A data asynchronous update method and system based on edge computing
By adopting the asynchronous data update method of edge computing in the Internet of Vehicles system, the problem of insufficient computing resources and time delay in the Internet of Vehicles system is solved, low-latency and efficient data transmission are achieved, and system stability and user experience are improved.
Patent Information
- Application Number
- CN202411266063.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-11
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-09-11
AI Technical Summary
The existing Internet of Vehicles systems cannot meet the needs of new vehicle applications in terms of computing resources and delays, resulting in a decrease in vehicle driving stability and safety. The synchronous update mode may cause the system to wait for network delays or equipment failures, affecting other Execution of the operation.
Adopt the asynchronous data update method based on edge computing, and by obtaining the data terminal data in the Internet of Vehicles, performing migration calculation and hierarchical storage, real-time monitoring of the data status, setting the asynchronous update trigger conditions, using data compression and encryption algorithms to prioritize and transmit data, dynamic The differential algorithm predicts the data to be updated, and the asynchronous transmission time prediction method determines the optimal update time, and adjusts the transmission strategy according to the network bandwidth prediction model.
It achieves low latency, reduces bandwidth consumption, improves system response speed and stability, optimizes network resource utilization, improves user experience, supports flexible data processing strategies, and ensures efficient and reliable data transmission.
Smart Images

Figure CN118802986B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of asynchronous data updating, and in particular to an asynchronous data updating method and system based on edge computing. Background Art
[0002] With the continuous development of Internet of Vehicles technology, the amount of data generated by vehicles continues to surge, including data information such as vehicle status, driving routes, driving behavior, and environmental information. A large amount of storage and computing resources are needed to accurately process and analyze these data in a timely manner to achieve vehicle intelligence and automation, and higher requirements are placed on the service quality of the communication network. However, the existing Internet of Vehicles cannot meet the computing resource and latency requirements of vehicle applications. Slow data processing will lead to reduced stability and safety of vehicle driving. Moreover, in the synchronous update mode of data, if the data update operation is blocked due to network delays, equipment failures, etc., the entire system may fall into a waiting state, affecting the execution of other operations. Asynchronous updates allow the system to continue to perform other tasks when blocked, thereby enhancing the stability of the system.
[0003] Therefore, the use of edge computing to meet the computing resource and latency requirements of new in-vehicle applications can achieve low latency and reduce bandwidth consumption; at the same time, asynchronous data updates can achieve a more efficient and reliable data transmission and update mechanism by improving system response speed, optimizing network resource utilization, enhancing system stability, improving user experience, and supporting flexible data processing strategies. Summary of the invention
[0004] (1) Technical issues to be resolved
[0005] The purpose of the present invention is to provide a data asynchronous update method and system based on edge computing to solve the problem that data cannot be updated in a timely and effective manner due to complex factors, including network delays, equipment failures, etc.
[0006] (2) Technical solution
[0007] To achieve the above object, on the one hand, the present invention provides a data asynchronous update method based on edge computing, the method comprising:
[0008] S1. Acquire data from an Internet of Vehicles terminal, transfer the data from the Internet of Vehicles terminal to an edge computing node according to a migration computing method, and perform hierarchical storage through a data caching method; monitor the data status in real time through the edge computing node, wherein the data status includes a timestamp of the data, data changes, and cache capacity;
[0009] S2. Setting a trigger condition for asynchronous update according to the data status; obtaining data whose data status satisfies the trigger condition for asynchronous update and recording it as data to be updated; obtaining the importance and update frequency of the data to be updated, and performing priority sorting by a statistical analysis method according to the importance and update frequency of the data to be updated to obtain a sorting result of the data to be updated; and performing asynchronous update according to the sorting result;
[0010] S3, transmitting the data to be updated to the receiving party according to the asynchronous update after passing the data compression algorithm and the data encryption algorithm; after the receiving party receives the data to be updated, the data is verified; when the data to be updated fails to be transmitted, the data is retransmitted according to the preset retry times and retry interval;
[0011] S4. When the monitoring edge computing node updates data, the cache status is updated. The updated cache status includes updating the management cache space and updating the cache timestamp.
[0012] Further, the triggering condition of asynchronous update is set according to the data state; data whose data state satisfies the triggering condition of asynchronous update is obtained and recorded as data to be updated; the importance and update frequency of the data to be updated are obtained, and priority sorting is performed according to the importance and update frequency of the data to be updated by a statistical analysis method to obtain a sorting result of the data to be updated; the method of performing asynchronous update according to the sorting result includes:
[0013] The trigger conditions include data expiration, data change, and cache capacity reaching a threshold; data expiration means that the current time of the data exceeds the data valid timestamp plus a preset time threshold; data change means that the monitored data change exceeds the preset change threshold; cache capacity reaching a threshold means that the current cache capacity occupancy rate of the edge computing node exceeds the preset cache capacity occupancy rate threshold;
[0014] By acquiring historical data received by the edge computing node, the historical data includes data transmission time, data type, data volume, and transmission frequency; obtaining analysis results through trend analysis of the historical data, and adjusting the triggering conditions of asynchronous update according to the analysis results;
[0015] Get the set of edge computing nodes and record it as , the first edge computing nodes at time Data to be updated The data to be updated The calculation formula is:
[0016] ;
[0017] in Indicates time The data status, For the previous time The data status, is the attenuation factor, is an exponential decay function, The time when the data was last updated;
[0018] Asynchronous updating is performed according to the sorting result and in accordance with an asynchronous updating time, wherein the asynchronous updating time is obtained by an asynchronous transmission time prediction method.
[0019] Further, the method of performing asynchronous updating according to the sorting result and according to the asynchronous updating time, wherein the asynchronous updating time is obtained by an asynchronous transmission time prediction method, includes:
[0020] Obtain network bandwidth usage, establish a prediction model for network bandwidth changes over time through time series analysis, and obtain , Represents network bandwidth over time The changes are detected in real time by network performance analysis method. , Indicates The real-time transmission rate from the edge computing node to the receiver; the asynchronous update time is obtained by the asynchronous transmission time prediction method ; The asynchronous update time The calculation formula is:
[0021] ;
[0022] in, To calculate the time interval Internal network bandwidth The cumulative amount minus the cumulative difference in the amount of data required to be transmitted by all edge computing nodes; Indicates that all edge computing nodes are at time The time required to transmit per second; Indicates the average transmission rate of all edge computing nodes The cumulative amount of network bandwidth required to complete data transmission; It is The incremental data to be updated that the edge computing nodes need to transmit is The calculation formula is:
[0023] ;
[0024] For the The average transmission rate of edge computing nodes is obtained by obtaining the average transmission rate of edge computing nodes within a preset time period. The average transmission rate is obtained by the mean calculation method. The average transmission rate The calculation formula is:
[0025] ;
[0026] in, is the preset time window, is the preset start time.
[0027] Furthermore, the method of transmitting the to-be-updated data to the receiving party according to the asynchronous update time after using a data compression algorithm and a data encryption algorithm includes:
[0028] The data to be updated is passed through a data compression algorithm to obtain first compressed data, and the first compressed data is passed through a compression efficiency algorithm to obtain a compression efficiency evaluation value , through the compression efficiency evaluation value Select a data compression algorithm; the compression efficiency evaluation value The calculation formula is:
[0029] * ;
[0030] in, is the amount of data to be updated; is the amount of the first compressed data, Indicates the compression efficiency of the data volume before and after compression; Indicates compression efficiency index; Indicates the compression efficiency index adjustment factor; for The use influence coefficient is according to The experimental data were obtained through statistical analysis; express Utilization rate, the Obtained through system performance monitoring methods;
[0031] Select different data encryption algorithms to encrypt the data according to the importance of the data to be updated.
[0032] Furthermore, the method of transmitting the data of the Internet of Vehicles terminal to the edge computing node according to the migration computing method and performing hierarchical storage through the data caching method includes:
[0033] Obtain the distance, bandwidth, processing capacity, and load of the edge computing node and record them as first data, obtain second data by normalizing the first data, and obtain a comprehensive weight value by using a linear weighted algorithm for the second data; obtain the comprehensive weight value from the Internet of Vehicles terminal to all edge computing nodes, and transmit the data of the Internet of Vehicles terminal to the edge computing node corresponding to the maximum value of the comprehensive weight value;
[0034] The transmission strategy is adjusted by real-time monitoring of the load of the edge computing node, the growth rate of the load of the edge computing node is obtained, and the saturation time when the load of the edge computing node reaches a preset load saturation threshold is predicted according to the growth rate; the completion time of data transmission to the edge computing node is obtained, and when the completion time is after the saturation time, the edge computing node is reselected in descending order according to the comprehensive weight value;
[0035] The data caching method is to perform hierarchical storage according to the access frequency and importance of data; the cache level of the hierarchical storage is preset according to the access speed and cache capacity.
[0036] Based on the same inventive concept, the present invention also provides a data asynchronous update system based on edge computing, the system comprising: a data acquisition and management module, a data asynchronous update trigger module, a data asynchronous update management module, and a data cache management module, and each module is sequentially connected in communication;
[0037] A data acquisition and management module, which is used to acquire data from an Internet of Vehicles terminal, transmit the data from the Internet of Vehicles terminal to an edge computing node according to a migration calculation method, and store the data in layers through a data caching method; and monitor the data status in real time through the edge computing node, wherein the data status includes a timestamp of the data, data changes, and cache capacity;
[0038] A data asynchronous update trigger module is used to set a trigger condition for asynchronous update according to the data state; obtain data whose data state meets the trigger condition for asynchronous update and record it as data to be updated; obtain the importance and update frequency of the data to be updated, and perform priority sorting by a statistical analysis method according to the importance and update frequency of the data to be updated to obtain a sorting result of the data to be updated; and perform asynchronous update according to the sorting result;
[0039] A data asynchronous update management module, used to transmit the data to be updated to the receiving party according to the asynchronous update after passing the data compression algorithm and the data encryption algorithm; when the receiving party receives the data to be updated, it performs data verification; when monitoring the data to be updated fails to be transmitted, it retransmits according to the preset retry times and retry interval;
[0040] The data cache management module is used to update the cache status after the monitoring edge computing node updates the data. The updated cache status includes updating the management cache space and updating the cache timestamp.
[0041] Furthermore, the data asynchronous update trigger module includes:
[0042] A trigger condition management module, for trigger conditions including data expiration, data change, and cache capacity reaching a threshold; data expiration means that the current time of the data exceeds the data valid timestamp plus a preset time threshold; data change means that the data change exceeds the preset change threshold; cache capacity reaching the threshold means that the current cache capacity occupancy rate of the edge computing node exceeds the preset cache capacity occupancy rate threshold; by acquiring historical data received by the edge computing node, the historical data includes data transmission time, data type, data volume, and transmission frequency; through the historical data, an analysis result is obtained by trend analysis, and the trigger condition of asynchronous update is adjusted according to the analysis result;
[0043] The data calculation module to be updated is used to obtain the set of edge computing nodes and record it as , the first edge computing nodes at time Data to be updated The data to be updated The calculation formula is:
[0044] ;
[0045] in Indicates time The data status, For the previous time The data status, is the attenuation factor, is an exponential decay function, The time when the data was last updated;
[0046] The asynchronous update execution module is used to perform asynchronous update according to the sorting result and the asynchronous update time, and the asynchronous update time is obtained by an asynchronous transmission time prediction method.
[0047] Furthermore, the asynchronous update execution module includes the following steps:
[0048] Obtain network bandwidth usage, establish a prediction model for network bandwidth changes over time through time series analysis, and obtain , Represents network bandwidth over time The changes are detected in real time by network performance analysis method. , Indicates The real-time transmission rate from the edge computing node to the receiver; the asynchronous update time is obtained by the asynchronous transmission time prediction method ; The asynchronous update time The calculation formula is:
[0049] ;
[0050] in, To calculate the time interval Internal network bandwidth The cumulative amount minus the cumulative difference in the amount of data required to be transmitted by all edge computing nodes; Indicates that all edge computing nodes are at time The time required to transmit per second; Indicates the average transmission rate of all edge computing nodes The cumulative amount of network bandwidth required to complete data transmission; It is The incremental data to be updated that the edge computing nodes need to transmit is The calculation formula is:
[0051] ;
[0052] For the The average transmission rate of edge computing nodes is obtained by obtaining the average transmission rate of edge computing nodes within a preset time period. The average transmission rate is obtained by the mean calculation method. The average transmission rate The calculation formula is:
[0053] ;
[0054] in, is the preset time window, is the preset start time.
[0055] Furthermore, the data asynchronous update management module includes the following steps:
[0056] The data to be updated is passed through a data compression algorithm to obtain first compressed data, and the first compressed data is passed through a compression efficiency algorithm to obtain a compression efficiency evaluation value , through the compression efficiency evaluation value Select a data compression algorithm; the compression efficiency evaluation value The calculation formula is:
[0057] * ;
[0058] in, is the amount of data to be updated; is the amount of the first compressed data, Indicates the compression efficiency of the data volume before and after compression; Indicates compression efficiency index; Indicates the compression efficiency index adjustment factor; for The use influence coefficient is according to The experimental data were obtained through statistical analysis; express Utilization rate, the Obtained through system performance monitoring methods;
[0059] Select different data encryption algorithms to encrypt the data according to the importance of the data to be updated.
[0060] Furthermore, the data acquisition and management module includes the following steps:
[0061] Obtain the distance, bandwidth, processing capacity, and load of the edge computing node and record them as first data, obtain second data by normalizing the first data, and obtain a comprehensive weight value by using a linear weighted algorithm for the second data; obtain the comprehensive weight value from the Internet of Vehicles terminal to all edge computing nodes, and transmit the data of the Internet of Vehicles terminal to the edge computing node corresponding to the maximum value of the comprehensive weight value;
[0062] The transmission strategy is adjusted by real-time monitoring of the load of the edge computing node, the growth rate of the load of the edge computing node is obtained, and the saturation time when the load of the edge computing node reaches a preset load saturation threshold is predicted according to the growth rate; the completion time of data transmission to the edge computing node is obtained, and when the completion time is after the saturation time, the edge computing node is reselected in descending order according to the comprehensive weight value;
[0063] The data caching method is to perform hierarchical storage according to the access frequency and importance of data; the cache level of the hierarchical storage is preset according to the access speed and cache capacity.
[0064] (3) Beneficial effects
[0065] Compared with the prior art, the present invention has the following beneficial effects:
[0066] 1. By setting the trigger conditions for asynchronous updates and prioritizing the data based on its importance and update frequency, the intelligent and efficient asynchronous update of the data of the connected vehicle terminals is realized. The data to be updated is predicted by the dynamic differential algorithm, and the optimal update time is determined by the asynchronous transmission time prediction method, which further optimizes the data transmission strategy, reduces unnecessary network bandwidth usage, and improves the overall data transmission efficiency.
[0067] 2. The updated data is processed by data compression and encryption algorithms, which not only reduces the time and bandwidth required for data transmission, but also improves the security of data during transmission. By comprehensively considering the distance, bandwidth, processing power and load of edge computing nodes, the most suitable nodes are selected for data caching and hierarchical storage, further enhancing the efficiency and reliability of asynchronous data update processing. BRIEF DESCRIPTION OF THE DRAWINGS
[0068] Figure 1 This is a flowchart of a data asynchronous update method based on edge computing according to Embodiment 1 of the present invention;
[0069] Figure 2 This is a schematic diagram of the module composition of a data asynchronous update system based on edge computing according to Example 2 of the present invention. DETAILED DESCRIPTION
[0070] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0071] Before giving examples, it is necessary to explain the application scenarios of the present invention. The present invention is a data asynchronous update method and system based on edge computing. In the intelligent transportation system, the Internet of Vehicles terminal is an important node for the interaction of vehicle and external information. It is widely deployed on various vehicles to provide drivers and passengers with rich driving experience and services, such as real-time traffic information, driving records, non-contact payment, remote monitoring (remote sentinel), etc. However, the traditional data transmission mode directly uploads all the data of the Internet of Vehicles terminal to the cloud for processing and response, which not only puts great pressure on network resources, but also has a significant delay, especially in the case of network congestion or instability, which may cause the lag of key information, thereby affecting driving safety and user experience. In order to overcome these problems, a data asynchronous update method based on edge computing is introduced, which makes full use of the distribution and computing power of edge computing nodes, processes, analyzes and stores the data generated by the Internet of Vehicles terminal nearby, greatly reduces the burden on the cloud, and significantly reduces the delay of data transmission. At the same time, by improving the system response speed, optimizing the utilization of network resources, enhancing system stability, improving user experience and supporting flexible data processing strategies, a more efficient and reliable data transmission and update mechanism is achieved.
[0072] Example 1: Figure 1 As shown, this embodiment provides a data asynchronous update method based on edge computing, and the method includes:
[0073] S1. Acquire the data of the Internet of Vehicles terminal, transmit the data of the Internet of Vehicles terminal to the edge computing node according to the migration calculation method, and store the data in layers through the data caching method; monitor the data status in real time through the edge computing node, and the data status includes the timestamp of the data, the data change, and the cache capacity; the data of the Internet of Vehicles terminal is acquired through the camera, built-in GPS, Beidou chip, etc. of the Internet of Vehicles terminal, and the data is transmitted through the 4G communication module and the Internet of Things SIM card, etc.
[0074] S2. Set the triggering condition of asynchronous update according to the data status; obtain the data whose data status meets the triggering condition of asynchronous update and record it as the data to be updated; obtain the importance and update frequency of the data to be updated, and prioritize the data to be updated by statistical analysis method according to the importance and update frequency of the data to be updated to obtain the sorting result of the data to be updated; perform asynchronous update according to the sorting result; the importance of the data depends on the business category to which the data belongs, for example, the data for driving navigation is more important, while the data for senseless payment can be relatively less important. The higher the data update frequency, the more frequently the data is used, so it is necessary to ensure the real-time nature of the data. The weighted scoring method is used to comprehensively consider the importance and update frequency of the data. For the data packets to be updated, the sorting result shows its comprehensive urgency, so the data to be updated of the edge computing node needs to be updated asynchronously according to the sorting result, avoiding the update of all the data to be updated together, which is easy to cause network congestion. When the triggering condition is met, the data to be updated will be calculated, and at this time it only means that the data meets the asynchronous update condition, but when to transmit it needs to consider the actual situation of the network.
[0075] S3. The data to be updated is transmitted to the receiver according to the asynchronous update after passing the data compression algorithm and the data encryption algorithm; when the receiver receives the data to be updated, the data is verified; when the monitoring data transmission fails, the data is retransmitted according to the preset number of retries and retry interval; the data size is reduced by the data compression algorithm, thereby reducing the consumption of network bandwidth and speeding up the data transmission speed. To ensure the privacy and security of the user's data on the Internet of Vehicles terminal, the data needs to be encrypted. Data encryption can prevent the data from being intercepted and tampered with by unauthorized third parties. The selection of encryption algorithms should take into account the balance of security, performance and compatibility. Common encryption algorithms include AES, RSA, TLS / SSL, etc. The receiver is a cloud server.
[0076] S4. When the monitoring edge computing node updates the data, the cache status is updated. The updating of the cache status includes updating the management cache space and updating the cache timestamp. Updating the cache status is the key to ensuring efficient operation of the system and data consistency. Updating the management cache space can release the cache space by deleting the transmitted data. It can also adjust and optimize the cache replacement strategy, cache size limit and other parameters according to the cache space, because too low cache space may cause the cache hit rate to decrease, increase the delay and cost of data transmission. The cache timestamp is used to record the last modification time of the cached data, so as to understand which data has been recently accessed or modified.
[0077] Further, the triggering condition of asynchronous update is set according to the data state; data whose data state satisfies the triggering condition of asynchronous update is obtained and recorded as data to be updated; the importance and update frequency of the data to be updated are obtained, and priority sorting is performed according to the importance and update frequency of the data to be updated by a statistical analysis method to obtain a sorting result of the data to be updated; the method of performing asynchronous update according to the sorting result includes:
[0078] The trigger conditions include data expiration, data change, and cache capacity reaching a threshold; data expiration means that the current time of the data exceeds the data valid timestamp plus a preset time threshold; data change means that the monitored data change exceeds the preset change threshold; cache capacity reaching a threshold means that the current cache capacity occupancy rate of the edge computing node exceeds the preset cache capacity occupancy rate threshold;
[0079] By obtaining historical data received by the edge computing node, the historical data includes data transmission time, data type, data volume, and transmission frequency; the analysis results are obtained through trend analysis of the historical data, and the triggering conditions of asynchronous updates are adjusted according to the analysis results; for example, if it is found that the access frequency of a certain type of data increases significantly within a specific time period, the system can correspondingly reduce the expiration threshold of this type of data to ensure the freshness of the data. Similarly, if the system load is high, the threshold of cache capacity occupancy can be adjusted to trigger asynchronous updates in advance, thereby updating the management cache space and avoiding cache overflow.
[0080] Get the set of edge computing nodes and record it as , the first edge computing nodes at time Data to be updated The data to be updated The calculation formula is:
[0081] ;
[0082] in Indicates time The data status, For the previous time The data status, is the attenuation factor, is an exponential decay function, The last data update time; the data to be updated at each edge computing node at each time point is obtained through the dynamic difference algorithm , which represents the data increment at the current moment. The last data update time is the timestamp that the system marks the data. . Each time the data is updated, the data to be updated will be cleared and the calculation will start again. Exponential decay function It means that as time goes by, the influence of old data on the current data status gradually weakens. It describes the reduction of the weight of old data in the decision-making or calculation process, rather than the reduction of the value of the old data itself in storage or recording, which is used to more accurately reflect the dynamic characteristics of data changing over time.
[0083] Asynchronous updating is performed according to the sorting result and in accordance with an asynchronous updating time, wherein the asynchronous updating time is obtained by an asynchronous transmission time prediction method.
[0084] Further, the method of performing asynchronous updating according to the sorting result and according to the asynchronous updating time, wherein the asynchronous updating time is obtained by an asynchronous transmission time prediction method, includes:
[0085] Obtain network bandwidth usage, establish a prediction model for network bandwidth changes over time through time series analysis, and obtain , Represents network bandwidth over time The changes are detected in real time by network performance analysis method. , Indicates The real-time transmission rate from the edge computing node to the receiver; the asynchronous update time is obtained by the asynchronous transmission time prediction method ; The asynchronous update time The calculation formula is:
[0086] ;
[0087] in, To calculate the time interval Internal network bandwidth The cumulative amount minus the cumulative difference in the amount of data required to be transmitted by all edge computing nodes; Indicates that all edge computing nodes are at time The time required to transmit per second; Indicates the average transmission rate of all edge computing nodes The cumulative amount of network bandwidth required to complete data transmission; It is The incremental data to be updated that the edge computing nodes need to transmit is The calculation formula is:
[0088] ;
[0089] This formula calculates the within, no. The cumulative increment of data to be updated for each edge computing node.
[0090] For the The average transmission rate of edge computing nodes is obtained by obtaining the average transmission rate of edge computing nodes within a preset time period. The average transmission rate is obtained by the mean calculation method. The average transmission rate The calculation formula is:
[0091] ;
[0092] in, is the preset time window, is the preset start time.
[0093] According to the characteristics of network bandwidth data (such as periodicity, trend, seasonality, etc.), select a suitable time series model, such as ARIMA (autoregressive integrated moving average model), seasonal ARIMA (SARIMA), LSTM, etc. This embodiment uses LSTM to predict changes in network bandwidth. Asynchronous update time The calculation formula is used to predict the optimal data transmission time in the network, that is, to predict when is the best time to start data transmission. Asynchronous update time The calculation formula is to find the minimum value, so that from 0 to The cumulative difference in time is greater than or equal to the total amount of data that all edge computing nodes need to transmit at the average transmission rate. The time 0 is the time when the first edge computing node meets the trigger condition after the last asynchronous update is completed. From this moment on, each edge computing node begins to generate data to be updated. The system will start to analyze and calculate all edge computing nodes according to the asynchronous transmission time prediction method to obtain the time for the next asynchronous update. It is considered to be the best time for data transmission because it ensures the efficient utilization of network bandwidth while meeting the data transmission needs of all edge computing nodes. It means the network is in time The total bandwidth available (e.g., in bits per second), and It represents all edge computing nodes at time The "equivalent bandwidth consumption" of the required transfer time (e.g. in seconds). This integral formula represents a cumulative difference between "capacity-demand". It does not calculate the cumulative amount of network bandwidth (for example, in bits), but rather calculates the accumulation of a time-related performance metric or deviation. This metric or deviation is determined by the difference between the actual bandwidth of the network and the "equivalent bandwidth occupancy" caused by the data transmission requirements of the edge computing nodes. In this integral, is used as a reference value, and It represents the "equivalent bandwidth" actually occupied due to data transmission requirements. The result of the integration represents the time interval The cumulative effect of this difference (i.e., the "remaining" or "available" portion of the network bandwidth) is not an accumulation of data in bits, but a comprehensive measure related to time, bandwidth, and data transmission requirements.
[0094] Furthermore, the method of transmitting the to-be-updated data to the receiving party according to the asynchronous update time after using a data compression algorithm and a data encryption algorithm includes:
[0095] The data to be updated is passed through a data compression algorithm to obtain first compressed data, and the first compressed data is passed through a compression efficiency algorithm to obtain a compression efficiency evaluation value , through the compression efficiency evaluation value Select a data compression algorithm; the compression efficiency evaluation value The calculation formula is:
[0096] * ;
[0097] in, is the amount of data to be updated; is the amount of the first compressed data, Indicates the compression efficiency of the data volume before and after compression; Indicates compression efficiency index; Indicates the compression efficiency index adjustment factor; for The use influence coefficient is according to The experimental data were obtained through statistical analysis; express Utilization rate, the Obtained through system performance monitoring method; the amount of data to be updated The data volume can be directly obtained through data management tools or monitoring tools. It is used to measure the performance of data compression algorithms, taking into account the compression ratio (i.e. the degree of reduction in the amount of data before and after compression) and the compression process. The amount of data to be updated is the original data amount. The compression efficiency index is calculated by dividing the compression efficiency by the original data amount. Thus, a compression efficiency index is obtained which is independent of the size of the original data. For adjustment The compression efficiency index adjustment factor is used to adjust the compression efficiency index. The impact of utilization rate. As the utilization rate increases, the compression efficiency index adjustment factor decreases, thereby reducing the overall compression efficiency evaluation value.
[0098] Select different data encryption algorithms to encrypt the data according to the importance of the data to be updated. Different data encryption algorithms need to be selected to encrypt the data according to the importance of the data to be updated. The importance of the data is based on factors such as the sensitivity of the data, security requirements, and transmission reliability. Select the appropriate encryption algorithm according to the importance of the data. For example, for highly sensitive data, you can select a high-strength encryption algorithm (such as AES-256); for less sensitive data, you can select a lower-strength encryption algorithm to save computing resources.
[0099] Furthermore, the method of transmitting the data of the Internet of Vehicles terminal to the edge computing node according to the migration computing method and performing hierarchical storage through the data caching method includes:
[0100] Obtain the distance, bandwidth, processing capacity, and load of the edge computing node and record them as the first data, obtain the second data by normalizing the first data, and obtain the comprehensive weight value of the second data by a linear weighted algorithm; obtain the comprehensive weight value from the Internet of Vehicles terminal to all edge computing nodes, and transmit the data of the Internet of Vehicles terminal to the edge computing node corresponding to the maximum value of the comprehensive weight value; for each edge computing node, collect its distance ,bandwidth , Processing Capacity and current load conditions As the first data. Since the dimensions and ranges of the data indicators are different, they are normalized to the same scale to obtain the normalized distance , Normalized bandwidth , Normalization processing capability , Normalized Load The linear weighted algorithm is used to calculate the comprehensive weight value of each edge computing node. Assume that the weight coefficients are (distance), (bandwidth), (processing power) and (load condition), and , then the comprehensive weight value for: ; For load conditions, take the complement of the normalized value , because lower load values indicate better performance.
[0101] By real-time monitoring of the load of edge computing nodes, the transmission strategy is adjusted, the growth rate of the load of edge computing nodes is obtained, and the saturation time when the load of edge computing nodes reaches the preset load saturation threshold is predicted based on the growth rate; the completion time of data transmission to the edge computing node is obtained, and when the completion time is after the saturation time, the edge computing nodes are reselected in order from large to small according to the comprehensive weight value; by predicting when the load saturation threshold is reached by the growth rate of the load, measures can be taken in advance to avoid service interruption or performance degradation. By avoiding transmitting data to edge computing nodes that are about to reach load saturation, the risk of service interruption and data loss caused by overload of edge computing nodes can be reduced.
[0102] The data caching method is to perform hierarchical storage according to the access frequency and importance of data; the cache level of the hierarchical storage is preset according to the access speed and cache capacity. The access frequency is the frequency of data access. High-frequency access data means that the data is often used in the system, so it should be stored in a cache level that can be accessed quickly. Even if the access frequency of important data is not high, it needs to be properly stored and protected to prevent data loss or damage from causing serious impact on the system. The cache level of the hierarchical storage is preset according to the access speed and cache capacity of the storage medium. First, the data is classified according to the access frequency and importance of the data. According to the classification results of the data, the data is allocated to different cache levels, for example, data with high frequency access and high importance should be stored in a fast access level, while data with low frequency access or low importance is stored in a slow access level. When a low-frequency access data suddenly becomes high-frequency access, it can be promoted to a faster cache level.
[0103] Embodiment 2: Based on the same inventive concept, this embodiment also provides a data asynchronous update system based on edge computing, including: a data acquisition and management module, a data asynchronous update trigger module, a data asynchronous update management module, and a data cache management module, and each module is sequentially communicated and connected;
[0104] A data acquisition and management module, which is used to acquire data from an Internet of Vehicles terminal, transmit the data from the Internet of Vehicles terminal to an edge computing node according to a migration calculation method, and store the data in layers through a data caching method; and monitor the data status in real time through the edge computing node, wherein the data status includes a timestamp of the data, data changes, and cache capacity;
[0105] A data asynchronous update trigger module is used to set a trigger condition for asynchronous update according to the data state; obtain data whose data state meets the trigger condition for asynchronous update and record it as data to be updated; obtain the importance and update frequency of the data to be updated, and perform priority sorting by a statistical analysis method according to the importance and update frequency of the data to be updated to obtain a sorting result of the data to be updated; and perform asynchronous update according to the sorting result;
[0106] A data asynchronous update management module, used to transmit the data to be updated to the receiving party according to the asynchronous update after passing the data compression algorithm and the data encryption algorithm; when the receiving party receives the data to be updated, it performs data verification; when monitoring the data to be updated fails to be transmitted, it retransmits according to the preset retry times and retry interval;
[0107] The data cache management module is used to update the cache status after the monitoring edge computing node updates the data. The updated cache status includes updating the management cache space and updating the cache timestamp.
[0108] Furthermore, the data asynchronous update trigger module includes:
[0109] A trigger condition management module, for trigger conditions including data expiration, data change, and cache capacity reaching a threshold; data expiration means that the current time of the data exceeds the data valid timestamp plus a preset time threshold; data change means that the data change exceeds the preset change threshold; cache capacity reaching the threshold means that the current cache capacity occupancy rate of the edge computing node exceeds the preset cache capacity occupancy rate threshold; by acquiring historical data received by the edge computing node, the historical data includes data transmission time, data type, data volume, and transmission frequency; through the historical data, an analysis result is obtained by trend analysis, and the trigger condition of asynchronous update is adjusted according to the analysis result;
[0110] The data calculation module to be updated is used to obtain the set of edge computing nodes and record it as , the first edge computing nodes at time Data to be updated The data to be updated The calculation formula is:
[0111] ;
[0112] in Indicates time The data status, For the previous time The data status, is the attenuation factor, is an exponential decay function, The time when the data was last updated;
[0113] The asynchronous update execution module is used to perform asynchronous update according to the sorting result and the asynchronous update time, and the asynchronous update time is obtained by an asynchronous transmission time prediction method.
[0114] Furthermore, the asynchronous update execution module includes the following steps:
[0115] Obtain network bandwidth usage, establish a prediction model for network bandwidth changes over time through time series analysis, and obtain , Represents network bandwidth over time The changes are detected in real time by network performance analysis method. , Indicates The real-time transmission rate from the edge computing node to the receiver; the asynchronous update time is obtained by the asynchronous transmission time prediction method ; The asynchronous update time The calculation formula is:
[0116] ;
[0117] in, To calculate the time interval Internal network bandwidth The cumulative amount minus the cumulative difference in the amount of data required to be transmitted by all edge computing nodes; Indicates that all edge computing nodes are at time The time required to transmit per second; Indicates the average transmission rate of all edge computing nodes The cumulative amount of network bandwidth required to complete data transmission; It is The incremental data to be updated that the edge computing nodes need to transmit is The calculation formula is:
[0118] ;
[0119] For the The average transmission rate of edge computing nodes is obtained by obtaining the average transmission rate of edge computing nodes within a preset time period. The average transmission rate is obtained by the mean calculation method. The average transmission rate The calculation formula is:
[0120] ;
[0121] in, is the preset time window, is the preset start time.
[0122] Furthermore, the data asynchronous update management module includes the following steps:
[0123] The data to be updated is passed through a data compression algorithm to obtain first compressed data, and the first compressed data is passed through a compression efficiency algorithm to obtain a compression efficiency evaluation value , through the compression efficiency evaluation value Select a data compression algorithm; the compression efficiency evaluation value The calculation formula is:
[0124] * ;
[0125] in, is the amount of data to be updated; is the amount of the first compressed data, Indicates the compression efficiency of the data volume before and after compression; Indicates compression efficiency index; Indicates the compression efficiency index adjustment factor; for The use influence coefficient is according to The experimental data were obtained through statistical analysis; express Utilization rate, the Obtained through system performance monitoring methods;
[0126] Select different data encryption algorithms to encrypt the data according to the importance of the data to be updated.
[0127] Furthermore, the data acquisition and management module includes the following steps:
[0128] Obtain the distance, bandwidth, processing capacity, and load of the edge computing node and record them as first data, obtain second data by normalizing the first data, and obtain a comprehensive weight value by using a linear weighted algorithm for the second data; obtain the comprehensive weight value from the Internet of Vehicles terminal to all edge computing nodes, and transmit the data of the Internet of Vehicles terminal to the edge computing node corresponding to the maximum value of the comprehensive weight value;
[0129] The transmission strategy is adjusted by real-time monitoring of the load of the edge computing node, the growth rate of the load of the edge computing node is obtained, and the saturation time when the load of the edge computing node reaches a preset load saturation threshold is predicted according to the growth rate; the completion time of data transmission to the edge computing node is obtained, and when the completion time is after the saturation time, the edge computing node is reselected in descending order according to the comprehensive weight value;
[0130] The data caching method is to perform hierarchical storage according to the access frequency and importance of data; the cache level of the hierarchical storage is preset according to the access speed and cache capacity.
[0131] It should be noted that, regarding the system in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.
[0132] Finally, it should be noted that: Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A data asynchronous update method based on edge computing, It is characterized in that The method comprises: Acquire data from a connected vehicle terminal, transfer the data from the connected vehicle terminal to an edge computing node according to a migration computing method, and perform hierarchical storage through a data caching method; monitor data status in real time through the edge computing node, wherein the data status includes a timestamp of the data, data changes, and cache capacity; Setting a trigger condition for asynchronous update according to the data state; obtaining data whose data state satisfies the trigger condition for asynchronous update and recording it as data to be updated; obtaining the importance and update frequency of the data to be updated, and performing priority sorting by a statistical analysis method according to the importance and update frequency of the data to be updated to obtain a sorting result of the data to be updated; performing asynchronous update according to the sorting result; The data to be updated is transmitted to the receiving party according to the asynchronous update after passing the data compression algorithm and the data encryption algorithm; after the receiving party receives the data to be updated, the data is verified; when the data to be updated fails to be transmitted, the data is retransmitted according to the preset retry times and retry interval; When the monitoring edge computing node updates the data, the cache status is updated, and the updated cache status includes updating the management cache space and updating the cache timestamp; The method of setting a trigger condition for asynchronous update according to the data state; obtaining data whose data state satisfies the trigger condition for asynchronous update and recording it as data to be updated; obtaining the importance and update frequency of the data to be updated, and performing priority sorting according to the importance and update frequency of the data to be updated by a statistical analysis method to obtain a sorting result of the data to be updated; and performing asynchronous update according to the sorting result includes: The trigger conditions include data expiration, data change, and cache capacity reaching a threshold; data expiration means that the current time of the data exceeds the data valid timestamp plus a preset time threshold; data change means that the monitored data change exceeds the preset change threshold; cache capacity reaching a threshold means that the current cache capacity occupancy rate of the edge computing node exceeds the preset cache capacity occupancy rate threshold; By acquiring historical data received by the edge computing node, the historical data includes data transmission time, data type, data volume, and transmission frequency; obtaining analysis results through trend analysis of the historical data, and adjusting the triggering conditions of asynchronous update according to the analysis results; Get the set of edge computing nodes and record it as (k 1 ,…,k i ,…,k n ), the data to be updated ΔD of the i-th edge computing node at time s is obtained through the dynamic difference algorithm i (s); the data to be updated ΔD i The calculation formula of (s) is: Among them, D i (s) represents the data state at time s, D i (s-1) is the data state at the previous time (s-1), λ is the attenuation factor, is an exponential decay function, t 0 The time when the data was last updated; Performing asynchronous updating according to the sorting result and in accordance with the asynchronous updating time, wherein the asynchronous updating time is obtained by an asynchronous transmission time prediction method; The method of performing asynchronous updating according to the sorting result and the asynchronous updating time, wherein the asynchronous updating time is obtained by an asynchronous transmission time prediction method, comprises: Obtain network bandwidth usage, establish a prediction model for network bandwidth changes over time through time series analysis, and obtain B(s) based on the prediction model, where B(S) represents the change of network bandwidth over time S; obtain R through real-time detection using a network performance analysis method i (S), the R i (s) represents the real-time transmission rate from the i-th edge computing node to the receiver; the asynchronous update time T is obtained by the asynchronous transmission time prediction method. w ; The asynchronous update time T w The calculation formula is: in, It is used to calculate the cumulative difference between the amount of network bandwidth B(s) and the amount of data required to be transmitted by all edge computing nodes in the time interval [0, t]; It represents the time required for all edge computing nodes to transmit per second at time s; Indicates the average transmission rate of all edge computing nodes The cumulative amount of network bandwidth required to complete data transmission under ΔD i is the incremental data to be updated that needs to be transmitted by the i-th edge computing node, and the incremental data to be updated ΔD i The calculation formula is: is the average transmission rate of the i-th edge computing node, and is obtained by obtaining R i (s) and obtain the average transmission rate by the mean calculation method The average transmission rate The calculation formula is: Where T is the preset time window, t' 0 is the preset start time.
2. According to the method for asynchronously updating data based on edge computing in claim 1, It is characterized in that The method of transmitting the to-be-updated data to the receiving party according to the asynchronous update time after using a data compression algorithm and a data encryption algorithm comprises: The data to be updated is passed through a data compression algorithm to obtain first compressed data, the first compressed data is passed through a compression efficiency algorithm to obtain a compression efficiency evaluation value C, and a data compression algorithm is selected according to the compression efficiency evaluation value C; the calculation formula of the compression efficiency evaluation value C is: Among them, L 1 is the amount of data to be updated; L 2 is the amount of data of the first compressed data, log 2 L 1 -log 2 L 2 Indicates the compression efficiency of the data volume before and after compression; Indicates compression efficiency index; represents the compression efficiency index adjustment factor; μ is the CPU usage impact coefficient, and the μ is obtained through statistical analysis based on CPU experimental data; CPU 1 Indicates CPU usage. 1 Obtained through system performance monitoring methods; Select different data encryption algorithms to encrypt the data according to the importance of the data to be updated.
3. According to the method for asynchronously updating data based on edge computing in claim 1, It is characterized in that The method of transmitting the data of the Internet of Vehicles terminal to the edge computing node according to the migration computing method and storing it in layers through the data caching method includes: Obtain the distance, bandwidth, processing capacity, and load of the edge computing node and record them as first data, obtain second data by normalizing the first data, and obtain a comprehensive weight value by using a linear weighted algorithm for the second data; obtain the comprehensive weight value from the Internet of Vehicles terminal to all edge computing nodes, and transmit the data of the Internet of Vehicles terminal to the edge computing node corresponding to the maximum value of the comprehensive weight value; The transmission strategy is adjusted by real-time monitoring of the load of the edge computing node, the growth rate of the load of the edge computing node is obtained, and the saturation time when the load of the edge computing node reaches a preset load saturation threshold is predicted according to the growth rate; the completion time of data transmission to the edge computing node is obtained, and when the completion time is after the saturation time, the edge computing node is reselected in descending order according to the comprehensive weight value; The data caching method is to perform hierarchical storage according to the access frequency and importance of data; the cache level of the hierarchical storage is preset according to the access speed and cache capacity.
4. A data asynchronous update system based on edge computing, It is characterized in that The system includes: a data acquisition and management module, a data asynchronous update trigger module, a data asynchronous update management module, and a data cache management module, and each module is sequentially connected to each other; A data acquisition and management module, which is used to acquire data from an Internet of Vehicles terminal, transmit the data from the Internet of Vehicles terminal to an edge computing node according to a migration calculation method, and store the data in layers through a data caching method; and monitor the data status in real time through the edge computing node, wherein the data status includes a timestamp of the data, data changes, and cache capacity; A data asynchronous update trigger module is used to set a trigger condition for asynchronous update according to the data state; obtain data whose data state meets the trigger condition for asynchronous update and record it as data to be updated; obtain the importance and update frequency of the data to be updated, and perform priority sorting by a statistical analysis method according to the importance and update frequency of the data to be updated to obtain a sorting result of the data to be updated; and perform asynchronous update according to the sorting result; A data asynchronous update management module, used to transmit the data to be updated to the receiving party according to the asynchronous update after passing the data compression algorithm and the data encryption algorithm; when the receiving party receives the data to be updated, it performs data verification; when monitoring the data to be updated fails to be transmitted, it retransmits according to the preset retry times and retry interval; A data cache management module, used to update the cache status after the monitoring edge computing node performs data update, wherein the updated cache status includes updating the management cache space and updating the cache timestamp; The data asynchronous update trigger module includes: A trigger condition management module, for trigger conditions including data expiration, data change, and cache capacity reaching a threshold; data expiration means that the current time of the data exceeds the data valid timestamp plus a preset time threshold; data change means that the data change exceeds the preset change threshold; cache capacity reaching the threshold means that the current cache capacity occupancy rate of the edge computing node exceeds the preset cache capacity occupancy rate threshold; by acquiring historical data received by the edge computing node, the historical data includes data transmission time, data type, data volume, and transmission frequency; through the historical data, an analysis result is obtained by trend analysis, and the trigger condition of asynchronous update is adjusted according to the analysis result; The data calculation module to be updated is used to obtain the set of edge computing nodes and record it as (k 1 ,…,k i ,…,k n ), the data to be updated ΔD of the i-th edge computing node at time s is obtained through the dynamic difference algorithm i (s); the data to be updated ΔD i The calculation formula of (s) is: Where D i (s) represents the data state at time s, D i (s-1) is the data state at the previous time (s-1), λ is the attenuation factor, is an exponential decay function, t 0 The time when the data was last updated; An asynchronous update execution module, used for performing asynchronous update according to the sorting result and according to the asynchronous update time, wherein the asynchronous update time is obtained by an asynchronous transmission time prediction method; The asynchronous update execution module includes the following steps: Obtain network bandwidth usage, establish a prediction model for network bandwidth changes over time through time series analysis, and obtain B(s) based on the prediction model, where B(s) represents the change of network bandwidth over time s; obtain R in real time through network performance analysis method i (s), said R i (s) represents the real-time transmission rate from the i-th edge computing node to the receiver; the asynchronous update time T is obtained by the asynchronous transmission time prediction method. w ; The asynchronous update time T w The calculation formula is: in, It is used to calculate the cumulative difference between the amount of network bandwidth B(s) and the amount of data required to be transmitted by all edge computing nodes in the time interval [0, t]; It represents the time required for all edge computing nodes to transmit per second at time s; Indicates the average transmission rate of all edge computing nodes The cumulative amount of network bandwidth required to complete data transmission under ΔD i is the incremental data to be updated that needs to be transmitted by the i-th edge computing node, and the incremental data to be updated ΔD i The calculation formula is: is the average transmission rate of the i-th edge computing node, and is obtained by obtaining R i (s) and obtain the average transmission rate by the mean calculation method The average transmission rate The calculation formula is: Where T is the preset time window, t' 0 is the preset start time.
5. According to the data asynchronous update system based on edge computing according to claim 4, It is characterized in that The data asynchronous update management module includes the following steps: The data to be updated is passed through a data compression algorithm to obtain first compressed data, the first compressed data is passed through a compression efficiency algorithm to obtain a compression efficiency evaluation value C, and a data compression algorithm is selected according to the compression efficiency evaluation value C; the calculation formula of the compression efficiency evaluation value C is: Among them, L 1 is the amount of data to be updated; L 2 is the amount of data of the first compressed data, log 2 L 1 -log 2 L 2 Indicates the compression efficiency of the data volume before and after compression; Indicates compression efficiency index; represents the compression efficiency index adjustment factor; μ is the CPU usage impact coefficient, and the μ is obtained through statistical analysis based on CPU experimental data; CPU 1 Indicates CPU usage. 1 Obtained through system performance monitoring methods; Select different data encryption algorithms to encrypt the data according to the importance of the data to be updated.
6. According to claim 4, a data asynchronous update system based on edge computing, It is characterized in that The data acquisition and management module includes the following steps: Obtain the distance, bandwidth, processing capacity, and load of the edge computing node and record them as first data, obtain second data by normalizing the first data, and obtain a comprehensive weight value by using a linear weighted algorithm for the second data; obtain the comprehensive weight value from the Internet of Vehicles terminal to all edge computing nodes, and transmit the data of the Internet of Vehicles terminal to the edge computing node corresponding to the maximum value of the comprehensive weight value; The transmission strategy is adjusted by real-time monitoring of the load of the edge computing node, the growth rate of the load of the edge computing node is obtained, and the saturation time when the load of the edge computing node reaches a preset load saturation threshold is predicted according to the growth rate; the completion time of data transmission to the edge computing node is obtained, and when the completion time is after the saturation time, the edge computing node is reselected in descending order according to the comprehensive weight value; The data caching method is to perform hierarchical storage according to the access frequency and importance of data; the cache level of the hierarchical storage is preset according to the access speed and cache capacity.
Citation Information
Patent Citations
Multilayer center asynchronous federal learning method applied to Internet of Vehicles
CN114827198A
Configuration management, performance management, and fault management to support edge computing
US20220086072A1
