Vehicle-mounted network transmission monitoring method and device, terminal, vehicle and storage medium

Abstract

The invention provides a vehicle-mounted network transmission monitoring method and device, a terminal, a vehicle and a storage medium. The method comprises the following steps: acquiring a vehicle-mounted Ethernet data sample; inputting the data sample into a to-be-trained feature dimension reduction model, and generating an initial low-dimensional feature through an encoder; performing recursive optimization on the initial low-dimensional features through a feature adjustment module, adjusting the weight of each initial low-dimensional feature, and generating a target low-dimensional feature; re-mapping the target low-dimensional feature back to the high-dimensional space through a decoder to obtain target vehicle-mounted Ethernet data; performing iterative training on the feature dimension reduction model to obtain a target feature dimension reduction model; and performing feature dimension reduction on the vehicle-mounted Ethernet data of the target vehicle through the target feature dimension reduction model, and classifying the vehicle-mounted Ethernet data after dimension reduction to obtain a vehicle-mounted network transmission state result. According to the invention, efficient management and rapid feedback of the transmission state of the vehicle-mounted network are realized, and the stability of data transmission and the safety of vehicle operation are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle network transmission monitoring, and in particular to a vehicle-mounted network transmission monitoring method, device, terminal, vehicle and storage medium. Background Art

[0002] With the development of automobile intelligence and networking, in-vehicle Ethernet, as a key technology for internal vehicle communication, faces the challenges of surging data volume and increasing real-time requirements. Traditional in-vehicle network data management systems lack efficient data processing and analysis capabilities, making it difficult to cope with the growing data volume and real-time requirements, resulting in the inability to monitor vehicle status in a timely and accurate manner, affecting vehicle safety and reliability. In addition, existing technologies have limitations in data collection, feature extraction, dimensionality reduction processing, and state feedback, and are unable to effectively process and analyze high-dimensional data, resulting in the loss of key information and low data processing efficiency.

[0003] Therefore, the existing technology has deficiencies in system stability and reliability in the management of in-vehicle Ethernet data transmission, and is unable to effectively process and analyze high-dimensional data. These problems lead to incomplete vehicle status monitoring, insufficiently fast and accurate data processing, possible neglect of key information, insufficiently timely response to communication problems and hardware failures, and the system is difficult to adapt to complex and changeable operating environments, thus affecting the stability of in-vehicle network data transmission and the level of vehicle intelligence. Summary of the invention

[0004] The main purpose of the present invention is to provide a vehicle-mounted network transmission monitoring method, device, terminal, vehicle and storage medium, aiming to solve the technical problems that the prior art has insufficient system stability and reliability in vehicle-mounted Ethernet data transmission management, and cannot effectively process and analyze high-dimensional data.

[0005] In a first aspect, the present invention provides a vehicle network transmission monitoring method, comprising:

[0006] Get the vehicle Ethernet data sample;

[0007] Inputting the vehicle Ethernet data sample into a feature dimension reduction model to be trained, wherein the feature dimension reduction model includes an encoder, a decoder and a feature adjustment module, and mapping the vehicle Ethernet data sample to a low-dimensional feature space through the encoder to generate an initial low-dimensional feature;

[0008] Recursively optimizing the initial low-dimensional features through the feature adjustment module, adjusting the weights of each of the initial low-dimensional features, and generating target low-dimensional features;

[0009] Re-mapping the target low-dimensional features back to a high-dimensional space through the decoder to obtain target vehicle Ethernet data;

[0010] Iteratively training the feature dimensionality reduction model according to a preset stop iteration condition to obtain a target feature dimensionality reduction model;

[0011] The target feature dimension reduction model is used to perform feature dimension reduction on the collected vehicle Ethernet data of the target vehicle, and the reduced vehicle Ethernet data is classified by the classifier model to obtain the vehicle network transmission status result of the target vehicle.

[0012] In a specific embodiment, the calculation formula of the initial low-dimensional feature is expressed as:

[0013] Z r =Sig enc (W r X r +b r )

[0014] In the formula, Z r represents the initial low-dimensional features, W r is the weight matrix of the encoder, X r is the vehicle Ethernet data sample input to the encoder, b r is the bias vector of the encoder, Sig enc () is the multi-layer Sigmoid activation function of the encoder.

[0015] In a specific embodiment, recursively optimizing the initial low-dimensional features through the feature adjustment module, adjusting the weights of each of the initial low-dimensional features, and generating target low-dimensional features includes:

[0016] Generating, by the feature adjustment module, initial weights of the initial low-dimensional features according to the feature importance of the initial low-dimensional features, and generating adjusted low-dimensional features according to the initial weights;

[0017] The adjusted low-dimensional features are recursively optimized through the feature adjustment module, so that in each round of iteration, the weights of the adjusted low-dimensional features are adjusted according to the performance of the adjusted low-dimensional features in the previous round, and the target low-dimensional features are generated.

[0018] In a specific embodiment, the calculation method of the matrix of the initial weights is expressed as:

[0019] A r =diag(α r )

[0020] In the formula, A r is the initial weight matrix; α r is the feature weight vector, α r Each element α inr,i Initialized to the same value, indicating that all the initial low-dimensional features have the same importance in the initial stage; diag() is a function for extracting the diagonal elements of the initial weight matrix;

[0021] The calculation method of adjusting the low-dimensional features is expressed as:

[0022] Z′ r =A r Z r

[0023] In the formula, Z′ r To adjust the low-dimensional features, Z r is the initial low-dimensional feature.

[0024] In a specific embodiment, in each round of iteration, adjusting the weight of each of the adjusted low-dimensional features according to the performance of the adjusted low-dimensional features in the previous round includes:

[0025] Assume that in the tth iteration, the weight calculation formula is shown as:

[0026]

[0027] In the formula, represents the feature weight vector of the t+1th iteration, represents the feature weight vector of the tth iteration, η r is the learning rate of the feature dimensionality reduction model, L r () is the loss function of the feature dimensionality reduction model, Y r is the label data, Represents the gradient of the loss function with respect to the feature weight vector.

[0028] In a specific embodiment, the calculation formula of the target vehicle Ethernet data is expressed as:

[0029] X′ r =Sig dec (W′ r Z′ r +b′ r )

[0030] In the formula, X′ r is the target vehicle Ethernet data reconstructed by the decoder, W′ r is the weight matrix of the decoder, b′ r is the bias vector of the decoder, Sig dec () is the multi-layer Sigmoid activation function of the decoder.

[0031] In a second aspect, the present invention provides a vehicle-mounted network transmission monitoring device, comprising:

[0032] A data sample acquisition module, used to acquire vehicle Ethernet data samples;

[0033] An initial low-dimensional feature generation module, used for inputting the vehicle Ethernet data sample into a feature dimension reduction model to be trained, wherein the feature dimension reduction model includes an encoder, a decoder and a feature adjustment module, and mapping the vehicle Ethernet data sample to a low-dimensional feature space through the encoder to generate an initial low-dimensional feature;

[0034] A target low-dimensional feature generation module, used for recursively optimizing the initial low-dimensional features through the feature adjustment module, adjusting the weights of each of the initial low-dimensional features, and generating target low-dimensional features;

[0035] A target vehicle Ethernet data generation module, used for remapping the target low-dimensional features back to the high-dimensional space through the decoder to obtain target vehicle Ethernet data;

[0036] An iterative training module, used for iteratively training the feature dimensionality reduction model according to a preset stop iteration condition to obtain a target feature dimensionality reduction model;

[0037] The vehicle network transmission status generation module is used to perform feature dimension reduction on the collected vehicle Ethernet data of the target vehicle through the target feature dimension reduction model, and classify the reduced vehicle Ethernet data through the classifier model to obtain the vehicle network transmission status result of the target vehicle.

[0038] In a third aspect, the present invention provides a terminal comprising: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements a vehicle network transmission monitoring method as described in the first aspect.

[0039] In a fourth aspect, the present invention provides a vehicle, comprising the vehicle-mounted network transmission monitoring device as described in the second aspect or the terminal as described in the third aspect.

[0040] In a fifth aspect, the present invention provides a computer-readable storage medium, which stores a computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute a vehicle network transmission monitoring method as described in the first aspect.

[0041] Compared with the prior art, the beneficial effects of the present invention are: monitoring the vehicle status in real time and collecting multi-dimensional data through the vehicle-mounted Ethernet, then using the autoencoder neural network based on feature refinement to train the feature dimensionality reduction model to obtain the target feature dimensionality reduction model, and then using the target feature dimensionality reduction model to perform feature dimensionality reduction on the vehicle-mounted Ethernet data of the target vehicle, optimizing the data while retaining key information, and improving the efficiency and accuracy of vehicle network transmission data processing; in addition, the state of the reduced data is classified by the classifier model, and through this state feedback mechanism, the communication status of the vehicle network can be identified and responded to in real time, and potential communication problems and hardware failures can be discovered and warned in time, realizing efficient management and rapid feedback of the vehicle network transmission status, thereby improving the stability and reliability of the vehicle network transmission data and enhancing the vehicle's intelligence level and driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a flowchart of a vehicle network transmission monitoring method provided by an embodiment of the present invention;

[0043] Figure 2 It is a structural schematic diagram of a vehicle network transmission monitoring device provided by an embodiment of the present invention;

[0044] Figure 3 It is a schematic diagram of the structure of a terminal provided by an embodiment of the present invention.

[0045] in:

[0046] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0047] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0048] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0049] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, a direct connection, or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0050] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may include that the first and second features are in direct contact, or may include that the first and second features are not in direct contact but are in contact through another feature between them. Moreover, a first feature being "above", "above" and "above" a second feature includes that the first feature is directly above and obliquely above the second feature, or simply indicates that the first feature is higher in level than the second feature. A first feature being "below", "below" and "below" a second feature includes that the first feature is directly below and obliquely below the second feature, or simply indicates that the first feature is lower in level than the second feature.

[0051] See also Figure 1 , Figure 1 It is a flowchart of a vehicle network transmission monitoring method provided by an embodiment of the present invention.

[0052] A vehicle network transmission monitoring method according to an embodiment of the present invention comprises the following steps:

[0053] S100: Obtain vehicle Ethernet data samples.

[0054] The data collection source of the present invention is the vehicle-mounted Ethernet in the actual vehicle operation, and the vehicle-mounted Ethernet data is collected through the internal network interface of the vehicle to ensure the real-time and relevance of the data. The vehicle-mounted Ethernet data collection method adopts a method combining real-time monitoring and periodic capture, so that the vehicle status, sensor output and network status data can be efficiently collected without interfering with the normal operation of the vehicle.

[0055] In one embodiment, the attributes of the vehicle Ethernet data include:

[0056] Ra is the vehicle speed, indicating the current speed of the vehicle;

[0057] Da is the acceleration sensor data, which records the instantaneous value of vehicle acceleration or deceleration;

[0058] Ta is the temperature sensor data, which records the temperature conditions inside and outside the vehicle;

[0059] Fa is the fuel consumption, which shows the fuel consumption of the vehicle;

[0060] Sa is the vehicle status indication, such as engine status, transmission status, etc.;

[0061] Va is the vehicle voltage, which monitors the voltage status of the vehicle electrical system;

[0062] Pa is an indicative parameter, such as the door switch status, light switch status, etc.;

[0063] Ka is the network communication quality, which records the transmission quality and delay of the vehicle network;

[0064] Ha is the hardware status information, such as the operating status of various sensors and controllers;

[0065] Za is a custom event marker used to identify specific driving behaviors or network events.

[0066] It should be noted that this embodiment is only used to illustrate one data format and type of the present invention. In actual applications, the attributes of vehicle Ethernet data are usually more than ten attributes, and the number of attributes of the data may reach dozens or even hundreds.

[0067] Furthermore, the collected vehicle Ethernet data is labeled, and the labeling method of the embodiment of the present invention is manual labeling.

[0068] In one embodiment, the marked categories include:

[0069] Normal communication: The state in which the vehicle network operates normally within standard performance parameters.

[0070] Communication delay: There is a large data transmission delay in the vehicle network.

[0071] Data loss: Data packets are lost during transmission in the vehicle network.

[0072] Hardware failure: Vehicle hardware equipment fails, affecting data collection and transmission.

[0073] Abnormal behavior: The vehicle exhibits abnormal driving behavior or network usage behavior.

[0074] This embodiment achieves comprehensive monitoring of the vehicle's operating status through an efficient data collection and labeling mechanism.

[0075] The collected data is input into the feature dimensionality reduction model for model training, wherein the present invention adopts an autoencoding neural network based on feature refinement as the feature dimensionality reduction model, and the autoencoding neural network based on feature refinement includes an encoder, a decoder and a feature adjustment module. The encoder is used to map high-dimensional input data to a low-dimensional feature space, and the decoder is used to reconstruct the reduced-dimensional features back to the original space to ensure the reversibility of the dimensionality reduction process. The feature adjustment module dynamically adjusts the reduced-dimensional feature space through recursive feature adaptive optimization, so that important features can be strengthened and secondary features are gradually weakened, so that the reduced-dimensional feature representation has better simplicity while retaining data information.

[0076] Specifically, the training process of the feature dimension reduction model based on the autoencoder neural network algorithm of feature refinement is described by the following steps.

[0077] S200: Input the vehicle Ethernet data sample into the feature dimension reduction model to be trained, and map the vehicle Ethernet data sample to a low-dimensional feature space through the encoder to generate initial low-dimensional features.

[0078] In this embodiment, the encoder uses a multi-layer nonlinear mapping structure to map high-dimensional in-vehicle Ethernet data samples to an initial low-dimensional feature space.

[0079] Specifically, the calculation formula of the initial low-dimensional features is expressed as:

[0080] Z r =Sig enc (W r X r +b r )

[0081] In the formula, Z r represents the initial low-dimensional features, W r is the weight matrix of the encoder, X r is the vehicle Ethernet data sample input to the encoder, b r is the bias vector of the encoder, Sig enc() is the multi-layer Sigmoid activation function of the encoder.

[0082] S300: recursively optimize the initial low-dimensional features through the feature adjustment module, adjust the weights of each of the initial low-dimensional features, and generate target low-dimensional features.

[0083] In this embodiment, after the initial low-dimensional features are generated, the feature adjustment module automatically generates feature weights according to the feature importance in the current feature space.

[0084] In a specific embodiment, the step S300 recursively optimizes the initial low-dimensional features through the feature adjustment module, adjusts the weights of each of the initial low-dimensional features, and generates target low-dimensional features, including the following steps:

[0085] S310: Generate initial weights of the initial low-dimensional features according to the feature importance of the initial low-dimensional features through the feature adjustment module, and generate adjusted low-dimensional features according to the initial weights.

[0086] When the feature adjustment module is initialized, all initial low-dimensional features are given the same initial weights so that the subsequent steps can gradually adjust the weights according to the feature contributions.

[0087] In a specific embodiment, the calculation method of the matrix of the initial weights is expressed as:

[0088] A r =diag(α r )

[0089] In the formula, A r is the initial weight matrix; α r is the feature weight vector, α r Each element α in r,i Initialized to the same value, indicating that all the initial low-dimensional features have the same importance in the initial stage; diag() is a function for extracting the diagonal elements of the initial weight matrix;

[0090] The calculation method of adjusting the low-dimensional features is expressed as:

[0091] Z′ r =A r Z r

[0092] In the formula, Z′ r To adjust the low-dimensional features, Z r is the initial low-dimensional feature.

[0093] S320, recursively optimizing the adjusted low-dimensional features through the feature adjustment module, so as to adjust the weights of each of the adjusted low-dimensional features according to the performance of the adjusted low-dimensional features in the previous round in each iteration, and generate target low-dimensional features.

[0094] In this embodiment, the feature adjustment module recursively optimizes the adjusted low-dimensional features. In each round of iteration, the feature adjustment module adjusts the weights of each adjusted low-dimensional feature according to the performance of the previous round of adjusted low-dimensional features, gradually enhances the features with important influence, and gradually weakens the redundant or noise features.

[0095] Specifically, the weight update rules are as follows:

[0096] Assume that in the tth iteration, the weight calculation formula is shown as:

[0097]

[0098] In the formula, represents the feature weight vector of the t+1th iteration, represents the feature weight vector of the tth iteration, η r is the learning rate of the feature dimensionality reduction model, L r () is the loss function of the feature dimensionality reduction model, Y r is the label data, Represents the gradient of the loss function with respect to the feature weight vector.

[0099] Optionally, the loss function L of the autoencoder neural network is r () is the reconstruction error loss function.

[0100] Optionally, n r Set to 0.01.

[0101] S400: re-map the target low-dimensional features back to a high-dimensional space through the decoder to obtain target vehicle Ethernet data.

[0102] In this embodiment, in order to ensure the effectiveness of the dimensionality reduction process, the decoder remaps the target low-dimensional features back to the high-dimensional space to ensure that no important information is lost in the dimensionality reduction process.

[0103] In a specific embodiment, the calculation formula of the target vehicle Ethernet data is expressed as:

[0104] X′ r =Sig dec (W′ r Z′ r +b′ r )

[0105] In the formula, X′r is the target vehicle Ethernet data reconstructed by the decoder, W′ r is the weight matrix of the decoder, b′ r is the bias vector of the decoder, Sig dec () is the multi-layer Sigmoid activation function of the decoder.

[0106] In a specific embodiment, a loss value is calculated based on the target vehicle Ethernet data, the true label and the loss function, and the feature dimensionality reduction model is adjusted according to the loss value to reduce the training error and gradually approach the optimal solution.

[0107] S500: Iteratively train the feature dimensionality reduction model according to a preset stop iteration condition to obtain a target feature dimensionality reduction model.

[0108] The above steps S200 to S400 are iterated repeatedly until the preset stop iteration condition is met, which means that the feature dimensionality reduction model training is completed, and the optimal model parameters are obtained, thereby obtaining the target feature dimensionality reduction model.

[0109] In a specific embodiment, the preset condition for stopping iteration is reaching a preset maximum number of iterations. Preferably, the preset maximum number of iterations is set to 1000 times.

[0110] The embodiment of the present invention trains the feature dimension reduction model by using an autoencoder neural network algorithm based on feature refinement, thereby optimizing the vehicle Ethernet data processing flow, reducing data redundancy, and improving the efficiency and accuracy of vehicle Ethernet data processing.

[0111] S600, performing feature dimension reduction on the collected in-vehicle Ethernet data of the target vehicle through the target feature dimension reduction model, and classifying the reduced in-vehicle Ethernet data through the classifier model to obtain the in-vehicle network transmission status result of the target vehicle.

[0112] The target vehicle's on-board Ethernet collects the corresponding on-board Ethernet data, and collects data through the target vehicle's internal network interface to ensure the real-time and relevance of the data. The data collection method adopts a method that combines real-time monitoring with periodic capture, so that the vehicle status, sensor output and network status data can be efficiently collected without interfering with the normal operation of the vehicle. It can be understood that the attributes of the on-board Ethernet data sample can cover the newly collected on-board Ethernet data, for example, the on-board Ethernet data includes vehicle speed, acceleration sensor data, temperature sensor data, fuel consumption, vehicle status indication, vehicle voltage, indicative parameters, network communication quality, hardware status information, and one or more custom event markers.

[0113] The trained target feature dimensionality reduction model is used to perform feature dimensionality reduction on the collected in-vehicle Ethernet data of the target vehicle, and then the reduced in-vehicle Ethernet data is classified through the classifier model to obtain the target in-vehicle Ethernet data transmission management status feedback, that is, the in-vehicle network transmission status result, which realizes efficient management and rapid feedback of the in-vehicle network transmission status, thereby improving the stability of in-vehicle network data transmission and the safety of vehicle operation.

[0114] In one embodiment, the classification categories include:

[0115] Normal communication: The vehicle network is operating normally within standard performance parameters;

[0116] Communication delay: The vehicle network has a large data transmission delay;

[0117] Data loss: Data packets are lost during transmission in the vehicle network;

[0118] Hardware failure: Vehicle hardware equipment fails, affecting data collection and transmission;

[0119] Abnormal behavior: The vehicle exhibits abnormal driving behavior or network usage behavior.

[0120] In one embodiment, the classifier model includes a support vector machine classifier.

[0121] In summary, an embodiment of the present invention provides a vehicle network transmission monitoring method, which monitors the vehicle status in real time and collects multi-dimensional data through the vehicle Ethernet, and then uses the autoencoder neural network based on feature refinement to train the feature dimensionality reduction model to obtain the target feature dimensionality reduction model, and then uses the target feature dimensionality reduction model to perform feature dimensionality reduction on the vehicle Ethernet data of the target vehicle, optimizes the data while retaining key information, and improves the efficiency and accuracy of vehicle network transmission data processing; in addition, the state of the reduced data is classified by the classifier model, and through this state feedback mechanism, it can identify and respond to the communication status of the vehicle network in real time, and timely discover and warn potential communication problems and hardware failures, thereby realizing efficient management and rapid feedback of the vehicle network transmission status, thereby improving the stability and reliability of the vehicle network transmission data, and enhancing the vehicle's intelligence level and driving safety.

[0122] See also Figure 2 , Figure 2 It is a structural schematic diagram of a vehicle-mounted network transmission monitoring device provided by one embodiment of the present invention.

[0123] An in-vehicle network transmission monitoring device according to an embodiment of the present invention includes:

[0124] A data sample acquisition module, used to acquire vehicle Ethernet data samples;

[0125] An initial low-dimensional feature generation module, used for inputting the vehicle Ethernet data sample into a feature dimension reduction model to be trained, wherein the feature dimension reduction model includes an encoder, a decoder and a feature adjustment module, and mapping the vehicle Ethernet data sample to a low-dimensional feature space through the encoder to generate an initial low-dimensional feature;

[0126] A target low-dimensional feature generation module, used for recursively optimizing the initial low-dimensional features through the feature adjustment module, adjusting the weights of each of the initial low-dimensional features, and generating target low-dimensional features;

[0127] A target vehicle Ethernet data generation module, used for remapping the target low-dimensional features back to the high-dimensional space through the decoder to obtain target vehicle Ethernet data;

[0128] An iterative training module, used for iteratively training the feature dimensionality reduction model according to a preset stop iteration condition to obtain a target feature dimensionality reduction model;

[0129] The vehicle network transmission status generation module is used to perform feature dimension reduction on the collected vehicle Ethernet data of the target vehicle through the target feature dimension reduction model, and classify the reduced vehicle Ethernet data through the classifier model to obtain the vehicle network transmission status result of the target vehicle.

[0130] An in-vehicle network transmission monitoring device provided in an embodiment of the present invention can execute all steps and functions of an in-vehicle network transmission monitoring method provided in any of the above embodiments, and the specific functions of the device are not described in detail herein.

[0131] See also Figure 3 , Figure 3 It is a schematic diagram of the structure of a terminal provided by an embodiment of the present invention.

[0132] The terminal includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above-mentioned vehicle network transmission monitoring method are implemented, for example: Figure 1 Alternatively, the processor implements the functions of each module in the above-mentioned device embodiments when executing the computer program.

[0133] Exemplarily, the computer program may be divided into one or more modules, and the one or more modules are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal. For example, the computer program may be divided into several modules, and the specific functions of each module have been described in detail in a vehicle network transmission monitoring method provided in any of the above embodiments, and the specific functions of the device will not be repeated here.

[0134] The terminal may be a computing device such as a desktop computer, a notebook, a PDA, and a cloud server. The terminal may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the schematic diagram is merely an example of a terminal and does not constitute a limitation on a terminal. The terminal may include more or fewer components than shown in the figure, or may combine certain components, or different components. For example, the terminal may also include input and output devices, network access devices, buses, etc.

[0135] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal, and uses various interfaces and lines to connect various parts of the entire terminal.

[0136] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the vehicle network transmission monitoring method by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0137] An embodiment of the present invention further provides a vehicle, which includes the above-mentioned vehicle-mounted network transmission monitoring device or the above-mentioned terminal.

[0138] An embodiment of the present invention further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned vehicle network transmission monitoring method is implemented.

[0139] If the terminal integrated module 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 present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0140] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A vehicle network transmission monitoring method, characterized in that: include: Get the vehicle Ethernet data sample; Inputting the vehicle Ethernet data sample into a feature dimension reduction model to be trained, wherein the feature dimension reduction model includes an encoder, a decoder and a feature adjustment module, and mapping the vehicle Ethernet data sample to a low-dimensional feature space through the encoder to generate initial low-dimensional features; Recursively optimizing the initial low-dimensional features through the feature adjustment module, adjusting the weights of each of the initial low-dimensional features, and generating target low-dimensional features; Re-mapping the target low-dimensional features back to a high-dimensional space through the decoder to obtain target vehicle Ethernet data; Iteratively training the feature dimensionality reduction model according to a preset stop iteration condition to obtain a target feature dimensionality reduction model; The target feature dimension reduction model is used to perform feature dimension reduction on the collected vehicle Ethernet data of the target vehicle, and the reduced vehicle Ethernet data is classified by the classifier model to obtain the vehicle network transmission status result of the target vehicle.

2. A vehicle network transmission monitoring method according to claim 1, characterized in that: The calculation formula of the initial low-dimensional feature is expressed as: Z r =Sig enc (W r X r +b r ) In the formula, Z r represents the initial low-dimensional features, W r is the weight matrix of the encoder, X r is the vehicle Ethernet data sample input to the encoder, b r is the bias vector of the encoder, Sig enc () is the multi-layer Sigmoid activation function of the encoder.

3. The vehicle network transmission monitoring method according to claim 1, characterized in that: The recursive optimization of the initial low-dimensional features by the feature adjustment module, adjusting the weights of each of the initial low-dimensional features, and generating target low-dimensional features includes: Generating, by the feature adjustment module, initial weights of the initial low-dimensional features according to the feature importance of the initial low-dimensional features, and generating adjusted low-dimensional features according to the initial weights; The adjusted low-dimensional features are recursively optimized through the feature adjustment module, so that in each round of iteration, the weights of the adjusted low-dimensional features are adjusted according to the performance of the adjusted low-dimensional features in the previous round, and the target low-dimensional features are generated.

4. A vehicle network transmission monitoring method according to claim 3, characterized in that: The calculation method of the matrix of the initial weight is expressed as: A r =diag(a r ) In the formula, A r is the initial weight matrix; α r is the feature weight vector, α r Each element α in r,i Initialized to the same value, indicating that all the initial low-dimensional features have the same importance in the initial stage; diag() is a function for extracting the diagonal elements of the initial weight matrix; The calculation method of adjusting the low-dimensional features is expressed as: Z′ r =A r From r In the formula, Z′ r To adjust the low-dimensional features, Z r is the initial low-dimensional feature.

5. The vehicle network transmission monitoring method according to claim 3 is characterized in that: In each round of iteration, according to the performance of the adjusted low-dimensional features in the previous round, the weights of the adjusted low-dimensional features are adjusted, including: Assume that in the tth iteration, the weight calculation formula is shown as: In the formula, represents the feature weight vector of the t+1th iteration, represents the feature weight vector of the tth iteration, η r is the learning rate of the feature dimensionality reduction model, L r () is the loss function of the feature dimensionality reduction model, Y r is the label data, Represents the gradient of the loss function with respect to the feature weight vector.

6. The vehicle network transmission monitoring method according to claim 1, characterized in that: The calculation formula of the target vehicle Ethernet data is expressed as: X r ′=Sig dec (W r ′Z r ′+b r ′) Where, X r ′ is the target vehicle Ethernet data reconstructed by the decoder, W r ′ is the weight matrix of the decoder, b r ' is the bias vector of the decoder, Sig dec () is the multi-layer Sigmoid activation function of the decoder.

7. A vehicle-mounted network transmission monitoring device, characterized in that: include: A data sample acquisition module, used to acquire vehicle Ethernet data samples; An initial low-dimensional feature generation module, used for inputting the vehicle Ethernet data sample into a feature dimension reduction model to be trained, wherein the feature dimension reduction model includes an encoder, a decoder and a feature adjustment module, and mapping the vehicle Ethernet data sample to a low-dimensional feature space through the encoder to generate an initial low-dimensional feature; A target low-dimensional feature generation module, used for recursively optimizing the initial low-dimensional features through the feature adjustment module, adjusting the weights of each of the initial low-dimensional features, and generating target low-dimensional features; A target vehicle Ethernet data generation module, used for remapping the target low-dimensional features back to the high-dimensional space through the decoder to obtain target vehicle Ethernet data; An iterative training module, used for iteratively training the feature dimensionality reduction model according to a preset stop iteration condition to obtain a target feature dimensionality reduction model; The vehicle network transmission status generation module is used to perform feature dimension reduction on the collected vehicle Ethernet data of the target vehicle through the target feature dimension reduction model, and classify the reduced vehicle Ethernet data through the classifier model to obtain the vehicle network transmission status result of the target vehicle.

8. The vehicle-mounted network transmission monitoring device according to claim 7, characterized in that: The calculation formula of the initial low-dimensional feature is expressed as: Z r =Sig enc (W r X r +b r ) In the formula, Z r represents the initial low-dimensional features, W r is the weight matrix of the encoder, X r is the vehicle Ethernet data sample input to the encoder, b r is the bias vector of the encoder, Sig enc () is the multi-layer Sigmoid activation function of the encoder.

9. The vehicle-mounted network transmission monitoring device according to claim 7, characterized in that: The target low-dimensional feature generation module is used to: Generating, by the feature adjustment module, initial weights of the initial low-dimensional features according to the feature importance of the initial low-dimensional features, and generating adjusted low-dimensional features according to the initial weights; The adjusted low-dimensional features are recursively optimized through the feature adjustment module, so that in each round of iteration, the weights of the adjusted low-dimensional features are adjusted according to the performance of the adjusted low-dimensional features in the previous round, and the target low-dimensional features are generated.

10. The vehicle-mounted network transmission monitoring device according to claim 9, characterized in that: The calculation method of the matrix of the initial weight is expressed as: A r =diag(a r ) In the formula, A r is the initial weight matrix; α r is the feature weight vector, α r Each element α in r,i Initialized to the same value, indicating that all the initial low-dimensional features have the same importance in the initial stage; diag() is a function for extracting the diagonal elements of the initial weight matrix; The calculation method of adjusting the low-dimensional features is expressed as: From r ′=A r From r In the formula, Z r ′ is the adjusted low-dimensional feature, Z r is the initial low-dimensional feature.

11. The vehicle network transmission monitoring device according to claim 9, characterized in that: The target low-dimensional feature generation module is used to: Assume that in the tth iteration, the weight calculation formula is shown as: In the formula, represents the feature weight vector of the t+1th iteration, represents the feature weight vector of the tth iteration, η r is the learning rate of the feature dimensionality reduction model, L r () is the loss function of the feature dimensionality reduction model, Y r is the label data, Represents the gradient of the loss function with respect to the feature weight vector.

12. The vehicle-mounted network transmission monitoring device according to claim 7, characterized in that: The calculation formula of the target vehicle Ethernet data is expressed as: X r ′=Sig dec (W r ′Z r ′+b r ′) Where, X r ′ is the target vehicle Ethernet data reconstructed by the decoder, W r ′ is the weight matrix of the decoder, b r ' is the bias vector of the decoder, Sig dec () is the multi-layer Sigmoid activation function of the decoder.

13. A terminal, characterized in that: include: A processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, a vehicle network transmission monitoring method as described in any one of claims 1 to 6 is implemented.

14. A vehicle, characterized in that: The vehicle includes the in-vehicle network transmission monitoring device as claimed in claim 7 or the terminal as claimed in claim 13.

15. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute a vehicle network transmission monitoring method as described in any one of claims 1 to 6.

Images