Task scheduling methods and devices, access nodes and task scheduling systems

By using a blockchain-based task scheduling method in edge computing to dynamically calculate the reputation value of edge nodes, the problem of insufficient trust guarantee mechanism in edge computing is solved, and multi-dimensional trust assessment and robustness improvement of edge nodes are achieved.

CN123123445APending Publication Date: 2026-11-17CHINA TELECOM CORP LTD TECHNOLOGY INNOVATION CENTER +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611433242.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-15
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

The lack of a unified trust guarantee mechanism in edge computing leads to information asymmetry and conflicts of interest among computing resource providers, task schedulers, and task executors, making it difficult to identify malicious behaviors of edge nodes, such as false results and resource abuse.

Method used

By using a blockchain-based task scheduling method, the reputation value of edge nodes is dynamically calculated. Taking into account the expected reliability, expected behavior, and trust volatility, a multi-dimensional trust assessment is provided to dynamically identify and punish malicious behavior and improve system robustness.

Benefits of technology

It enables multi-dimensional trust assessment of edge nodes, dynamically identifies and punishes malicious behavior, and improves the trust guarantee and robustness of edge computing systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123123445A_ABST
    Figure CN123123445A_ABST
Patent Text Reader

Abstract

This disclosure provides a task scheduling method and apparatus, an access node, and a task scheduling system. The task scheduling method includes: parsing a received task request to obtain a scheduling request for scheduling tasks; sending the scheduling request to a task scheduling apparatus so that the task scheduling apparatus can allocate tasks to edge nodes for processing tasks according to scheduling decisions; receiving task execution status sent by the edge nodes; determining the reputation value of the edge nodes based on the task execution status; and sending the reputation value to a blockchain node.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This disclosure relates to the field of blockchain technology, and in particular to a task scheduling method and apparatus, access node and task scheduling system. Background Technology

[0002] The edge computing market has many participants, and edge nodes are usually managed by different organizations or individuals. They may engage in malicious activities such as providing false data, abusing resources, leaking privacy, and tampering with results. There is a lack of necessary trust between end users and unfamiliar edge servers.

[0003] Currently, edge computing lacks a unified trust guarantee mechanism. Information asymmetry and conflicts of interest often exist among computing resource providers, task schedulers, and task executors, posing a significant challenge to ensuring the reliability of services. Summary of the Invention

[0004] The inventors noted that in related technologies, blockchain-based edge computing task scheduling systems often only focus on the trusted transmission of data to address the insecurity challenges of data sharing caused by edge servers. However, such solutions can only guarantee that data is tamper-proof and traceable during transmission, but they are insufficient to constrain malicious behavior on the edge nodes themselves, and cannot identify problems such as edge nodes outputting false results, abusing resources, or exhibiting multiple levels of abnormal behavior.

[0005] Accordingly, this disclosure provides a task scheduling method that dynamically calculates and updates the reputation value of edge nodes based on multi-source dynamic information during task execution, and propagates it through a blockchain network. This provides information support for selecting trustworthy edge nodes with good performance and high reliability, thereby ensuring the trustworthiness of the service.

[0006] In a first aspect of this disclosure, a task scheduling method is provided, executed by an access node, comprising: parsing a received task request to obtain a scheduling request for scheduling tasks; sending the scheduling request to a task scheduling device so that the task scheduling device allocates the task to an edge node for processing the task according to a scheduling decision; receiving task execution status sent by the edge node; determining the reputation value of the edge node based on the task execution status; and sending the reputation value to a blockchain node.

[0007] In some embodiments, determining the reputation value of an edge node includes: determining the expected reliability value, the expected behavioral performance value, and the trust volatility of the edge node. The reputation value of the edge node is obtained based on the expected reliability value, the expected behavioral performance value, and the trust volatility of the edge node.

[0008] In some embodiments, determining the expected reliability value of an edge node includes obtaining the expected reliability value of the edge node based on the number of successful and failed task executions of the edge node.

[0009] In some embodiments, obtaining the expected reliability value of an edge node based on the number of successful and failed task executions includes: weighting and accumulating the task execution results of the current task period and the historical number of successful executions of the task in the previous task period, and using the weighted accumulation result as the first parameter of the Beta distribution; weighting and accumulating the task execution results of the current task period and the historical number of failed executions of the task in the previous task period, and using the weighted accumulation result as the second parameter of the Beta distribution; and obtaining the expected reliability value of the edge node based on the first and second parameters.

[0010] In some embodiments, determining the expected reliability value of an edge node based on a first parameter and a second parameter includes: calculating the sum of the first parameter and the second parameter to obtain a first intermediate value; and calculating the ratio of the first parameter to the first intermediate value to obtain the expected reliability value of the edge node.

[0011] In some embodiments, weighting and accumulating the task execution result of the current task period and the historical success count of the task in the previous task period, and using the weighted accumulation result as the first parameter of the Beta distribution, includes: calculating the product of the historical success count of the task in the previous task period and a first discount factor to obtain a second intermediate value; and calculating the sum of the second intermediate value and the task execution result to obtain the first parameter.

[0012] In some embodiments, weighting and accumulating the task execution result of the current task period and the historical failure count of the task in the previous task period, and using the weighted accumulation result as the second parameter of the Beta distribution, includes: calculating the product of the historical failure count of the task in the previous task period and a first discount factor to obtain a third intermediate value; calculating the difference between a first predetermined parameter value and the task execution result to obtain a fourth intermediate value; and calculating the sum of the third and fourth intermediate values ​​to obtain the second parameter.

[0013] In some embodiments, determining the expected behavior of an edge node includes obtaining the expected behavior of the edge node based on different levels of behavior generated by the edge node during task execution.

[0014] In some embodiments, obtaining the expected performance value of an edge node based on different levels of behavior generated during task execution includes: weighting and accumulating the behavior quality score of the current task period and the historical positive behavior of the task in the previous task period, and using the weighted accumulation result as the third parameter of the Beta distribution; weighting and accumulating the behavior quality score of the current task period and the historical negative behavior of the task in the previous task period, and using the weighted accumulation result as the fourth parameter of the Beta distribution; and obtaining the expected performance value of the edge node based on the third and fourth parameters.

[0015] In some embodiments, obtaining the expected behavior of the edge node based on the third and fourth parameters includes: calculating the sum of the third and fourth parameters to obtain a fifth intermediate value; and calculating the ratio of the third parameter to the fifth intermediate value to obtain the expected behavior of the edge node.

[0016] In some embodiments, weighting and accumulating the behavioral quality score of the current task period and the historical positive behavior of the task in the previous task period, and using the weighted accumulation result as the third parameter of the Beta distribution, includes: calculating the product of the historical positive behavior of the task in the previous task period and the second discount factor to obtain a sixth intermediate value; calculating the product of the behavioral quality score and the scaling factor to obtain a seventh intermediate value; and calculating the sum of the sixth and seventh intermediate values ​​to obtain the third parameter.

[0017] In some embodiments, weighting and accumulating the behavior quality score of the current task period and the historical negative behavior of the task in the previous task period, and using the weighted accumulation result as the fourth parameter of the Beta distribution, includes: calculating the product of the historical negative behavior of the task in the previous task period and the second discount factor to obtain the eighth intermediate value; calculating the difference between the second predetermined parameter value and the behavior quality score to obtain the ninth intermediate value; calculating the product of the ninth intermediate value and the scaling factor to obtain the tenth intermediate value; and calculating the sum of the eighth and tenth intermediate values ​​to obtain the fourth parameter.

[0018] In some embodiments, determining the trust volatility of an edge node includes obtaining the trust volatility of the edge node based on the expected reliability value and the expected behavioral performance value of the edge node.

[0019] In some embodiments, obtaining the trust volatility of an edge node based on its expected reliability and expected behavior includes: calculating a weighted sum of the expected reliability and expected behavior to obtain a preliminary reputation value for the current task period; and obtaining the trust volatility for the current task period based on the preliminary reputation value for the current task period, the preliminary reputation value for the previous task period, the trust volatility for the previous task period, and a smoothing factor.

[0020] In some embodiments, obtaining the trust volatility of the current task period based on the preliminary credit value of the current task period, the preliminary credit value of the previous task period, the trust volatility of the previous task period, and a smoothing factor includes: calculating the difference between the preliminary credit value of the current period and the preliminary credit value of the previous task period to obtain an eleventh intermediate value; calculating the square of the eleventh intermediate value to obtain a twelfth intermediate value; and calculating the weighted sum of the trust volatility of the previous task period and the twelfth intermediate value to obtain the trust volatility of the current task period, wherein the weight of the trust volatility of the previous task period is the smoothing factor, and the weight of the twelfth intermediate value is the difference between a third predetermined parameter value and the smoothing factor.

[0021] In some embodiments, obtaining the reputation value of an edge node based on its expected reliability, expected behavior, and trust volatility includes: calculating the square root of the trust volatility for the current task period to obtain the thirteenth intermediate value; calculating the product of the penalty coefficient and the thirteenth intermediate value to obtain the fourteenth intermediate value; and calculating the difference between the initial reputation value for the current task period and the fourteenth intermediate value to obtain the reputation value of the edge node.

[0022] In some embodiments, after receiving the task execution result, the execution result is verified. If the verification is successful, the execution result is sent to the task requester.

[0023] In a second aspect of this disclosure, an access node is provided, comprising: a memory configured to store instructions; and a processor coupled to the memory, the processor being configured to execute the method described in the first aspect based on the instructions stored in the memory.

[0024] In a third aspect of this disclosure, a task scheduling method is provided, executed by a task scheduling device, comprising: receiving a scheduling request for scheduling tasks sent by an access node; sending the scheduling request to a blockchain node so that the blockchain node triggers a smart contract to generate a scheduling decision based on the task status, the status of candidate edge nodes, and the reputation value of the candidate edge nodes; receiving the scheduling decision sent by the blockchain node; and assigning the task to the edge node specified in the scheduling decision.

[0025] In some embodiments, after receiving scheduling decisions from blockchain nodes, task execution monitoring is established to periodically report execution status and resource usage to blockchain nodes.

[0026] In a fourth aspect of this disclosure, a task scheduling apparatus is provided, comprising: a memory configured to store instructions; and a processor coupled to the memory, the processor being configured to execute the method described in the third aspect based on the instructions stored in the memory.

[0027] In a fifth aspect of this disclosure, a task scheduling system is provided, comprising: an access node as described in the second aspect, a task scheduling device as described in the fourth aspect, a blockchain node, and edge nodes. The blockchain node is configured to trigger a smart contract based on a request sent by the task scheduling device, generate a scheduling decision based on the task status, the status of candidate edge nodes, and the reputation value of the candidate edge nodes, and send the scheduling decision to the task scheduling device. The edge nodes are configured to process tasks assigned by the task scheduling device.

[0028] In a sixth aspect of this disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions that, when executed by a processor, implement the method as described in any of the above embodiments.

[0029] In a seventh aspect of this disclosure, a computer program product is provided, including computer instructions, wherein when executed by a processor, the computer instructions implement the method as described in any of the above embodiments.

[0030] Other features and advantages of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a flowchart illustrating a task scheduling method according to an embodiment of the present disclosure;

[0033] Figure 2 This is a schematic diagram of the structure of an access node according to an embodiment of the present disclosure;

[0034] Figure 3 This is a flowchart illustrating a task scheduling method according to another embodiment of the present disclosure;

[0035] Figure 4 This is a schematic diagram of the structure of a task scheduling device according to an embodiment of the present disclosure;

[0036] Figure 5 This is a schematic diagram of the structure of a task scheduling system according to an embodiment of the present disclosure;

[0037] Figure 6 This is a flowchart illustrating a task scheduling method according to yet another embodiment of the present disclosure. Detailed Implementation

[0038] The technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit this disclosure or its application or use. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without creative effort are within the scope of protection of this disclosure.

[0039] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of this disclosure.

[0040] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0041] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0042] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0043] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0044] Figure 1 This is a task scheduling method according to one embodiment of the present disclosure. In some embodiments, the following task scheduling method is executed by the access node, including steps 101-105.

[0045] In step 101, the received task request is parsed to obtain the scheduling request used to schedule the task.

[0046] For example, an access node receives a task request, parses the task's basic information, including the task type, required resources, and real-time requirements, and encapsulates it into a standardized scheduling request.

[0047] In step 102, a scheduling request is sent to the task scheduling device so that the task scheduling device can allocate the task to the edge node for processing the task based on the scheduling decision.

[0048] For example, the access node pushes the scheduling request to the task scheduling device through the blockchain transaction interface.

[0049] In step 103, the task execution status sent by the edge node is received.

[0050] In step 104, the reputation value of the edge node is determined based on the task execution status.

[0051] In some embodiments, the expected reliability value, expected behavior value, and trust volatility of the edge node are determined. Based on the expected reliability value, expected behavior value, and trust volatility of the edge node, the reputation value of the edge node is obtained.

[0052] Therefore, by comprehensively considering the expected reliability, expected behavior, and trust volatility of edge nodes, the reputation value of edge nodes can be obtained. This enables a joint evaluation of the multi-dimensional trust attributes of edge nodes, providing a quantitative basis for node screening for task scheduling smart contracts on the blockchain, effectively reducing the impact of malicious nodes, dynamically identifying and punishing bad and malicious behaviors, and improving the robustness of the system.

[0053] In some embodiments, the expected reliability value of an edge node is obtained based on the number of successful and failed task executions.

[0054] Therefore, by obtaining the expected reliability value based on the number of successful and failed tasks executed by edge nodes, the success or failure of task execution by edge nodes can be quantitatively evaluated, which can objectively reflect the reliability level of edge nodes in completing tasks and provide a reliability dimension basis for the reputation evaluation of edge nodes.

[0055] In some embodiments, the task execution result of the current task period and the historical number of successes of the task in the previous task period are weighted and accumulated, and the weighted accumulation result is used as the first parameter of the Beta distribution. The task execution result of the current task period and the historical number of failures of the task in the previous task period are weighted and accumulated, and the weighted accumulation result is used as the second parameter of the Beta distribution.

[0056] For example, during task period t, the task execution result of edge node i is represented by a binary variable. This indicates that success is recorded as 1 and failure as 0, representing the reliability parameter of edge node i. It follows a Beta distribution, as shown in formula (1).

[0057] (1)

[0058] In formula (1), Let be the reliability parameter of the i-th edge node. The historical number of successful task executions for the i-th edge node is used as the first parameter of the Beta distribution. The historical number of failures in executing tasks for the i-th edge node is used as the second parameter of the Beta distribution.

[0059] In some embodiments, the task execution result of the current task cycle and the historical number of successes of the task in the previous task cycle are weighted and accumulated, and the weighted accumulation result is used as the first parameter of the Beta distribution.

[0060] For example, to reflect the time decay effect, a first discount factor is introduced. This is used to reduce the weight of past task records.

[0061] In some embodiments, the product of the historical success count of the task in the previous task cycle and the first discount factor is calculated to obtain the second intermediate value.

[0062] For example, the second intermediate value is shown in formula (2).

[0063] Second median value (2)

[0064] In formula (2), This represents the historical success count for the (t-1)th task cycle. As the first discount factor, and .

[0065] In some embodiments, the sum of the second intermediate value and the task execution result is calculated to obtain the first parameter.

[0066] For example, the first parameter is shown in formula (3).

[0067] (3)

[0068] In formula (3), The first parameter represents the historical number of successes in the t-th task cycle. This represents the historical success count for the (t-1)th task cycle. As the first discount factor, and , This represents the task execution result for the current task cycle.

[0069] This reduces the weight of historical success counts recorded in past task cycles, thus improving the accuracy of edge node reliability assessment.

[0070] In some embodiments, the task execution result of the current task cycle and the number of historical failures of the task in the previous task cycle are weighted and accumulated, and the weighted accumulation result is used as the second parameter of the Beta distribution.

[0071] For example, to reflect the time decay effect, a first discount factor is introduced. This is used to reduce the weight of past task records.

[0072] In some embodiments, the product of the historical number of failures of the task in the previous task cycle and the first discount factor is calculated to obtain a third intermediate value.

[0073] For example, the third intermediate value is shown in formula (4).

[0074] Third median (4)

[0075] In formula (4), This represents the historical failure count for the (t-1)th task cycle. As the first discount factor, and .

[0076] In some embodiments, the difference between the first predetermined parameter value and the task execution result is calculated to obtain a fourth intermediate value.

[0077] For example, the fourth intermediate value is shown in formula (5).

[0078] Fourth intermediate value (5)

[0079] In formula (5), The first predetermined parameter value, This represents the task execution result for the current task cycle.

[0080] For example, The value of is 1.

[0081] In some embodiments, the sum of the third intermediate value and the fourth intermediate value is calculated to obtain the second parameter.

[0082] For example, the second parameter is shown in formula (6).

[0083] (6)

[0084] In formula (6), The second parameter represents the historical number of failures in the t-th task cycle. This represents the historical number of failures in the (t-1)th task cycle. As the first discount factor, and , The first predetermined parameter value, This represents the task execution result for the current task cycle.

[0085] For example, in When the value of is 1, the second parameter is as shown in formula (7).

[0086] (7)

[0087] In formula (7), The second parameter represents the historical number of failures in the t-th task cycle. This represents the historical number of failures in the (t-1)th task cycle. As the first discount factor, and , This represents the task execution result for the current task cycle.

[0088] Therefore, by reducing the weight of historical failure counts recorded in past task cycles, the accuracy of edge node reliability assessment can be improved.

[0089] In some embodiments, the expected reliability value of the edge node is obtained based on the first parameter and the second parameter.

[0090] In some embodiments, the sum of the first parameter and the second parameter is calculated to obtain a first intermediate value.

[0091] For example, the first intermediate value is shown in formula (8).

[0092] First Intermediate Value (8)

[0093] In formula (8), The first parameter represents the historical number of successes in the t-th task cycle. The second parameter represents the number of historical failures in the t-th task cycle.

[0094] In some embodiments, the ratio of the first parameter to the first intermediate value is calculated to obtain the expected reliability value of the edge node.

[0095] For example, the expected reliability of edge nodes is shown in formula (9).

[0096] (9)

[0097] In formula (9), This represents the expected reliability value of the i-th edge node during the current task period t. The first parameter represents the historical number of successes in the t-th task cycle. The second parameter represents the number of historical failures in the t-th task cycle.

[0098] Therefore, the posterior expectation value is obtained based on the shape parameter of the Beta distribution in the reliability dimension, thereby quantifying the reliability of edge nodes in performing tasks.

[0099] It should be noted that by updating the data based on the number of successful and failed tasks performed by the edge nodes and using the task execution results of the current task cycle, historical reliability and short-term reliability can be integrated to achieve joint modeling of the historical and instantaneous reliability of the edge nodes, thereby improving the accuracy and anti-interference capability of the reliability assessment results.

[0100] In some embodiments, the expected performance value of the edge node is obtained based on the different levels of behavior generated by the edge node during task execution.

[0101] Therefore, by obtaining expected values ​​of behavior based on different levels of behavior generated during the execution of tasks by edge nodes, and by quantitatively evaluating the different levels of behavior generated by edge nodes during the execution process, we can objectively reflect the behavioral performance of edge nodes in completing tasks and provide a behavioral performance dimension basis for the reputation evaluation of edge nodes.

[0102] In some embodiments, a weighted sum is calculated between the behavior quality score of the current task period and the historical positive behavior of the task in the previous task period, and the weighted sum is used as the third parameter of the Beta distribution. Similarly, a weighted sum is calculated between the behavior quality score of the current task period and the historical negative behavior of the task in the previous task period, and the weighted sum is used as the fourth parameter of the Beta distribution.

[0103] It should be noted that the third and fourth parameters are the two shape parameters of the Beta distribution under the behavioral performance dimension, while the first and second parameters mentioned above are the two shape parameters of the Beta distribution under the reliability dimension. It should be understood that the third and fourth parameters are merely global designations to distinguish different dimensions; they themselves are still the two shape parameters of the Beta distribution.

[0104] For example, the behavior quality score for the current task cycle is determined as follows: During task cycle t, edge node i may produce different levels of behavior during task execution, ranging from minor to severe anomalies, and is divided into several levels, denoted as . The preset behavior weight coefficient is , is used to represent the relative credibility of each behavior level. The behavior level is mapped to the behavior quality score, as shown in formula (10).

[0105] (10)

[0106] In formula (10), The behavior quality score for the i-th edge node. is the behavior weight coefficient of the i-th edge node.

[0107] For example, when K = 4, the behavior weights of the four levels of behavior are set sequentially, as shown in formula (11).

[0108] (11)

[0109] For example, according to formula (11), when edge node i exhibits slightly abnormal behavior, its behavior quality score is 0.95; when edge node i exhibits generally abnormal behavior, its behavior quality score is 0.8; when edge node i exhibits abnormal behavior, its behavior quality score is 0.5; and when edge node i exhibits severely abnormal behavior, its behavior quality score is 0.1.

[0110] Behavioral parameters of edge node i It follows a Beta distribution, as shown in formula (12).

[0111] (12)

[0112] In formula (12), Let i be the behavior parameter of the i-th edge node. The historical positive behaviors generated during the task execution at the i-th edge node are used as the third parameter of the Beta distribution. The historical negative behaviors generated during the execution of tasks at the i-th edge node are used as the fourth parameter of the Beta distribution.

[0113] In some embodiments, the behavior quality score of the current task cycle and the historical positive behavior of the task in the previous task cycle are weighted and accumulated, and the weighted accumulation result is used as the third parameter of the Beta distribution.

[0114] For example, a second discount factor is introduced to reflect the time decay effect. This is used to reduce the weight of past task records.

[0115] In some embodiments, the sixth intermediate value is obtained by multiplying the historical positive behavior of the task in the previous task cycle and the second discount factor.

[0116] For example, the sixth intermediate value is shown in formula (13).

[0117] Sixth Intermediate Value (13)

[0118] In formula (13), For the historical positive behavior generated by the i-th edge node during task execution in the (t-1)-th task cycle, As the second discount factor, and .

[0119] In some embodiments, the product of the behavior quality score for the current task cycle and the scaling factor is calculated to obtain the seventh intermediate value.

[0120] For example, the seventh intermediate value is shown in formula (14).

[0121] Seventh Intermediate Value (14)

[0122] In formula (14), s is a scaling factor used to control the strength of the influence of a single behavior quality score on the reputation value, and , The behavior quality score of the i-th edge node in the t-th task cycle.

[0123] In some embodiments, the sum of the sixth intermediate value and the seventh intermediate value is calculated to obtain the third parameter.

[0124] For example, the third parameter is shown in formula (15).

[0125] (15)

[0126] In formula (15), The third parameter represents the historical positive behavior in the t-th task cycle. This represents the historical positive behavior during the (t-1)th task cycle. As the second discount factor, and s is a scaling factor used to control the strength of the impact of a single behavior's quality score on the reputation score, and , This is the behavioral quality score for the current task cycle.

[0127] This reduces the influence weight of historical positive behaviors recorded in past task cycles, thereby improving the accuracy of reliable assessment of edge node behavior.

[0128] In some embodiments, the behavior quality score of the current task cycle and the historical negative behavior of the task in the previous task cycle are weighted and accumulated, and the weighted accumulation result is used as the fourth parameter of the Beta distribution.

[0129] For example, a second discount factor is introduced to reflect the time decay effect. This is used to reduce the weight of past task records.

[0130] In some embodiments, the product of the historical negative behavior of the task in the previous task cycle and the second discount factor is calculated to obtain the eighth intermediate value.

[0131] For example, the eighth intermediate value is shown in formula (16).

[0132] Eighth intermediate value (16)

[0133] In formula (16), This represents the historical negative behavior during the (t-1)th task cycle. As the second discount factor, and .

[0134] In some embodiments, the difference between the second predetermined parameter value and the behavior quality score of the current task cycle is calculated to obtain a ninth intermediate value.

[0135] For example, the ninth intermediate value is shown in formula (17).

[0136] Ninth Intermediate Value (17)

[0137] In formula (17), For the second predetermined parameter value, This is the behavioral quality score for the current task cycle.

[0138] For example, The value of is 1.

[0139] In some embodiments, the product of the ninth intermediate value and the scaling factor is calculated to obtain the tenth intermediate value.

[0140] For example, the tenth intermediate value is shown in formula (18).

[0141] Tenth Intermediate Value (18)

[0142] In formula (18), s is a scaling factor used to control the strength of the influence of a single behavior quality score on the reputation value, and , For the second predetermined parameter value, The behavior quality score of the i-th edge node in the t-th task cycle.

[0143] In some embodiments, the sum of the eighth intermediate value and the tenth intermediate value is calculated to obtain the fourth parameter.

[0144] For example, the fourth parameter is shown in formula (19).

[0145] (19)

[0146] In formula (19), The fourth parameter represents the historical negative behavior in the t-th task cycle. This represents the historical negative behavior during the (t-1)th task cycle. As the second discount factor, and , Here, s is the second predetermined parameter value, and s is the scaling factor. , This is the behavioral quality score for the current task cycle.

[0147] For example, in When the value of is 1, the fourth parameter is as shown in formula (20).

[0148] (20)

[0149] In formula (20), The fourth parameter represents the historical negative behavior in the t-th task cycle. This represents the historical negative behavior during the (t-1)th task cycle. As the second discount factor, and s is the scaling factor, and , This is the behavioral quality score for the current task cycle.

[0150] This reduces the weight of negative historical behaviors recorded in past task cycles, thus improving the accuracy of reliable assessment of edge node behavior.

[0151] In some embodiments, the expected behavior of the edge node is obtained based on the third and fourth parameters.

[0152] In some embodiments, the sum of the third and fourth parameters is calculated to obtain the fifth intermediate value.

[0153] For example, the fifth intermediate value is shown in formula (21).

[0154] Fifth intermediate value (twenty one)

[0155] In formula (21), The third parameter represents the historical positive behavior in the t-th task cycle. The fourth parameter represents the historical negative behavior in the t-th task cycle.

[0156] In some embodiments, the ratio of the third parameter to the fifth intermediate value is calculated to obtain the expected value of the behavior of the edge node.

[0157] For example, the expected value of the behavior of edge nodes is shown in formula (22).

[0158] (twenty two)

[0159] In formula (22), This represents the expected performance of the i-th edge node during the current task period t. The third parameter represents the historical positive behavior in the t-th task cycle. The fourth parameter represents the historical negative behavior in the t-th task cycle.

[0160] Therefore, the posterior expectation value is obtained based on the shape parameter of the Beta distribution in the behavioral performance dimension, thereby quantifying the accuracy of the reliable assessment of the behavior of edge nodes performing tasks.

[0161] It should be noted that by combining historical positive and negative behaviors generated during the execution of tasks by edge nodes and updating the behavior quality score of the current task cycle, it is possible to integrate historical behavior performance with short-term behavior performance, thereby achieving joint modeling of historical and instantaneous behavior performance of edge nodes and improving the accuracy and anti-interference ability of behavior performance evaluation results.

[0162] In some embodiments, the trust volatility of an edge node is obtained based on the expected reliability of the edge node and the expected behavior of the edge node.

[0163] Therefore, based on the expected reliability and expected behavior of edge nodes, the stability of edge nodes over time is quantitatively assessed, providing a basis for the stability dimension of edge node reputation evaluation.

[0164] In some embodiments, a weighted sum of the reliability expectation and the behavioral performance expectation is calculated to obtain a preliminary reputation value for the current task cycle.

[0165] For example, the initial reputation value for the current task cycle is shown in formula (23).

[0166] (twenty three)

[0167] In formula (23), Let be the initial reputation value of the i-th edge node at the current task period t. Let be the expected reliability value of the i-th edge node during the current task period t. This represents the expected performance of the i-th edge node during the current task period t. The weights are the expected values ​​of reliability. The weights are the expected values ​​of the behavioral performance, where, .

[0168] In some embodiments, the trust volatility of the current task period is obtained based on the initial credit value of the current task period, the initial credit value of the previous task period, the trust volatility of the previous task period, and the smoothing factor.

[0169] For example, the difference between the initial reputation value of the current period and the initial reputation value of the previous task period is calculated to obtain the eleventh intermediate value.

[0170] For example, the eleventh intermediate value is shown in formula (24).

[0171] Eleventh Intermediate Value (twenty four)

[0172] In formula (24), Let be the initial reputation value of the i-th edge node at the current task period t. This is the initial reputation value of the i-th edge node at the time of the previous task cycle t-1.

[0173] Calculate the square of the eleventh median to obtain the twelfth median.

[0174] For example, the twelfth intermediate value is shown in formula (25).

[0175] Twelfth median (25)

[0176] In formula (25), Let be the initial reputation value of the i-th edge node at the current task period t. This is the initial reputation value of the i-th edge node at the time of the previous task cycle t-1.

[0177] The weighted sum of the trust volatility of the previous task period and the twelfth median value is calculated to obtain the trust volatility of the current task period. The weight of the trust volatility of the previous task period is the smoothing factor, and the weight of the twelfth median value is the difference between the third predetermined parameter value and the smoothing factor.

[0178] For example, the trust volatility of the current task cycle is shown in formula (26).

[0179] (26)

[0180] In formula (26), Let be the trust volatility of the i-th edge node in the current task cycle. As a smoothing factor, Let be the trust volatility of the i-th edge node in the previous task cycle. For the third predetermined parameter, Let be the initial reputation value of the i-th edge node at the current task period t. This is the initial reputation value of the i-th edge node at the time of the previous task cycle t-1.

[0181] For example, The value of is 1.

[0182] For example, in When the value of is 1, the trust volatility of the current task cycle is shown in formula (27).

[0183] (27)

[0184] In formula (27), Let be the trust volatility of the i-th edge node in the current task cycle. As a smoothing factor, Let be the trust volatility of the i-th edge node in the previous task cycle. Let be the initial reputation value of the i-th edge node at the current task period t. This is the initial reputation value of the i-th edge node at the time of the previous task cycle t-1.

[0185] It should be noted that the smoothing factor is used to control the sensitivity of confidence volatility to short-term changes, and the value of the smoothing factor ranges from 0 to 1.

[0186] Therefore, it is possible to characterize the stability of edge nodes over time. The higher the trust volatility, the more frequently the reputation value of the edge node changes and the poorer the stability, thus objectively representing the degree of oscillation of the trust state of the edge node.

[0187] In some embodiments, the reputation value of an edge node is obtained based on the expected reliability value, the expected behavior value, and the trust volatility of the edge node.

[0188] For example, the arithmetic square root of the trust volatility of the current task cycle is used to obtain the thirteenth intermediate value.

[0189] For example, the thirteenth intermediate value is shown in formula (28).

[0190] Thirteenth Intermediate Value (28)

[0191] In formula (28), Let be the trust volatility of the i-th edge node in the current task cycle.

[0192] The product of the penalty coefficient and the thirteenth median is used to obtain the fourteenth median.

[0193] For example, the fourteenth intermediate value is shown in formula (29).

[0194] Fourteenth Intermediate Value (29)

[0195] In formula (29), It is the penalty coefficient, and , Let be the trust volatility of the i-th edge node in the current task cycle.

[0196] The reputation value of the edge node is obtained by calculating the difference between the initial reputation value of the current task cycle and the fourteenth intermediate value.

[0197] For example, the reputation value of an edge node is shown in formula (30).

[0198] (30)

[0199] In formula (30), Let be the reputation value of the i-th edge node in the current task cycle. Let be the initial reputation value of the i-th edge node at the current task period t. It is the penalty coefficient, and , Let be the trust volatility of the i-th edge node in the current task cycle.

[0200] It's important to note that when the trust volatility of edge nodes is high, the system automatically lowers their final reputation score to prevent short-term manipulation. This volatility penalty mechanism suppresses drastic reputation fluctuations, improving the stability and resilience of the assessment results.

[0201] In step 105, the reputation value is sent to the blockchain node.

[0202] For example, the final reputation value, after normalization, is limited to the range [0, 1], and can be further divided into different levels based on a set threshold, such as high trust, medium trust, low trust, and untrustworthy. Reputation value-related information is uploaded by the access node to the blockchain node for storage, and is used as input for subsequent task scheduling decisions.

[0203] Therefore, by introducing a multi-dimensional dynamic evaluation mechanism that comprehensively considers three dimensions of edge nodes—reliability, behavioral performance, and trust volatility—and combining a discount factor to achieve a dynamic balance between historical trust and short-term trust, we can more accurately depict the reputation change trend of edge nodes at different time scales, thereby improving the timeliness, stability, and robustness of trust evaluation.

[0204] In some embodiments, after receiving the task execution result, the execution result is verified. If the verification is successful, the execution result is sent to the task requester.

[0205] Therefore, by verifying the execution results, the accuracy of the data received by the task requester can be improved.

[0206] Figure 2 This is a schematic diagram of the structure of an access node according to an embodiment of the present disclosure.

[0207] like Figure 2As shown, the access node 200 can be represented in the form of a general computing device. The access node 200 includes a memory 201, a processor 202, and a bus 203 connecting different system components.

[0208] The memory 201 may include, for example, system memory, non-volatile storage media, etc. The system memory may store, for example, an operating system, application programs, a boot loader, and other programs. The system memory may include volatile storage media, such as random access memory (RAM) and / or cache memory. The non-volatile storage media may store, for example, instructions for executing corresponding embodiments of at least one task scheduling method of this disclosure. Non-volatile storage media include, but are not limited to, disk storage, optical storage, flash memory, etc.

[0209] The processor 202 can be implemented using a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete hardware components such as discrete gates or transistors. Accordingly, each module, such as the acquisition module, the calculation module, and the adjustment module, can be implemented by executing instructions in the central processing unit (CPU) running memory to perform the corresponding steps, or by implementing dedicated circuits that perform the corresponding steps.

[0210] For example, processor 202 is configured as a memory-based instruction execution implementation such as Figure 1 The method involved in any of the embodiments.

[0211] Bus 203 can use any of the various bus architectures. For example, bus architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MCA) bus, and Peripheral Component Interconnect (PCI) bus.

[0212] The interfaces 204, 205, and 206 of access node 200, as well as memory 201 and processor 202, can be connected via bus 203. Input / output interface 204 provides a connection interface for input / output devices such as monitors, mice, and keyboards. Network interface 205 provides a connection interface for various networked devices. Storage interface 206 provides a connection interface for external storage devices such as floppy disks, USB flash drives, and SD cards.

[0213] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus, and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations thereof, can be implemented by computer-readable program instructions.

[0214] These computer-readable program instructions are provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable device to produce a machine, such that execution of the instructions by the processor produces means for implementing the functions specified in one or more boxes of the flowchart and / or block diagram.

[0215] These computer-readable program instructions may also be stored in a computer-readable storage medium. These instructions cause a computer to work in a particular manner to produce an article of manufacture, including instructions that implement the functions specified in one or more boxes in a flowchart and / or block diagram.

[0216] This disclosure may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects.

[0217] This disclosure also provides a computer-readable storage medium storing computer instructions that, when executed by a processor, implement... Figure 1 The method involved in any of the embodiments.

[0218] This disclosure also provides a computer program product, including computer instructions, wherein the computer instructions, when executed by a processor, implement as follows: Figure 1 The method involved in any of the embodiments.

[0219] Figure 3 This is a flowchart illustrating a task scheduling method according to another embodiment of the present disclosure. In some embodiments, the following task scheduling method is executed by a task scheduling device, including steps 301-304.

[0220] In step 301, a scheduling request for scheduling tasks is received from the access node.

[0221] In step 302, a scheduling request is sent to the blockchain node so that the blockchain node can trigger a smart contract to generate a scheduling decision based on the task status, the status of the candidate edge nodes, and the reputation value of the candidate edge nodes.

[0222] In step 303, the scheduling decision sent by the blockchain node is received.

[0223] In step 304, the task is assigned to the edge node specified by the scheduling decision.

[0224] For example, after receiving the scheduling decision determined on the chain, the task is specifically scheduled according to the node allocation scheme output by the smart contract.

[0225] Therefore, by leveraging smart contracts to generate reliable scheduling decisions, and by generating scheduling decisions based on the reputation values ​​of candidate edge nodes, the task scheduling device can allocate tasks, thereby improving the security and rationality of task allocation.

[0226] In some embodiments, after receiving scheduling decisions from blockchain nodes, task execution monitoring is established to periodically report execution status and resource usage to blockchain nodes.

[0227] Therefore, the task scheduling process, task execution status, and edge node reputation updates are all recorded through blockchain nodes, thus achieving immutability and traceability, facilitating post-event auditing and dispute resolution.

[0228] This enables blockchain nodes to obtain timely information on the execution status of edge nodes, facilitating subsequent updates to reputation values ​​and dynamic adjustments to scheduling decisions, thereby improving the adaptability of task scheduling.

[0229] Figure 4 This is a schematic diagram of the structure of a task scheduling device according to an embodiment of the present disclosure.

[0230] like Figure 4 As shown, the task scheduling device 400 can be represented in the form of a general computing device. The task scheduling device 400 includes a memory 401, a processor 402, and a bus 403 connecting different system components.

[0231] The memory 401 may include, for example, system memory, non-volatile storage media, etc. The system memory may store, for example, an operating system, application programs, a boot loader, and other programs. The system memory may include volatile storage media, such as random access memory (RAM) and / or cache memory. The non-volatile storage media may store, for example, instructions for executing corresponding embodiments of at least one task scheduling method of this disclosure. Non-volatile storage media include, but are not limited to, disk storage, optical storage, flash memory, etc.

[0232] Processor 402 can be implemented using a general-purpose processor, digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA) or other programmable logic devices, discrete hardware components such as discrete gates or transistors. Accordingly, each module, such as the acquisition module, calculation module, and adjustment module, can be implemented by executing instructions in the central processing unit (CPU) running memory to perform the corresponding steps, or by implementing dedicated circuits that perform the corresponding steps.

[0233] For example, processor 402 is configured for memory-based instruction execution implementation such as Figure 3 The method involved in any of the embodiments.

[0234] Bus 403 can use any of the various bus architectures. For example, bus architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MCA) bus, and Peripheral Component Interconnect (PCI) bus.

[0235] The interfaces 404, 405, and 406 of the task scheduling device 400, as well as the memory 401 and processor 402, can be connected via bus 403. Input / output interface 404 provides a connection interface for input / output devices such as monitors, mice, and keyboards. Network interface 405 provides a connection interface for various networked devices. Storage interface 406 provides a connection interface for external storage devices such as floppy disks, USB flash drives, and SD cards.

[0236] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus, and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations thereof, can be implemented by computer-readable program instructions.

[0237] These computer-readable program instructions are provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable device to produce a machine, such that execution of the instructions by the processor produces means for implementing the functions specified in one or more boxes of the flowchart and / or block diagram.

[0238] These computer-readable program instructions may also be stored in a computer-readable storage medium. These instructions cause a computer to work in a particular manner to produce an article of manufacture, including instructions that implement the functions specified in one or more boxes in a flowchart and / or block diagram.

[0239] This disclosure may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects.

[0240] This disclosure also provides a computer-readable storage medium storing computer instructions that, when executed by a processor, implement... Figure 3 The method involved in any of the embodiments.

[0241] This disclosure also provides a computer program product, including computer instructions, wherein the computer instructions, when executed by a processor, implement as follows: Figure 3 The method involved in any of the embodiments.

[0242] Figure 5 This is a schematic diagram of the structure of a task scheduling system according to an embodiment of the present disclosure.

[0243] like Figure 5 As shown, the task scheduling system includes an access node 200, a task scheduling device 400, an edge node 501, and a blockchain node 502. The access node 200 is as follows: Figure 2 The access node involved in any of the embodiments. The task scheduling device 400 is as follows: Figure 4 The task scheduling device involved in any of the embodiments.

[0244] Edge node 501 is configured to handle tasks assigned by task scheduler 400.

[0245] Blockchain node 502 is configured to trigger a smart contract based on a request sent by task scheduling device 400, generate a scheduling decision based on task status, candidate edge node status, and candidate edge node reputation value, and send the scheduling decision to task scheduling device 400.

[0246] Figure 6 This is a flowchart illustrating a task scheduling method according to another embodiment of the present disclosure, including steps 601-610.

[0247] In step 601, the access node parses the received task request to obtain a scheduling request for scheduling the task.

[0248] For example, an access node receives a task request, parses the task's basic information, including the task type, required resources, and real-time requirements, and encapsulates it into a standardized scheduling request.

[0249] In step 602, the access node sends a scheduling request to the task scheduling device.

[0250] For example, the access node pushes the scheduling request to the task scheduling device through the blockchain transaction interface.

[0251] In step 603, the task scheduling device sends the scheduling request to the blockchain node.

[0252] In step 604, the blockchain node triggers a smart contract based on the request sent by the task scheduling device, and generates a scheduling decision based on the task status, the status of the candidate edge nodes, and the reputation value of the candidate edge nodes.

[0253] This enables multiple edge nodes to participate in task scheduling transparently and fairly in environments lacking a foundation of trust, avoiding single points of failure and data tampering.

[0254] In step 605, the blockchain node sends the scheduling decision to the task scheduling device.

[0255] For example, scheduling decisions are confirmed through a consensus mechanism. Verified scheduling decisions are published on the chain in the form of transactions and automatically sent to the task scheduling device. Their metadata is recorded in the distributed ledger.

[0256] Therefore, a decision verification system based on smart contracts is constructed. The scheduling scheme that has been confirmed and verified through the consensus mechanism will be packaged into a blockchain transaction to ensure that the scheduling decision is open, transparent and verifiable.

[0257] In step 606, the task is assigned to the edge node specified by the scheduling decision based on the scheduling decision.

[0258] For example, after receiving the scheduling decision confirmed on-chain, the task scheduling device schedules the tasks according to the node allocation scheme output by the smart contract. Simultaneously, the task scheduling device establishes task execution monitoring, periodically reporting the execution status and resource usage to the on-chain.

[0259] Therefore, the task scheduling process, task execution status, and edge node reputation updates are all recorded through blockchain nodes, thus achieving immutability and traceability, facilitating post-event auditing and dispute resolution.

[0260] In step 607, the edge node processes the tasks assigned by the task scheduling device.

[0261] In step 608, the edge node sends the task execution result to the access node.

[0262] In step 609, the access node determines the reputation value of the edge node based on the task execution status.

[0263] For example, the reputation value of an edge node can be obtained based on its expected reliability, expected behavior, and trust volatility.

[0264] Therefore, by comprehensively considering the expected reliability, expected behavior, and trust volatility of edge nodes, the reputation value of edge nodes can be obtained. This enables a joint evaluation of the multi-dimensional trust attributes of edge nodes, providing a quantitative basis for node screening for task scheduling smart contracts on the blockchain, effectively reducing the impact of malicious nodes, dynamically identifying and punishing bad and malicious behaviors, and improving the robustness of the system.

[0265] In some embodiments, the process of determining the reputation value of an edge node is as described in step 104.

[0266] In step 610, the access node sends the reputation value to the blockchain node.

[0267] For example, the access node calculates a reputation value based on the task execution status and the historical performance of the edge node, taking into account both historical and short-term trust, and uploads it to the blockchain node for on-chain storage, maintenance, and dissemination throughout the network.

[0268] By implementing the above embodiments of this disclosure, the following beneficial effects can be obtained:

[0269] (1) A blockchain-based trusted task scheduling system is proposed. In related technologies, blockchain-based edge computing task scheduling systems often only focus on the trusted transmission of data, while this disclosure uses a blockchain network to transmit data, providing reliable data support for edge computing task scheduling.

[0270] Furthermore, by building a task scheduling decision-making system based on smart contracts on the blockchain platform, it is ensured that the scheduling decisions in the task scheduling process are not only open and transparent, but also have strong anti-tampering capabilities, realizing the transparency and verifiability of task scheduling and improving the credibility and security of the entire task scheduling process.

[0271] (2) A trust management mechanism based on blockchain is designed. In related technologies, trust management schemes have the risk of data leakage and single point of failure. This disclosure uses the security characteristics of blockchain to store, update and disseminate trust information, which not only solves the above problems and realizes a global view of trust information, but also saves the limited resources of IoT devices by eliminating the need for heavy trust calculations.

[0272] Furthermore, a multi-dimensional dynamic evaluation mechanism is introduced, comprehensively considering three dimensions: node reliability, behavioral performance, and trust volatility, and combining a time decay factor to achieve a dynamic balance between historical trust and short-term trust. Unlike existing Bayesian trust models that are based solely on static reliability or single behavioral features, this disclosure, through joint modeling and dynamic updating of multi-dimensional trust features within a Bayesian inference framework, can more accurately characterize the reputation change trends of edge nodes at different time scales, thereby improving the timeliness, stability, and robustness of trust evaluation.

[0273] The above embodiments of this disclosure can be applied in the following scenarios:

[0274] (1) For edge computing service markets with multiple third parties participating, this disclosure can provide a unified blockchain platform to store reputation, task status and node status information. At the same time, the scheduling process is transparent on the chain, ensuring that scheduling decisions and data are tamper-proof, realizing the transparency and auditability of task scheduling among multiple parties, and reducing trust costs.

[0275] (2) In fine-grained computing-power collaboration at the edge, edge nodes provide both computing resources and consume power resources. Using the task scheduling method described in this disclosure, the matching between task scheduling and power supply is transparent, and dynamic optimization of task scheduling can be achieved in an environment where multiple parties do not fully trust each other, thereby improving resource utilization, reducing energy consumption, and enhancing system security.

[0276] In some embodiments, the functional units described above may be implemented as general-purpose processors, programmable logic controllers (PLCs), digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or any suitable combination thereof for performing the functions described herein.

[0277] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0278] The description in this disclosure is provided for illustrative and descriptive purposes only and is not intended to be exhaustive or to limit the disclosure to its forms. Many modifications and variations will be apparent to those skilled in the art. The embodiments were chosen and described in order to better illustrate the principles and practical application of this disclosure and to enable those skilled in the art to understand this disclosure and to design various embodiments with various modifications suitable for a particular purpose.

Claims

1. A task scheduling method, executed by an access node, comprising: The received task request is parsed to obtain the scheduling request used to schedule the task; The scheduling request is sent to the task scheduling device so that the task scheduling device can allocate the task to the edge node for processing the task according to the scheduling decision. Receive the task execution status sent by the edge node; Based on the task execution status, determine the reputation value of the edge node; The reputation value is sent to the blockchain node.

2. The task scheduling method according to claim 1, wherein, The determination of the reputation value of the edge node includes: Determine the expected reliability of the edge node, the expected behavior of the edge node, and the trust volatility of the edge node; The reputation value of the edge node is obtained based on the expected reliability value, the expected behavior value, and the trust volatility of the edge node.

3. The task scheduling method according to claim 2, wherein, The determination of the expected reliability value of the edge node includes: The expected reliability value of the edge node is obtained based on the number of successful and failed attempts at executing the task.

4. The task scheduling method according to claim 3, wherein, The step of obtaining the expected reliability value of the edge node based on the number of successful and failed executions of the task by the edge node includes: The task execution result of the current task cycle and the historical success count of the task in the previous task cycle are weighted and accumulated, and the weighted accumulation result is used as the first parameter of the Beta distribution. The task execution result of the current task cycle and the number of historical failures of the task in the previous task cycle are weighted and accumulated, and the weighted accumulation result is used as the second parameter of the Beta distribution. Based on the first parameter and the second parameter, the expected reliability value of the edge node is obtained.

5. The task scheduling method according to claim 4, wherein, The step of obtaining the expected reliability value of the edge node based on the first parameter and the second parameter includes: Calculate the sum of the first parameter and the second parameter to obtain the first intermediate value; The ratio of the first parameter to the first intermediate value is calculated to obtain the expected reliability value of the edge node.

6. The task scheduling method according to claim 4, wherein, The step of weighting and accumulating the task execution result of the current task cycle and the historical success count of the task in the previous task cycle, and using the weighted accumulation result as the first parameter of the Beta distribution, includes: The second intermediate value is obtained by multiplying the historical success count of the task in the previous task cycle by the first discount factor. The first parameter is obtained by summing the second intermediate value and the task execution result.

7. The task scheduling method according to claim 6, wherein, The step of weighting and accumulating the task execution result of the current task cycle and the historical failure count of the task in the previous task cycle, and using the weighted accumulation result as the second parameter of the Beta distribution, includes: The third intermediate value is obtained by multiplying the historical failure count of the task in the previous task cycle by the first discount factor. The difference between the first predetermined parameter value and the task execution result is calculated to obtain the fourth intermediate value; The second parameter is obtained by summing the third intermediate value and the fourth intermediate value.

8. The task scheduling method according to claim 2, wherein, The determination of the expected behavior of the edge node includes: Based on the different levels of behavior generated by the edge node during the execution of the task, the expected value of the edge node's behavior performance is obtained.

9. The task scheduling method according to claim 8, wherein, The step of obtaining the expected value of the edge node's behavior based on different levels of behavior generated during the execution of the task by the edge node includes: The behavior quality score of the current task cycle and the historical positive behavior of the task in the previous task cycle are weighted and accumulated, and the weighted accumulation result is used as the third parameter of the Beta distribution. The behavior quality score of the current task cycle and the historical negative behavior of the task in the previous task cycle are weighted and accumulated, and the weighted accumulation result is used as the fourth parameter of the Beta distribution. Based on the third and fourth parameters, the expected behavior of the edge node is obtained.

10. The task scheduling method according to claim 9, wherein, The step of obtaining the expected behavioral performance value of the edge node based on the third parameter and the fourth parameter includes: The sum of the third parameter and the fourth parameter is calculated to obtain the fifth intermediate value; The ratio of the third parameter to the fifth intermediate value is calculated to obtain the expected value of the behavior of the edge node.

11. The task scheduling method according to claim 9, wherein, The step of weighting and accumulating the behavior quality score of the current task cycle and the historical positive behavior of the task in the previous task cycle, and using the weighted accumulation result as the third parameter of the Beta distribution, includes: The sixth intermediate value is obtained by calculating the product of the historical positive behavior of the task in the previous task cycle and the second discount factor. The product of the behavior quality score and the scaling factor is calculated to obtain the seventh intermediate value; The third parameter is obtained by summing the sixth intermediate value and the seventh intermediate value.

12. The task scheduling method according to claim 11, wherein, The step of weighting and accumulating the behavior quality score of the current task cycle and the historical negative behavior of the task in the previous task cycle, and using the weighted accumulation result as the fourth parameter of the Beta distribution, includes: The product of the historical negative behavior of the task in the previous task cycle and the second discount factor is calculated to obtain the eighth intermediate value; The difference between the second predetermined parameter value and the behavior quality score is calculated to obtain the ninth intermediate value; Calculate the product of the ninth intermediate value and the scaling factor to obtain the tenth intermediate value; The fourth parameter is obtained by summing the eighth intermediate value and the tenth intermediate value.

13. The task scheduling method according to claim 2, wherein, Determining the trust volatility of the edge node includes: The trust volatility of the edge node is obtained based on the expected reliability value and the expected behavior value of the edge node.

14. The task scheduling method according to claim 13, wherein, The process of obtaining the trust volatility of the edge node based on its expected reliability and expected behavior includes: Calculate the weighted sum of the reliability expectation value and the behavioral performance expectation value to obtain the preliminary reputation value for the current task cycle; The trust volatility of the current task period is obtained based on the initial credit value of the current task period, the initial credit value of the previous task period, the trust volatility of the previous task period, and the smoothing factor.

15. The task scheduling method according to claim 14, wherein, The step of obtaining the trust volatility of the current task period based on the preliminary credit value of the current task period, the preliminary credit value of the previous task period, the trust volatility of the previous task period, and the smoothing factor includes: The difference between the initial reputation value of the current period and the initial reputation value of the previous task period is calculated to obtain the eleventh intermediate value; Calculate the square of the eleventh median to obtain the twelfth median; The trust volatility of the previous task period is calculated by weighting the trust volatility of the previous task period and the twelfth intermediate value to obtain the trust volatility of the current task period, wherein the weight of the trust volatility of the previous task period is the smoothing factor, and the weight of the twelfth intermediate value is the difference between the third predetermined parameter value and the smoothing factor.

16. The task scheduling method according to claim 15, wherein, The process of obtaining the reputation value of an edge node based on its expected reliability, expected behavior, and trust volatility includes: Calculate the arithmetic square root of the trust volatility of the current task cycle to obtain the thirteenth intermediate value; Calculate the product of the penalty coefficient and the thirteenth intermediate value to obtain the fourteenth intermediate value; The reputation value of the edge node is obtained by calculating the difference between the initial reputation value of the current task cycle and the fourteenth intermediate value.

17. The task scheduling method according to claim 1, further comprising: After receiving the task execution result, the execution result is verified; After successful verification, the execution result will be sent to the task requester.

18. An access node, comprising: The memory is configured to store instructions; A processor, coupled to a memory, configured to implement the method as described in any one of claims 1-17 based on memory-stored instruction execution.

19. A task scheduling method, executed by a task scheduling device, comprising: Receive scheduling requests for scheduling tasks sent by access nodes; The scheduling request is sent to the blockchain node so that the blockchain node can trigger a smart contract to generate a scheduling decision based on the task status, the status of the candidate edge node, and the reputation value of the candidate edge node. Receive the scheduling decision sent by the blockchain node; The task is assigned to the edge node specified by the scheduling decision.

20. The task scheduling method according to claim 19, further comprising: After receiving the scheduling decision sent by the blockchain node, a task execution monitoring system is established to periodically report the execution status and resource usage to the blockchain node.

21. A task scheduling device, comprising: The memory is configured to store instructions; A processor, coupled to a memory, configured to implement the method as described in any one of claims 19-20 based on memory-stored instruction execution.

22. A task scheduling system, comprising: The access node as described in claim 18; The task scheduling device as described in claim 21; A blockchain node is configured to trigger a smart contract based on a request sent by the task scheduling device, generate a scheduling decision based on the task status, the status of the candidate edge nodes, and the reputation value of the candidate edge nodes, and send the scheduling decision to the task scheduling device. Edge nodes are configured to process tasks assigned by the task scheduling device.

23. A computer-readable storage medium, wherein, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the method as described in any one of claims 1-17 and 19-20.

24. A computer program product comprising computer instructions, wherein the computer instructions, when executed by a processor, implement the method as described in any one of claims 1-17, 19-20.