Edge computing cooperative task unloading reputation evaluation method and system, medium and product

Abstract

The invention discloses an edge computing collaborative task unloading reputation evaluation method and system, a medium and a product, and relates to the field of edge computing. The method comprises the following steps: acquiring a task completion condition of an edge server participating in collaborative task unloading and running state data of the edge server; determining a direct reputation value of the edge server according to the task completion condition, and determining an indirect reputation value of the edge server according to the running state data; based on the direct reputation value and the indirect reputation value, a comprehensive reputation evaluation value of the edge server is determined, and the comprehensive reputation evaluation value is used for the service request end to determine a target edge server for task unloading. According to the method, the credible edge server can be effectively identified and stimulated, and the risk of task unloading failure caused by hardware or software exception and intentional or accidental damage caused by uncredibility of the edge server is reduced.

Description

Technical Field

[0001] The present invention relates to the field of edge computing technology, and in particular to an edge computing collaborative task offloading reputation evaluation method, system, medium and product. Background Art

[0002] With the explosive growth of mobile and IoT devices, online applications are becoming more and more complex, such as interactive online games, virtual reality, video analysis, and natural language processing. The resources required to execute these applications are increasing, and the terminal devices are limited by their physical size and do not have enough resources to execute these complex programs. Edge computing is developed on the basis of cloud computing. By pushing powerful computing and storage capabilities from the remote cloud to the edge of the network, it can help users complete tasks in a shorter time. Therefore, offloading tasks generated by terminal devices to edge nodes is an effective solution. In a real edge environment, the computing power of servers deployed at the edge of the network varies. Although these servers have computing power far exceeding that of terminal devices, they are ultimately limited. Faced with a large number of users' task offloading requests, a single edge server is difficult to meet the requirements.

[0003] Collaborative task offloading is one of the solutions to this problem. Adjacent edge servers deployed in a specific area can transmit data through high-speed links to form an edge service network. Through this network, edge servers can transfer tasks offloaded to themselves to other idle edge servers for computing.

[0004] However, in a dynamic and distributed environment, it is difficult for edge servers to perform internal maintenance like cloud servers. This results in edge servers not being completely trustworthy and being subject to intentional or accidental damage caused by various events. For example, during task offloading, hackers may delete or tamper with the calculation results of the task; edge servers may also encounter sudden hardware or software anomalies, resulting in task offloading failure. Summary of the invention

[0005] In view of the above-mentioned technical problems and defects, the purpose of the present invention is to provide an edge computing collaborative task offloading reputation evaluation method, system, medium and product, which can effectively identify and motivate trusted edge servers, reduce the risk of intentional or accidental damage due to untrustworthy edge servers, and task offloading failure caused by hardware or software anomalies.

[0006] To achieve the above-mentioned purpose, in a first aspect, the present invention provides an edge computing collaborative task offloading reputation evaluation method, which is applied to a reputation management unit of a task offloading collaborative system, wherein the task offloading collaborative system also includes a service request end and an edge server, wherein the service request end is connected to the edge server, and the edge server is connected to the reputation management unit; the method includes: obtaining the task completion status of the edge server participating in the collaborative task offloading, and the running status data of the edge server; determining the direct reputation value of the edge server according to the task completion status, and determining the indirect reputation value of the edge server according to the running status data; determining a comprehensive reputation evaluation value of the edge server based on the direct reputation value and the indirect reputation value, and the comprehensive reputation evaluation value is used by the service request end to determine the target edge server for task offloading.

[0007] The present invention can comprehensively consider the performance and stability of each server by obtaining the task completion status and operation status data of the edge server, thereby calculating the direct reputation value and the indirect reputation value. This comprehensive evaluation mechanism not only improves the prediction accuracy of the edge server behavior, but also enhances the anti-attack capability of the task offloading collaboration system, because the unreliable edge server will reduce the chance of being selected due to the low reputation value. And by determining the comprehensive reputation evaluation value, a clear indicator is provided for the service request end to select the best edge server for task offloading. This not only improves the success rate of task offloading, but also reduces the calculation delay and resource waste caused by edge server failure or poor performance. The reputation evaluation mechanism also encourages the edge server to improve its own service quality and stability to obtain a higher reputation value and more task offloading opportunities, reduce the risk of intentional or accidental damage caused by untrustworthy edge servers, and task offloading failure caused by hardware or software anomalies, thereby improving the performance of the entire edge service network.

[0008] In combination with some embodiments of the first aspect, in some embodiments, after the step of determining the comprehensive reputation evaluation value of the edge server based on the direct reputation value and the indirect reputation value, it also includes: determining the current time window of the comprehensive reputation evaluation value based on the comprehensive reputation evaluation value; and updating the comprehensive reputation evaluation value of the edge server based on the current time window.

[0009] By adopting the technical solution of this embodiment, the concept of time window is introduced after determining the comprehensive reputation evaluation value of the edge server, which significantly improves the dynamics and adaptability of the reputation evaluation. This method allows the system to adjust the evaluation frequency according to the reputation performance of the server and optimize resource usage. By setting the time window, the system can respond more flexibly to changes in the behavior of the edge server and quickly adapt to fluctuations in server performance. This real-time update mechanism ensures that the reputation evaluation value always reflects the latest service level, so that the service request end can always make task offloading decisions based on the latest reputation information, enhancing the responsiveness and user satisfaction of the entire system.

[0010] In combination with some embodiments of the first aspect, in some embodiments, the step of updating the comprehensive reputation evaluation value of the edge server based on the current time window includes: determining an update rate of the comprehensive reputation evaluation value according to the current time window; calculating the updated comprehensive reputation evaluation value through the update rate and the reputation value update formula, and the reputation value update formula includes: Rep3=(1-ρ)·Rep1+ρ·Rep2; Wherein, ρ represents the update rate, Rep1 represents the comprehensive reputation evaluation value in the previous time window, Rep2 represents the comprehensive reputation evaluation value in the current time window, and Rep3 represents the updated comprehensive reputation evaluation value.

[0011] The technical solution of this embodiment is used to explain in detail how to update the comprehensive reputation evaluation value of the edge server according to the current time window. This step is the key to ensure that the evaluation value reflects the latest performance of the server in a timely manner. By calculating the update rate and applying the reputation value update formula, the system can smoothly transition the reputation value and reduce the impact caused by instantaneous fluctuations. This method not only improves the accuracy of the evaluation, but also enhances the stability and fairness of the evaluation mechanism by weighing historical reputation and current performance. In addition, this update strategy also encourages edge servers to continue to provide high-quality services because their improvements can be quickly recognized by the system and reflected in the reputation value.

[0012] In conjunction with some embodiments of the first aspect, in some embodiments, the step of determining the update rate of the comprehensive reputation evaluation value according to the current time window includes: calculating the update rate of the comprehensive reputation evaluation value according to a time rate formula, and the time rate formula includes: Wherein, ρ represents the update rate, T represents the size of the current time window, T0 is the size of the reference time window, and k is the adjustment coefficient.

[0013] By adopting the technical solution of this embodiment, the time rate formula is introduced to determine the update rate, which further improves the flexibility and accuracy of the reputation evaluation system. This formula takes into account the size of the time window, allowing the system to dynamically adjust the update rate based on the size of the current time window relative to the reference window. This adaptive update mechanism enables the system to more accurately control the speed of change of the reputation value and better balance the impact of historical data and the latest data. In this way, it is possible to respond more effectively to changes in the behavior of the edge server, ensuring that the update of the reputation value is both rapid and accurate, thereby improving the reliability of the entire edge computing system and the trust of users.

[0014] In combination with some embodiments of the first aspect, in some embodiments, the task completion status includes task completion time, data integrity, and task completion degree; determining the direct reputation value of the edge server according to the task completion status includes: calculating the direct reputation value through a direct reputation value formula, and the direct reputation value formula includes: Among them, Rep. D represents the direct reputation value, φ T represents the estimated value of the task completion time, φ I Represents the evaluation value of the data integrity, φ C It represents the evaluation value of task completion, and ω1, ω2, and ω3 are weight factors respectively.

[0015] The technical solution of this embodiment is adopted to enhance the meticulousness and accuracy of reputation evaluation by describing in detail how to determine the direct reputation value of the edge server based on the task completion situation. Task completion time, data integrity and task completion are used as key indicators, and weights are assigned to them, making the evaluation results more comprehensive and objective. This method not only ensures that the evaluation value of the edge server can truly reflect its task execution capability, but also improves the level of refinement of the evaluation by considering the importance of different task attributes. This helps the service request end to more accurately judge the execution capability of the edge server, thereby making more reasonable task offloading decisions.

[0016] In combination with some embodiments of the first aspect, in some embodiments, the operating status data includes connection status, access status, and communication status; and determining the indirect reputation value of the edge server according to the operating status data includes: calculating the indirect reputation value by an indirect reputation value formula, and the indirect reputation value formula includes: Among them, Rep. I represents the indirect reputation value, φ L Indicates the calculated value of the connection condition, φ A Indicates the calculated value of the access situation, φM represents the calculated value of the communication situation, and ω4, ω5, and ω6 are weight factors respectively.

[0017] The technical solution of this embodiment is adopted to enhance the evaluation of the overall service quality of the server by describing how to determine the indirect reputation value of the edge server based on the operating status data. The connection situation, access situation and communication situation are taken into account, so that the evaluation results more comprehensively reflect the operating stability and performance of the server. This comprehensive evaluation method not only helps to identify key issues that may affect the quality of service, but also allows the system to adjust the importance of different status data according to actual conditions through the introduction of weight factors, thereby improving the flexibility and accuracy of the evaluation. This enables the service requester to have a more comprehensive understanding of the performance of the edge server, so as to make more informed task offloading choices.

[0018] In combination with some embodiments of the first aspect, in some embodiments, the task offloading collaboration system also includes a gateway, which is respectively connected to the reputation management unit, the service request end and the edge server, and the gateway is used to obtain the task completion status and the operation status data.

[0019] By adopting the technical solution of this embodiment, the efficiency and reliability of data collection are improved by introducing a gateway to obtain task completion status and operation status data. The gateway acts as a bridge between the reputation management unit, the service request end, and the edge server, ensuring the stability and security of data transmission. This structure not only optimizes the data flow, but also reduces the complexity of the system by centrally managing the data collection process. The existence of the gateway enables the reputation management unit to obtain the performance data of the edge server more accurately, thereby performing a more accurate reputation assessment. This not only improves the operating efficiency of the entire system, but also enhances the stability of the system and the trust of users.

[0020] In a second aspect, an embodiment of the present invention provides a task offloading collaboration system, comprising a reputation management unit, a service request end and an edge server, wherein the service request end is connected to the edge server, and the edge server is connected to the reputation management unit, and the reputation management unit comprises: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, and the computer program code comprises computer instructions, and the one or more processors call the computer instructions to enable the reputation management unit to execute the method described in the first aspect or the second aspect, and any possible implementation manner in the first aspect or the second aspect.

[0021] In a third aspect, the present invention provides a computer-readable storage medium comprising instructions, which, when executed on a reputation management unit of a task offloading collaboration system, causes the reputation management unit to execute the method described in the first aspect or the second aspect, and any possible implementation of the first aspect or the second aspect.

[0022] In a fourth aspect, the present invention provides a computer program product comprising instructions, which, when the above-mentioned computer program product runs on a reputation management unit of a task offloading collaboration system, enables the above-mentioned reputation management unit to execute the method described in the first aspect or the second aspect, and any possible implementation method of the first aspect or the second aspect.

[0023] It is understandable that the task offloading cooperation system provided in the second aspect, the storage medium provided in the third aspect, and the computer program product provided in the fourth aspect are all used to execute the method provided by the present invention. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, which will not be repeated here.

[0024] One or more technical solutions provided by the present invention have at least the following technical effects or advantages: 1. Improve the decision quality of task offloading: By comprehensively evaluating the direct and indirect reputation values ​​of edge servers, a more accurate and comprehensive method is provided to determine the reliability and performance of edge servers. This evaluation mechanism enables the service requester to make more informed task offloading decisions based on the comprehensive reputation evaluation value of the edge server. By considering the task completion status and operating status data, the present invention can identify edge servers with excellent or poor performance, thereby optimizing resource allocation and improving the success rate and efficiency of task offloading.

[0025] 2. Enhance the security of the edge computing environment: By establishing a reputation management unit and implementing dynamic reputation evaluation, the present invention can timely detect and reduce the risk of intentional attacks or accidental damage to edge servers. The reputation evaluation mechanism can identify potentially unsafe or unreliable edge servers and avoid offloading sensitive tasks to these servers, thereby protecting data from hacker attacks and tampering. In addition, by incentivizing edge servers to maintain a high level of service quality, the present invention helps to build a more secure and trusted edge computing environment.

[0026] 3. Realize the dynamics and adaptability of reputation evaluation: The present invention introduces the concepts of time window and update rate, so that the reputation evaluation value can be dynamically updated according to the latest performance of the edge server. This dynamic update mechanism can not only quickly reflect the changes in edge server performance, but also adapt to the changing needs and conditions in the edge computing environment. By adjusting the reputation evaluation value in real time, the present invention can more effectively respond to emergencies such as changes in edge server load, hardware failures or software anomalies, and ensure the continuity and stability of task offloading. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The drawings herein are incorporated into and constitute a part of the specification, showing embodiments consistent with the present invention, and together with the specification, are used to explain the principles of the present invention. Obviously, the drawings described below are only some embodiments of the present invention, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work. In the drawings: Figure 1 is a schematic diagram of the architecture of a task offloading collaboration system according to an embodiment of the present invention; Figure 2 It is a flow chart of a method for evaluating reputation of edge computing collaborative task offloading according to an embodiment of the present invention; Figure 3 It is a flow chart of another edge computing collaborative task offloading reputation evaluation method according to an embodiment of the present invention; Figure 4 It is a schematic diagram of the architecture of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The terms used in the following embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to be limiting of the present invention. As used in the specification of the present invention, the singular expressions "a", "a", "above", "the" and "this" are intended to also include plural expressions, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in the present invention refers to any or all possible combinations comprising one or more of the listed items.

[0029] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood as implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present invention, unless otherwise specified, the meaning of "plurality" is two or more.

[0030] It should also be noted that, unless otherwise clearly specified and limited, in the embodiments of the present invention, the terms such as "setting" and "connection" should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two components; it can be a wired communication connection or a wireless communication connection. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances. The embodiments of the present invention are described in detail below.

[0031] The present invention provides a task offloading cooperation system, such as Figure 1 As shown, it includes a reputation management unit 101 , a service request end 102 and an edge server 103 , the service request end 102 is connected to the edge server 103 , and the edge server 103 is connected to the reputation management unit 101 .

[0032] The reputation management unit 101 is a core component in the task offloading collaborative system, responsible for maintaining the reputation architecture of the entire system and ensuring the credibility of the edge server. As a trusted third party, it monitors and quantifies the performance of the edge server by implementing the edge computing collaborative task offloading reputation evaluation method.

[0033] The reputation management unit 101 collects data from edge servers, including key indicators such as task completion time, data integrity, task completion, connection status, access status, and communication status. Based on this data, the reputation management unit can calculate the direct reputation value and indirect reputation value of each edge server, and then combine these reputation values ​​to form a comprehensive reputation score for each server. In addition, the reputation management unit is also responsible for dynamically updating the reputation value, ensuring the accuracy and timeliness of the reputation value by analyzing the performance of the server in real time and adjusting the update rate. It also provides reputation data to service requesters to help them make wise task offloading decisions, thereby improving the efficiency and reliability of the entire system.

[0034] The service request end 102 may include mobile devices, IoT devices, enterprise applications, or other clients that require edge computing resources. The main function of the service request end is to initiate a task offloading request and send the task to be processed to the edge server. It first evaluates the reputation values ​​of different edge servers, and then decides which edge server to offload the task to based on these reputation values ​​to ensure that the task can be completed efficiently and safely. The service request end is also responsible for interacting with users, collecting user feedback, and sending this feedback to the reputation management unit as a reference for evaluating the service quality of the edge server. In addition, the service request end also has the functions of error detection and exception reporting, which can promptly detect problems in the task offloading process, and work with the reputation management unit to ensure that the problems are quickly resolved.

[0035] The edge servers 103 are deployed at the edge of the network, close to the user and service request end. They are equipped with certain computing and storage resources and can execute various task offloading requests. The main task of the edge server is to receive tasks from the service request end, use its computing power to quickly process these tasks, and return the results to the service request end. In order to improve the quality of service, the edge server needs to maintain a good connection state, efficient data processing capabilities and stable communication performance. They also maintain a connection with the reputation management unit and regularly upload task completion and service status data so that the reputation management unit can evaluate its reputation value. The edge server also needs to continuously optimize its own services and improve its reputation value based on the feedback and guidance of the reputation management unit to obtain more task offloading opportunities and resource allocation.

[0036] In some embodiments, the task offloading collaboration system further includes a gateway 104, which is respectively connected to the reputation management unit 101, the service request end 102 and the edge server 103, and the gateway 104 is used to obtain the task completion status of the edge server 103 participating in the collaborative task offloading, and the operating status data of the edge server 103.

[0037] Specifically, the service request end 102 and the edge server 103 are connected by a high-speed link, and the gateway 104 is connected to the high-speed link and can upload the detected data to the reputation management unit 101 .

[0038] The gateway 104 is a hub connecting the reputation management unit, the service request end and the edge server, and is responsible for the collection, transmission and preliminary processing of data. The existence of the gateway enables the system to efficiently monitor and manage the task execution and operation status of the edge server. It collects the task completion data of the edge server in real time, including key indicators such as task completion time, data integrity and task completion, and transmits this information to the reputation management unit for reputation evaluation. At the same time, the gateway also monitors the operation status of the edge server, such as connection stability, access frequency and communication efficiency, to ensure that these data can be fed back to the reputation management unit in a timely manner for calculating the indirect reputation value. In addition, the gateway is also responsible for processing the task offloading requests initiated by the service request end and forwarding these requests to the appropriate edge server. It also participates in the preliminary processing and verification of the data to ensure the accuracy and integrity of the data. Through the efficient work of the gateway, the task offloading collaboration system can realize real-time monitoring and dynamic management of the edge server, thereby improving the reliability and efficiency of the entire system.

[0039] The embodiment of the present invention provides an edge computing collaborative task offloading reputation evaluation method, which is applied to the task offloading collaborative system provided in the above embodiment. By obtaining the task completion status and operation status data of the edge server, the performance and stability of each server can be comprehensively considered, so as to calculate the direct reputation value and the indirect reputation value. This comprehensive evaluation mechanism not only improves the prediction accuracy of the edge server behavior, but also enhances the anti-attack capability of the task offloading collaborative system, because the unreliable edge server will reduce the chance of being selected due to the low reputation value. And by determining the comprehensive reputation evaluation value, a clear indicator is provided for the service request end to select the best edge server for task offloading. This not only improves the success rate of task offloading, but also reduces the computing delay and resource waste caused by edge server failure or poor performance. The reputation evaluation mechanism also encourages the edge server to improve its own service quality and stability to obtain higher reputation values ​​and more task offloading opportunities, reduce the risk of intentional or accidental damage caused by untrustworthy edge servers, and task offloading failure caused by hardware or software anomalies, thereby improving the performance of the entire edge service network.

[0040] In a dynamic and distributed edge computing environment, the method of this embodiment ensures that the system can quickly respond to changes in edge server behavior by updating the reputation value in real time, thereby improving the adaptability and flexibility of the system. Ultimately, this method not only enhances the security and credibility of edge computing, but also provides users with a more efficient and reliable service experience, promoting the practical application and widespread application of edge computing technology.

[0041] Combine the following Figure 2 , specifically introduce the edge computing collaborative task offloading reputation evaluation method of this embodiment, including the following steps: Step 201: Obtain the task completion status of the edge server participating in the collaborative task offloading and the operation status data of the edge server.

[0042] First, when the edge server completes or offloads a task, it will generate a detailed task log, including the task start and end time, data transmission integrity, task execution success rate and other information, and send this data to the reputation management unit in real time. At the same time, the edge server will also monitor its own operating status, including key performance indicators such as CPU and memory usage, network connection stability, and service response time, and periodically report this status data to the reputation management unit.

[0043] In some embodiments, the reputation management unit can also deploy a dedicated data collection agent or use a gateway or other device that can monitor network traffic and the interaction between edge servers in real time and collect performance data during task offloading. The reputation management unit can obtain the above task completion status and operation status data through these devices.

[0044] Step 202: determining a direct reputation value of the edge server according to the task completion status, and determining an indirect reputation value of the edge server according to the operation status data.

[0045] The reputation management unit determines the direct reputation value of edge servers by analyzing their task completion. Specifically, it first collects detailed data about offloaded tasks from edge servers, including task completion, completion time, and integrity of data transmission. Using this data, the reputation management unit uses a preset algorithm model to evaluate the performance of each task. For example, it may assign positive scores to tasks completed on time, give extra rewards to tasks completed ahead of schedule, and impose penalties on delayed or unfinished tasks. In addition, the integrity of the data and the quality of task execution are also key factors in the scoring. All these factors are taken into account and the direct reputation value of each edge server is calculated according to certain weights. This value directly reflects the server's ability and reliability to complete specific tasks.

[0046] At the same time, the reputation management unit also determines the indirect reputation value of the edge server based on the operating status data of the edge server. This involves monitoring the general operating status of the edge server, including key performance indicators such as the edge server's connection status, the edge server's access status, and the edge server's communication status. By tracking and analyzing these indicators in real time, the reputation management unit is able to assess the overall health and service quality of the edge server. For example, a high access rate may indicate that the edge server is heavily loaded, while a stable network connection and a fast response time indicate that the edge server is running well. The calculation of the indirect reputation value also takes into account historical data and trend analysis to identify early signs of performance degradation or service interruption. In this way, the indirect reputation value comprehensively reflects the stability of the edge server and its continued contribution to collaborative task offloading.

[0047] Step 203: Determine a comprehensive reputation evaluation value of the edge server based on the direct reputation value and the indirect reputation value.

[0048] The comprehensive reputation evaluation value is used by the service request end to determine the target edge server for task offloading.

[0049] Specifically, the reputation management unit weights and fuses the direct reputation value and indirect reputation value of the edge server. The direct reputation value reflects the performance of the edge server in executing a specific task, while the indirect reputation value considers the operating status of the edge server. The reputation management unit combines these two parts of the reputation value according to a certain weight ratio through a pre-set algorithm to calculate a comprehensive reputation evaluation value. This comprehensive value not only reflects the performance of the server in a single task, but also takes into account its long-term stability and reliability, providing a comprehensive reference indicator for the service request end.

[0050] Based on this comprehensive reputation evaluation value, the service requester can select an edge server with a relatively high reputation as the target edge server for task offloading to ensure that the task can be completed efficiently and safely. This method not only improves the success rate of task offloading, but also optimizes resource allocation, improving the operating efficiency and user satisfaction of the entire edge computing network.

[0051] The edge computing collaborative task offloading reputation evaluation method of the embodiment of the present invention provides an effective solution to the challenges in the edge computing environment to enhance the reliability and efficiency of task offloading. In the context of multiple user requests and limited server resources, this embodiment optimizes the decision-making process of task offloading by accurately evaluating the reputation of the edge server.

[0052] In some embodiments, the definition of the edge server is as follows: 1) Edge server: Use the set S = {s1, s2, ...s n} to indicate that s i Defined as a triple, expressed as ID i Edge servers i Identification; Edge servers i The direct reputation value ranges from 0 to 1 and is calculated based on the task completion of the edge server; Edge servers i The indirect reputation value ranges from 0 to 1 and is calculated based on the status of the edge server itself.

[0053] 2) Edge network: Use matrix E to represent the connection between edge servers. If e ij The value of 1 indicates that the edge server s i and j A connection link exists, otherwise it does not exist.

[0054] 3) Direct reputation: Use set A to represent the completion of the collaborative offloading task of the edge server. For each edge server, there is A(s i )={a T ,a I ,a C},a T Indicates the task completion time; a I Indicates the integrity of task transmission data; a C Indicates the task completion status. The task completion status is collected by the server requesting collaboration and then transmitted to the trusted third-party database for storage.

[0055] 4) Indirect reputation: Use set B to represent the status of edge servers. For each edge server, there is B(s i )={B L ,B A ,B M}, where B L Indicates the server connection status, B A Indicates the access status of the server, B M Indicates the communication status of the server. is the set of all connection behaviors, such as Represents resource acquisition behavior. It is the collection of all access situations. It is the set of all communication situations. X, Y, and Z represent the number of connection behaviors, access behaviors, and communication behaviors. These sets are collected by the gateway and transmitted to the database of the trusted third party for storage.

[0056] Based on the above definition of edge servers, Figure 3, to further introduce the edge computing collaborative task offloading reputation evaluation method of this embodiment, including the following steps: Step 301: Determine the direct reputation value of the edge server according to the task completion status of the edge server participating in the collaborative task offloading.

[0057] The task completion status includes task completion time, data integrity and task completion degree, and these data can be obtained from the task log records of the edge server through the gateway.

[0058] Task completion time refers to the total time required from the time the edge server starts processing a task to the time the task is completed. This indicator reflects the efficiency of the edge server in processing tasks and is one of the key factors in measuring server performance. In reputation evaluation, a shorter task completion time usually means a higher processing capacity and therefore receives a better evaluation.

[0059] Data integrity refers to the state in which data remains undamaged, lost, or tampered with during transmission and processing during task offloading. It ensures the accuracy and reliability of task input and output data. In reputation evaluation, data integrity is an important indicator for measuring the reliability of edge servers, because any data corruption may invalidate task results or produce wrong decisions.

[0060] Task completion refers to the thoroughness and quality of task execution, including whether all scheduled subtasks have been completed and whether the results meet the expected standards. This indicator not only considers whether the task is completed, but also the quality of completion. In reputation evaluation, high task completion indicates that the edge server can provide high-quality services and meet the needs of the service requester.

[0061] In some embodiments, step 301 specifically includes: The direct reputation value is calculated by a direct reputation value formula, and the direct reputation value formula includes: Rep D =f[A(s i )]; Among them, Rep. D Indicates the direct reputation value.

[0062] f() represents the direct reputation value function, A(s i ) represents the edge server s i completion of tasks.

[0063] φ TThe evaluation value of the task completion time can be determined by measuring the time required for the edge server to complete the task. Generally, the shorter this time is, the stronger the processing power of the server is. T The evaluation criteria may set an ideal time to complete the task and compare the actual completion time to it to calculate the evaluation value. For example, if the actual completion time is close to the ideal time, a higher score may be given; if it is far beyond the ideal time, the score will be lower.

[0064] φ I The evaluation value of the data integrity reflects whether the data remains intact during the task execution. It is usually determined by checking the consistency of the data before and after processing, which may involve error detection and verification mechanisms. If the data is not damaged or lost during the processing, then φ I If the data integrity is compromised, the score will be lower.

[0065] φ C The evaluation value of task completion can measure whether the task is completed according to the expected goal. It not only considers whether the task is completed, but also the quality of completion. C The value of may be determined based on an analysis of the task results compared to the expected results. If the task results fully meet or exceed expectations, the score will be higher; if the results do not meet expectations, the score will be reduced accordingly.

[0066] This setting facilitates unified calculation.

[0067] ω1, ω2, and ω3 are weight factors, respectively, satisfying ω1+ω2+ω3=1, and ω1,ω2,ω3∈(0,1). These weight factors reflect the importance of different evaluation indicators in the overall reputation evaluation. The choice of weights is usually based on the importance that system designers attach to different evaluation indicators and their impact on the success of task offloading.

[0068] In some embodiments, the evaluation value is calculated using the task completion status set A, specifically including: (1) Calculate the expected value of task completion using the β probability density function.

[0069] The beta probability density function is used to evaluate and quantify the performance of edge servers on task completion (such as task completion time, data integrity, or task completion). The expected performance of edge servers on these attributes can be calculated based on the number of positive and negative behaviors.

[0070] According to the Beta Reputation System (BRS), the positive and negative behaviors of edge servers over a period of time can be comprehensively considered. The β probability density function beta(p|s,f) is shown in the following formula: Where p represents the probability that the edge server successfully completes the task, Γ is the gamma function, s represents positive behavior, and f represents the number of negative behaviors.

[0071] beta(p|s,f) can represent the performance of the edge server in terms of a certain task completion situation (such as task completion time, data integrity, or task completion degree).

[0072] The expected value of task completion is expressed as the expected value of beta E[beta(p|s,f)], which can be calculated by the following formula: (2) Calculate the evaluation value function based on the expected value of task completion.

[0073] Assuming that the task completion of the edge server follows the Beta distribution, the expected values ​​of the task completion time, data integrity, and completion degree of the edge server can be modeled as the expected value of the Beta distribution, as shown in the following formula: where φ is the expected value of the beta distribution. The s+f+2 in the denominator ensures that the formula remains valid even if there is no negative behavior, that is, f=0.

[0074] Moreover, the initial reputation value is set to 0.5, which takes into account the neutral attitude when there is no historical behavior.

[0075] (2) Add a penalty factor to the evaluation function.

[0076] In order to increase the impact of negative behavior on the reputation value, a penalty factor θ is added so that negative behavior has a more significant impact on the reputation value than positive behavior. Then, The value of can be calculated by the following formula Among them, θ T ,θ I ,θ C Represent the negative behavior penalty factors for task completion time, data integrity, and task completion, respectively. T ,θ I ,θ C ≥ 1. The penalty factor is used to amplify the impact of negative behavior, making the reputation value more sensitive to negative behavior.

[0077] S T ,SI ,S C : Represents the positive behavior counts of task completion time, data completeness, and task completion, respectively.

[0078] F T ,F I ,F C : Represents the negative behavior counts of task completion time, data completeness, and task completion, respectively.

[0079] Through the above steps, it can be ensured that the reputation evaluation method can comprehensively and fairly evaluate the performance of edge servers and effectively motivate servers to provide high-quality services, while ensuring the reliability and efficiency of task offloading.

[0080] Step 302: Determine the indirect reputation value of the edge server according to the running status data of the edge server.

[0081] The operation status data includes connection status, access status and communication status.

[0082] Connectivity refers to the ability of an edge server to establish and maintain connections with other network entities, including indicators such as connection stability, connection speed, and connection success rate.

[0083] The access situation reflects the frequency and pattern of edge servers being accessed by other users or systems, including the rate, time distribution, and source distribution of access requests.

[0084] The communication situation involves the efficiency and quality of data transmission between edge servers or between edge servers and service requesters, including transmission speed, data integrity, and error rate.

[0085] These operating status data are crucial for evaluating the performance and reliability of edge servers, as they directly affect the efficiency and success rate of task offloading. By real-time monitoring and analyzing these operating status data, the reputation management unit can calculate the indirect reputation value of the edge server, thereby providing important reference information for the service requester to help it make more reasonable task offloading decisions.

[0086] In some embodiments, this step may include: calculating the indirect reputation value through an indirect reputation value formula.

[0087] The indirect reputation value formula includes: Among them, Rep. I Represents the indirect reputation value. L Indicates the calculated value of the connection condition, φ A Indicates the calculated value of the access situation, φ MA calculated value representing the communication situation.

[0088] ω4, ω5, and ω6 are weight factors, corresponding to the importance of connection, access, and communication in calculating indirect reputation values. These weight factors allow the system to adjust the impact of different status indicators on the overall reputation evaluation according to specific application scenarios and requirements.

[0089] Specifically, φ can be obtained by the following formula: L ,φ A ,φ M The calculated value is: Therefore, the indirect reputation value formula can specifically include: In the above series of formulas: α x Represents the weight factor of the xth sub-behavior in the connection situation. In the connection evaluation of the edge server, there may be multiple different sub-behaviors to consider, such as connection stability, average connection speed, peak connection speed, response time of connection request, etc. Each sub-behavior may have different impacts on the overall connection evaluation. x It is the weight used to quantify the degree of this influence.

[0090] β y Represents the weight factor of the yth sub-behavior in the access situation. The access situation may include different sub-behaviors such as access frequency, access pattern, average response time, etc. Each of these sub-behaviors may have different contributions to the overall access situation evaluation of the edge server. y Used to measure the size of this contribution.

[0091] γ z Represents the weight factor of the zth sub-behavior in the communication situation. The communication situation may include sub-behaviors such as data transmission speed, transmission error rate, and packet loss rate. z This reflects the relative importance of these sub-behaviors in the evaluation of the communication situation.

[0092] Indicates the evaluation value of the xth sub-behavior of the connection situation. This sub-behavior refers to a specific aspect related to the connection, such as connection stability, connection speed, connection success rate, etc. is the positive behavior count of this subbehavior, is the negative behavior count, θ L is a penalty factor for connection conditions, used to amplify the impact of negative behavior.

[0093] Indicates the evaluation value of the yth sub-behavior of the access situation. This sub-behavior includes access frequency, access mode, average response time and other aspects related to the server being accessed. is the positive behavior count of this subbehavior, is the negative behavior count, θ A It is the penalty factor for access.

[0094] The evaluation value of the zth sub-behavior representing the communication situation. This sub-behavior includes aspects related to data communication such as data transmission speed, transmission error rate, and packet loss rate. is the positive behavior count of this subbehavior, is the negative behavior count, θ C is the penalty factor for communication conditions.

[0095] Step 303: Determine a comprehensive reputation evaluation value of the edge server based on the direct reputation value and the indirect reputation value.

[0096] Specifically, it can be calculated according to the following formula: Rep=λ1·Rep D +λ2·Rep I ; Among them, λ1, λ2 are weight factors, λ1+λ2=1, Rep D is the direct reputation value, Rep I It is an indirect reputation value.

[0097] Step 304: determine the current time window of the comprehensive reputation evaluation value according to the comprehensive reputation evaluation value.

[0098] The time window is designed to capture the latest changes in the comprehensive reputation evaluation value of the edge server, while reducing the reputation evaluation error caused by short-term fluctuations. If the comprehensive reputation evaluation value of the edge server changes rapidly, the time window may be shortened to reflect these changes more quickly; if the reputation of the server is stable, the time window may be extended to reduce the frequency of evaluation and the consumption of computing resources.

[0099] Specifically, the time window defines a specific time interval during which edge server performance data, such as task completion, service quality, and system stability, is collected and evaluated. The start and end points of the time window are used to determine which data is used to calculate and update the server reputation value, which allows the system to take a snapshot of the server's performance within a fixed time period and perform reputation scoring accordingly.

[0100] By setting a time window, you can control the frequency and timeliness of reputation value updates, ensuring that the reputation evaluation can reflect the latest performance of the server while avoiding instability caused by too frequent updates. The size of the time window can be dynamically adjusted based on system requirements, the speed of change of server behavior, and the reputation evaluation strategy.

[0101] In this embodiment, the process of determining the current time window of the comprehensive reputation evaluation value by the reputation management unit involves evaluating the historical reputation performance and recent reputation change trend of the edge server. First, the reputation management unit analyzes the server's past reputation records, including historical reputation values ​​and the change rate of the reputation values. Then, combined with the server's current task unloading load, task completion status, and user feedback and other real-time data, the reputation management unit calculates an adaptive time window length.

[0102] For edge servers with higher reputation values, it means that they have performed well in past task offloading, so they can be given a longer time window. A longer time window means that the reputation value is updated less frequently, which can reduce the frequent evaluation of these servers and save computing resources.

[0103] For edge servers with lower reputation values, shorter time windows are assigned, perhaps because of poor past performance or uncertainty. Shorter time windows allow reputation values ​​to be updated more frequently to quickly reflect the latest performance and behavior of these servers, ensuring the safety and efficiency of task offloading.

[0104] By adjusting the size of the time window, the number of reputation evaluation calculations for edge servers can be reduced. For servers with good reputation, reducing the number of evaluations means reducing the consumption of computing resources.

[0105] For servers with poor reputation, although the evaluation frequency is higher, due to their relatively small number, the overall resource consumption can still be controlled within a reasonable range, while ensuring close monitoring of these servers.

[0106] Step 305: Update the comprehensive reputation evaluation value of the edge server based on the current time window.

[0107] The process of updating the comprehensive reputation evaluation value of the edge server based on the current time window by the reputation management unit is a dynamic adjustment mechanism. At the end of each time window, the reputation management unit collects new data points, including the task completion status, operation status data or other relevant indicators of the edge server in the time window, and calculates the comprehensive reputation evaluation value in the current time window.

[0108] Then, the reputation management unit recalculates the comprehensive reputation evaluation value using the comprehensive reputation evaluation value in the current time window and the previous comprehensive reputation evaluation value.

[0109] In some embodiments, this step may specifically include: (1) Determine the update rate of the comprehensive reputation evaluation value according to the current time window.

[0110] The update rate is determined based on the length of the current time window and the frequency of behavior changes of the edge server. In the reputation management unit, a baseline update rate is usually set, which is inversely proportional to the length of the time window.

[0111] Specifically, if the time window is short, it means that the reputation value needs to be updated more frequently to capture the latest performance of the edge server, so a higher update rate is set. On the contrary, if the time window is long, it means that the reputation value can be updated at less frequent intervals, thereby reducing the computational overhead and increasing the stability of the reputation value, and a lower update rate is set.

[0112] In addition, if the reputation of the edge server changes dramatically or the server's task offload load suddenly increases, the reputation management unit may increase the update rate to quickly adapt to these changes. Conversely, if the server performs stably, the update rate may be reduced to avoid frequent adjustments to the reputation value due to short-term fluctuations.

[0113] In this way, the reputation management unit can flexibly adjust the update rate to ensure that the comprehensive reputation evaluation value can not only reflect the actual performance of the edge server in a timely manner, but also maintain the stability and efficiency of the system.

[0114] Specifically, the reputation management unit can calculate the update rate of the comprehensive reputation evaluation value according to the time rate formula. In some embodiments, the time rate formula includes: in: ρ represents the update rate, ranging from 0 to 1.

[0115] T represents the size of the current time window.

[0116] T0 represents the size of the reference time window, which can be determined based on system design or empirical values.

[0117] k represents an adjustment coefficient, which is generally a positive number and is used to control the influence of the time window on the update rate.

[0118] e is a natural constant.

[0119] If the current time window T is smaller (i.e., updated more frequently), e -k·(T-T0) The value of will be very small, making ρ close to 1, indicating that the update rate is very high and the new reputation value will be reflected faster.

[0120] If the current time window T is large (i.e., the update frequency is low), e -k·(T-T0) The value of will be close to 1, making ρ close to 0, indicating that the update rate is very low and the impact of the new reputation value on the reputation will be slowed down.

[0121] T0 can be set to the ideal time window size expected by the system, while k can be adjusted according to the actual response requirements of the system and the frequency of changes in edge server behavior. The larger the value of k, the more obvious the impact of the time window on the update rate, that is, the change of the time window will be reflected in the update rate more quickly.

[0122] This calculation method allows the system to dynamically adjust the update rate based on the actual task offloading and performance of the edge server to achieve more effective reputation management.

[0123] By adjusting the time window and update rate, it is possible to ensure that the reputation value is updated in a timely manner while avoiding the stability of the reputation evaluation due to frequent fluctuations.

[0124] (2) Calculate the updated comprehensive reputation evaluation value through the update rate and the reputation value update formula, and the reputation value update formula includes: Rep3=(1-ρ)·Rep1+ρ·Rep2; in: ρ represents the update rate, which is a parameter between 0 and 1 and is used to control the weight of the new and old reputation values ​​in the update process.

[0125] Rep1 represents the comprehensive reputation evaluation value in the previous time window, that is, the reputation value of the edge server at the end of the previous time window.

[0126] Rep2 represents the comprehensive reputation evaluation value within the current time window, that is, the reputation value calculated based on the performance of the edge server within the current time window.

[0127] Rep3 represents the updated comprehensive reputation evaluation value, that is, the new reputation value determined by the edge server at the end of the current time window is the updated comprehensive reputation evaluation value.

[0128] (1-ρ)·Rep1: This part represents the weight of the reputation value of the previous time window in the update. 1-ρ is the retention factor of the historical reputation value, which determines the influence of the historical reputation value in the calculation of the new reputation value. If ρ is small, the influence of the historical reputation value will be greater; if ρ is large, the influence of the historical reputation value will be reduced.

[0129] ρ·Rep2: This part represents the weight of the reputation value calculated in the current time window in the update. ρ is the update factor of the new reputation value, which determines the influence of the newly calculated reputation value in the update process. If ρ is large, the newly calculated reputation value will be reflected in the updated reputation value more quickly; if ρ is small, the updated reputation value will retain more historical reputation values.

[0130] The reputation management unit updates the reputation value of the edge server in real time according to its performance in the current time window through the reputation value update formula.

[0131] By adjusting the value of ρ, a balance can be achieved between the old and new reputation values. If the system needs to respond quickly to changes in the behavior of edge servers, the value of ρ can be increased; if the system needs to stabilize the reputation value and reduce the impact of short-term fluctuations, the value of ρ can be reduced.

[0132] Using this weighted average method can avoid drastic fluctuations in the reputation value and achieve a smooth transition of the reputation value.

[0133] Through the dynamic update mechanism, edge servers can be encouraged to continue to provide high-quality services because their performance will be reflected in the reputation value more quickly. It can also reduce the adverse impact on the reputation value of edge servers caused by short-term abnormal performance, thereby evaluating edge servers more fairly. The reputation management unit can adjust the value of ρ according to the actual operation situation to adapt to different operation strategies and goals.

[0134] Through the above-mentioned manner, this embodiment can realize dynamic and real-time updating of the reputation value of the edge server, ensuring that the reputation value can accurately reflect the latest performance and behavior of the edge server.

[0135] The embodiment of the present invention comprehensively considers the task completion status and operation status data of the edge server to evaluate its reputation, and guides the service request end to make a decision on task offloading based on the reputation.

[0136] The reputation management unit first collects the task completion status of the edge server, including task completion time, data integrity, and task completion. These data reflect the performance of the edge server in performing specific tasks. Based on these task completion status, the reputation management unit calculates the direct reputation value of the edge server through the direct reputation value formula, which takes into account the weight factors of different task completion status.

[0137] The reputation management unit also collects the running status data of the edge server, including connection status, access status and communication status. These data reflect the overall service quality and stability of the edge server. Using the indirect reputation value formula, the reputation management unit combines these running status data to calculate the indirect reputation value of the edge server. The formula also takes into account the weight factors of different running status data.

[0138] After obtaining the direct reputation value and the indirect reputation value, the reputation management unit combines them to determine the comprehensive reputation evaluation value of the edge server. This comprehensive reputation evaluation value is an important basis for the service request end to determine the target edge server for task offloading.

[0139] In order to dynamically adjust the reputation evaluation value, the reputation management unit also determines the current time window based on the comprehensive reputation evaluation value, and updates the comprehensive reputation evaluation value of the edge server based on the current time window. The update process includes calculating the update rate and obtaining the updated comprehensive reputation evaluation value through the reputation value update formula. The calculation of the update rate takes into account the size of the current time window and the size of the base time window, as well as an adjustment factor.

[0140] The present invention not only improves the efficiency and reliability of task offloading, but also enhances the stability and security of the edge computing environment through a comprehensive reputation evaluation method. By dynamically updating the reputation evaluation value, the system can adapt to changes in the edge computing environment and better meet the needs of the service requester.

[0141] The method provided in the above embodiment can be executed by a reputation management unit, which is an electronic device. The electronic device in the embodiment of the present invention is described below from the perspective of hardware processing. Figure 4 , is a schematic diagram of a physical device structure of an electronic device in an embodiment of the present invention.

[0142] It should be noted that Figure 4 The structure of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0143] like Figure 4As shown, the electronic device includes a central processing unit (CPU) 401, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 402 or the program loaded from the storage part 408 to the random access memory (RAM) 403, such as executing the method described in the above embodiment. In RAM 403, various programs and data required for system operation are also stored. CPU 401, ROM 402 and RAM 403 are connected to each other through bus 404. Input / output (I / O) interface 405 is also connected to bus 404.

[0144] The following components are connected to the I / O interface 405: an input section 406 including an audio input device, a button switch, etc.; an output section 407 including a liquid crystal display (LCD) and an audio output device, an indicator light, etc.; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the I / O interface 405 as needed. A removable medium 411, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 410 as needed so that a computer program read therefrom is installed into the storage section 408 as needed.

[0145] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 409, and / or installed from the removable medium 411. When the computer program is executed by the central processing unit (CPU) 401, various functions defined in the present invention are performed.

[0146] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0147] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram may represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box may also occur in an order different from that marked in the accompanying drawings.

[0148] Specifically, the electronic device of this embodiment includes a processor and a memory, the memory is coupled to one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and one or more processors call the computer instructions to enable the electronic device to execute the method provided by the above embodiment.

[0149] As another aspect, the present invention further provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiment; or may exist independently without being assembled into the electronic device. The above storage medium carries one or more computer programs, and when the above one or more computer programs are executed by a processor of the electronic device, the electronic device implements the method provided in the above embodiment.

[0150] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention.

[0151] As used in the above embodiments, the term "when..." may be interpreted to mean "if..." or "after..." or "in response to determining..." or "in response to detecting...", depending on the context. Similarly, the phrases "upon determining..." or "if (the stated condition or event) is detected" may be interpreted to mean "if determining..." or "in response to determining..." or "upon detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)", depending on the context.

[0152] Those skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by computer programs to instruct related hardware, and the programs can be stored in computer-readable storage media. When the programs are executed, they can include the processes of the above-mentioned method embodiments. The aforementioned storage media include: ROM or random access memory RAM, magnetic disk or optical disk and other media that can store program codes.

Claims

1. A reputation evaluation method for edge computing collaborative task offloading, characterized in that: A reputation management unit applied to a task offloading collaboration system, wherein the task offloading collaboration system further comprises a service request end and an edge server, wherein the service request end is connected to the edge server, and the edge server is connected to the reputation management unit; the method comprises: Obtaining task completion status of edge servers participating in collaborative task offloading and operation status data of the edge servers; Determining a direct reputation value of the edge server according to the task completion status, and determining an indirect reputation value of the edge server according to the operation status data; Based on the direct reputation value and the indirect reputation value, a comprehensive reputation evaluation value of the edge server is determined, and the comprehensive reputation evaluation value is used by the service request end to determine a target edge server for task offloading.

2. The method according to claim 1, characterized in that After the step of determining the comprehensive reputation evaluation value of the edge server based on the direct reputation value and the indirect reputation value, the method further includes: Determining a current time window of the comprehensive reputation evaluation value according to the comprehensive reputation evaluation value; The comprehensive reputation evaluation value of the edge server is updated based on the current time window.

3. The method according to claim 2, characterized in that The step of updating the comprehensive reputation evaluation value of the edge server based on the current time window includes: Determining an update rate of the comprehensive reputation evaluation value according to the current time window; The updated comprehensive reputation evaluation value is calculated by using the update rate and the reputation value update formula, wherein the reputation value update formula includes: Rep3=(1-ρ)·Rep1+ρ·Rep2; Wherein, ρ represents the update rate, Rep1 represents the comprehensive reputation evaluation value in the previous time window, Rep2 represents the comprehensive reputation evaluation value in the current time window, and Rep3 represents the updated comprehensive reputation evaluation value.

4. The method according to claim 3, characterized in that The step of determining the update rate of the comprehensive reputation evaluation value according to the current time window includes: The update rate of the comprehensive reputation evaluation value is calculated according to the time rate formula, and the time rate formula includes: Wherein, ρ represents the update rate, T represents the size of the current time window, T0 is the size of the reference time window, and k is the adjustment coefficient.

5. The method according to any one of claims 1 to 4, characterized in that: The task completion status includes task completion time, data integrity and task completion degree; and determining the direct reputation value of the edge server according to the task completion status includes: The direct reputation value is calculated by a direct reputation value formula, and the direct reputation value formula includes: Among them, Rep. D represents the direct reputation value, φ T represents the estimated value of the task completion time, φ I represents the evaluation value of the data integrity, φ C It represents the evaluation value of task completion, and ω1, ω2, and ω3 are weight factors respectively.

6. The method according to claim 5, characterized in that The operation status data includes connection status, access status and communication status; and determining the indirect reputation value of the edge server according to the operation status data includes: The indirect reputation value is calculated by an indirect reputation value formula, and the indirect reputation value formula includes: Among them, Rep. I represents the indirect reputation value, φ L represents the calculated value of the connection condition, φ A represents the calculated value of the access situation, φ M represents the calculated value of the communication situation, and ω4, ω5, and ω6 are weight factors respectively.

7. The method according to any one of claims 1 to 4, characterized in that: The task offloading cooperation system further includes a gateway, which is respectively connected to the reputation management unit, the service request end and the edge server, and is used to obtain the task completion status and the operation status data.

8. A task offloading collaboration system, characterized in that: It includes a reputation management unit, a service request end and an edge server, wherein the service request end is connected to the edge server, the edge server is connected to the reputation management unit, and the reputation management unit includes one or more processors and memories; The memory is coupled to the one or more processors, and the memory is used to store computer program codes, wherein the computer program codes include computer instructions, and the one or more processors call the computer instructions to enable the reputation management unit to perform the method according to any one of claims 1 to 7.

9. A computer-readable storage medium storing computer instructions, characterized in that: When the computer instructions are executed on a reputation management unit of the task offloading cooperation system, the reputation management unit is caused to execute the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that When the computer program product runs on a reputation management unit of a task offloading cooperation system, the reputation management unit is enabled to execute the method according to any one of claims 1 to 7.

Images