Internet of vehicles abnormal signal detection and data processing method, system and device

Abstract

The invention discloses an Internet of Vehicles abnormal signal detection and data processing method, system and device. The method comprises the steps that S1, Internet of Vehicles terminal equipment collects signal data and sends the signal data to a big data center; s2, streaming data processing: after the data in the step S1 is generated in real time and continuously arrives, dividing time windows according to a certain rule for continuous processing, and enabling the streaming data to enter a pre-configured streaming signal processing module for processing; step S3, batch data processing: the data is uniformly processed after being accumulated to a certain amount, generally one hour or one day; and the batch data enters a pre-configured batch signal processing module to be processed. According to the invention, the big data calculation capability of the Internet of Vehicles cloud is fully utilized, the effective combination of streaming calculation and batch calculation is realized, and the problems of missed calculation, wrong calculation and the like of vehicle anomaly detection caused by data acquisition interruption, acquisition delay, disordered acquisition sequence and the like of the Internet of Vehicles are solved.

Description

Technical Field

[0001] The present invention belongs to the field of Internet of Vehicles big data, and in particular, relates to an Internet of Vehicles abnormal signal detection and data processing method, system and device. Background Art

[0002] Smart cars are becoming more and more popular, and vehicle driving safety is the focus of people's attention. The Internet of Vehicles uses data collected by various sensors on the vehicle body to perform real-time analysis and real-time calculation, discover bad driving habits or predict potential faults, and then send an alarm to the user. However, in actual operation, when a vehicle enters a tunnel, underground garage or encounters other network anomalies, data collection interruptions, collection delays, and chaotic collection order often occur, resulting in problems such as omissions or miscalculations in the algorithms and models related to vehicle anomaly detection. In addition, the communication between vehicles (V2V), vehicles and infrastructure (V2I), and vehicles and the cloud (V2C) is becoming more and more frequent and complex. The Internet of Vehicles system collects and transmits a large amount of vehicle operation data, environmental data, and user behavior data in real time through sensors, communication modules, and computing platforms. These data include not only basic information such as vehicle speed, acceleration, and location, but may also involve complex information such as engine status, battery status, and driving behavior. Real-time processing and anomaly detection of these data are of great significance to ensuring vehicle safety, improving traffic efficiency, and optimizing user experience.

[0003] However, the processing of Internet of Vehicles data faces the following major challenges:

[0004] Large data volume and high real-time requirements: IoV terminal devices (such as vehicle sensors, GPS modules, etc.) generate a large amount of data every second, which needs to be transmitted to the cloud in real time for processing. Traditional batch data processing methods cannot meet real-time requirements, and simple streaming processing may lead to inaccurate detection due to data delay or loss.

[0005] Data delay and disorder problem: Due to the instability of network transmission, IoV data may be delayed or arrive out of order during transmission. How to correctly handle delayed and disordered data in stream processing and ensure the accuracy of anomaly detection is an important technical challenge.

[0006] Complexity of abnormal signals: Abnormal signals in the Internet of Vehicles may appear as transient abnormalities (such as sudden failure of a sensor) or persistent abnormalities (such as long-term failure of a component). Transient abnormalities can be detected in real time through streaming processing, while persistent abnormalities require batch analysis combined with historical data. How to effectively combine streaming processing and batch processing to achieve accurate detection of different types of abnormal signals is a key issue in Internet of Vehicles data processing.

[0007] Optimal utilization of computing resources: The processing of Internet of Vehicles data needs to be carried out in the cloud. How to reasonably allocate computing resources to ensure the efficient operation of streaming and batch processing while reducing computing costs is an important consideration in the design of Internet of Vehicles systems.

[0008] Disadvantages of existing technology

[0009] Existing IoV data processing methods usually use a single stream processing or batch processing method, which is difficult to meet the requirements of real-time and accuracy at the same time. Specifically:

[0010] Single stream processing: Although it can process data in real time, its ability to process delayed or out-of-order data is limited, which can easily lead to missed or false detection of abnormal signals.

[0011] Single batch processing: Although it can process a large amount of historical data, it cannot meet the real-time requirements and cannot respond to instantaneous abnormal signals in time.

[0012] In addition, existing anomaly detection models usually only target specific types of abnormal signals and lack the ability to comprehensively process different types of abnormal signals. Therefore, how to design an IoV abnormal signal detection method that can effectively combine streaming processing and batch processing has become an important research direction in the current technical field. Summary of the invention

[0013] The purpose of the invention is to provide a method, system and device for abnormal signal detection and data processing in the Internet of Vehicles, so as to solve the problems of data collection interruption, collection delay, collection sequence confusion, etc. in the Internet of Vehicles, which lead to the inability to accurately calculate the vehicle abnormal detection or related data analysis models.

[0014] Technical solution, a method, system and device for detecting abnormal signals and processing data in the Internet of Vehicles, based on the big data computing capability of the Internet of Vehicles cloud, realizes the effective combination of real-time computing of streaming data and offline computing of batch data, including the following steps:

[0015] Step S1, the Internet of Vehicles terminal device collects signal data and sends it to the big data center;

[0016] Step S2, streaming data processing: After the data of step S1 is generated in real time and arrives continuously, it is processed continuously in time windows divided according to certain rules, and the streaming data enters the pre-configured streaming signal processing module for processing; the specific steps are as follows:

[0017] Step S21, first define two times: one is the time when data is generated, called event time, whose value is automatically generated when data is collected and is sent out at the same time as the signal value measured by the collection device; the other is the time when data arrives at the big data cloud signal processing module, called reception time;

[0018] Step S21, first sorting by receiving time, then dividing the data into time windows according to the receiving time, and the data is processed continuously according to the time windows;

[0019] Step S22: re-order the data by event time within the time window of step S21, compare the data receiving time with the event time, and determine the delayed data; if the delay time does not exceed a certain waiting time, and the waiting time does not exceed one tenth of the above time window, the data is processed in the current time window, otherwise the delayed data is discarded;

[0020] Step S23: Calculate the data index in the above-divided time window according to the pre-configured detection model to determine whether it is an abnormal signal;

[0021] Step S24: if an abnormal signal is found, trigger the downstream pre-configured abnormal message construction module to generate an abnormal message, and send the abnormal message through the pre-configured abnormal message sending module, and save the abnormal message to the data warehouse;

[0022] Step S25: The streaming signal processing process is continuously performed while the collected data is sent to the big data center;

[0023] Step S3, batch data processing: after a certain amount of data is accumulated, it is processed uniformly, usually for one hour or one day; the batch data enters the pre-configured batch signal processing module for processing; the specific steps are as follows:

[0024] Step S31, first define two times: one is the time when data is generated, called event time, whose value is automatically generated when data is collected and is sent out at the same time as the signal value measured by the collection device; the other is the time when data arrives at the big data cloud signal processing module, called reception time;

[0025] Step S32: After the batch data accumulates to a certain amount, it is partitioned and stored in the data warehouse according to the receiving time, with the partitions being stored in units of hours or days;

[0026] Step S33: repartition and store the batch data according to the event time, and read the data partitioned by the receiving time for multiple days to obtain a complete set of data partitioned by the event time;

[0027] Step S34: Calculate indicators based on the re-partitioned data and use a pre-configured detection model to detect abnormal signals. For continuous abnormal signals across partitions, it is necessary to integrate data from multiple partitions for calculation;

[0028] Step S35: compare the abnormal signals detected by batch computing with the abnormal signals detected by streaming computing stored in the data warehouse, and use the batch computing as the standard to filter out abnormal signals that are not found or not accurately calculated by streaming computing;

[0029] Step S36: using the screened abnormal signal to trigger the downstream pre-configured abnormal message construction module to generate an abnormal message, and sending the abnormal message through the pre-configured abnormal message sending module, and saving these abnormal messages to the data warehouse;

[0030] Step S37, batch signal processing is performed once every hour or day after the collected data is sent to the big data center and saved in the data warehouse.

[0031] According to a further improvement of the present invention, the preconfigured detection model can select input variables, set detection models, and set output indicators from thousands of Internet of Vehicles signals according to business needs; the model is compiled into a pickle file to define and execute judgment rules, and can also support the compilation of complex machine learning models, while supporting cross-platform operation, ensuring that both streaming signal processing modules and batch signal processing modules are easy to parse and calculate.

[0032] According to a further improvement of the present invention, the preconfigured exception message construction module mainly performs the following operations:

[0033] Read the IoV signal interpretation table from the data warehouse and convert IoV signal codes that are difficult for general users to identify into Chinese content;

[0034] The exception message content is written in markdown language to ensure that the structure of the output message has good readability.

[0035] According to a further improvement of the present invention, the abnormal message sending module is a standardized API data interface. By triggering the interface, the abnormal message can be sent to the APP terminal, Web monitoring page, DingTalk message or WeChat service account as needed.

[0036] Beneficial effects: The present invention makes full use of the big data computing capabilities of the Internet of Vehicles cloud, realizes the effective combination of streaming computing and batch computing, and solves the problems of missed calculations and wrong calculations of vehicle anomaly detection when data collection is interrupted, delayed, or disordered in the order of collection in the Internet of Vehicles. The solution and system ensure the real-time calculation of the anomaly detection model as well as the accuracy and completeness.

[0037] In practice, this solution and system have played a positive role in improving driving safety, reducing the risk of traffic accidents, detecting potential faults in advance, and extending the service life of vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is the overall flow chart of abnormal signal detection in Internet of Vehicles of the present invention.

[0039] Figure 2It is a flow chart of the streaming signal processing module of the present invention.

[0040] Figure 3 It is a flow chart of the batch signal processing module of the present invention. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution in the embodiment of the present invention is clearly and completely described. Obviously, the described embodiment is a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] like Figure 1 As shown, a method, system and device for detecting abnormal signals and processing data in an Internet of Vehicles (IoV) is provided, which realizes an effective combination of real-time computing of streaming data and offline computing of batch data based on the big data computing capability of the IoV cloud, and includes the following steps:

[0043] Step S1, the Internet of Vehicles terminal device collects signal data and sends it to the big data center;

[0044] Step S2, streaming data processing (such as Figure 2 As shown): After the data of step S1 is generated in real time and arrives continuously, it is processed continuously in time windows divided according to certain rules, and the streaming data enters the pre-configured streaming signal processing module for processing; the specific steps are as follows:

[0045] Step S21, first define two times: one is the time when data is generated, called event time, whose value is automatically generated when data is collected and is sent out at the same time as the signal value measured by the collection device; the other is the time when data arrives at the big data cloud signal processing module, called reception time;

[0046] Step S21, first sorting by receiving time, then dividing the data into time windows according to the receiving time, and the data is processed continuously according to the time windows;

[0047] Step S22: re-order the data by event time within the time window of step S21, compare the data receiving time with the event time, and determine the delayed data; if the delay time does not exceed a certain waiting time, and the waiting time does not exceed one tenth of the above time window, the data is processed in the current time window, otherwise the delayed data is discarded;

[0048] Step S23: Calculate the data index in the above-divided time window according to the pre-configured detection model to determine whether it is an abnormal signal;

[0049] Step S24: if an abnormal signal is found, trigger the downstream pre-configured abnormal message construction module to generate an abnormal message, and send the abnormal message through the pre-configured abnormal message sending module, and save the abnormal message to the data warehouse;

[0050] Step S25: The streaming signal processing process is continuously performed while the collected data is sent to the big data center;

[0051] Step S3, batch data processing (such as Figure 3 As shown in the figure): After a certain amount of data is accumulated, it is processed uniformly, usually for one hour or one day; batch data enters the pre-configured batch signal processing module for processing; the specific steps are as follows:

[0052] Step S31, first define two times: one is the time when data is generated, called event time, whose value is automatically generated when data is collected and is sent out at the same time as the signal value measured by the collection device; the other is the time when data arrives at the big data cloud signal processing module, called reception time;

[0053] Step S32: After the batch data accumulates to a certain amount, it is partitioned and stored in the data warehouse according to the receiving time, with the partitions being stored in units of hours or days;

[0054] Step S33: repartition and store the batch data according to the event time, and read the data partitioned by the receiving time for multiple days to obtain a complete set of data partitioned by the event time;

[0055] Step S34: Calculate indicators based on the re-partitioned data and use a pre-configured detection model to detect abnormal signals. For continuous abnormal signals across partitions, it is necessary to integrate data from multiple partitions for calculation;

[0056] Step S35: compare the abnormal signals detected by batch computing with the abnormal signals detected by streaming computing stored in the data warehouse, and use the batch computing as the standard to filter out abnormal signals that are not found or not accurately calculated by streaming computing;

[0057] Step S36: using the screened abnormal signal to trigger the downstream pre-configured abnormal message construction module to generate an abnormal message, and sending the abnormal message through the pre-configured abnormal message sending module, and saving these abnormal messages to the data warehouse;

[0058] Step S37, batch signal processing is performed once every hour or day after the collected data is sent to the big data center and saved in the data warehouse.

[0059] The present invention provides an efficient and accurate method, system and device for detecting and processing abnormal signals in the Internet of Vehicles by combining streaming data processing and batch data processing. Its technical effects are mainly reflected in the following aspects: real-time detection of instantaneous abnormal signals, timely generation and sending of abnormal messages; in-depth analysis of historical data, detection of continuous abnormal signals; effective processing of delayed and disordered data, reducing false detection rate and missed detection rate; optimizing computing resource utilization, improving system processing efficiency; providing flexible and scalable modular design to adapt to different application scenarios. Specifically:

[0060] Real-time streaming processing: Through the streaming data processing module, the signal data collected by the Internet of Vehicles terminal devices can be transmitted to the cloud in real time and processed within the pre-configured time window. This real-time processing mechanism can quickly detect instantaneous abnormal signals (such as sudden sensor failures, vehicle collisions, etc.) and trigger the generation and sending of abnormal messages in a timely manner, ensuring that relevant parties can take countermeasures in the first place.

[0061] Batch processing accuracy: Through the batch data processing module, the system can conduct in-depth analysis of the accumulated historical data and detect persistent abnormal signals (such as long-term failures of vehicle components, abnormal patterns of driving behavior, etc.). Batch processing can integrate multiple days of data to avoid false detection or missed detection due to the limitations of single streaming processing, thereby improving the accuracy of anomaly detection.

[0062] Distinguishing between event time and reception time: By defining event time (data generation time) and reception time (data arrival time at the cloud), the present invention can effectively identify and process delayed or disordered data caused by network transmission.

[0063] Flexible division of time windows: In streaming processing, the system divides the time windows by the receiving time and reorders the data by event time within the window. For data whose delay does not exceed a certain waiting time, the system will still process it within the current time window to ensure the integrity and real-time nature of the data; for data with timeout delay, the system will discard it to avoid excessive occupation of system resources and ensure computing efficiency.

[0064] Repartitioning of batch data: In batch processing, after the system stores data by receiving time partition, it further repartitions it by event time to ensure that continuous abnormal signals across partitions can be comprehensively detected, avoiding detection omissions due to discontinuous data partitions and avoiding false detections due to data disorder.

[0065] Real-time detection of transient abnormal signals: Through the streaming processing module, the system can detect transient abnormal signals (such as sudden changes in sensor data, sudden braking of the vehicle, etc.) in real time and immediately trigger the generation and sending of abnormal messages.

[0066] In-depth analysis of persistent abnormal signals: Through the batch processing module, the system can conduct a comprehensive analysis of historical data and detect persistent abnormal signals (such as long-term engine overheating, continuous battery performance degradation, etc.). This combination of streaming and batch processing can cover various types of abnormal signals in the Internet of Vehicles and ensure comprehensive detection.

[0067] Comparison and screening of abnormal signals: The system will compare the abnormal signals detected by batch processing with those detected by streaming processing, and filter out abnormal signals that were not found or not accurately calculated by streaming processing based on the batch processing results. This mechanism further improves the reliability of anomaly detection.

[0068] Low latency and high throughput of stream processing: The stream processing module can achieve high throughput while ensuring low latency by dividing time windows and continuously processing data, making full use of cloud computing resources.

[0069] Offline optimization of batch processing: The batch processing module processes the data uniformly after a certain amount of data has accumulated (such as every hour or every day), avoiding excessive occupation of system resources by real-time computing. At the same time, batch processing uses partition storage and re-partitioning to optimize the efficiency of data reading and computing.

[0070] Dynamic balance of resource allocation: Through the combination of streaming processing and batch processing, the system can dynamically allocate resources according to the amount of data and computing needs, ensuring efficient use of computing resources and reducing overall computing costs.

[0071] Generation of real-time exception messages: In streaming processing, once an abnormal signal is detected, the system will immediately trigger the exception message construction module to generate an exception message, and send it to relevant parties (such as drivers, vehicle maintenance centers, traffic management departments, etc.) through the exception message sending module.

[0072] Supplementation of batch exception messages: In batch processing, the system will filter out exception signals that are not found or not accurately calculated by stream processing, and generate corresponding exception messages for supplementary sending. This dual guarantee mechanism ensures the integrity and accuracy of exception messages.

[0073] Persistent storage of exception messages: All detected exception messages will be saved in the data warehouse for subsequent query, analysis and tracing, providing data support for vehicle maintenance, accident analysis, etc.

[0074] Modular design: The streaming processing module and batch processing module of the present invention are pre-configured, and users can flexibly configure parameters such as detection model, time window size, partition storage strategy, etc. according to actual needs to adapt to the application requirements of different scenarios.

[0075] Support large-scale data processing: Based on the big data computing capabilities of the Internet of Vehicles cloud, the system can process massive amounts of Internet of Vehicles data and support the simultaneous access and data processing of large-scale vehicles.

[0076] Compatible with multiple anomaly detection models: The system supports pre-configured detection models. Users can select or develop different anomaly detection algorithms according to specific needs, which are suitable for different types of Internet of Vehicles application scenarios.

[0077] Real-time verification of streaming processing: Through data reordering and delayed data processing mechanisms within the time window, the system can effectively reduce the false detection rate caused by data delay or disorder.

[0078] Deep verification of batch processing: Through comprehensive analysis of historical data through batch processing, the system can detect abnormal signals that may be missed by streaming processing and reduce the missed detection rate.

[0079] Comparison and screening of abnormal signals: By comparing the results of streaming processing and batch processing, the system can further screen out accurate abnormal signals to avoid false detection and missed detection.

[0080] These technical effects make the present invention have important application value in the field of abnormal signal detection and data processing in the Internet of Vehicles, and can significantly improve the security, reliability and user experience of the Internet of Vehicles system.

[0081] According to a further improvement of the present invention, the preconfigured detection model can select input variables from thousands of Internet of Vehicles signals, set the detection model, and set the output indicators according to business needs; for example:

[0082] Table 1 Detection model examples

[0083]

[0084] The model is compiled into a pickle file, which is used to define and execute judgment rules. It can also support the compilation of complex machine learning models and cross-platform operation, ensuring that both streaming signal processing modules and batch signal processing modules are easy to parse and calculate.

[0085] According to a further improvement of the present invention, the preconfigured exception message construction module mainly performs the following operations:

[0086] Read the IoV signal interpretation table from the data warehouse and convert IoV signal codes that are difficult for general users to identify into Chinese content;

[0087] The exception message content is written in markdown language to ensure that the structure of the output message has good readability.

[0088] According to a further improvement of the present invention, the abnormal message sending module is a standardized API data interface. By triggering the interface, the abnormal message can be sent to the APP terminal, Web monitoring page, DingTalk message or WeChat service account as needed.

[0089] As described above, although the present invention has been shown and described with reference to specific preferred embodiments, it should not be construed as limiting the present invention itself. Various changes may be made to it in form and detail without departing from the spirit and scope of the present invention as defined in the appended claims.

Claims

1. A method for detecting and processing abnormal signals in an Internet of Vehicles, characterized in that: The following steps are involved: Step S1, the Internet of Vehicles terminal device collects signal data and sends it to the big data center; Step S2, streaming data processing: after the data of step S1 is generated in real time and arrives continuously, it is processed continuously in time windows divided according to certain rules, and the streaming data enters the pre-configured streaming signal processing module for processing; The specific steps are as follows: Step S21, first define two times: one is the time when data is generated, called event time, whose value is automatically generated when data is collected and is sent out at the same time as the signal value measured by the collection device; the other is the time when data arrives at the big data cloud signal processing module, called reception time; Step S21, first sorting by receiving time, then dividing the data into time windows according to the receiving time, and the data is processed continuously according to the time windows; Step S22: re-order the data by event time within the time window of step S21, compare the data receiving time with the event time, and determine the delayed data; if the delay time does not exceed a certain waiting time, and the waiting time does not exceed one tenth of the above time window, the data is processed in the current time window, otherwise the delayed data is discarded; Step S23: Calculate the data index in the above-divided time window according to the pre-configured detection model to determine whether it is an abnormal signal; Step S24: if an abnormal signal is found, trigger the downstream pre-configured abnormal message construction module to generate an abnormal message, and send the abnormal message through the pre-configured abnormal message sending module, and save the abnormal message to the data warehouse; Step S25: The streaming signal processing process is continuously performed while the collected data is sent to the big data center; Step S3, batch data processing: after a certain amount of data is accumulated, it is processed uniformly, usually for one hour or one day; the batch data enters the pre-configured batch signal processing module for processing; The specific steps are as follows: Step S31, first define two times: one is the time when data is generated, called event time, whose value is automatically generated when data is collected and is sent out at the same time as the signal value measured by the collection device; the other is the time when data arrives at the big data cloud signal processing module, called reception time; Step S32: After the batch data accumulates to a certain amount, it is partitioned and stored in the data warehouse according to the receiving time, with the partitions being stored in units of hours or days; Step S33: repartition and store the batch data according to the event time, and read the data partitioned by the receiving time for multiple days to obtain a complete set of data partitioned by the event time; Step S34: Calculate indicators based on the re-partitioned data and use a pre-configured detection model to detect abnormal signals. For continuous abnormal signals across partitions, it is necessary to integrate data from multiple partitions for calculation; Step S35: compare the abnormal signals detected by batch computing with the abnormal signals detected by streaming computing stored in the data warehouse, and use the batch computing as the standard to filter out abnormal signals that are not found or not accurately calculated by streaming computing; Step S36: using the screened abnormal signal to trigger the downstream pre-configured abnormal message construction module to generate an abnormal message, and sending the abnormal message through the pre-configured abnormal message sending module, and saving these abnormal messages to the data warehouse; Step S37, batch signal processing is performed once every hour or day after the collected data is sent to the big data center and saved in the data warehouse.

2. The method for detecting and processing abnormal signals in an Internet of Vehicles according to claim 1, characterized in that: The preconfigured detection model can select input variables, set detection models, and set output indicators in the Internet of Vehicles signal according to business needs; the model is compiled into a pickle file to define and execute judgment rules, and can also support the compilation of complex machine learning models. It also supports cross-platform operation, ensuring that both the streaming signal processing module and the batch signal processing module are easy to parse and calculate.

3. The method for detecting and processing abnormal signals in an Internet of Vehicles according to claim 1, characterized in that: The preconfigured exception message construction module mainly performs the following operations: Read the IoV signal interpretation table from the data warehouse and convert IoV signal codes that are difficult for general users to identify into Chinese content; The exception message content is written in markdown language to ensure that the structure of the output message has good readability.

4. The method for detecting and processing abnormal signals in an Internet of Vehicles according to claim 1, characterized in that: The abnormal message sending module is a standardized API data interface. By triggering the interface, the abnormal message can be sent to the APP terminal, Web monitoring page, DingTalk message or WeChat service account as needed.

5. A vehicle network abnormal signal detection and data processing system, characterized in that: include: Data acquisition module: IoV terminal equipment collects signal data; Streaming signal processing module: The streaming data is processed continuously in time windows divided according to certain rules. The streaming data enters the pre-configured streaming signal processing module for processing; Batch signal processing module: After a certain amount of data is accumulated, it is processed uniformly, usually one hour or one day; batch data enters the pre-configured batch signal processing module for processing; Detection model configuration module: select input variables from the Internet of Vehicles signal, set the detection model, and set the output indicators according to business needs; Abnormal message construction module: reads the IoV signal interpretation table from the data warehouse and converts IoV signal codes that are difficult for general users to identify into Chinese content abnormalities; Abnormal message sending module: It is a standardized API data interface. By triggering this interface, abnormal messages can be sent to the APP terminal, Web monitoring page, DingTalk message or letter service number as needed.

6. A vehicle network abnormal signal detection and data processing device, characterized in that: include: at least one processor; And at least one memory communicatively connected to the processor; wherein the memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement a vehicle network abnormal signal detection and data processing method as described in any one of claims 1 to 4.

Images