of the vehicle. The present application does not limit the configuration information of the vehicle obtained. Any information that can be used to determine the network function of the vehicle can be used as the configuration information of the vehicle provided by the present application. Optionally, the configuration information of the vehicle includes but is not limited to the hardware configuration information of the vehicle and the network configuration information of the vehicle.

[0055] Among them, this application does not limit the network functions of the vehicle. Optionally, the network functions of the vehicle may include: user functions and supplier functions, wherein user functions refer to functions that can be realized by vehicle users through the vehicle network, and supplier functions refer to functions that can be realized by vehicle suppliers through the vehicle network. Optionally, user functions include but are not limited to: driving data collection, vehicle fault monitoring, vehicle condition report, remote unlocking of doors and windows, remote start and stop of the vehicle, remote air conditioning, remote seat adjustment, information sharing, communication, security and anti-theft functions, etc. Optionally, supplier functions include but are not limited to: remote car system upgrade, vehicle condition diagnosis, etc. Among them, Table 1 is an example table of network functions of the vehicle.

[0056] Table 1

[0057]

[0058]

[0059] Among them, the vehicle's functional types include vehicle management functions, digital key functions, vehicle monitoring functions and other functions.

[0060] Step S202: Acquire network operation data corresponding to each of the multiple network functions of the vehicle.

[0061] Specifically, after determining the multiple network functions of the vehicle, the network operation data corresponding to each of the multiple network functions of the vehicle can be obtained. Among them, the present application does not limit the process of obtaining the network operation data corresponding to each of the multiple network functions of the vehicle. Any process that can obtain the network operation data corresponding to each of the multiple network functions of the vehicle can be used as the process of obtaining the network operation data corresponding to each of the multiple network functions of the vehicle provided by the present application. Optionally, based on the description of the network operation data in step S101, the data transmission path corresponding to the network function can be determined according to the description of the network function, and then the corresponding data acquisition method can be determined according to the determined data transmission path, and then the corresponding network operation data can be obtained according to the corresponding data acquisition method.

[0062] Step S203: Perform feature extraction processing on the network operation data corresponding to each network function to obtain vehicle data features corresponding to each network function.

[0063] Specifically, feature extraction processing can be performed on the network operation data corresponding to each network function to obtain the vehicle data features corresponding to each network function. Among them, the present application does not limit the process of performing feature extraction processing on the network operation data corresponding to each network function to obtain the vehicle data features corresponding to each network function. Any process that can perform feature extraction processing on the network operation data corresponding to each network function to obtain the vehicle data features corresponding to each network function can be used as the process provided by the present application for performing feature extraction processing on the network operation data corresponding to each network function to obtain the vehicle data features corresponding to each network function. Optionally, the network threat scenario corresponding to each network function can be determined first, and then according to the network threat scenario corresponding to each network function, feature extraction can be performed on the corresponding network operation data to obtain the vehicle data features corresponding to each network function.

[0064] Step S204: Summarize and process the vehicle data features corresponding to each network function to obtain a preset data feature table.

[0065] Specifically, the vehicle data features corresponding to each network function obtained in step S203 are summarized and processed to obtain a preset data feature table. Among them, the present application does not limit the process of summarizing and processing the vehicle data features corresponding to each network function to obtain the preset data feature table. Optionally, the vehicle data features corresponding to each network function can be classified and summarized, for example: the vehicle data features corresponding to each network function are divided into Ethernet connection basic class features, Ethernet content connection class features, time-based network traffic statistics class features, CAN bus class features, vehicle control application message class features, driving conditions and management class features, etc. Among them, the present application does not limit the types of vehicle data features included in each type of data features, for example: basic features of Ethernet connection, which may include: connection duration features, protocol type features, network service type features of the target host, network service type features of the target host, byte count features of data from the source host to the target host, byte count features of data from the target host to the source host, number of error segmentation features, etc. For example: time-based network traffic statistics features, which may include: connection features with the same destination address as the current connection in the past 2 seconds, number of connections with SYN errors, number of connections with REJ errors, number of connections with the same service, number of connections with different services, number of connections with the same service as the current connection in the past 2 seconds, number of connections with SYN errors, number of connections with REJ errors, number of connections with different hosts, etc. For example: CAN bus features, which may include: message length detection features , message cycle detection features, message context consistency detection features, CAN data frame flooding attack features, CANID forgery, no identity authentication features, CAN data frame replay attack features, priority error features, broadcast error features, UDS service error statistics features and other features. For example, vehicle control application message features can include: CONNECT control message features, control header contains subject name and message identifier features, control message-QoS2 level protocol exchange features, control message-QoS2 level protocol exchange features, subscription control message features, client wants to subscribe to the topic features, session status usage features and other features. For example, driving conditions and management features can include: vehicle speed features, vehicle steering wheel angle features, vehicle heading angle change rate features, accelerator pedal opening features, brake pedal opening features, main driving door status (open, closed) features, ignition signal features, engine status features, remote control request features, remote control result features, vehicle lock status (unlocked, locked), air conditioning remote status features and other features.

[0066] The process of obtaining a preset data feature table provided in an embodiment of the present application determines multiple network functions of a vehicle, obtains network operation data corresponding to each of the multiple network functions of the vehicle, performs feature extraction processing on the network operation data corresponding to each network function to obtain vehicle data features corresponding to each network function, and summarizes the vehicle data features corresponding to each network function to obtain a preset data feature table, wherein, based on the multiple network functions of the vehicle, the accuracy of the obtained preset data feature table can be improved, and the accuracy of feature extraction based on the preset data feature table can be further improved, thereby improving the accuracy of vehicle network threats.

[0067] Figure 3 Schematic diagram of the process of identifying vehicle network threats provided by this application Figure 3 ,like Figure 3 As shown, in this embodiment Figure 1 or Figure 2 On the basis of the illustrated embodiment, a process of extracting features from network operation data according to a preset data feature table to obtain network data features of a vehicle is described in detail. The method includes:

[0068] Step S301: extracting network operation sub-data corresponding to each vehicle data feature included in a preset data feature table from network operation data.

[0069] Specifically, the network operation sub-data corresponding to each vehicle data feature included in the preset data feature table can be extracted from the network operation data. The description of the preset data feature table can refer to the above description and will not be repeated here. The network operation sub-data is the data in the network operation data used to extract the network data feature corresponding to the vehicle data feature.

[0070] The present application does not limit the process of extracting network operation sub-data corresponding to each vehicle data feature included in the preset data feature table from the network operation data. Any process that can extract network operation sub-data corresponding to each vehicle data feature included in the preset data feature table from the network operation data can be used as the process provided by the present application for extracting network operation sub-data corresponding to each vehicle data feature included in the preset data feature table from the network operation data. Figure 2 The description of the network operation data and vehicle data characteristics in the illustrated embodiment can obtain the mapping relationship between the preset vehicle data characteristics and the network operation sub-data, and then extract the network operation sub-data corresponding to each vehicle data feature included in the preset data feature table from the network operation data based on the mapping relationship between the preset vehicle data characteristics and the network operation sub-data.

[0071] Step S302: Determine a feature extraction method set corresponding to each vehicle data feature according to the network operation sub-data corresponding to each vehicle data feature.

[0072] Specifically, the feature extraction method set corresponding to each vehicle data feature can be determined according to the network operation sub-data corresponding to each vehicle data feature obtained in step S301. Among them, the feature extraction method set includes at least one feature extraction method. The feature extraction method is a method for extracting features from data. Among them, the present application does not limit the feature extraction methods included in the feature extraction method set, and any method that can extract features from data can be used as the feature extraction method provided by the present application. Optionally, the feature extraction method can be any one of the algorithms based on statistics, frequency-based algorithms, time window-based algorithms, deep learning-based algorithms, principal component analysis algorithms, linear discriminant analysis algorithms, and the like. Among them, the statistics-based algorithm is used to calculate the statistical characteristics of network connections, such as the mean, standard deviation, minimum, maximum, etc. These characteristics can help identify abnormal network connections; among them, the frequency-based algorithm is used to calculate the frequency of different types of network connections, such as the number of connections per second, the number of connections per IP address, etc. These frequency characteristics can be used to detect large-scale malicious behavior; among them, the time window-based algorithm is used to calculate the characteristics of network connections within a given time window. For example, the number of connections in the past 10 minutes can be calculated to discover abnormal activities in a short period of time; among them, algorithms such as algorithms based on deep learning are used to extract data features from complex network data. The deep learning-based algorithm is implemented through a deep neural network, and the deep neural network includes but is not limited to networks such as convolutional neural networks (CNN) and recurrent neural networks (RNN); among them, Principal Component Analysis (PCA for short), when random variables have a strong linear correlation between each other, they contain more common information. If the common information is extracted without losing too much information of the original variables, the purpose of simplifying the problem can be achieved. Its purpose is to reduce the number of features and turn multiple indicators into comprehensive indicators; among them, Linear Discriminant Analysis (Linear Discriminant Analysis) LDA, also known as Fisher Linear Discriminant Analysis (FLD), achieves the purpose of dimensionality reduction by projecting data to the direction with the best classification performance. Linear Discriminant Analysis is a supervised learning method.

[0073] Among them, the present application does not limit the process of determining the set of feature extraction methods corresponding to each vehicle data feature based on the network operation sub-data corresponding to each vehicle data feature. Any process that can determine the set of feature extraction methods corresponding to each vehicle data feature based on the network operation sub-data corresponding to each vehicle data feature can be used as the process provided by the present application of determining the set of feature extraction methods corresponding to each vehicle data feature based on the network operation sub-data corresponding to each vehicle data feature. Optionally, the characteristics of the network operation sub-data, the requirements for feature extraction, and the computing resources for feature extraction can be determined based on the network operation sub-data corresponding to each vehicle data feature, and then the set of feature extraction methods corresponding to each vehicle data feature can be determined based on the characteristics of the determined network operation sub-data, the requirements for feature extraction, and the computing resources for feature extraction.

[0074] Step S303: According to the determined feature extraction method set for each vehicle data feature, feature extraction processing is performed on the corresponding network operation sub-data to obtain network data features.

[0075] Specifically, according to the feature extraction method set of each vehicle data feature determined in step S202, feature extraction processing is performed on the corresponding network operation sub-data to obtain network data features. In this application, the process of performing feature extraction processing on the corresponding network operation sub-data to obtain network data features according to the feature extraction method set determined for each vehicle data feature is not limited. In this application, the process of performing feature extraction processing on the corresponding network operation sub-data to obtain network data features according to the feature extraction method set determined for each vehicle data feature is related to the type of network operation sub-data and the type of feature extraction method included in the feature extraction method set. When the type of network operation sub-data and / or the type of feature extraction method included in the feature extraction method set are different, the process of performing feature extraction processing on the corresponding network operation sub-data to obtain network data features according to the feature extraction method set determined for each vehicle data feature is also different.

[0076] The embodiment of the present application provides a process for performing feature extraction processing on network operation data according to a preset data feature table to obtain network data features of a vehicle, by extracting network operation sub-data corresponding to each vehicle data feature included in the preset data feature table from the network operation data, determining a feature extraction method set corresponding to each vehicle data feature according to the network operation sub-data corresponding to each vehicle data feature, and performing feature extraction processing on the corresponding network operation sub-data according to the determined feature extraction method set for each vehicle data feature to obtain network data features, wherein by extracting the network operation sub-data and determining the corresponding feature extraction method set, the network data features of the vehicle can be obtained more accurately, thereby further improving the accuracy of vehicle network threats.

[0077] In a possible embodiment, before determining the identification result of the network threat according to the network data features and the pre-built network threat identification model set, the vehicle network threat identification method also includes: according to the preset feature screening method, the network data features are subjected to dimensionality reduction processing. Among them, the preset feature screening method refers to the preset method for screening network data features. The present application does not limit the preset feature screening method. Any method that can screen network data features can be used as the feature screening method provided by the present application. Optionally, the feature screening method can be a filtering method. Specifically, the multiple features included in the network data features can be filtered according to the variance threshold. Among them, the variance threshold is a class that screens features by the variance of the feature itself. For example: if the variance of a feature itself is very small, it means that the sample has basically no difference in this feature. It is possible that most of the values ​​in the feature are the same, or even the values ​​of the entire feature are the same. Then, this feature has no effect on sample differentiation, so no matter what the next feature engineering is to do, it is necessary to give priority to eliminating features whose variance is less than or equal to the variance threshold. Specifically, if the feature screening method is the filtering method described above, the variance of each feature in the network data feature can be calculated first, and then the features whose corresponding variance is less than or equal to the preset variance threshold are filtered out, and the filtered network feature data is determined as the network feature data after dimensionality reduction processing. Among them, before determining the identification result of the network threat based on the network data features and the pre-built network threat identification model set, the network data features are subjected to dimensionality reduction processing, which can reduce the identification cost and improve the identification efficiency while ensuring the identification accuracy.

[0078] Figure 4 Schematic diagram of the process of identifying vehicle network threats provided by this application Figure 4 ,like Figure 4 As shown, in this embodiment Figure 1 or Figure 2 or Figure 3On the basis of the illustrated embodiment, if the pre-built network threat identification model set includes multiple pre-built network threat identification models, a process of determining the identification result of the network threat according to the network data characteristics and the pre-built network threat identification model set is described in detail, and the method includes:

[0079] Step S401: input network data features into each pre-constructed network threat identification model included in the pre-constructed network threat identification model set, and obtain a sub-identification result of the network threat corresponding to each pre-constructed network threat identification model.

[0080] Specifically, the network data features are respectively input into each pre-built network threat identification model included in the pre-built network threat identification model set, and a sub-identification result of the network threat corresponding to each pre-built network threat identification model can be obtained. The description of the pre-built network threat identification model can refer to the description in step S103, which will not be repeated here.

[0081] Step S402: Obtain a model weight set of a pre-built network threat identification model set.

[0082] Specifically, a model weight set of a pre-constructed network threat identification model set can be obtained, wherein the model weight set includes the weight value of each pre-constructed network threat identification model included in the pre-constructed network threat identification model set. The weight value of each pre-constructed network threat identification model can be obtained through the error parameter of each pre-constructed network threat identification model. For example, if the pre-constructed network threat identification model set includes two pre-constructed network threat identification models, such as a first pre-constructed network threat identification model and a second pre-constructed network threat identification model, wherein the error parameter of the first pre-constructed network threat identification model is ε1 and the error parameter of the second pre-constructed network threat identification model is ε2, then the weight value of the first pre-constructed network threat identification model is The weight of the second pre-built network threat identification model is

[0083] Step S403: According to the weight value of each pre-constructed network threat identification model included in the pre-constructed network threat identification model set included in the model weight set, the sub-identification results of the network threats corresponding to each pre-constructed network threat identification model are fused to obtain the identification result of the network threat.

[0084] Specifically, according to the weight value of each pre-constructed network threat identification model included in the pre-constructed network threat identification model set included in the model weight set obtained in step S402, the sub-identification results of the network threats corresponding to each pre-constructed network threat identification model obtained in step S401 are fused to obtain the identification result of the network threat.

[0085] For example, if the pre-constructed network threat identification model set includes two pre-constructed network threat identification models, such as a first pre-constructed network threat identification model and a second pre-constructed network threat identification model, and the weight of the first pre-constructed network threat identification model is α1, and the weight of the second pre-constructed network threat identification model is α2, the sub-identification result corresponding to the first pre-constructed network threat identification model is f 1t , the sub-identification result corresponding to the second pre-built network threat identification model is f 2t , then the identification result of network threat f t f t =α1f 1t +α2f 2t .

[0086] Optionally, if the pre-constructed network threat identification model set includes a pre-constructed network threat identification model, determining the network threat identification result according to the network data feature and the pre-constructed network threat identification model set includes: inputting the network data feature into a pre-constructed network threat identification model included in the pre-constructed network threat identification model set, and outputting the network threat identification result. For example: if the pre-constructed network threat identification model set includes a pre-constructed network threat identification model, such as the first pre-constructed network threat identification model, then the network threat identification result f t f t =f 1t For example, if the set of pre-built network threat identification models includes a pre-built network threat identification model, such as the second pre-built network threat identification model, then the identification result of the network threat f t f t =f 2t .

[0087] The embodiment of the present application provides a process for determining the identification result of a network threat based on network data features and a pre-constructed network threat identification model set, by respectively inputting the network data features into each pre-constructed network threat identification model included in the pre-constructed network threat identification model set, obtaining a sub-identification result of the network threat corresponding to each pre-constructed network threat identification model, obtaining a model weight set of the pre-constructed network threat identification model set, and fusing the sub-identification result of the network threat corresponding to each pre-constructed network threat identification model according to the weight value of each pre-constructed network threat identification model included in the pre-constructed network threat identification model set included in the model weight set to obtain the identification result of the network threat, wherein the process of obtaining the sub-identification result of the network threat based on the pre-constructed network threat identification model and fusing the sub-identification result of the network threat can improve the accuracy of the obtained identification result of the network threat.

[0088] Figure 5 Schematic diagram of the process of identifying vehicle network threats provided by this application Figure 5 ,like Figure 5 As shown, in this embodiment Figure 1 or Figure 2 or Figure 3 or Figure 4 Based on the illustrated embodiment, the process of constructing a pre-built network threat identification model is described in detail. The method includes:

[0089] Step S501: Obtain historical network operation data, and construct a model training set and a model test set according to a preset data feature table and the historical network operation data.

[0090] Specifically, historical network operation data can be obtained, and a model training set and a model test set can be constructed based on a preset data feature table and the historical network operation data. Specifically, feature extraction processing can be performed on the historical network operation data according to the preset data feature table to obtain multiple samples. The specific process can refer to the description above and will not be repeated here; then, label the multiple samples obtained according to the real network threat data corresponding to the historical network operation data; then, the multiple labeled samples obtained are divided according to a preset ratio to obtain a model training set and a model test set, wherein the model training set includes multiple training samples, and the model test set includes multiple test samples.

[0091] Optionally, the historical network operation data may include a first historical network operation sub-data and a second historical network operation sub-data. The first historical network operation sub-data indicates a first network threat set. The second historical network operation sub-data indicates a second network threat set. The first network threat set includes M network threat data. The second network threat set includes N network threat data among the M network threat data included in the first network threat set. M and N are positive integers, and M is greater than N. For example: the first network threat set includes 28 network threat data, and the second network threat set includes 25 network threat data. Optionally, the process of obtaining historical network operation data and constructing a model training set and a model test set according to a preset data feature table and the historical network operation data may also include: constructing a model training set according to a preset data feature table and the second historical network operation sub-data. Constructing a model test set according to a preset data feature table and the first historical network operation sub-data. Specifically, the feature extraction process can be performed on the second historical network operation data according to the preset data feature table to obtain multiple samples. The specific process can refer to the description above and will not be repeated here; then, according to the real network threat data corresponding to the second historical network operation data, the multiple samples obtained are labeled to obtain a model training set. Specifically, the feature extraction process can be performed on the first historical network operation data according to the preset data feature table to obtain multiple samples. The specific process can refer to the description above and will not be repeated here; then, according to the real network threat data corresponding to the first historical network operation data, the multiple samples obtained are labeled to obtain a model test set.

[0092] Among them, by performing model training based on a model training set obtained based on the second historical network operation sub-data, and then performing model testing based on a model test set obtained based on the first historical network operation sub-data, the generalization ability of the obtained pre-constructed network threat identification model can be improved, that is, the ability of the pre-constructed network threat identification model to identify network threats of unknown attack types can be improved.

[0093] Step S502: Based on the model training set and the model test set, the initial network threat identification model is subjected to model training processing and model test processing to obtain a pre-built network threat identification model.

[0094] Specifically, based on the model training set and the model test set obtained in step S501, the initial network threat identification model can be subjected to model training processing and model test processing to obtain a pre-constructed network threat identification model. Optionally, the initial network threat identification model can be subjected to model training processing based on the model training set to obtain a trained initial network threat identification model; and then the trained initial network threat identification model can be subjected to model test processing based on the model test set. If the trained initial network threat identification model passes the test, the trained initial network threat identification model is determined as the pre-constructed network threat identification model. Otherwise, the process of performing model training processing on the initial network threat identification model based on the model training set and performing model test processing on the trained initial network threat identification model based on the model test set is repeated until the trained initial network threat identification model passes the test.

[0095] Among them, the present application does not limit the process of model testing the trained initial network threat identification model based on the model test set. Optionally, in the process of model testing the trained initial network threat identification model based on the model test set, the performance parameter of the root mean square error of the trained initial network threat identification model can be calculated. If the calculated performance parameter of the root mean square error of the trained initial network threat identification model is less than a preset threshold, it is determined that the performance parameter of the root mean square error of the trained initial network threat identification model has passed the test. Otherwise, it is determined that the performance parameter of the root mean square error of the trained initial network threat identification model has not passed the test. Among them, the performance parameter ε of the root mean square error is RMSE The calculation formula is as follows:

[0096]

[0097] Among them, p(i) and are the actual value corresponding to the sample label and the predicted value of the initial network threat identification model after training; n is the number of test samples; i is the sequence number of the test sample.

[0098] Optionally, if the pre-built network identification model is a network threat identification model pre-built based on a decision tree. Then the initial network threat identification model can be an initial model built based on a CART regression tree model and an XGBoost model. Among them, the XGBoost algorithm generates a new tree to fit the residual of the previous tree through continuous iteration. As the number of iterations increases, the accuracy continues to improve. Therefore, the XGBoost algorithm can better fit the risk quantity data, thereby reducing the prediction error and achieving a higher prediction accuracy. Among them, the formula of the XGBoost model is as follows:

[0099]

[0100] in, is the predicted value, Π is the number of trees, f t is a function in the function space F, F is the set of all possible CARTs, x i is the i-th data input. Each iteration does not affect the model, that is, the original model remains unchanged, and a new function is added to the model. A function corresponds to a tree, and the newly generated tree fits the residual of the last prediction. The iterative process is as follows:

[0101]

[0102] in, is the predicted value after the tth superposition, where the objective function X of XGBoost is obj As shown below:

[0103]

[0104] in, Used to measure the gap between the predicted score and the actual score; is a regularization term. T is the number of leaf nodes; ω is the score of leaf nodes; γ is used to control the number of leaf nodes; λ ensures that the score of leaf nodes is not too large.

[0105] The process of constructing a pre-constructed network threat identification model provided by the embodiment of the present application obtains historical network operation data, and constructs a model training set and a model test set according to a preset data feature table and the historical network operation data, and performs model training processing and model test processing on the initial network threat identification model based on the model training set and the model test set to obtain a pre-constructed network threat identification model, wherein the model training set and the model test set constructed according to the preset data feature table can accurately characterize the characteristics of the historical network operation data, thereby improving the accuracy of the obtained pre-constructed network threat identification model, and further improving the accuracy of identifying network threats.

[0106] Figure 6 A schematic diagram of the structure of the vehicle network threat identification device provided by this application, such as Figure 6 As shown, the vehicle network threat identification device 60 provided in this embodiment includes:

[0107] The acquisition module 601 is used to acquire the network operation data of the vehicle and a preset data feature table; wherein the network operation data is the network data generated during the operation of the vehicle; the preset data feature table includes a plurality of vehicle data features, and the vehicle data features are data features determined based on the network function of the vehicle;

[0108] A feature extraction module 602 is used to perform feature extraction processing on the network operation data according to a preset data feature table to obtain network data features of the vehicle; wherein the network data features correspond to the vehicle data features included in the preset data feature table;

[0109] The identification module 603 is used to determine the identification result of the network threat according to the network data characteristics and the pre-constructed network threat identification model set; wherein the pre-constructed network threat identification model set includes at least one pre-constructed network threat identification model.

[0110] In a possible embodiment, the acquisition module 601 is specifically used to obtain a preset data feature table, including: determining multiple network functions of the vehicle; obtaining network operation data corresponding to each of the multiple network functions of the vehicle; performing feature extraction processing on the network operation data corresponding to each network function to obtain vehicle data features corresponding to each network function; and summarizing the vehicle data features corresponding to each network function to obtain a preset data feature table.

[0111] In a possible embodiment, the feature extraction module 602 is specifically used to perform feature extraction processing on the network operation data according to a preset data feature table to obtain the network data features of the vehicle, including: extracting the network operation sub-data corresponding to each vehicle data feature included in the preset data feature table from the network operation data; determining a feature extraction method set corresponding to each vehicle data feature according to the network operation sub-data corresponding to each vehicle data feature; wherein the feature extraction method set includes at least one feature extraction method; and performing feature extraction processing on the corresponding network operation sub-data according to the determined feature extraction method set for each vehicle data feature to obtain the network data features.

[0112] In a possible embodiment, the identification module 603 is specifically used to determine the identification result of the network threat according to the network data features and the preconstructed network threat identification model set if the preconstructed network threat identification model set includes multiple preconstructed network threat identification models, including: inputting the network data features into each preconstructed network threat identification model included in the preconstructed network threat identification model set, and obtaining the sub-identification result of the network threat corresponding to each preconstructed network threat identification model; obtaining the model weight set of the preconstructed network threat identification model set; wherein the model weight set includes the weight value of each preconstructed network threat identification model included in the preconstructed network threat identification model set; according to the weight value of each preconstructed network threat identification model included in the preconstructed network threat identification model set included in the model weight set, the sub-identification result of the network threat corresponding to each preconstructed network threat identification model is fused to obtain the identification result of the network threat.

[0113] In a possible embodiment, the identification module 603 is also specifically used to obtain historical network operation data, and construct a model training set and a model test set based on a preset data feature table and the historical network operation data; based on the model training set and the model test set, the initial network threat identification model is subjected to model training processing and model testing processing to obtain a pre-constructed network threat identification model.

[0114] In a possible embodiment, the historical network operation data includes first historical network operation sub-data and second historical network operation sub-data; wherein the first historical network operation sub-data indicates a first network threat set; the second historical network operation sub-data indicates a second network threat set; the first network threat set includes M network threat data; the second network threat set includes N network threat data among the M network threat data included in the first network threat set; M and N are positive integers, and M is greater than N; the identification module 603 is further specifically used to construct a model training set according to a preset data feature table and the second historical network operation sub-data; and to construct a model test set according to the preset data feature table and the first historical network operation sub-data.

[0115] In a possible embodiment, the feature extraction module 602 is further specifically configured to perform dimensionality reduction processing on network data features according to a preset feature screening method.

[0116] The vehicle network threat identification device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effects are similar, and this embodiment will not be described in detail here.

[0117] Figure 7 This is a schematic diagram of the structure of the electronic device provided in this application. Figure 7 As shown, the electronic device 70 provided in this embodiment includes: at least one processor 701 and a memory 702. Optionally, the device 70 further includes a communication component 703. The processor 701, the memory 702 and the communication component 703 are connected via a bus 704.

[0118] In a specific implementation process, at least one processor 701 executes the computer-executable instructions stored in the memory 702, so that at least one processor 701 executes the above method.

[0119] The specific implementation process of the processor 701 can be found in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.

[0120] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the invention may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0121] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (Non-volatile Memory, NVM), such as at least one disk memory.

[0122] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of the present application is not limited to only one bus or one type of bus.

[0123] The present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0124] The present application also provides a vehicle, comprising an electronic device for implementing the above method.

[0125] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above method is implemented.

[0126] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special-purpose computer.

[0127] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (Application Specific Integrated Circuits, referred to as: ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0128] The division of units is only a logical function division, and there may be other divisions in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0129] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0130] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0131] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0132] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0133] Finally, it should be noted that those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention, which follow the general principles of the present invention and include common knowledge or customary technical means in the art not disclosed by the present invention, are not limited to the precise structure described above and shown in the drawings, and may be modified and changed in various ways without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A method for identifying vehicle network threats, characterized in that: include: Acquire network operation data of the vehicle and a preset data feature table; wherein the network operation data is network data generated during the operation of the vehicle; the preset data feature table includes a plurality of vehicle data features, and the vehicle data features are data features determined based on the network functions of the vehicle; Performing feature extraction processing on the network operation data according to the preset data feature table to obtain network data features of the vehicle; wherein the network data features correspond to the vehicle data features included in the preset data feature table; The network threat identification result is determined according to the network data characteristics and a pre-constructed network threat identification model set; wherein the pre-constructed network threat identification model set includes at least one pre-constructed network threat identification model.

2. The method according to claim 1, characterized in that: The process of obtaining the preset data feature table includes: Identify multiple network functions of the vehicle; Acquire network operation data corresponding to each of a plurality of network functions of the vehicle; Perform feature extraction processing on the network operation data corresponding to each network function to obtain the vehicle data features corresponding to each network function; The vehicle data features corresponding to each network function are aggregated and processed to obtain the preset data feature table.

3. The method according to claim 1, characterized in that: The network operation data is subjected to feature extraction processing according to the preset data feature table to obtain the network data features of the vehicle, including: Extracting network operation sub-data corresponding to each vehicle data feature included in the preset data feature table from the network operation data; Determine a feature extraction method set corresponding to each vehicle data feature according to the network operation sub-data corresponding to each vehicle data feature; wherein the feature extraction method set includes at least one feature extraction method; According to the determined feature extraction method set for each vehicle data feature, feature extraction processing is performed on the corresponding network operation sub-data to obtain the network data feature.

4. The method according to claim 1, characterized in that: If the pre-constructed network threat identification model set includes a plurality of the pre-constructed network threat identification models, determining the identification result of the network threat according to the network data feature and the pre-constructed network threat identification model set includes: Inputting the network data features into each of the pre-constructed network threat identification models included in the pre-constructed network threat identification model set, respectively, to obtain a sub-identification result of the network threat corresponding to each of the pre-constructed network threat identification models; Obtaining a model weight set of the pre-constructed network threat identification model set; wherein the model weight set includes a weight value of each pre-constructed network threat identification model included in the pre-constructed network threat identification model set; According to the weight value of each pre-constructed network threat identification model included in the pre-constructed network threat identification model set included in the model weight set, the sub-identification results of the network threats corresponding to each of the pre-constructed network threat identification models are fused to obtain the identification result of the network threat.

5. The method according to claim 1, characterized in that The process of building the pre-built network threat identification model includes: Acquire historical network operation data, and construct a model training set and a model test set according to the preset data feature table and the historical network operation data; Based on the model training set and the model test set, model training processing and model testing processing are performed on the initial network threat identification model to obtain the pre-constructed network threat identification model.

6. The method according to claim 5, characterized in that The historical network operation data includes first historical network operation sub-data and second historical network operation sub-data; wherein the first historical network operation sub-data indicates a first network threat set; the second historical network operation sub-data indicates a second network threat set; the first network threat set includes M network threat data; the second network threat set includes N network threat data among the M network threat data included in the first network threat set; M and N are positive integers, and M is greater than N; According to the preset data feature table and the historical network operation data, a model training set and a model test set are constructed, including: Constructing the model training set according to the preset data feature table and the second historical network operation sub-data; The model test set is constructed according to the preset data feature table and the first historical network operation sub-data.

7. The method according to any one of claims 1 to 6, characterized in that Before determining the identification result of the network threat according to the network data characteristics and the pre-built network threat identification model set, the method further includes: According to a preset feature screening method, the network data features are subjected to dimensionality reduction processing.

8. A vehicle network threat identification device, characterized in that: include: An acquisition module, used to acquire network operation data of the vehicle and a preset data feature table; wherein the network operation data is network data generated during the operation of the vehicle; the preset data feature table includes a plurality of vehicle data features, and the vehicle data features are data features determined based on the network functions of the vehicle; A feature extraction module, used to perform feature extraction processing on the network operation data according to the preset data feature table to obtain network data features of the vehicle; wherein the network data features correspond to the vehicle data features included in the preset data feature table; An identification module is used to determine the identification result of the network threat according to the network data characteristics and a pre-constructed network threat identification model set; wherein the pre-constructed network threat identification model set includes at least one pre-constructed network threat identification model.

9. An electronic device, characterized in that: include: Memory, processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 7 when executed by a processor.

11. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

12. A vehicle, characterized in that: The invention comprises an electronic device for implementing the method according to any one of claims 1 to 7.

Citation Information

Patent Citations

  • Network application flow recognition method and apparatus and network application flow management apparatus

    CN101505276A

  • Network data feature processing method and device, equipment and storage medium

    CN113315721A

  • Network threat monitoring method and device, electronic equipment and readable storage medium

    CN114285617A

  • Internet of vehicles security situation prediction method, electronic equipment and readable storage medium

    CN116502078A

  • Vehicle-mounted network security threat detection method and system based on multi-feature fusion

    CN118054939A