CN116567719B - Data transmission method, vehicle-mounted system, device and storage medium - Google Patents

Data transmission method, vehicle-mounted system, device and storage medium Download PDF

Info

Publication number
CN116567719B
CN116567719B CN202310822509.3A CN202310822509A CN116567719B CN 116567719 B CN116567719 B CN 116567719B CN 202310822509 A CN202310822509 A CN 202310822509A CN 116567719 B CN116567719 B CN 116567719B
Authority
CN
China
Prior art keywords
data
vehicle
noise reduction
application data
mounted application
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202310822509.3A
Other languages
Chinese (zh)
Other versions
CN116567719A (en
Inventor
赵建智
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Jidu Technology Co Ltd
Original Assignee
Beijing Jidu Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Jidu Technology Co Ltd filed Critical Beijing Jidu Technology Co Ltd
Priority to CN202310822509.3A priority Critical patent/CN116567719B/en
Publication of CN116567719A publication Critical patent/CN116567719A/en
Application granted granted Critical
Publication of CN116567719B publication Critical patent/CN116567719B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H04W28/06—Optimizing the usage of the radio link, e.g. header compression, information sizing, discarding information
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/16—Central resource management; Negotiation of resources or communication parameters, e.g. negotiating bandwidth or QoS [Quality of Service]
    • H04W28/24—Negotiating SLA [Service Level Agreement]; Negotiating QoS [Quality of Service]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
    • H04W4/48—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P] for in-vehicle communication
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W8/00—Network data management
    • H04W8/22—Processing or transfer of terminal data, e.g. status or physical capabilities
    • H04W8/24—Transfer of terminal data

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Quality & Reliability (AREA)
  • Databases & Information Systems (AREA)
  • Stored Programmes (AREA)

Abstract

The embodiment of the invention provides a data transmission method, a vehicle-mounted system, equipment and a storage medium, wherein the method comprises the following steps: acquiring performance data and vehicle-mounted application data of a vehicle-mounted system in a current period; and determining classification results corresponding to the performance data and the vehicle-mounted application data by using the classification model. And finally, transmitting the vehicle-mounted application data to a corresponding receiving end according to the target transmission parameters corresponding to the classification result corresponding to the current period. In the scheme, the overall performance and the running state of the vehicle-mounted system can be known more timely by collecting the performance data and the vehicle-mounted application data of the vehicle-mounted system in real time, the target transmission parameters suitable for the current period are dynamically determined according to the classification result output by the classification model, and the safety and reliability of the vehicle-mounted application data can be ensured to be transmitted by using the target transmission parameters.

Description

Data transmission method, vehicle-mounted system, device and storage medium
Technical Field
The present invention relates to the field of artificial intelligence technologies, and in particular, to a data transmission method, a vehicle-mounted system, a device, and a storage medium.
Background
With the vigorous development and wide application of the internet of vehicles, the more and more functions can be provided by the vehicle-mounted system, such as vehicle-mounted entertainment functions, automatic driving functions, intelligent diagnosis and the like. Then, as the functions supportable by the vehicle-mounted system are more and the volume of data generated by the vehicle-mounted system is more and more, the requirements on reliability and real-time performance of the data are higher.
However, as the functions supportable by the vehicle-mounted system are more and more, the number of application programs in the vehicle-mounted system is increased continuously, and when a certain application program in the vehicle-mounted system fails, the condition of overhigh load, blocking and the like of the vehicle-mounted system may be caused, so that the reliability and instantaneity of data transmission are affected. Therefore, how to improve the reliability of each data transmission in the vehicle-mounted system is a problem to be solved.
Disclosure of Invention
The embodiment of the invention provides a data transmission method, a training method, a vehicle-mounted system, equipment and a storage medium, which are used for improving the data transmission performance in the vehicle-mounted system.
In a first aspect, an embodiment of the present invention provides a data transmission method, where the method includes:
acquiring performance data and vehicle-mounted application data of a vehicle-mounted system in a current period;
determining classification results corresponding to the performance data and the vehicle-mounted application data by using a classification model, wherein the classification results are reflected in the current period, and the vehicle-mounted system transmits the capability level of the vehicle-mounted application data;
and transmitting the vehicle-mounted application data to a corresponding receiving end according to a target transmission parameter corresponding to the classification result, wherein the target transmission parameter comprises a quality of service Qos strategy and a data block length of single transmission.
In a second aspect, an embodiment of the present invention provides an in-vehicle system, the system including:
the acquisition module is used for acquiring performance data and vehicle-mounted application data of the vehicle-mounted system in the current period;
the determining module is used for determining a classification result corresponding to the performance data and the vehicle-mounted application data by using a classification model, wherein the classification result is reflected in the current period, and the vehicle-mounted system transmits the capability level of the vehicle-mounted application data;
and the transmission module is used for transmitting the vehicle-mounted application data to a corresponding receiving end according to a target transmission parameter corresponding to the classification result, wherein the target transmission parameter comprises a quality of service Qos strategy and a data block length of single transmission.
In a third aspect, an embodiment of the present invention provides an electronic device, including: a memory, a processor, a communication interface; wherein the memory has stored thereon executable code which, when executed by the processor, causes the processor to perform the data transmission method according to the first aspect.
In a fourth aspect, embodiments of the present invention provide a non-transitory machine-readable storage medium having stored thereon executable code, which when executed by a processor of an electronic device, causes the processor to at least implement a data transmission method as described in the first aspect.
In the data transmission scheme provided by the embodiment of the invention, the performance data and the vehicle-mounted application data of the vehicle-mounted system in the current period are collected. And then determining classification results corresponding to the performance data and the vehicle-mounted application data by using the classification model. And finally, transmitting the vehicle-mounted application data to a corresponding receiving end according to the target transmission parameters corresponding to the classification result corresponding to the current period.
In the scheme, the overall performance and the running state of the vehicle-mounted system can be known more timely by collecting the performance data and the vehicle-mounted application data of the vehicle-mounted system in real time, the target transmission parameters suitable for the current period are dynamically determined according to the classification result output by the classification model, and the safety and reliability of the vehicle-mounted application data can be ensured to be transmitted by using the target transmission parameters.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a data transmission method according to an embodiment of the present invention;
FIG. 2 is a flowchart for determining a target transmission parameter according to performance data and vehicle-mounted application data according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of determining a target transmission parameter according to performance data and vehicle-mounted application data according to an embodiment of the present invention;
FIG. 4a is a schematic diagram of a training noise reduction self-encoder;
FIG. 4b is a schematic diagram of another training noise reduction self-encoder;
fig. 5 is an application schematic diagram of a data transmission method according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of a training method for classification models according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of another training method for classification models according to an embodiment of the present invention;
FIG. 8 is a schematic diagram of another classification model training method according to an embodiment of the present invention;
fig. 9 is a schematic structural diagram of a vehicle-mounted system according to an embodiment of the present invention;
fig. 10 is a schematic structural diagram of an electronic device according to the present embodiment.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention. In addition, the sequence of steps in the method embodiments described below is only an example and is not strictly limited.
It should be noted that, the user information (including but not limited to user equipment information, user personal information, etc.) and the data (including but not limited to data for analysis, stored data, presented data, etc.) related to the embodiments of the present invention are information and data authorized by the user or fully authorized by each party, and the collection, use and processing of the related data need to comply with the related laws and regulations and standards of the related country and region, and provide corresponding operation entries for the user to select authorization or rejection.
With the popularization of vehicle-mounted systems, the requirements on the functions of the vehicle-mounted systems are increasing, and the functions such as vehicle navigation, communication, multimedia video entertainment, vehicle control functions, over-the-Air Technology (OTA for short) functions, vehicle diagnosis and the like can be integrated into the vehicle-mounted systems in the form of application programs through network technologies at present. However, when an application program in the vehicle-mounted system fails, the system performance is most likely to reach a critical value in advance, so that the vehicle-mounted system has the problems of system performance such as overhigh load, overhigh cpu occupancy, overhigh memory occupancy and the like, and the vehicle-mounted system has the problems of clamping and the like. In addition, as the functions supportable by the vehicle-mounted system are more and more, the number of application programs in the vehicle-mounted system is increased continuously, so that hidden dangers of faults of the application programs in the vehicle-mounted system are also greater and greater, and the vehicle-mounted system may frequently have the conditions of clamping, etc.
In the existing data transmission scheme, a fixed transmission parameter, namely a fixed quality of service (Quality of Service, qos for short) strategy and a fixed data block size are adopted for data transmission, so that when an application program in a vehicle-mounted system fails, the vehicle-mounted system may have the problems of overhigh load, overhigh cpu occupancy, overhigh memory occupancy and other system performances, so that the vehicle-mounted system is blocked, the transmission is continued with the current fixed transmission parameter, and the problems of data loss, data delay transmission and the like may occur, thereby seriously affecting the safety, reliability, instantaneity and data transmission efficiency of the data transmission in each application scene in the vehicle-mounted system.
In order to solve the problem of reduced data transmission performance caused by the occurrence of faults of application programs in a vehicle-mounted system, the embodiment of the invention provides a novel data transmission scheme. In the data transmission scheme, the performance data and the vehicle-mounted application data of the vehicle-mounted system are collected according to a preset period, and the target transmission parameters are dynamically adjusted according to the performance data and the data volume of the vehicle-mounted application data of the vehicle-mounted system in the current period so as to better adapt to the current running state of the vehicle-mounted system, even if the vehicle-mounted system is in a cartoon or other abnormal condition, the abnormality of the vehicle-mounted system can be timely perceived, and the transmission parameters are timely and dynamically adjusted so as to ensure that the vehicle-mounted application data can be transmitted to a corresponding receiving end safely and reliably.
Some embodiments of the present invention are described in detail below with reference to the accompanying drawings. In the case where there is no conflict between the embodiments, the following embodiments and features in the embodiments may be combined with each other.
Fig. 1 is a flowchart of a data transmission method according to an embodiment of the present invention, as shown in fig. 1, where the method includes the following steps:
101. and collecting performance data and vehicle-mounted application data of the vehicle-mounted system in the current period.
102. And determining classification results corresponding to the performance data and the vehicle-mounted application data by using the classification model, wherein the classification results are reflected in the current period, and the vehicle-mounted system transmits the capability level of the vehicle-mounted application data.
103. And transmitting the vehicle-mounted application data to a corresponding receiving end according to a target transmission parameter corresponding to the classification result, wherein the target transmission parameter comprises a quality of service Qos strategy and a data block length of single transmission.
The data transmission scheme provided by the embodiment of the invention can be applied to various vehicle-mounted systems, and the vehicle-mounted system is a vehicle-mounted integrated information processing system which can be arranged in a vehicle or a cloud-mounted integrated information processing system, so that a series of applications including three-dimensional navigation, real-time road conditions, auxiliary driving, fault detection, vehicle information, vehicle body control, wireless communication, entertainment functions and the like can be realized, and the level of electronization, networking and intellectualization of the vehicle is greatly improved.
When providing various application functions for users, the vehicle-mounted system often needs to transmit a large amount of application data when aiming at the application functions such as real-time synchronous playing of multiple paths of high-definition videos in a vehicle, vehicle-mounted karaoke and the like. In order to ensure the transmission safety, reliability and real-time performance of the application data corresponding to the different application functions in the vehicle-mounted system, the performance data and the vehicle-mounted application data of the vehicle-mounted system can be collected regularly, and the transmission parameters can be dynamically adjusted according to the performance data and the vehicle-mounted application data of the vehicle-mounted system.
The vehicle-mounted system generally comprises a plurality of functional modules, each functional module in the vehicle-mounted system can be uniformly controlled by the central processing unit, and the performance data of the vehicle-mounted system refer to various data capable of reflecting the overall performance of the vehicle-mounted system, for example: cpu resource occupancy, disk throughput, memory usage, network throughput, bandwidth, etc. The vehicle-mounted application data refers to data to be transmitted when the vehicle-mounted system realizes each application function, namely, the data to be transmitted when the vehicle-mounted system is in various application scenes. For example: when the vehicle-mounted system is in a vehicle OTA scene, the vehicle-mounted application data comprises effective information such as the downloading state (idle, in downloading, ready to update, decrypted and checked signature, downloading completed and the like) of the OTA, the size of the downloaded file and the like. The OTA refers to an over-the-air technology, and can update and upgrade software in a vehicle-mounted system. When the vehicle-mounted system is in a vehicle diagnosis scene, the vehicle-mounted application data comprise state diagnosis information (pre-programmed state, transmission data, downloading completion and the like) corresponding to a diagnosis fault code (Diagnostic Trouble Code, DTC for short) of the vehicle, wherein the DTC refers to the diagnosis fault code, and the fault code state is used for reflecting the diagnosis state of the fault in real time. The in-vehicle application data may include image data, audio data, etc. when the in-vehicle system is in an in-vehicle entertainment scene. When the in-vehicle system is in an autopilot scenario, the in-vehicle application data may include point cloud data, navigational map data, and the like.
The vehicle-mounted system is mainly used for providing various application functions for users, in practical application, the requirements of the users can change at any time, so that the application functions provided by the vehicle-mounted system also change at any time, and then the performance of the vehicle-mounted system and the vehicle-mounted application data transmitted by each functional module in the vehicle-mounted system also change at any time. For example, the preset acquisition period may be 1s, 2s, 4s, etc., which may be specifically set according to actual requirements.
The application data to be transmitted and the application scene where each functional module in the vehicle-mounted system is located are different, and the performance data and the vehicle-mounted application data of the vehicle-mounted system acquired in different acquisition periods are different, so that the performance data and the vehicle-mounted application data of the vehicle-mounted system in the current period can be acquired when the target transmission parameters corresponding to the current moment are determined.
The embodiment of the invention does not limit the specific mode for collecting the performance data and the vehicle-mounted application data of the vehicle-mounted system, and can select the specific collecting mode according to actual conditions. Preferably, the performance data of the vehicle-mounted system can be filtered through the linux pipeline by using a vmstat tool, and then the filtered performance data is stored in a data hard disk in the vehicle-mounted system or in a data hard disk corresponding to a cloud end so as to be analyzed and processed based on the acquired data.
Preferably, the vehicle diagnosis data in the vehicle-mounted application data can collect diagnosis information of the vehicle from a vehicle-mounted automatic diagnosis interface (OBD port), and finally the collected diagnosis information is recorded in a data hard disk in the vehicle-mounted system. The vehicle OTA application data, the vehicle entertainment application data, the automatic driving application data and the like in the vehicle application data can be acquired in a mode of subscribing corresponding interface description languages (Interface description language, idl for short), the acquired vehicle application data can be stored in a corresponding memory key value pair container (map container), and finally the acquired vehicle application data is stored in a corresponding data hard disk after being compressed from a memory. Unstructured data in vehicle-mounted application data such as point cloud data, image data and the like can be acquired in a manner of recording data packets by the ros2, and finally the acquired application data are stored in a data hard disk in the form of data packets.
After the performance data and the vehicle-mounted application data of the vehicle-mounted system in the current period are obtained, the classification model can be used for classifying the performance data and the vehicle-mounted application data of the vehicle-mounted system. And then, according to the target transmission parameters corresponding to the classification result, transmitting the vehicle-mounted application data to the corresponding receiving end. The classification result is reflected in the current period, and the vehicle-mounted system transmits the capability level of the vehicle-mounted application data.
The target transmission parameters may specifically include parameters such as Qos policy and a data block length of a single transmission. Alternatively, qos policies may include a service level, a delay BUDGET (latency_buffer) policy, a transmission PRIORITY (transmission PRIORITY) policy, and so on. The preset data block length tag may comprise any one of 1M, 10M, 50M, 100M.
Wherein, the relation between the Qos strategy and the performance of the vehicle-mounted system can comprise:
optionally, the lower the CPU resource occupancy rate of the vehicle-mounted system is, the higher the service level of the QoS strategy corresponding to the classification result is, the longer the data block length is, and the target transmission parameter is suitable for transmission of important data, can improve the data transmission efficiency, improve the utilization rate of the system and avoid resource waste.
Alternatively, the higher the fluency of the in-vehicle system, the smaller the delay budget, and the higher the fluency of the in-vehicle system, so that the quality of service can be improved. The fluency of the vehicle-mounted system can be determined by the memory utilization rate and the disk throughput function.
Optionally, under the condition of considering throughput of the vehicle-mounted system, the transmission priority can be adjusted, and important data is sent preferentially, so that the data transmission efficiency is improved, and the service quality is ensured. Optionally, the lower the CPU resource occupancy rate of the vehicle-mounted system is, the longer the data block length can be, so that the data transmission efficiency and the utilization rate of the system can be improved, and the resource waste is avoided.
Optionally, the determining manner of the target transmission parameters may combine the combination of different performance data and vehicle-mounted application data with the mapping relationship between different transmission parameters, so as to perform classification processing on the performance data and vehicle-mounted application data of the vehicle-mounted system, so as to further obtain the target transmission parameters corresponding to the classification result.
In practical application, the application function provided by the vehicle-mounted system in the current period may be changed with the application function provided by the vehicle-mounted system in the previous acquisition period, so that the collected performance data of the vehicle-mounted system and the vehicle-mounted application data to be transmitted are changed greatly, then the classification result corresponding to the performance data and the vehicle-mounted application data can be determined by utilizing the classification model according to the currently collected performance data and the vehicle-mounted application data, and the target transmission parameter which corresponds to the classification result and is suitable for the current period is obtained, so that the target transmission parameter which is determined in the previous acquisition period can be adjusted in time, and the problems that the currently used transmission parameter is not suitable for the current data transmission, the data transmission performance is reduced and the like are avoided.
In addition, in practical application, if the application function provided by the vehicle-mounted system in the current period is the same as the application function provided by the vehicle-mounted system in the last acquisition period and the performance data of the vehicle-mounted system and the vehicle-mounted application data to be transmitted are basically unchanged, the determined target transmission parameters in the current period can be the same as the target transmission parameters in the last acquisition period, namely, the transmission parameters in the data transmission process are not adjusted.
From the above description, it is clear that: the overall performance of the vehicle-mounted system and the current running state of the vehicle-mounted system can be known more timely by periodically collecting the performance data and the vehicle-mounted application data of the vehicle-mounted system, and the data transmission parameters can be adjusted timely according to the collected performance data and the vehicle-mounted application data of the vehicle-mounted system in the current period.
And finally, transmitting the vehicle-mounted application data acquired in the current period to a corresponding receiving end according to the target transmission parameters corresponding to the classification result. For example, when the vehicle-mounted entertainment scene is currently in, the receiving end can be a vehicle-mounted entertainment module in the vehicle-mounted system, and entertainment audio data is sent to the corresponding vehicle-mounted entertainment module, so that the module can realize audio playing based on the entertainment audio data.
In the data transmission method provided by the embodiment of the invention, the classification model is utilized to determine the classification result corresponding to the performance data and the vehicle-mounted application data. And finally, transmitting the vehicle-mounted application data to a corresponding receiving end according to the target transmission parameters corresponding to the classification result corresponding to the current period. In the scheme, the overall performance and the running state of the vehicle-mounted system can be known more timely by collecting the performance data and the vehicle-mounted application data of the vehicle-mounted system in real time, the target transmission parameters suitable for the current period are dynamically determined according to the classification result output by the classification model, and the safety and reliability of the vehicle-mounted application data can be ensured to be transmitted by using the target transmission parameters.
In order to facilitate understanding of the data transmission scheme provided by the embodiment of the present invention, an example is illustrated in connection with a specific application scenario. The vehicle is provided with a vehicle-mounted system, the vehicle-mounted system comprises a plurality of vehicle-mounted application programs, and the vehicle-mounted application programs can transmit vehicle-mounted application data to a central control module in the vehicle-mounted system through a network provided by the vehicle, or the vehicle-mounted application programs mutually transmit the vehicle-mounted application data so as to provide corresponding services for users. The user may send an application request to the in-vehicle system based on a user interaction interface or a voice recognition function or a gesture recognition function of the in-vehicle system, or based on a user mobile terminal connected to bluetooth of the in-vehicle system.
After receiving an application request sent by a user, the vehicle-mounted system can call a corresponding functional module in the vehicle-mounted system based on the request to obtain vehicle-mounted application data, and the vehicle-mounted application data is transmitted to a control module in the vehicle-mounted system, so that the control module can control a corresponding application program based on the data to realize application functions required by the user. And when the data transmission process is carried out, acquiring performance data and vehicle-mounted application data of the vehicle-mounted system in the current period, determining a classification result corresponding to the performance data and the vehicle-mounted application data by utilizing a classification model according to the current performance data of the vehicle-mounted system and the vehicle-mounted application data to be transmitted in the current period, obtaining a target transmission parameter corresponding to the classification result, dynamically adjusting the transmission parameter of the current data transmission if the currently obtained target transmission parameter is inconsistent with the previous transmission parameter, and transmitting the vehicle-mounted application data to the control module according to the adjusted target transmission parameter. By adopting the data transmission method, the data transmission parameters can be adjusted in real time, so that the current data transmission parameters are more suitable for the running state of the current vehicle-mounted system, the data transmission efficiency can be improved, and the safety, reliability and instantaneity of data transmission can be ensured.
In the above embodiment, the vehicle-mounted system is disposed in the vehicle, and may also be disposed in the cloud, and when the user sends an application request to the vehicle-mounted system in the cloud through the vehicle terminal, to call each function module in the vehicle-mounted system in the cloud, obtain vehicle-mounted application data, and send the vehicle-mounted application data to the vehicle terminal, so as to implement an application function required by the user at the vehicle-mounted terminal. In the data transmission process, the data transmission method provided by the embodiment of the invention can also be adopted to collect the performance data and the vehicle-mounted application data of the vehicle-mounted system in the current period, determine the classification result corresponding to the performance data and the vehicle-mounted application data by using the classification model according to the current performance data of the vehicle-mounted system and the vehicle-mounted application data to be transmitted in the current period, and transmit the vehicle-mounted application data to the vehicle terminal according to the target transmission parameters corresponding to the classification result.
In addition, the data transmission method provided by the embodiment of the invention can also be executed in the cloud, a plurality of computing nodes (cloud servers) can be deployed in the cloud, and each computing node has processing resources such as computation, storage and the like. At the cloud, the data transmission service may be provided by a plurality of computing nodes, although one computing node may also provide the service.
It has been described in the embodiment shown in fig. 1 that the classification model may be used to classify the performance data and the vehicle application data of the vehicle-mounted system to obtain the classification result. To facilitate a specific implementation of determining the target transmission parameters based on the classification model, an exemplary implementation may be described in conjunction with fig. 2 and 3.
FIG. 2 is a flowchart for determining a target transmission parameter according to performance data and vehicle-mounted application data according to an embodiment of the present invention; FIG. 3 is a schematic diagram of determining a target transmission parameter according to performance data and vehicle-mounted application data according to an embodiment of the present invention; on the basis of the above embodiment, referring to fig. 2 to 3, the method comprises the following steps:
201. and combining the performance data of the vehicle-mounted system and the vehicle-mounted application data to obtain combined data.
202. And classifying the merged data by using the classification model to output classification results by the classification model. 203. And determining a target transmission parameter corresponding to the classification result.
In the embodiment of the invention, after the performance data and the vehicle-mounted application data of the vehicle-mounted system in the current period are acquired, the performance data and the vehicle-mounted application data of the vehicle-mounted system can be combined to obtain combined data.
Since the performance of the data transmission is not only related to the performance data of the vehicle-mounted system but also to the vehicle-mounted application data to be transmitted, it is not accurate to determine the target transmission parameters based on the performance data alone or the vehicle-mounted application data alone. In order to improve the accuracy of the target transmission parameters, the performance data of the vehicle-mounted system and the vehicle-mounted application data may be combined before the target transmission parameters are determined, so as to obtain combined data. The above-described merging process may be regarded as a splicing process. And inputting the combined data into a classification model to output a classification result by the classification model, and finally, determining a target transmission parameter corresponding to the combined data based on the corresponding relation between the classification result and the reference transmission parameter.
The classification model can learn the mapping relation between the combination of different performance data and vehicle-mounted application data and a plurality of classification results. The classification result and the transmission parameters also have a preset corresponding relation.
Alternatively, the classification model may determine the classification result by means of a softmax function. In particular, the formula can be utilizedAnd determining the probability that the merged data sample belongs to the ith reference classification result, and determining the reference classification result corresponding to the maximum probability as the reference classification result corresponding to the merged data. Wherein (1) >Refers to the ith reference classification junctionFruit of (Bu)>For the j-th reference classification result, +.>The probability corresponding to the ith reference classification result is indicated, and N is the number of types of the reference classification result.
In the embodiment of the invention, the performance data and the vehicle-mounted application data of the vehicle-mounted system acquired in real time are combined to obtain the combined data. And then, classifying the merged data by using a classification model so as to output a classification result by the classification model. And determining a target transmission parameter corresponding to the combined data based on the correspondence between the classification result and the reference transmission parameter. The embodiment of the invention considers the performance data of the vehicle-mounted system and the vehicle-mounted application data to be transmitted simultaneously so as to accurately determine the target transmission parameters, and the safe and reliable transmission of the vehicle-mounted application data can be ensured by using the accurate target transmission parameters.
In practice, the merged data may contain noise data that would affect the classification result of the classification model. Optionally, the combined data may also be denoised using a denoising model prior to entering the classification model. The noise reduction model may specifically comprise an encoding layer and a decoding layer.
Specifically, the combined data is input into a coding layer in the noise reduction model, so that the coding layer performs dimension reduction processing on the combined data to output hidden variables. Then, inputting the hidden variable output by the coding layer into a decoding layer in a noise reduction model, so that the decoding layer carries out dimension lifting processing on the hidden vector to output a noise reduction result of the combined data; and inputting the noise reduction result corresponding to the combined data into a classification model so as to output a classification result by the classification model.
Wherein the encoding layer is arranged to convert the input data into a representation of hidden variables, and to obtain a representation of the input data, which is typically a nonlinear affine transformation. The decoding layer maps the hidden variable identification obtained by the encoder to a feature vector of the input space, which is an affine transformation, optionally followed by a nonlinear transformation. The purpose of the conversion is to filter out noise data, leaving only the primary information of the original input data to obtain a valid vector representation of the original input data.
Alternatively, the noise reduction model may be embodied as a depth auto-encoder. At this time, the training process for the encoding layer and the decoding layer of the noise reduction model may be as shown in fig. 4. An alternative training pattern is shown in fig. 4a, where x is the original input data, and the decoding layer sets the value of the input layer node to 0 with a certain probability, so as to obtain data containing noise (i.e. corrupted data). Then use the data containing noise +.>The weighted data y is calculated, the data y is subjected to inverse weighted mapping to obtain reconstruction data z, error iteration is carried out on the reconstruction data z and the original data x, and an error function is obtained>This noise data is learned through these continuous learning training processes. An alternative training method is shown in FIG. 4b, in which x is the original input data, the data y is obtained after weighted mapping, and the value of the input layer node is set to 0 with a certain probability based on the data y, thereby obtaining data +. >. Then use the data containing noise +.>The weighted data y 'is calculated, the data y' is subjected to inverse weighted mapping to obtain reconstruction data z ', error iteration is carried out on the reconstruction data z' and the original data x, and an error function is obtained>This noise data is learned through these continuous learning training processes.
In summary, in the embodiment of the invention, by performing merging processing on performance data and vehicle-mounted application data of a vehicle-mounted system to obtain merged data, then performing noise reduction processing on the merged data by using a noise reduction model, inputting a noise reduction result into a classification model, outputting the classification result by the classification model, and determining a transmission parameter corresponding to the classification result as a target transmission parameter, the accuracy of the determined target transmission parameter can be improved, so that the determined target transmission parameter is more suitable for data transmission of the current vehicle-mounted system, and the best data transmission performance is obtained.
In practice, since the values of the vehicle-mounted application data and the performance data of the vehicle-mounted system contained in the combined data may have a large difference, the noise reduction effect may be affected by directly inputting the combined data into the noise reduction model. Then, in order to avoid the influence of the difference of the values of the data values on the noise reduction effect, the combined data is optionally normalized.
Where Normalization processing generally refers to a method of converting data features to the same scale (i.e., range of values), such as mapping data features into intervals of [0, 1] or [ -1, 1], or into a distribution subject to a mean of 0 and a variance of 1.
In an alternative embodiment, the combined data may be processed using a maximum-minimum normalization method. Wherein the minimum maximum normalization (Min-Max Normalization) normalizes the range of values of each feature to [0, 1] by scaling]Or [ -1, 1]Between them. In particular, the formula can be utilizedAnd calculating the normalized value of each combined data. Wherein (1)>For the j-th characteristic value, < >>Refers to->Normalized value->Refers to the eigenvalue of the ith sample, +.>Refers to->Minimum value of column in +.>Refers to->The maximum value of the column.
The above embodiments introduce a specific implementation process for processing the collected performance data and the vehicle application data of the vehicle-mounted system by using the noise reduction model and the classification model to obtain the target transmission parameters. In order to facilitate understanding of the whole data transmission process, the data transmission method provided by the invention is applied to various vehicle-mounted systems by way of illustration in conjunction with fig. 5. In particular application, reference is made to fig. 5.
Firstly, collecting performance data and vehicle-mounted application data of a vehicle-mounted system in a current period, and carrying out combination processing on the performance data and the vehicle-mounted application data to obtain combination data. Then, the hidden variable of the combined data, namely, the coding result, is obtained through the coding layer in the noise reduction model, the coding result is input into the decoding layer in the noise reduction model, the decoding layer carries out dimension-lifting processing, namely, noise reduction processing on the coding result, the noise reduction result of the combined data is output, and the noise reduction result is input into the classification model to output the classification result. Furthermore, the target transmission parameter can be obtained according to the mapping relation between the classification result and the reference transmission parameter. And finally, according to the target transmission parameters, transmitting the vehicle-mounted application data to a corresponding receiving end. The specific implementation process involved in the embodiment of the present invention may refer to the content in the foregoing embodiment, and will not be described herein again.
The above embodiment describes the determination of the target transmission parameters by means of a classification model. In addition, in this embodiment, a specific implementation manner of determining the target transmission parameter according to the performance data and the vehicle-mounted application data of the vehicle-mounted system is not limited, and a person skilled in the art may set the target transmission parameter according to specific application requirements and design requirements. In order to facilitate understanding of the working principle of the classification model, the embodiment of the invention also provides a classification model training method.
FIG. 6 is a schematic diagram of a training method for classification models according to an embodiment of the present invention; referring to fig. 6, the present embodiment provides a classification model training method, and the execution subject of the method may be a classification model training apparatus, and it is understood that the model training apparatus may be implemented as software, or a combination of software and hardware. Specifically, the method may include:
601. and acquiring a training sample of the vehicle-mounted system in the history period, wherein the training sample comprises historical performance data and historical vehicle-mounted application data of the vehicle-mounted system.
602. And combining the historical performance data and the historical vehicle-mounted application data to obtain a combined data sample.
603. And obtaining a reference classification result of the combined data sample, wherein the reference classification result corresponds to the reference transmission parameter.
604. The merged data samples are input into a classification model to train the classification model based on the loss calculation result between the predicted classification result and the reference classification result output by the classification model.
In training the classification model used in the embodiment shown in fig. 2, training samples of the vehicle-mounted system during the history period are first collected. The training samples comprise performance data of the historical vehicle-mounted system and historical vehicle-mounted application data. And then, combining the historical performance data and the historical vehicle-mounted application data to obtain a combined data sample. After the merged data sample is obtained, the merged data sample is input into a classification model to output a predictive classification result of the merged data sample from the classification model. Wherein the reference classification result corresponds to the reference transmission parameter. Finally, the classification model can be trained according to the loss calculation result between the prediction classification result and the reference classification result output by the classification model. The reference classification result of the merged data sample can be manually or automatically divided in advance.
In addition, the training process of the classification model adopts a continuous iteration mechanism, and continuously collects performance data and vehicle-mounted application data of the vehicle-mounted system, continuously trains and learns the classification model by utilizing the continuously updated performance data and vehicle-mounted application data of the vehicle-mounted system, updates network parameters in the model, and realizes more intelligent self-adaptive determination of the classification result, thereby self-adaptively determining Qos strategy and self-adaptive determination of the length of a data block for single transmission based on the classification result.
According to the embodiment of the invention, by means of the corresponding relation between the reference classification result and the transmission parameters, the historical performance data and the historical vehicle-mounted application data of the vehicle-mounted system are used as training data, and the classification model trained by taking the reference classification result as the supervision information can accurately learn the internal relation among the performance data, the vehicle-mounted application data and the transmission parameters, so that the accurate transmission parameters can be determined, and the reliability of data transmission is ensured.
The above embodiments mainly describe a specific implementation manner of determining the target transmission parameters according to the performance data and the vehicle-mounted application data, regardless of whether the vehicle-mounted system is abnormal or not. In practical application, when an abnormal condition occurs in the vehicle-mounted system, the abnormal condition is usually caused by the occurrence of a fault in a certain application program in the vehicle-mounted system, and when the abnormal condition of the vehicle-mounted system is sensed, the transmission parameters can be dynamically adjusted in time so as to ensure that the vehicle-mounted application data can be transmitted to the corresponding receiving end safely and reliably. In order to improve the accuracy of determining the target transmission parameters, the situation that the vehicle-mounted application data cannot be safely and reliably transmitted to the corresponding receiving end is avoided, and then the target transmission parameters can be determined by combining the fault types of the application programs.
Specifically, in an alternative embodiment, each application program in the vehicle-mounted system is monitored, and if a fault exists in a target application program in the vehicle-mounted system, a classification model is utilized to determine the fault type and performance data of the target application program and a classification result corresponding to the vehicle-mounted application data generated by the target application program.
Optionally, according to the fault type, the performance data and the vehicle-mounted application data of the target application program, the specific implementation manner of determining the target transmission parameter may be: and carrying out merging processing on the fault type, the performance data and the vehicle-mounted application data of the target application program to obtain merged data, carrying out sorting processing on the merged data by utilizing a sorting model to obtain a sorting result corresponding to the merged data, and obtaining the target transmission parameters corresponding to the merged data according to the mapping relation between the sorting result and the reference transmission parameters. The classification model learns the mapping relation between the combination of the data quantity of different fault types, performance data and vehicle-mounted application data and the classification result.
In this embodiment, when determining the transmission parameters, the situation that the application program fails is further considered on the basis of the performance data and the vehicle-mounted application data, so that the determined transmission parameters are more accurate, and the reliability of data transmission is ensured. In addition, the specific implementation process involved in the embodiment of the present invention may refer to the content in the foregoing embodiment, and will not be described herein.
Alternatively, the classification model training method provided by the embodiment shown in FIG. 6 may be used when the training samples include historical in-vehicle system performance data and historical in-vehicle application data. In correspondence with the above-described embodiments,
the method of fig. 7 may be performed to train the classification model when determining the target transmission parameters also taking into account the type of failure. Fig. 7 is a schematic diagram of another classification model training method according to an embodiment of the present invention, where the method specifically may include:
701. and acquiring a training sample of the vehicle-mounted system in the history period, wherein the training sample comprises the history performance data of the vehicle-mounted system, the history vehicle-mounted application data and the fault type of the history application program.
702. And merging the historical performance data, the historical vehicle-mounted application data and the fault type of the historical application program to obtain merged data samples.
703. And obtaining a reference classification result of the combined data sample, wherein the reference classification result corresponds to the reference transmission parameter.
704. The merged data samples are input into a classification model to train the classification model based on the loss calculation result between the predicted classification result and the reference classification result output by the classification model.
First, training samples of the on-board system during a history period are collected. The training samples comprise historical performance data of the vehicle-mounted system, historical vehicle-mounted application data and fault types of the historical application program. And then, merging the historical performance data, the historical vehicle-mounted application data and the fault type of the historical application program to obtain merged data samples. After the merged data sample is obtained, the merged data sample is input into a classification model to output a predictive classification result of the merged data sample from the classification model. Wherein the reference classification result corresponds to the reference transmission parameter. Finally, the classification model can be trained according to the loss calculation result between the prediction classification result and the reference classification result output by the classification model. The reference classification result of the merged data sample can be manually or automatically divided in advance.
In the embodiment of the invention, by means of the corresponding relation between the reference classification result and the transmission parameters, the historical performance data, the historical vehicle-mounted application data and the fault type of the historical application program of the vehicle-mounted system are taken as training data, and the classification model trained by taking the reference classification result as supervision information can accurately learn the internal relation among the performance data, the vehicle-mounted application data, the fault type of the application program and the transmission parameters, so that more accurate transmission parameters can be determined, and the reliability of data transmission is ensured.
In addition, the specific implementation process and the technical effects that can be achieved in the embodiment of the present invention may refer to the content in the embodiment shown in fig. 6, which is not described herein again.
In practice, when the classification model classifies, the performance data and the vehicle-mounted application data of the vehicle-mounted system are considered, and the data quantity of the vehicle-mounted application data can be considered.
At this time, the flow of the classification model training may be as shown in fig. 8. Fig. 8 is a schematic diagram of another method for training a classification model according to an embodiment of the present invention, where the method specifically may include:
801. and acquiring a training sample of the vehicle-mounted system in the history period, wherein the training sample comprises historical performance data, historical vehicle-mounted application data and data quantity of the historical vehicle-mounted application data of the vehicle-mounted system.
802. And combining the historical performance data, the historical vehicle-mounted application data and the data volume of the historical vehicle-mounted application data to obtain a combined data sample.
803. And obtaining a reference classification result of the combined data sample, wherein the reference classification result corresponds to the reference transmission parameter.
804. The merged data samples are input into a classification model to train the classification model based on the loss calculation result between the predicted classification result and the reference classification result output by the classification model.
First, training samples of the on-board system during a history period are collected. The training samples comprise historical performance data of the vehicle-mounted system, historical vehicle-mounted application data and data quantity of the historical vehicle-mounted application data. Wherein the data amount of the historical vehicle-mounted application data is determined according to the historical vehicle-mounted application data.
And then, combining the historical performance data, the historical vehicle-mounted application data and the data quantity of the historical vehicle-mounted application data to obtain a combined data sample. After the merged data sample is obtained, the merged data sample is input into a classification model to output a predictive classification result of the merged data sample from the classification model. Finally, the classification model can be trained according to the loss calculation result between the prediction classification result and the reference classification result output by the classification model. The reference classification result of the merged data sample can be manually or automatically divided in advance.
In an alternative embodiment, a specific implementation process for automatically determining the reference classification result of the merged data sample may include: acquiring a plurality of groups of transmission parameters and respective parameter identifiers of the plurality of groups of transmission parameters, wherein any group of transmission parameters comprises a preset quality of service Qos strategy and a preset data block length; according to the historical performance data in the combined data sample and the data quantity of the historical vehicle-mounted application data, determining a plurality of first value ranges corresponding to the historical performance data and a plurality of second value ranges corresponding to the data quantity, wherein the number of the first value ranges and the number of the second value ranges are the same as the number of the groups of the transmission parameters; respectively determining the value ranges of the historical performance data and the data quantity of the historical vehicle-mounted application data in any one of the combined data samples in the first value range and the second value range; according to the value range of each of the historical performance data and the data quantity of the historical vehicle-mounted application data in any merged data sample, the corresponding relation between the first value range and a preset Qos strategy, the corresponding relation between the second value range and the data block length, and determining a target Qos strategy and a target data block length corresponding to any merged data sample; and determining the parameter identification of the reference transmission parameters comprising the target Qos strategy and the target data block length as the reference classification result of any combined data sample.
For example, assume that there are 5 sets of transmission parameters, the 5 sets of transmission parameters are respectively a preset QOS policy of 1, a preset data block length of 1M, a preset QOS policy of 2, a preset data block length of 2M, a preset QOS policy of 3, a preset data block length of 3M, a preset QOS policy of 4, a preset data block length of 4M, a preset QOS policy of 5, a preset data block length of 5M, and parameter identifiers corresponding to the 5 sets of transmission parameters are respectively 1, 2, 3, 4, and 5. Meanwhile, 5 merged data samples are obtained, wherein performance data in each merged data sample are respectively 40% of cpu utilization rate, 50% of cpu utilization rate, 48% of cpu utilization rate, 30% of cpu utilization rate and 80% of cpu utilization rate, and vehicle-mounted application data in each merged data sample are respectively 10M, 40M, 20M and 60M.
According to the performance data in the 5 combined data samples, determining that the numerical range corresponding to the performance data is 30% -80%, and according to the number of groups of transmission parameters, dividing the performance data into 5 first value ranges, namely, 30% -40%, 41% -50%, 51% -60%, 61% -70% and 71% -80% of the 5 first value ranges respectively. According to the data quantity of the vehicle-mounted application data in the 5 combined data samples, determining that the numerical range corresponding to the data quantity is 10M-60M, and dividing the data quantity into 5 second value ranges according to the group number of the transmission parameters, namely, the 5 second value ranges are respectively 10M-20M, 21M-30M, 31M-40M, 41M-50M and 51M-60M.
Establishing a corresponding relation between each first value range and each preset Qos strategy, namely when the first value range is 30% -40%, the corresponding preset Qos strategy is 1; when the first value range is 41% -50%, the corresponding preset QOS strategy is 2; when the first value range is 51% -60%, the corresponding preset QOS strategy is 3; when the first value range is 61% -70%, the corresponding preset QOS strategy is 4; when the first value range is 71% -80%, the corresponding preset QOS strategy is 5. Simultaneously establishing a corresponding relation between each second value range and each preset data block length, namely when the second value range is 10M-20M, the corresponding preset data block length is 1M; when the second value range is 21M-30M, the length of the corresponding preset data block is 2M; when the second value range is 31M-40M, the corresponding preset data block length is 3M; when the second value range is 41M-50M, the length of the corresponding preset data block is 4M; and when the second value range is 51M-60M, the corresponding preset data block length is 5M.
Then, determining a value range of historical performance data and data quantity of historical vehicle-mounted application data in any combined data sample, wherein the value range of the performance data cpu in the first combined data sample is 30% -40%, the target Qos policy corresponding to the first combined data sample is 1, the value range of the data quantity of the vehicle-mounted application data in the first combined data sample is 10M-20M, and the target data block length corresponding to the first combined data sample is 1M. Since the methods for determining the target Qos policy and the target data block length of each combined data sample are consistent, the determination process of other combined data samples can refer to the above process.
And then, according to the target Qos strategy and the target data block length corresponding to each merged data sample, determining the parameter identification corresponding to each merged data sample, and determining the parameter identification containing the reference transmission parameters of the target Qos strategy and the target data block length as the reference classification result of any merged data sample. The preset QOS policy is 1, the preset data block length is 1M, and the corresponding reference identifier is 1, and then the parameter identifier corresponding to the first merged data sample is 1, so that the reference classification result corresponding to the first merged data sample is 1. Since the methods for determining the parameter identification of each merged data sample are consistent, the determination process of other merged data samples can refer to the above process.
The method for determining the reference classification result can realize automatic labeling of the sample, thereby reducing labor cost and time cost of classification model training and improving training efficiency of the classification model.
Optionally, the normalized combined data samples may also be input into a classification model to further complete training of the classification model.
After the classification model is trained, the classification model can be deployed into a vehicle-mounted system, and when data transmission is carried out, a classification result can be obtained by using the trained classification model, a target transmission parameter is determined based on a mapping relation between the classification result and a reference transmission parameter, and the data transmission parameter is automatically adjusted so as to realize self-adaptive data transmission. According to the embodiment of the invention, by means of the corresponding relation between the reference classification result and the transmission parameters, the historical performance data, the historical vehicle-mounted application data and the data quantity of the historical vehicle-mounted application data of the vehicle-mounted system are used as training data, and the classification model trained by taking the reference classification result as the supervision information can learn the internal relation among the performance data, the vehicle-mounted application data, the data quantity of the vehicle-mounted application data and the transmission parameters more accurately, so that the more accurate transmission parameters can be finally determined, and the reliability of data transmission is ensured.
In addition, the specific implementation procedure and technical effects of the steps in this embodiment are similar to those in the embodiment, and reference is specifically made to the above description, which is not repeated here.
Optionally, the classification model may also use the performance data and the vehicle-mounted application data of the vehicle-mounted system, the data amount of the vehicle-mounted application data, and the fault type of the application program to classify at the same time, and obtain the target transmission parameter. At this time, optionally, the training sample may include both the historical performance data and the historical vehicle-mounted application data of the vehicle-mounted system, and the failure type of the historical application program and the data amount of the historical vehicle-mounted application data. At this time, the specific process and the technical effects that can be achieved for the training of the classification model are similar to those of the above embodiment, and will not be described herein.
An in-vehicle system of one or more embodiments of the present invention will be described in detail below. Those skilled in the art will appreciate that these on-board systems may be configured using commercially available hardware components through the steps taught by the present solution.
Fig. 9 is a schematic structural diagram of a vehicle-mounted system according to an embodiment of the present invention, as shown in fig. 9, where the vehicle-mounted system includes: the device comprises an acquisition module 11, a determination module 12 and a transmission module 13.
And the acquisition module 11 is used for acquiring the performance data and the vehicle-mounted application data of the vehicle-mounted system in the current period.
And the determining module 12 is configured to determine a classification result corresponding to the performance data and the vehicle-mounted application data by using a classification model, where the classification result is reflected in the current period, and the vehicle-mounted system transmits the capability level of the vehicle-mounted application data.
And the transmission module 13 is configured to transmit the vehicle-mounted application data to a corresponding receiving end according to a target transmission parameter corresponding to the classification result, where the target transmission parameter includes a quality of service Qos policy and a data block length of a single transmission.
Alternatively, the determining module 12 may specifically be configured to: combining the performance data and the vehicle-mounted application data to obtain combined data; the merged data is input into the classification model to output classification results from the classification model.
Optionally, the vehicle-mounted system further comprises: a noise reduction module 14, configured to input the combined data into the noise reduction model, so that a coding layer in the noise reduction model performs a dimension reduction process on the combined data to output hidden variables; inputting the hidden variable into a decoding layer in the noise reduction model, so that the decoding layer carries out dimension lifting processing on the hidden variable to output a noise reduction result of the combined data.
Wherein the noise reduction model comprises a depth auto encoder.
The determining module 12 may specifically be configured to: and inputting the noise reduction result into the classification model. Optionally, the determining module 12 may be further specifically configured to: if the application program generating the vehicle-mounted application data in the vehicle-mounted system is detected to have faults, determining the fault type of the application program, the performance data and the classification result corresponding to the vehicle-mounted application data by using the classification model.
Optionally, the determining module 12 is configured to determine, using a classification model, the performance data, the vehicle-mounted application data, and a classification result corresponding to a data amount of the vehicle-mounted application data.
Optionally, the apparatus may further include a training module 15, which may be specifically configured to: collecting training samples of the vehicle-mounted system in a history period, wherein the training samples comprise historical performance data and historical vehicle-mounted application data of the vehicle-mounted system; combining the historical performance data and the historical vehicle-mounted application data to obtain a combined data sample; obtaining a reference classification result of the combined data sample, wherein the reference classification result corresponds to a reference transmission parameter; the merged data samples are input into the classification model to train the classification model based on a loss calculation result between a predicted classification result output by the classification model and the reference classification result.
Optionally, the training sample further includes a data amount of historical vehicle-mounted application data;
the training module 15 may be further configured to: acquiring a plurality of groups of transmission parameters and respective parameter identifiers of the plurality of groups of transmission parameters, wherein any group of transmission parameters comprises a preset Qos strategy and a preset data block length; determining a plurality of first value ranges corresponding to the performance data and a plurality of second value ranges corresponding to the data quantity according to the performance data in the combined data sample and the data quantity of the vehicle-mounted application data, wherein the number of the first value ranges and the number of the second value ranges are the same as the number of the groups of the transmission parameters; determining a target Qos policy and a target data block length corresponding to any sample according to a value range to which the data quantity of the performance data and the vehicle-mounted application data corresponding to any sample in the combined data samples respectively belong, a corresponding relation between the first value range and a preset Qos policy, and a corresponding relation between the second value range and the data block length; and determining a parameter identification of the reference transmission parameter comprising the target Qos strategy and a target data block length as a reference classification result of the any sample.
The apparatus shown in fig. 9 may perform the steps in the data transmission method in the foregoing embodiment, and the detailed performing process and technical effects are referred to the description in the foregoing embodiment, which is not repeated herein.
An embodiment of the present invention further provides an electronic device, as shown in fig. 10, where the electronic device may include: a processor 21, a memory 22, a communication interface 23. Wherein the memory 22 has stored thereon executable code which, when executed by the processor 21, causes the processor 21 to implement the data transmission method as in the previous embodiments.
In addition, embodiments of the present invention provide a non-transitory machine-readable storage medium having stored thereon executable code, which when executed by a processor of an electronic device, causes the processor to at least implement a data transmission method as provided in the previous embodiments.
The apparatus embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separate. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by adding necessary general purpose hardware platforms, or may be implemented by a combination of hardware and software. Based on such understanding, the foregoing aspects, in essence and portions contributing to the art, may be embodied in the form of a computer program product, which may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (10)

1. A data transmission method, comprising:
acquiring performance data and vehicle-mounted application data of a vehicle-mounted system in a current period;
inputting the combined data obtained by combining the performance data and the vehicle-mounted application data into a noise reduction model to obtain a noise reduction result of the combined data;
inputting the noise reduction result into a classification model to output a classification result corresponding to the combined data by the classification model, wherein the classification result is reflected in the current period, and the vehicle-mounted system transmits the capability level of the vehicle-mounted application data;
and transmitting the vehicle-mounted application data to a corresponding receiving end according to a target transmission parameter corresponding to the classification result, wherein the target transmission parameter comprises a quality of service Qos strategy and a data block length of single transmission.
2. The method of claim 1, wherein inputting the combined data obtained by combining the performance data and the vehicle-mounted application data into a noise reduction model to obtain a noise reduction result of the combined data comprises:
inputting the combined data into a noise reduction model, and performing dimension reduction processing on the combined data by a coding layer in the noise reduction model to output hidden variables;
Inputting the hidden variable into a decoding layer in the noise reduction model, so that the decoding layer carries out dimension lifting processing on the hidden variable to output a noise reduction result of the combined data.
3. The method of claim 2, wherein the noise reduction model comprises a depth auto encoder.
4. The method according to claim 1, wherein inputting the combined data obtained by combining the performance data and the vehicle-mounted application data into a noise reduction model to obtain a noise reduction result of the combined data, includes:
if the application program generating the vehicle-mounted application data in the vehicle-mounted system is detected to have faults, the combination result obtained by combining the fault type of the application program, the performance data and the vehicle-mounted application data is input into the noise reduction model, so that the noise reduction result of the combination data is obtained.
5. The method according to claim 1, wherein inputting the combined data obtained by combining the performance data and the vehicle-mounted application data into a noise reduction model to obtain a noise reduction result of the combined data, includes:
and inputting the combined data obtained by combining the performance data, the vehicle-mounted application data and the data quantity of the vehicle-mounted application data into a noise reduction model to obtain a noise reduction result of the combined data.
6. The method according to claim 2, wherein the method further comprises:
collecting training samples of the vehicle-mounted system in a history period, wherein the training samples comprise historical performance data and historical vehicle-mounted application data of the vehicle-mounted system;
combining the historical performance data and the historical vehicle-mounted application data to obtain a combined data sample;
obtaining a reference classification result of the combined data sample, wherein the reference classification result corresponds to a reference transmission parameter;
the merged data samples are input into the classification model to train the classification model based on a loss calculation result between a predicted classification result output by the classification model and the reference classification result.
7. The method of claim 6, wherein the training sample further comprises a data volume of historical vehicle-mounted application data;
the obtaining the reference classification result of the merged data sample includes:
acquiring a plurality of groups of transmission parameters and respective parameter identifiers of the plurality of groups of transmission parameters, wherein any group of transmission parameters comprises a preset Qos strategy and a preset data block length;
according to the historical performance data in the combined data sample and the data quantity of the historical vehicle-mounted application data, determining a plurality of first value ranges corresponding to the historical performance data and a plurality of second value ranges corresponding to the data quantity, wherein the number of the first value ranges and the number of the second value ranges are the same as the number of the groups of the transmission parameters;
Determining a target Qos policy and a target data block length corresponding to any sample according to a value range to which data amounts of historical performance data and historical vehicle-mounted application data corresponding to any sample in the merged data sample respectively belong, a corresponding relation between the first value range and a preset Qos policy, and a corresponding relation between the second value range and the data block length;
and determining a parameter identification of the reference transmission parameter comprising the target Qos strategy and a target data block length as a reference classification result of the any sample.
8. A vehicle-mounted system, comprising:
the acquisition module is used for acquiring the performance data and the vehicle-mounted application data of the vehicle-mounted system in the current period;
the determining module is used for inputting the combined data obtained by combining the performance data and the vehicle-mounted application data into a noise reduction model so as to obtain a noise reduction result of the combined data; inputting the noise reduction result into a classification model to output a classification result corresponding to the combined data by the classification model, wherein the classification result is reflected in the current period, and the vehicle-mounted system transmits the capability level of the vehicle-mounted application data;
And the transmission module is used for transmitting the vehicle-mounted application data to a corresponding receiving end according to a target transmission parameter corresponding to the classification result, wherein the target transmission parameter comprises a quality of service Qos strategy and a data block length of single transmission.
9. An electronic device, comprising: a memory, a processor, a communication interface; wherein the memory has stored thereon executable code which, when executed by the processor, causes the processor to perform the data transmission method according to any of claims 1 to 7.
10. A non-transitory machine-readable storage medium having stored thereon executable code which, when executed by a processor of an electronic device, causes the processor to perform the data transmission method of any of claims 1 to 7.
CN202310822509.3A 2023-07-05 2023-07-05 Data transmission method, vehicle-mounted system, device and storage medium Active CN116567719B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310822509.3A CN116567719B (en) 2023-07-05 2023-07-05 Data transmission method, vehicle-mounted system, device and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310822509.3A CN116567719B (en) 2023-07-05 2023-07-05 Data transmission method, vehicle-mounted system, device and storage medium

Publications (2)

Publication Number Publication Date
CN116567719A CN116567719A (en) 2023-08-08
CN116567719B true CN116567719B (en) 2023-11-10

Family

ID=87488251

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310822509.3A Active CN116567719B (en) 2023-07-05 2023-07-05 Data transmission method, vehicle-mounted system, device and storage medium

Country Status (1)

Country Link
CN (1) CN116567719B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109800875A (en) * 2019-01-08 2019-05-24 华南理工大学 Chemical industry fault detection method based on particle group optimizing and noise reduction sparse coding machine
CN113158544A (en) * 2021-02-03 2021-07-23 大连理工大学 Edge pre-caching strategy based on federal learning under vehicle-mounted content center network
CN113992707A (en) * 2021-09-18 2022-01-28 华人运通(上海)云计算科技有限公司 Data transmission method and device, vehicle-mounted Ethernet equipment and storage medium
CN115661549A (en) * 2022-11-21 2023-01-31 南京码极客科技有限公司 Fine-grained classification denoising training method based on prediction confidence

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113128781B (en) * 2021-04-30 2021-12-10 大连理工大学 Distributed industrial energy operation optimization platform for automatically constructing intelligent model and algorithm

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109800875A (en) * 2019-01-08 2019-05-24 华南理工大学 Chemical industry fault detection method based on particle group optimizing and noise reduction sparse coding machine
CN113158544A (en) * 2021-02-03 2021-07-23 大连理工大学 Edge pre-caching strategy based on federal learning under vehicle-mounted content center network
CN113992707A (en) * 2021-09-18 2022-01-28 华人运通(上海)云计算科技有限公司 Data transmission method and device, vehicle-mounted Ethernet equipment and storage medium
CN115661549A (en) * 2022-11-21 2023-01-31 南京码极客科技有限公司 Fine-grained classification denoising training method based on prediction confidence

Also Published As

Publication number Publication date
CN116567719A (en) 2023-08-08

Similar Documents

Publication Publication Date Title
CN112435469A (en) Vehicle early warning control method and device, computer readable medium and electronic equipment
US11620200B2 (en) System and method for integration testing
CN113688957A (en) Target detection method, device, equipment and medium based on multi-model fusion
CN116453343A (en) Intelligent traffic signal control optimization algorithm, software and system based on flow prediction in intelligent networking environment
CN113096405B (en) Construction method of prediction model, and vehicle accident prediction method and device
CN116567719B (en) Data transmission method, vehicle-mounted system, device and storage medium
CN108770014B (en) Calculation evaluation method, system and device of network server and readable storage medium
CN114332520B (en) Abnormal driving behavior recognition model construction method based on deep learning
CN114915940B (en) Vehicle-road communication link matching method, system, equipment and medium
CN111027859A (en) Driving risk prevention method and system based on motor vehicle state monitoring data mining
CN115049151A (en) Navigation route recommendation method, device, equipment and storage medium
CN115631457A (en) Man-machine cooperation abnormity detection method and system in building construction monitoring video
CN112801156B (en) Business big data acquisition method and server for artificial intelligence machine learning
CN112200052B (en) Track deviation recognition and vehicle running analysis method, device, equipment and medium
CN114418189A (en) Water quality grade prediction method, system, terminal device and storage medium
CN116229128A (en) Clustering method and device for entity images, electronic equipment and storage medium
CN115694937B (en) Method for training malicious flow detection system by constructing virtual abnormal flow
CN117132958B (en) Road element identification method and related device
Hong et al. SLICE-based Trustworthiness Analysis system
CN117035297B (en) Campus intelligent task allocation method and system based on big data
RU2724596C1 (en) Method, apparatus, a central device and a system for recognizing a distribution shift in the distribution of data and / or features of input data
CN115271001B (en) Vehicle driving condition identification method and device, vehicle and storage medium
CN111401224B (en) Target detection method and device and electronic equipment
CN114446042B (en) Method, device, equipment and storage medium for early warning traffic accidents
CN112135172A (en) Weak network-based audio and video processing method and system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant