Vehicle-end CPU anomaly identification method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of vehicles, and discloses a vehicle end CPU abnormity identification method and device, computer equipment and a storage medium, and the method is applied to a cloud end, and comprises the steps: obtaining a statistical demand for vehicle end CPU usage data, and determining a statistical dimension according to the statistical demand; screening out a plurality of pieces of pre-stored CPU use data from a preset database based on the statistical dimension, and packaging the pre-stored CPU use data according to a preset data structure to obtain target CPU use data; analyzing the target CPU use data to generate a corresponding CPU performance chart; and determining abnormal data existing in the CPU performance chart by using an abnormal analysis strategy corresponding to the statistical dimension, analyzing the abnormal data to obtain an abnormal reason, and feeding back the abnormal reason to the vehicle end. The problems that in the prior art, CPU data processing efficiency is low, analysis depth and breadth are insufficient, and multi-dimensional comprehensive consideration of application CPU statistics, single-vehicle and ROM version CPU performance statistics and the like is lacked are solved.

Description

Technical Field

[0001] The present invention relates to the field of vehicle technology, and in particular to a method, device, computer equipment and storage medium for identifying abnormalities of a vehicle-side CPU. Background Art

[0002] CPU usage is an important indicator to measure how busy the CPU is in processing tasks within a specific period of time, and it directly reflects the activity of the CPU. Under normal use conditions, CPU usage will vary with the fluctuation of the task volume. It is usually low when idle, but will increase significantly when processing resource-intensive tasks. However, if the CPU usage remains at a high level, it will easily cause the system to respond slowly, or even freeze or become unresponsive, seriously affecting the user experience and system stability.

[0003] Although existing technologies have made some progress in collecting CPU usage, the accuracy and real-time performance in a multi-process environment still need to be improved. At the same time, there are delays in CPU data sorting, storage, and statistical analysis methods, and the depth and breadth of analysis are limited. Application CPU statistics lack multi-dimensional in-depth analysis, and single-vehicle and ROM version CPU performance statistics are only based on a single dimension, which cannot achieve comprehensive monitoring and management. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a method, device, computer equipment and storage medium for identifying vehicle-side CPU anomalies, so as to solve the problems in the prior art such as low CPU data processing efficiency, insufficient analysis depth and breadth, and lack of comprehensive consideration of multiple dimensions such as application CPU statistics, single-vehicle and ROM version CPU performance statistics.

[0005] In a first aspect, an embodiment of the present invention provides a method for identifying an abnormality of a vehicle-side CPU, which is applied to a cloud. The method includes:

[0006] Obtain statistical requirements for vehicle-side CPU usage data, and determine statistical dimensions based on the statistical requirements;

[0007] Filtering out a plurality of pre-stored CPU usage data in a preset database based on the statistical dimension, and encapsulating the pre-stored CPU usage data according to a preset data structure to obtain target CPU usage data;

[0008] Parsing the target CPU usage data and generating a corresponding CPU performance chart;

[0009] The abnormal data in the CPU performance chart is determined using the abnormal analysis strategy corresponding to the statistical dimension, the abnormal data is analyzed, the cause of the abnormality is obtained, and the cause of the abnormality is fed back to the vehicle end.

[0010] In an optional implementation of the present application, before obtaining the statistical requirements for vehicle-side CPU usage data, the method further includes:

[0011] Receiving the CPU data acquisition structure sent by the vehicle end;

[0012] Parsing the CPU data collection structure to obtain CPU usage data and data type;

[0013] The CPU usage data and the data type are stored in the cloud as pre-stored CPU usage data.

[0014] In an optional implementation of the present application, a plurality of pre-stored CPU usage data are screened out in a preset database based on the statistical dimension, and the pre-stored CPU usage data are packaged according to a preset data structure to obtain target CPU usage data, including:

[0015] If the statistical dimension is an application dimension, obtaining a target application identifier and a first indicator in the statistical requirement;

[0016] Extracting a plurality of pre-stored CPU usage data associated with the target application identifier in the first indicator from a preset database;

[0017] The pre-stored CPU usage data is encapsulated into a first data structure corresponding to the application dimension to obtain target CPU usage data.

[0018] In an optional implementation of the present application, a plurality of pre-stored CPU usage data are screened out in a preset database based on the statistical dimension, and the pre-stored CPU usage data are packaged according to a preset data structure to obtain target CPU usage data, including:

[0019] If the statistical dimension is a single-vehicle dimension, obtaining the target vehicle frame number and the second index in the statistical requirement;

[0020] Extracting a plurality of pre-stored CPU usage data associated with the target vehicle frame number in the second index from a preset database;

[0021] The pre-stored CPU usage data is arranged in chronological order and encapsulated into a second data structure corresponding to the single-vehicle dimension to obtain the target CPU usage data.

[0022] In an optional implementation of the present application, a plurality of pre-stored CPU usage data are screened out in a preset database based on the statistical dimension, and the pre-stored CPU usage data are packaged according to a preset data structure to obtain target CPU usage data, including:

[0023] If the statistical dimension is the ROM version dimension, then obtaining the target ROM version number and the third indicator in the statistical requirement;

[0024] Extracting a plurality of pre-stored CPU usage data associated with the target ROM version number in the third indicator from a preset database;

[0025] The pre-stored CPU usage data is encapsulated into a third data structure corresponding to the ROM version dimension to obtain target CPU usage data.

[0026] In an optional implementation manner of the present application, parsing the target CPU usage data and generating a corresponding CPU performance chart includes:

[0027] Parsing the target CPU usage data to obtain multiple data items;

[0028] Calculate the statistical parameters of the target CPU usage data according to the data items;

[0029] Generate a corresponding CPU performance chart based on the statistical parameters and the chart type corresponding to the statistical dimension.

[0030] In an optional implementation manner of the present application, the abnormal data existing in the CPU performance chart is determined by using the abnormal analysis strategy corresponding to the statistical dimension, and the abnormal data is analyzed to obtain the abnormal cause, including:

[0031] Based on the mapping relationship between the preset statistical dimension and the preset abnormal analysis strategy, determine the abnormal analysis strategy corresponding to the statistical dimension, wherein the abnormal analysis strategy includes an abnormal screening method and an analysis rule;

[0032] Using the abnormal screening method to screen the CPU performance chart to obtain abnormal data;

[0033] The characteristic information of the abnormal data is analyzed using the analysis rules to obtain the corresponding abnormal cause.

[0034] In an optional implementation manner of the present application, the analyzing the characteristic information of the abnormal data using the analysis rule to obtain the corresponding abnormal cause includes:

[0035] Extracting key features of the CPU performance chart and extracting feature information of the abnormal data;

[0036] Analyzing the difference characteristics between the key characteristics and the characteristic information by using the analysis rule;

[0037] Traversing a plurality of preset exception types, wherein each of the preset exception types comprises specific preset features;

[0038] The difference feature is matched with a preset feature associated with each preset exception type. If there is at least one target feature matching the difference feature among the plurality of preset features, an exception cause is generated based on the preset exception type corresponding to the target feature.

[0039] In a second aspect, an embodiment of the present invention provides a method for collecting vehicle-side CPU data, which is applied to the vehicle-side, and the method includes:

[0040] Calling the CPU collection program of the vehicle end to periodically collect CPU usage data;

[0041] Analyze the CPU usage data to determine a data type of the CPU usage data, wherein the data type includes a single-process type and a multi-process type;

[0042] Encapsulating the CPU usage data according to the data type to generate a corresponding CPU data collection structure;

[0043] Sending the CPU data acquisition structure to the cloud, so that the cloud can identify the abnormality of the CPU data acquisition structure and obtain the cause of the abnormality;

[0044] Receive the abnormal reason sent by the cloud, and perform monitoring operations on each application process and thread on the vehicle side according to the abnormal reason.

[0045] In a third aspect, an embodiment of the present invention provides a vehicle-side CPU data collection device, which is applied to a vehicle-side, and the device includes:

[0046] A collection module, used to call the CPU collection program of the vehicle end to periodically collect CPU usage data;

[0047] An analysis module, configured to analyze the CPU usage data and determine a data type of the CPU usage data, wherein the data type includes a single-process type and a multi-process type;

[0048] A packaging module, used for packaging the CPU usage data according to the data type and generating a corresponding CPU data collection structure;

[0049] A sending module, used for sending the CPU data acquisition structure to the cloud, so that the cloud can identify the abnormality of the CPU data acquisition structure and obtain the abnormality cause;

[0050] The receiving module is used to receive the abnormal reason sent by the cloud, and perform monitoring operations on each application process and thread on the vehicle side according to the abnormal reason.

[0051] In a fourth aspect, an embodiment of the present invention provides an abnormality identification device for a vehicle-side CPU, which is applied to a cloud, and the device includes:

[0052] An acquisition module, used to acquire statistical requirements for vehicle-side CPU usage data and determine statistical dimensions according to the statistical requirements;

[0053] A screening module, used for screening out a plurality of pre-stored CPU usage data in a preset database based on the statistical dimension, and encapsulating the pre-stored CPU usage data according to a preset data structure to obtain target CPU usage data;

[0054] A parsing module, used to parse the target CPU usage data and generate a corresponding CPU performance chart;

[0055] A determination module is used to determine the abnormal data in the CPU performance chart using the abnormal analysis strategy corresponding to the statistical dimension, analyze the abnormal data, obtain the cause of the abnormality, and feed back the cause of the abnormality to the vehicle end.

[0056] In a fifth aspect, an embodiment of the present invention provides a vehicle, comprising one or more processors; a memory for storing executable instructions of the one or more processors; wherein the one or more processors are configured to: execute the above-mentioned method for collecting vehicle-side CPU data.

[0057] The method provided in the embodiment of the present application has the following beneficial effects:

[0058] The method provided in the embodiment of the present application first clarifies the statistical requirements to ensure that the collection and analysis of CPU usage data are targeted and practical. By determining the statistical dimensions, a clear direction is provided for subsequent data screening and analysis. Pre-stored CPU usage data is screened and packaged to ensure data accuracy and facilitate subsequent processing. A CPU performance chart is generated to intuitively display usage and performance, so that users can quickly understand the vehicle-side CPU status and trends. Abnormal analysis strategies are used to accurately identify and analyze abnormal data to provide support for troubleshooting and performance optimization.

[0059] The embodiments of the present application obtain the target identification, frame number, ROM version number and corresponding indicators respectively according to the requirements of different statistical dimensions (application, bicycle, ROM version), so as to ensure the accuracy and practicality of data screening and processing. Accurately extract the data associated with the statistical dimension in the preset database, and provide a solid foundation for subsequent analysis. The screened data is packaged according to the data structure corresponding to the different dimensions to facilitate processing, analysis and visual display, and improve efficiency. According to the statistical requirements of different dimensions, the data screening, extraction and packaging strategies are flexibly adjusted. This method has good adaptability and scalability, and can cope with the CPU usage data monitoring and management requirements in different scenarios.

[0060] The method provided in the embodiment of the present application realizes accurate identification of abnormal data in the chart by extracting key features of the CPU performance chart and feature information of abnormal data, and analyzing the difference features between the two using analysis rules. By traversing multiple preset abnormal types and matching the difference features with the preset features associated with each preset abnormal type, the cause of the abnormality is accurately located, and the efficiency of troubleshooting is improved. The analysis of abnormal data is matched with the preset abnormal type, which avoids the complexity of the abnormal cause analysis process and improves the analysis efficiency. By presetting multiple abnormal types and corresponding features, the flexibility of dealing with various abnormal situations is enhanced.

[0061] The method provided in the embodiment of the present application ensures the timeliness and continuity of the data by periodically collecting CPU usage data on the vehicle side, providing a basis for subsequent analysis. By analyzing and determining the data type, the data can be processed more specifically to improve the accuracy and efficiency of anomaly identification. The data is packaged according to the type to facilitate cloud parsing and processing. The abnormal reasons sent by the cloud are received to monitor the application processes and threads on the vehicle side, improve the system performance and efficiency, and ensure the stability and security of the system. The cloud side realizes data interaction by receiving the CPU data collection structure sent by the vehicle side, providing a basis for subsequent analysis and anomaly identification. By parsing the data structure, the cloud side can accurately obtain the CPU usage and type on the vehicle side, providing precise support for subsequent processing and anomaly identification. At the same time, these data are stored as a baseline to facilitate subsequent anomaly detection and comparative analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0063] Figure 1 is a flow chart of a method for identifying abnormality of a vehicle-side CPU according to some embodiments of the present invention;

[0064] Figure 2 is a schematic diagram of a process of cloud data processing according to an embodiment of the present invention;

[0065] Figure 3 is a schematic diagram of a process for counting CPU usage data in the application dimension according to an embodiment of the present invention;

[0066] Figure 4 is a schematic diagram of a process for calculating CPU usage data statistics in a bicycle dimension according to an embodiment of the present invention;

[0067] Figure 5 This is a schematic diagram of a flow chart of CPU usage data statistics in the ROM version dimension according to an embodiment of the present invention;

[0068] Figure 6 is a flow chart of a method for collecting vehicle-side CPU data according to an embodiment of the present invention;

[0069] Figure 7 is a flow chart of a vehicle-side single-process acquisition program according to an embodiment of the present invention;

[0070] Figure 8 is a flowchart of a vehicle-side multi-process acquisition program according to an embodiment of the present invention;

[0071] Fig. 9 is a structural block diagram of a vehicle-side CPU data collection device according to an embodiment of the present invention;

[0072] Fig.10 is a structural block diagram of an abnormality identification device for a vehicle-side CPU according to an embodiment of the present invention;

[0073] Fig.11 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0074] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0075] According to an embodiment of the present invention, a method, apparatus, computer device and storage medium for identifying abnormalities of a vehicle-side CPU are provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0076] In this embodiment, a method for identifying abnormalities of a vehicle-side CPU is provided, which is applied to the cloud. Figure 1 is a flow chart of a method for identifying abnormality of a vehicle-side CPU according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0077] Step S11, obtaining statistical requirements for vehicle-side CPU usage data, and determining statistical dimensions based on the statistical requirements.

[0078] In an embodiment of the present application, the vehicle-side CPU usage data includes the total CPU usage rate, the CPU usage rate of each process and thread, which is obtained through periodic collection and is used to monitor and manage the vehicle CPU performance and anomalies. Statistical requirements involve the query, analysis and display of CPU usage data in different dimensions, such as specific vehicle models, ROM versions, applications or time dimensions. The statistical dimension is the classification standard for CPU usage data statistics, including vehicle model, ROM version, application and time, etc., which is used to subdivide and classify data to more accurately analyze CPU performance and identify anomalies. For example, the CPU usage rate of different vehicle models can be counted by vehicle model, or the CPU usage in different time periods can be counted by time.

[0079] In the embodiment of the present application, before step S11, it also includes: receiving a CPU data acquisition structure sent by the vehicle; parsing the CPU data acquisition structure to obtain CPU usage data and data type; and storing the CPU usage data and data type as pre-stored CPU usage data to the cloud.

[0080] Specifically, Figure 2 As shown, the cloud first receives the CPU data collection structure from the vehicle side, where the CPU data collection structure is periodically collected and reported by the single-process collection program and the multi-process collection program on the vehicle side. Secondly, these data are parsed to extract the specific CPU usage data and its type (for example, the CPU usage data of a specific application, or the CPU usage data of the vehicle as a whole and each process). After the parsing is completed, after sorting, the CPU usage data and the collection data structure corresponding to the data type are stored in a preset database as pre-stored CPU usage data. In this way, the cloud has a record of all CPU usage data uploaded by the vehicle side, which provides a basis for subsequent data statistical analysis.

[0081] Step S12, based on the statistical dimension, a plurality of pre-stored CPU usage data are screened out in a preset database, and the pre-stored CPU usage data are packaged according to a preset data structure to obtain target CPU usage data.

[0082] In the embodiment of the present application, the relevant pre-stored CPU usage data is screened out in the preset database according to the statistical dimensions (such as application, bicycle, ROM version, etc.), and these data are packaged according to the preset data structure to finally obtain the target CPU usage data. The specific implementation includes the following three cases:

[0083] Case 1, such as Figure 3 As shown, step S12 includes the following steps A1-A3:

[0084] Step A1: If the statistical dimension is an application dimension, obtain the target application identifier and the first indicator in the statistical requirement.

[0085] In the embodiment of the present application, if the statistical dimension is the application dimension, that is, the focus is on the CPU usage of a specific application, then it is necessary to obtain the identifier of the target application specified in the statistical requirement. This identifier can be the name, ID or other unique identifier of the application, which is used to accurately identify the application in the database. At the same time, it is also necessary to determine the first statistical indicator, that is, to focus on the specific indicators of the application in terms of CPU usage, such as CPU occupancy, CPU usage time, CPU usage times, etc.

[0086] Step A2: extracting a plurality of pre-stored CPU usage data associated with the target application identifier in the first indicator from a preset database.

[0087] In an embodiment of the present application, a preset database is used to perform precise queries based on a specific target application identifier and a first indicator. A large amount of CPU usage data is stored in the database, and these data are organized in order according to multiple dimensions such as application, timestamp, CPU usage indicator, etc. By performing a query operation, all pre-stored CPU usage data associated with the target application identifier within the first indicator range are extracted. These data cover different time points or time periods, and fully reflect the CPU usage of the application at different times. The extracted data will be further used for subsequent packaging and display work.

[0088] Step A3: encapsulate the pre-stored CPU usage data into a first data structure corresponding to the application dimension to obtain target CPU usage data.

[0089] In the embodiment of the present application, the first data structure encapsulated into the application dimension means organizing and sorting the data by application to clearly reflect the CPU usage of each application within a specified time, including total usage, process usage, and thread usage. The data structure can be in JSON, XML, and other formats, and can be further processed, such as calculating statistical indicators or sorting and filtering. The target data finally obtained will be used for subsequent analysis, display, or storage.

[0090] Case 2: Figure 4 As shown, step S12 includes the following steps B1-B3:

[0091] Step B1, if the statistical dimension is a single-vehicle dimension, obtain the target frame number and the second indicator in the statistical requirement.

[0092] In an embodiment of the present application, if the statistical dimension is a single-vehicle dimension, that is, the focus is on the CPU usage of a specific vehicle, it is necessary to obtain the frame number of the target vehicle specified in the statistical requirements. The frame number is a unique identifier for the vehicle, which is used to accurately identify the vehicle in the database. At the same time, it is also necessary to determine the second statistical indicator, that is, to focus on the specific indicators of the vehicle in terms of CPU usage. Similar to the statistics of the application dimension, the second indicator can be CPU occupancy, CPU usage time, CPU usage times, etc., but here, these indicators will be counted for the entire vehicle, not a single application.

[0093] Step B2, extracting a plurality of pre-stored CPU usage data associated with the target vehicle frame number in the second index from a preset database.

[0094] In an embodiment of the present application, a preset database is accessed, which stores multi-dimensional CPU usage data, including vehicle frame number, timestamp, CPU usage index, etc. Query conditions for specific vehicle frame numbers and CPU usage indicators (such as CPU occupancy) are constructed as required. The query is executed to extract data related to the target vehicle frame number within the specified CPU usage index range. After extraction, data preprocessing is performed, such as cleaning and format conversion. Finally, a result set containing all data that meets the conditions is constructed for subsequent packaging, display or analysis.

[0095] Step B3, arrange the pre-stored CPU usage data in chronological order and encapsulate it into a second data structure corresponding to the single-vehicle dimension to obtain the target CPU usage data.

[0096] In an embodiment of the present application, multiple pre-stored CPU usage data associated with the target frame number in the second indicator are extracted from a preset database, and first need to be sorted by timestamp to ensure the timing of the data, so as to facilitate subsequent analysis and identification of anomalies. The sorted data needs to be encapsulated into a second data structure of the bicycle dimension, which clearly indicates the CPU usage of the bicycle at different time points, including the total CPU usage rate, the CPU usage rate of each process, etc. The data structure can be in the format of JSON, XML, etc., and contains fields such as timestamp, CPU usage rate, process information, etc. The sorted and encapsulated data is the target CPU usage data, which is used for subsequent display, analysis or processing to monitor and manage the CPU performance of the bicycle.

[0097] Case three, if Figure 5 As shown, step S12 includes the following steps C1-C3:

[0098] Step C1, if the statistical dimension is the ROM version dimension, obtain the target ROM version number and the third indicator in the statistical requirement.

[0099] In the embodiment of the present application, the ROM version dimension refers to the statistics of CPU usage data classified by the ROM version of the vehicle system to analyze its impact on CPU performance. The target ROM version number is the specific ROM identifier that the user wants to query and analyze, and is used to filter relevant data. The ROM version affects the function, performance and optimization of the vehicle system, so the CPU usage may vary. The third indicator is the specific CPU usage indicator that the user is concerned about, such as occupancy rate, usage time and number of times, which is used to measure and compare the CPU performance of different ROM versions.

[0100] Step C2, extracting multiple pre-stored CPU usage data associated with the target ROM version number in the third indicator from the preset database.

[0101] In the embodiment of the present application, in order to collect the usage of the vehicle CPU under a specific ROM version, so as to perform performance analysis and abnormality identification later. Specifically, based on the target ROM version number and the third indicator, the filtering conditions are constructed, and the data records that meet these conditions are retrieved from the database. These data records include CPU usage data of different vehicle models, different applications, and different time points, but they all share the same ROM version number and are within the range of the third indicator.

[0102] Step C3, encapsulating the pre-stored CPU usage data into a third data structure corresponding to the ROM version dimension to obtain the target CPU usage data.

[0103] In the embodiment of the present application, the pre-stored CPU usage data is first sorted out, including deduplication, error correction, and filling of missing values. Secondly, the data is classified according to the ROM version, and a data structure suitable for the ROM version dimension is defined to clearly indicate the CPU usage under different ROM versions, including the total usage rate, the usage rate of each process and thread, etc., and includes metadata such as ROM version and timestamp. The data is encapsulated according to the defined structure to ensure that the hierarchy and field naming meet the standards. Finally, the encapsulated data is verified, which can be achieved by comparing with the original data, performing integrity checks, etc.

[0104] It should be noted that in the data structures corresponding to the application dimension, single vehicle dimension, and ROM version dimension, their main differences lie in the different data dimensions and organizational methods they focus on. Specifically, the data structure of the application dimension mainly focuses on the CPU usage of specific applications, including detailed fields such as application identification and CPU occupancy, and is further subdivided by process and thread. The data structure of the single vehicle dimension integrates data across applications, focusing on the CPU usage of a specific vehicle, including fields such as frame number and total CPU occupancy, and is arranged in chronological order to analyze timing changes. The data structure of the ROM version dimension focuses on the CPU usage of different models and applications under a specific ROM version, including fields such as ROM version number and model, and is further subdivided by model and application to show the impact of ROM version on CPU performance under different conditions.

[0105] Step S13, parsing the target CPU usage data and generating a corresponding CPU performance chart.

[0106] In the application embodiment, step S13 includes the following steps D1-D3:

[0107] Step D1, parsing the target CPU usage data to obtain multiple data items.

[0108] As an example, import the psutil library in the Python code, "import psutil". Use the functions provided by the psutil library to obtain CPU usage data. For example, you can use "psutil.cpu_percent(interval=1)" to obtain the current CPU usage, where the "interval" parameter indicates the time interval for obtaining the usage. You can call this function multiple times to obtain usage data at different time points, thereby obtaining multiple data items. You can also use the "psutil.cpu_times()" function to obtain data items such as the CPU's running time and idle time. By calculating, you can obtain the CPU usage change, etc. Process and parse the obtained data to extract the required data items, such as the average, maximum, and minimum usage values ​​in different time periods.

[0109] Step D2, calculating statistical parameters of target CPU usage data according to the data items.

[0110] In an embodiment of the present application, multiple data items are first preprocessed, such as summarizing and analyzing CPU usage data in different time periods. Then, mathematical operations are used to determine statistical parameters, including calculating the average usage rate to reflect the overall level, searching for the highest usage rate value to show the peak condition, and calculating the minimum value to evaluate the CPU idle situation. These accurately calculated statistical parameters provide solid data support for the subsequent construction of performance charts. Taking a line chart as an example, the average value is used as a reference for the trend line, and the maximum and minimum values ​​are used as the upper and lower boundary determination benchmarks. In the data processing process, technologies such as mean smoothing and outlier detection are also used to improve data quality and reliability.

[0111] Step D3, generating a corresponding CPU performance chart according to the statistical parameters and the chart type corresponding to the statistical dimension.

[0112] In the embodiment of the present application, it is first necessary to clarify the chart type that matches the statistical dimension, such as choosing between line charts, bar charts, pie charts, etc. In the chart generation process, the functions and functions provided by the matplotlib library or the seaborn library are fully utilized for drawing and configuration operations. For example, the chart title, axis labels, colors, grid lines and other elements are set.

[0113] Specifically, if the statistical dimension is a bicycle dimension, it is advisable to select a line chart as the chart type. With time as the horizontal axis, the calculated statistical parameters such as the average, maximum, and minimum values ​​of the CPU usage are set as the vertical axis data, and a line is formed by connecting these data points to clearly present the trend of CPU usage over time. Among them, the line formed by the average value can reflect the overall development trend, while the maximum and minimum values ​​effectively define the range of fluctuations.

[0114] Specifically, if the statistical dimension is an application dimension, the chart type can be a bar chart, which is classified into different time periods or various performance indicators (such as average usage rate, maximum usage rate, etc.), and the corresponding statistical parameter value is set as the column height, which can intuitively compare and analyze the differences between different time periods or performance indicators.

[0115] Specifically, if the statistical dimension is the ROM version dimension, you can select a pie chart to display the proportion of CPU time in different working states (such as high utilization, medium utilization, and low utilization) in the size of the sector area, so as to clarify the proportional distribution of various CPU usage situations.

[0116] Step S14, using the abnormal analysis strategy corresponding to the statistical dimension to determine the abnormal data in the CPU performance chart, and analyze the abnormal data to obtain the cause of the abnormality, and feed back the cause of the abnormality to the vehicle end.

[0117] In the application embodiment, step S14 includes the following steps E1-E3:

[0118] Step E1, extracting key features of the CPU performance chart and extracting feature information of abnormal data.

[0119] In an embodiment of the present application, CPU performance chart data is obtained from a system performance monitoring tool, including but not limited to detailed data of indicators such as CPU usage, core frequency, temperature, process occupancy, etc. over a period of time. The data is cleaned and preprocessed to remove noise and missing values ​​to ensure the accuracy and completeness of the data. Statistical methods and data mining algorithms are used to extract key features of the CPU performance chart, for example: obtaining statistics such as the mean, median, and standard deviation of the CPU usage. Determine the highest and lowest values ​​of the core frequency. Identify the trend of changes in CPU usage, such as whether there are stages of continuous increase or decrease. Based on the set threshold or anomaly detection algorithm, extract the characteristic information of abnormal data, for example: when the CPU usage exceeds 80% and lasts for more than 5 minutes, mark it as abnormal data, and record its start time, duration, and peak value.

[0120] Step E2: Analyze the difference characteristics between the key characteristics and the characteristic information using analysis rules.

[0121] In an embodiment of the present application, a series of analysis rules can be predefined. For example, if the average CPU usage in the key feature is 50%, and the peak usage in the abnormal data reaches 90%, and the difference exceeds 40%, it is considered a significant difference. When the core frequency in the key feature is stable at 2.5GHz, and there is a sudden drop to 1.5GHz in the abnormal data, it is defined as an abnormal frequency difference. According to these rules, the key features and abnormal data features are compared and analyzed to calculate the difference features, including the difference value, change trend, duration, etc.

[0122] Step E3, traversing multiple preset exception types, wherein each preset exception type contains specific preset features.

[0123] In an embodiment of the present application, a preset exception type library can be established in advance, for example: the "software conflict" exception type, the preset feature is that a specific process occupies too much CPU resources and the fluctuation is large. The "hardware aging" exception type, the preset feature is that the CPU core frequency gradually decreases and the performance stability decreases. The "CPU overload" exception type, the preset feature is that the CPU usage rate remains high for a long time (such as more than 90%), and is accompanied by slow system response or jamming. Traverse all preset exception types, and match the previously calculated difference features with the preset features of each type.

[0124] Step E4, matching the difference feature with the preset feature associated with each preset abnormality type, if there is at least one target feature matching the difference feature among the multiple preset features, generating an abnormality cause based on the preset abnormality type corresponding to the target feature.

[0125] In the embodiment of the present application, when performing a matching operation, a detailed and comprehensive comparison is performed on the calculated difference features, not only a direct correspondence of simple values ​​or states, but also the comprehensive performance and change pattern of the difference features are considered.

[0126] Specifically, for the preset abnormal type of "software conflict", if the difference characteristics show that the CPU resource usage of a specific process is not only too high, exceeding 70%, but also its occupancy rate fluctuates drastically from 50% to 80% within 1 minute, showing irregular and large fluctuations, and also causes the response delay of other related processes to exceed 5 seconds, then these comprehensive performances match the preset characteristics of "software conflict". For the type of "hardware aging", if the difference characteristics show that the CPU core frequency not only gradually decreases from 3.0GHz to 2.5GHz within 1 week, but also is accompanied by frequent freezes and errors 3 times per hour during the decline, and the CPU temperature also abnormally rises from 50℃ to 80℃, but the cooling system fan speed works normally at 2000 rpm, these comprehensive conditions can be highly consistent with the preset characteristics of "hardware aging".

[0127] Specifically, if there is at least one target feature matching the difference feature among the multiple preset features, an abnormality cause is generated based on the preset abnormality type corresponding to the target feature, wherein the abnormality cause is generated by determining the fault information through a specific detection and analysis method based on the abnormality type.

[0128] For the exception type of "software conflict", the conflicting software can be located by monitoring the resource competition and communication between processes, and determining which software frequently seizes CPU resources and interferes with each other within a specific time period, so as to generate the corresponding exception cause: "Due to the fierce competition for CPU resources between software X and software Y when they are running at the same time, a conflict occurs between them, resulting in an imbalance in CPU resource allocation and affecting system performance."

[0129] For the "hardware aging" exception type, we can determine whether the transistors inside the CPU are aging or the capacitors are deteriorating by testing the performance of the CPU core components and analyzing the historical operation data, thereby generating the corresponding exception cause: "Due to the aging of some transistors inside the CPU after long-term high-frequency use, the core frequency has gradually decreased, affecting the overall performance."

[0130] In this embodiment, a method for collecting vehicle-side CPU data is provided, which is applied to the vehicle-side. Figure 6 is a flow chart of a method for identifying abnormality of a vehicle-side CPU according to an embodiment of the present invention. Figure 6 As shown, the process includes the following steps:

[0131] Step S21, calling the vehicle-side CPU collection program to periodically collect CPU usage data.

[0132] In the embodiment of the present application, the vehicle-side CPU collection program is started, and the program automatically collects CPU usage data according to a preset period (such as at regular intervals). These data include the overall CPU usage rate, the usage rate of each process, and the usage rate of each thread.

[0133] Step S22, analyzing the CPU usage data to determine a data type of the CPU usage data, wherein the data type includes a single-process type and a multi-process type.

[0134] In the embodiment of the present application, the collected CPU usage data is deeply analyzed to determine the data type of the data. The data type is mainly divided into two types: single process type and multi-process type.

[0135] like Figure 7 As shown, the single-process acquisition program on the vehicle side is responsible for collecting the CPU usage data of a specific application, including the CPU usage of the process and the CPU usage of each thread. The specific process is as follows: the single-process acquisition program starts automatically at startup and is responsible for collecting the raw CPU data of a specific application. The acquisition instructions are executed according to the preset cycle to obtain the CPU usage data of the process and thread. After parsing and processing the data, it is packaged into a specific data structure and reported to the cloud. After completing a collection, parsing, packaging and reporting, the program enters a waiting state until the next cycle arrives and repeats the above process.

[0136] like Figure 8 As shown, the multi-process acquisition program on the vehicle side is responsible for collecting the CPU usage data of the vehicle computer as a whole and each process. After these data are uploaded to the cloud, they can be further analyzed and processed according to the user's statistical needs and the determined statistical dimensions. The specific process is: the multi-process acquisition program starts automatically when the machine is turned on, and is responsible for collecting the CPU usage data of the vehicle computer as a whole and each process. Execute the acquisition instructions according to the preset cycle, obtain the CPU raw data, parse and process the data, encapsulate it into a multi-process CPU acquisition data structure and report it to the cloud. After completing a collection, parsing, packaging and reporting, the program enters a waiting state until the next cycle arrives, and repeats the above process.

[0137] Step S23, encapsulating the CPU usage data according to the data type, and generating a corresponding CPU data collection structure.

[0138] In the embodiment of the present application, according to the determined data type, the program will encapsulate the CPU usage data into the corresponding collection structure. For single-process type data, it will be encapsulated into a single-process CPU collection data structure; for multi-process type data, it will be encapsulated into a multi-process CPU collection data structure. These data structures contain rich information, such as ROM version number, vehicle model, frame number, timestamp, process number, process name, CPU occupancy, etc.

[0139] Step S24, sending the CPU data collection structure to the cloud, so that the cloud can identify the abnormality of the CPU data collection structure and obtain the cause of the abnormality.

[0140] In the embodiment of the present application, the packaged CPU data collection structure will be sent to the cloud for processing. The cloud will further analyze and process the data to identify possible anomalies or problems and provide the cause of the anomaly.

[0141] Step S25, receiving the abnormal reason sent by the cloud, and performing monitoring operations on each application process and thread on the vehicle side according to the abnormal reason.

[0142] In the embodiment of the present application, the vehicle receives the abnormal reasons sent by the cloud, and performs corresponding monitoring operations on each application process and thread of the vehicle according to these reasons. These operations may include adjusting process priority, closing certain processes, optimizing thread allocation, etc., to ensure stable operation and performance optimization of the vehicle.

[0143] In this embodiment, a vehicle-side CPU data collection device and a vehicle-side CPU abnormality identification device are also provided. The above two devices are used to implement the above embodiments and preferred implementation modes, and will not be repeated here. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0144] This embodiment provides a vehicle-side CPU data collection device, which is applied to the vehicle side, such as Fig. 9 As shown, including:

[0145] The collection module 91 is used to call the CPU collection program of the vehicle end to periodically collect CPU usage data;

[0146] An analysis module 92, configured to analyze the CPU usage data and determine a data type of the CPU usage data, wherein the data type includes a single-process type and a multi-process type;

[0147] The encapsulation module 93 is used to encapsulate the CPU usage data according to the data type and generate a corresponding CPU data collection structure;

[0148] A sending module 94 is used to send the CPU data acquisition structure to the cloud, so that the cloud can identify the abnormality of the CPU data acquisition structure and obtain the abnormality cause;

[0149] The receiving module 95 is used to receive the abnormal reason sent by the cloud, and perform monitoring operations on each application process and thread on the vehicle side according to the abnormal reason.

[0150] This embodiment provides a vehicle-side CPU abnormality recognition device, which is applied to the cloud. Fig.10 As shown, including:

[0151] The acquisition module 101 is used to obtain the statistical requirements for vehicle-side CPU usage data and determine the statistical dimensions according to the statistical requirements;

[0152] A screening module 102 is used to screen out a plurality of pre-stored CPU usage data in a preset database based on a statistical dimension, and encapsulate the pre-stored CPU usage data according to a preset data structure to obtain target CPU usage data;

[0153] The parsing module 103 is used to parse the target CPU usage data and generate a corresponding CPU performance chart;

[0154] The determination module 104 is used to determine the abnormal data in the CPU performance chart by using the abnormal analysis strategy corresponding to the statistical dimension, and analyze the abnormal data to obtain the abnormal cause.

[0155] In an optional embodiment of the present application, the device also includes: a receiving module for receiving a CPU data acquisition structure sent by the vehicle; parsing the CPU data acquisition structure to obtain CPU usage data and data type; and storing the CPU usage data and data type as pre-stored CPU usage data to the cloud.

[0156] In an optional implementation of the present application, the screening module 102 is used to obtain the target application identifier and the first indicator in the statistical requirement if the statistical dimension is the application dimension; extract multiple pre-stored CPU usage data associated with the target application identifier in the first indicator in a preset database; encapsulate the pre-stored CPU usage data into a first data structure corresponding to the application dimension to obtain the target CPU usage data.

[0157] In an optional implementation of the present application, the screening module 102 is used to obtain the target frame number and the second indicator in the statistical requirements if the statistical dimension is a single-vehicle dimension; extract multiple pre-stored CPU usage data associated with the target frame number in the second indicator from a preset database; arrange the pre-stored CPU usage data in chronological order, and encapsulate it into a second data structure corresponding to the single-vehicle dimension to obtain the target CPU usage data.

[0158] In an optional implementation of the present application, the screening module 102 is used to obtain the target ROM version number and the third indicator in the statistical requirements if the statistical dimension is the ROM version dimension; extract multiple pre-stored CPU usage data associated with the target ROM version number in the third indicator in the preset database; encapsulate the pre-stored CPU usage data into a third data structure corresponding to the ROM version dimension to obtain the target CPU usage data.

[0159] In an optional embodiment of the present application, the parsing module 103 is used to parse the target CPU usage data to obtain multiple data items; calculate the statistical parameters of the target CPU usage data based on the data items; and generate a corresponding CPU performance chart based on the statistical parameters and the chart type corresponding to the statistical dimensions.

[0160] In an optional implementation of the present application, a determination module 104 is used to determine an abnormality analysis strategy corresponding to a statistical dimension based on a mapping relationship between a preset statistical dimension and a preset abnormality analysis strategy, wherein the abnormality analysis strategy includes an abnormality screening method and an analysis rule; the CPU performance chart is screened using the abnormality screening method to obtain abnormal data; and the characteristic information of the abnormal data is analyzed using the analysis rule to obtain the corresponding abnormal cause.

[0161] In an optional embodiment of the present application, a determination module 104 is used to extract key features of a CPU performance chart and extract feature information of abnormal data; analyze difference features between key features and feature information using analysis rules; traverse multiple preset exception types, wherein each preset exception type contains specific preset features; match the difference features with preset features associated with each preset exception type, and if there is at least one target feature among the multiple preset features that matches the difference features, then generate an exception cause based on the preset exception type corresponding to the target feature.

[0162] An embodiment of the present application also provides a vehicle, which includes one or more processors; a memory for storing executable instructions of the one or more processors; wherein the one or more processors are configured to: execute a method for collecting vehicle-side CPU data of the above-mentioned embodiment.

[0163] See also Fig.11 , Fig.11is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Fig.11 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system).

[0164] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0165] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0166] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created by the use of a computer device based on the presentation of a small program landing page, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0167] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0168] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0169] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0170] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for identifying abnormality of a vehicle-side CPU, applied to the cloud, characterized in that: The method comprises: Obtain statistical requirements for vehicle-side CPU usage data, and determine statistical dimensions based on the statistical requirements; Filtering out a plurality of pre-stored CPU usage data in a preset database based on the statistical dimension, and encapsulating the pre-stored CPU usage data according to a preset data structure to obtain target CPU usage data; Parsing the target CPU usage data and generating a corresponding CPU performance chart; The abnormal data in the CPU performance chart is determined using the abnormal analysis strategy corresponding to the statistical dimension, the abnormal data is analyzed, the cause of the abnormality is obtained, and the cause of the abnormality is fed back to the vehicle end.

2. The method according to claim 1, characterized in that Before obtaining the statistical requirements for vehicle-side CPU usage data, the method further includes: Receiving the CPU data acquisition structure sent by the vehicle end; Parsing the CPU data collection structure to obtain CPU usage data and data type; The CPU usage data and the data type are stored in the cloud as pre-stored CPU usage data.

3. The method according to claim 1, characterized in that The step of screening out a plurality of pre-stored CPU usage data in a preset database based on the statistical dimension, and encapsulating the pre-stored CPU usage data according to a preset data structure to obtain target CPU usage data includes: If the statistical dimension is an application dimension, obtaining a target application identifier and a first indicator in the statistical requirement; Extracting a plurality of pre-stored CPU usage data associated with the target application identifier in the first indicator from a preset database; The pre-stored CPU usage data is encapsulated into a first data structure corresponding to the application dimension to obtain target CPU usage data.

4. The method according to claim 1, characterized in that: The step of screening out a plurality of pre-stored CPU usage data in a preset database based on the statistical dimension, and encapsulating the pre-stored CPU usage data according to a preset data structure to obtain target CPU usage data includes: If the statistical dimension is a single-vehicle dimension, obtaining the target vehicle frame number and the second index in the statistical requirement; Extracting a plurality of pre-stored CPU usage data associated with the target vehicle frame number in the second index from a preset database; The pre-stored CPU usage data is arranged in chronological order and encapsulated into a second data structure corresponding to the single-vehicle dimension to obtain the target CPU usage data.

5. The method according to claim 1, characterized in that The step of screening out a plurality of pre-stored CPU usage data in a preset database based on the statistical dimension, and encapsulating the pre-stored CPU usage data according to a preset data structure to obtain target CPU usage data includes: If the statistical dimension is the ROM version dimension, then obtaining the target ROM version number and the third indicator in the statistical requirement; Extracting a plurality of pre-stored CPU usage data associated with the target ROM version number in the third indicator from a preset database; The pre-stored CPU usage data is encapsulated into a third data structure corresponding to the ROM version dimension to obtain target CPU usage data.

6. The method according to claim 1, characterized in that The step of parsing the target CPU usage data and generating a corresponding CPU performance chart includes: Parsing the target CPU usage data to obtain multiple data items; Calculate the statistical parameters of the target CPU usage data according to the data items; Generate a corresponding CPU performance chart based on the statistical parameters and the chart type corresponding to the statistical dimension.

7. The method according to claim 1, characterized in that The determining the abnormal data in the CPU performance chart by using the abnormal analysis strategy corresponding to the statistical dimension, analyzing the abnormal data, and obtaining the abnormal cause includes: Based on the mapping relationship between the preset statistical dimension and the preset abnormal analysis strategy, determine the abnormal analysis strategy corresponding to the statistical dimension, wherein the abnormal analysis strategy includes an abnormal screening method and an analysis rule; Screening the CPU performance chart using the abnormal screening method to obtain abnormal data; The characteristic information of the abnormal data is analyzed using the analysis rules to obtain the corresponding abnormal cause.

8. The method according to claim 7, characterized in that The using the analysis rule to analyze the characteristic information of the abnormal data to obtain the corresponding abnormal cause includes: Extracting key features of the CPU performance chart and extracting feature information of the abnormal data; Analyzing the difference characteristics between the key characteristics and the characteristic information by using the analysis rule; Traversing a plurality of preset exception types, wherein each of the preset exception types comprises specific preset features; The difference feature is matched with a preset feature associated with each preset exception type. If there is at least one target feature matching the difference feature among the plurality of preset features, an exception cause is generated based on the preset exception type corresponding to the target feature.

9. A method for collecting vehicle-side CPU data, applied to the vehicle-side, characterized in that: The method comprises: Calling the CPU collection program of the vehicle end to periodically collect CPU usage data; Analyze the CPU usage data to determine a data type of the CPU usage data, wherein the data type includes a single-process type and a multi-process type; Encapsulating the CPU usage data according to the data type to generate a corresponding CPU data collection structure; Sending the CPU data acquisition structure to the cloud, so that the cloud can identify the abnormality of the CPU data acquisition structure and obtain the cause of the abnormality; Receive the abnormal reason sent by the cloud, and perform monitoring operations on each application process and thread on the vehicle side according to the abnormal reason.

10. A vehicle-side CPU data collection device, applied to the vehicle-side, characterized in that: The device comprises: A collection module, used to call the CPU collection program of the vehicle end to periodically collect CPU usage data; An analysis module, configured to analyze the CPU usage data and determine a data type of the CPU usage data, wherein the data type includes a single-process type and a multi-process type; A packaging module, used for packaging the CPU usage data according to the data type and generating a corresponding CPU data collection structure; A sending module, used for sending the CPU data acquisition structure to the cloud, so that the cloud can identify the abnormality of the CPU data acquisition structure and obtain the abnormality cause; The receiving module is used to receive the abnormal reason sent by the cloud, and perform monitoring operations on each application process and thread on the vehicle side according to the abnormal reason.

11. A vehicle-side CPU abnormality recognition device, applied to the cloud, characterized in that: The device comprises: An acquisition module, used to acquire statistical requirements for vehicle-side CPU usage data and determine statistical dimensions according to the statistical requirements; A screening module, used for screening out a plurality of pre-stored CPU usage data in a preset database based on the statistical dimension, and encapsulating the pre-stored CPU usage data according to a preset data structure to obtain target CPU usage data; A parsing module, used to parse the target CPU usage data and generate a corresponding CPU performance chart; A determination module is used to determine the abnormal data in the CPU performance chart using the abnormal analysis strategy corresponding to the statistical dimension, analyze the abnormal data, obtain the cause of the abnormality, and feed back the cause of the abnormality to the vehicle end.

12. A vehicle, characterized in that: include: one or more processors; A memory for storing the one or more processor executable instructions; wherein the one or more processors are configured to: execute a method for collecting vehicle-side CPU data as described in claim 9.

Images