CN120583086B - A blockchain edge computing resource collaborative scheduling system for IoT devices - Google Patents

Abstract

This invention discloses a blockchain edge computing resource collaborative scheduling system for IoT devices, relating to the field of IoT technology. It includes a device edge networking module for deploying edge nodes in densely populated IoT device areas and connecting these edge nodes to a blockchain network; a resource status initialization module for sending task information to the edge node layer after an IoT device generates a task; an intelligent task allocation module for calculating the optimal task allocation scheme after receiving a task and allocating the task to edge nodes for processing according to the optimal scheme; and a dynamic resource adjustment module for real-time monitoring of the resource load of edge nodes. When resource load imbalance occurs, resources are reallocated according to a dynamic resource adaptive adjustment strategy to ensure stable system operation. This achieves efficient collaborative scheduling and secure, reliable management of edge computing resources among IoT devices.

Description

A blockchain edge computing resource collaborative scheduling system for IoT devices

Technical Field

This invention relates to the field of Internet of Things (IoT) technology, and more specifically, to a blockchain edge computing resource collaborative scheduling system for IoT devices.

Background Technology

With the rapid development of IoT technology, the amount of data generated by a large number of IoT devices is exploding. Traditional cloud computing models face problems such as high network latency, high bandwidth pressure, and data privacy and security when processing IoT data. Edge computing, by deploying computing resources at the network edge, can process data generated by IoT devices locally, effectively reducing latency and bandwidth pressure. However, current edge computing resource scheduling suffers from low resource utilization and a lack of secure and reliable mechanisms. Blockchain technology, with its decentralized, tamper-proof, secure, and reliable characteristics, offers a new approach to solving these problems when combined with edge computing. However, a mature blockchain-based edge computing resource collaborative scheduling system for IoT devices is currently lacking.

Shortcomings of existing technology:

In traditional IoT systems, edge computing resource management is decentralized, with edge nodes lacking effective collaboration and making it difficult to flexibly allocate resources according to actual needs, resulting in low resource utilization. Data transmission and storage processes pose security risks; data is easily tampered with and forged, and there is a lack of effective traceability mechanisms. The introduction and application of blockchain technology are limited. Previous task allocation methods often lack comprehensive consideration of task characteristics and resource status, potentially leading to high processing latency and low efficiency. Existing systems struggle to monitor resource load in real time, and when resource load imbalances occur, timely adjustments cannot be made, impacting overall system performance and stability.

To address the above problems, this invention proposes a solution.

Summary of the Invention

To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a blockchain edge computing resource collaborative scheduling system for IoT devices. This system includes a device edge networking module, a resource status initialization module, an intelligent task allocation module, and a dynamic resource adjustment module, with connections between the modules.

The device edge networking module is used to deploy edge nodes in areas with dense IoT devices and connect the edge nodes to the blockchain network. IoT devices connect to the edge nodes through a wireless network to transmit data and submit tasks.

The resource status initialization module is used by each edge node to initialize the local resource scheduling ledger, record the initial resource status, and send the task information to the edge node layer after the IoT device generates a task.

The intelligent task allocation module is used to calculate the optimal task allocation scheme based on the intelligent task allocation algorithm, combined with its own resource status and the resource information of other nodes, after the edge node receives a task, and then allocate the task to the edge node for processing according to the optimal task allocation scheme.

The dynamic resource adjustment module is used to monitor the resource load of edge nodes in real time. When resource load imbalance occurs, resources are redistributed according to the dynamic resource adaptive adjustment strategy to achieve stable system operation, thereby solving the problems mentioned in the background technology.

To achieve the above objectives, the present invention provides the following technical solution:

A blockchain edge computing resource collaborative scheduling system for IoT devices includes a device edge networking module, a resource status initialization module, an intelligent task allocation module, and a dynamic resource adjustment module, with connections between the modules.

The device edge networking module is used to deploy edge nodes in areas with dense IoT devices and connect the edge nodes to the blockchain network. IoT devices connect to the edge nodes through a wireless network to transmit data and submit tasks.

The resource status initialization module is used by each edge node to initialize the local resource scheduling ledger, record the initial resource status, and send the task information to the edge node layer after the IoT device generates a task.

The intelligent task allocation module is used to calculate the optimal task allocation scheme based on the intelligent task allocation algorithm, combined with its own resource status and the resource information of other nodes, after the edge node receives a task, and then allocate the task to the edge node for processing according to the optimal task allocation scheme.

The dynamic resource adjustment module is used to monitor the resource load of edge nodes in real time. When resource load imbalance occurs, resources are redistributed according to the dynamic resource adaptive adjustment strategy to achieve stable system operation.

In a preferred embodiment, the process of obtaining the dense area of IoT devices is as follows:

Geographic Information System (GIS) technology is used to define the boundaries of densely populated areas of IoT devices, and the densely populated areas of IoT devices are abstracted into two-dimensional planar graphics to obtain their area S.

Divide the region into n grid cells, and count the number of IoT devices N _{_i} in each grid cell. Then, the formula for calculating the average device distribution density ρ of the region is:

In the formula, ρ is the average device distribution density in the region, _Ni is the number of IoT devices in each grid cell, n is the number of grid cells, and S is the area of the dense IoT device region.

The average device distribution density in densely populated IoT device areas is compared with a preset threshold. If the average device distribution density in densely populated IoT device areas is greater than or equal to the preset distribution density threshold, then edge nodes need to be deployed in densely populated IoT device areas.

If the average device density in a densely populated area of IoT devices is less than a preset density threshold, then it is not necessary to deploy edge nodes in such areas.

In a preferred embodiment, the process of deploying edge nodes is as follows:

Randomly assign m edge nodes to the grid cell boundaries and calculate the comprehensive allocation evaluation coefficient for each assignment. By comparing the comprehensive allocation evaluation coefficients of each assignment, the assignment with the largest comprehensive allocation evaluation coefficient is obtained, and the deployment location of the edge nodes is selected based on this assignment.

In a preferred embodiment, the process for obtaining the comprehensive allocation evaluation coefficients is as follows:

An evaluation index system is constructed based on the coverage function and data transmission delay to obtain the comprehensive allocation evaluation coefficient. The specific calculation formula is as follows:

In the formula, Score _q is the comprehensive allocation evaluation coefficient, C is the coverage function, T _lp is the data transmission delay, α ₁ is the coverage function weight coefficient, and α ₂ is the data transmission delay weight coefficient.

In a preferred embodiment, after receiving a task, the edge node calculates the optimal task allocation scheme based on its own resource status and the resource information of other nodes using an intelligent task allocation algorithm. The process is as follows:

After receiving the task data packet sent by the IoT device, the edge node parses the task information according to the encapsulation protocol and extracts key information.

Query the local resource scheduling ledger to check the current computing resource status of the edge node and obtain its computing capabilities.

If the resources available are insufficient to meet the task requirements, then the process enters the resource information exchange phase.

If its own resources meet the task requirements, it will participate in the subsequent optimal solution calculation as a task execution node;

Edge nodes interact with other edge nodes through the blockchain network. Each edge node broadcasts its current resource status information to other nodes and receives resource status information from other nodes.

Construct a task allocation model that satisfies the constraints that the task execution time is less than or equal to the set task completion deadline and that the resource usage of each edge node cannot exceed its available resources. Suppose there are k tasks in total, and x_{_ij} is a decision variable. When task j is allocated to edge node i, x_{_ij} = 1, otherwise x_{_ij} = 0.

For each task j assigned to edge node i, calculate the task execution time and solve for the optimal task allocation scheme based on the task execution time.

In a preferred embodiment, the process of obtaining the task execution time is as follows:

Calculate the data transmission time based on the data volume of the task and the network bandwidth between the edge node and the IoT device;

The task computation time is determined based on the estimated CPU computing resources required for the task and the computing capabilities of the edge nodes.

The task execution time is obtained by adding the data transmission time to the task computation time. The formula for calculating the task execution time is as follows:

In the formula, t is the task execution time, m is the number of edge nodes, k is the number of tasks, x _{_ij} is the decision variable, C_{_j} is the CPU computing resources expected to be required for the task, c _{_i} is the computing power of the edge node, b_{_i} is the network bandwidth between the edge node and the IoT device, and D _{_j} is the data volume of the task.

In a preferred embodiment, the constraint condition expression is as follows:

In the formula, t _{_ij} is the task execution time, x_{_ij} is the decision variable, m is the number of edge nodes, k is the number of tasks, T_{_deadlinej} is the set time limit for task completion, C_{_j} is the CPU computing resources expected to be required for the task, c_{_i} is the computing power of the edge node, b _{_i} is the network bandwidth between the edge node and the IoT device, and D_{_j} is the data volume of the task.

In a preferred embodiment, the coverage function and data transmission latency are obtained as follows:

Let r be the effective coverage radius of the edge node. Draw a circle with the candidate location as the center and r as the radius to construct the coverage area. To ensure that all IoT devices are effectively covered, a coverage function C is introduced:

In the formula, C is the coverage function, S is the area of the dense area of IoT devices, A _{covered, i} is the area covered by the edge node of the i-th grid cell, and n is the number of grid cells;

The total number of IoT devices is calculated based on the number of IoT devices in each grid cell, and the data transmission latency is calculated in conjunction with the signal transmission speed. The specific calculation formula is as follows:

In the formula, T _{_lp} is the data transmission delay, d _{_lpj} is the straight-line distance between each IoT device and the nearest edge node, the signal transmission speed is v, and N is the total number of IoT devices.

In a preferred embodiment, the process of real-time monitoring of the resource load of edge nodes is as follows:

Real-time monitoring of edge node resource load, obtaining the total CPU computing resources and CPU computing resource usage of edge nodes, and calculating the percentage of available CPU resources. The calculation formula is as follows:

In the formula, P _storage is the percentage of available CPU resources, S _total is the total CPU computing resources of the edge node, and S _used is the CPU computing resource usage.

In a preferred embodiment, when resource load imbalance occurs, the resource reallocation process according to the dynamic resource adaptive adjustment strategy is as follows:

By comparing the percentage of available CPU resources of each edge node, if the percentage of available CPU resources of a node exceeds 1.5 times the average system load, and at least one node has a load below 50% of the average load, then the computing resource load is determined to be unbalanced.

Resource reallocation is achieved using a priority-based preemptive scheduling algorithm. When a high-priority task requests resources, if the current node does not have sufficient resources, the low-priority task is temporarily interrupted, and CPU core computing resources are reallocated to ensure that the high-priority task is executed first.

The technical effects and advantages of the blockchain edge computing resource collaborative scheduling system for IoT devices of the present invention are as follows:

1. This invention uses an intelligent task allocation module to accurately assess task resource requirements and optimize allocation based on edge node resource status. Simultaneously, a dynamic resource adjustment module optimizes unbalanced resources, enabling the system to fully utilize edge computing resources, reducing waste and significantly improving resource utilization. Edge nodes are located close to IoT devices, performing data processing and task execution locally, reducing data transmission to the cloud and greatly lowering latency. The device-edge networking module ensures efficient data transmission, meeting the real-time requirements of IoT applications, such as real-time decision-making in industrial control and intelligent transportation scenarios. The application of blockchain technology ensures data is recorded in a distributed ledger, making it immutable and traceable. The data recording and consensus module ensures data consistency across nodes, effectively preventing malicious tampering and forgery, guaranteeing the security and trustworthiness of IoT data, and enhancing user trust in the system.

2. This invention uses a dynamic resource adjustment module to monitor resource load in real time, promptly identifying and resolving resource imbalances to prevent system crashes caused by excessive load on some nodes. Simultaneously, intelligent task allocation and data security mechanisms ensure smooth task execution and data integrity, thereby improving system stability and reliability and reducing system failures and service interruptions. The design of each module in the system offers good flexibility, adapting to densely populated areas with varying sizes and types of IoT devices. Whether adding new IoT devices or expanding edge nodes, the device-edge networking module and intelligent task allocation module can quickly adapt to changes, enabling resource reconfiguration and rational task scheduling, facilitating system expansion and upgrades.

Attached Figure Description

Figure 1 is a schematic diagram of the structure of a blockchain edge computing resource collaborative scheduling system for IoT devices according to the present invention.

Detailed Implementation

The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

Example 1, Figure 1 shows a blockchain edge computing resource collaborative scheduling system for IoT devices according to the present invention.

The device edge networking module is used to deploy edge nodes in areas with dense IoT devices and connect the edge nodes to the blockchain network. IoT devices connect to the edge nodes through a wireless network to transmit data and submit tasks.

It is necessary to assess areas with high density of IoT devices to determine the geographical extent of the area, the distribution density of IoT devices, the edge node coverage function, and the data transmission latency.

Based on the regional assessment results, the process of selecting suitable locations to deploy edge nodes is as follows:

Using Geographic Information System (GIS) technology, the boundaries of densely populated areas of IoT devices are defined, and the densely populated areas of IoT devices are abstracted into two-dimensional planar graphics (such as polygons) to obtain their area S.

Divide the region into n grid cells, and count the number of IoT devices N _{_i} in each grid cell. Then, the formula for calculating the average device distribution density ρ of the region is:

In the formula, ρ is the average device distribution density of the region, _Ni is the number of IoT devices in each grid cell, n is the number of grid cells, and S is the area of the dense IoT device region.

The average device distribution density in densely populated IoT device areas is compared with a preset threshold. If the average device distribution density in densely populated IoT device areas is greater than or equal to the preset distribution density threshold, then edge nodes need to be deployed in densely populated IoT device areas.

If the average device density in a densely populated area of IoT devices is less than a preset density threshold, then it is not necessary to deploy edge nodes in such areas.

Randomly assign m edge nodes to the grid cell boundaries and calculate the comprehensive allocation evaluation coefficient for each assignment. Select the deployment location of the edge nodes based on the comprehensive allocation evaluation coefficient for each assignment. The process of obtaining the comprehensive allocation evaluation coefficient is as follows:

Let r be the effective coverage radius of the edge node. Draw a circle with the candidate location as the center and r as the radius to construct the coverage area. To ensure that all IoT devices are effectively covered, a coverage function C is introduced:

In the formula, C is the coverage function, S is the area of the dense area of IoT devices, A _{covered, i} is the area covered by the edge node of the i-th grid cell, and n is the number of grid cells.

The system obtains the straight-line distance from each IoT device to each edge node, selects the nearest edge node for each IoT device by comparing the straight-line distances, and calculates the straight-line distance between each IoT device and its nearest edge node. It also calculates the total number of IoT devices based on the number of IoT devices in each grid cell, and calculates the data transmission delay by combining this with the signal transmission speed. The specific calculation formula is as follows:

In the formula, T _{_lp} is the data transmission delay, d _{_lpj} is the straight-line distance between each IoT device and the nearest edge node, the signal transmission speed is v, and N is the total number of IoT devices.

It's important to note that while tasks for various IoT devices may require processing by multiple edge nodes, this includes the nearest edge node; here, we only consider the nearest edge node. Focusing solely on the nearest edge node significantly simplifies the calculation of data transmission latency. It eliminates the need to consider the complex relationships between multiple edge nodes and IoT devices, as well as the impact of different transmission paths, reducing computational load and system complexity while improving computational efficiency and system operability. It also minimizes the distance data travels within the network, thereby reducing signal attenuation and latency during transmission. Since signal transmission speed is finite, shorter transmission distances mean shorter data arrival times at edge nodes, better meeting the real-time requirements of IoT applications.

An evaluation index system is constructed based on the coverage function and data transmission delay to obtain the comprehensive allocation evaluation coefficient. The specific calculation formula is as follows:

In the formula, Score _q is the comprehensive allocation evaluation coefficient, C is the coverage function, T _lp is the data transmission delay, α ₁ is the coverage function weight coefficient, and α ₂ is the data transmission delay weight coefficient.

It should be noted that _α1 and _α2 are both greater than 0, and are obtained by analyzing the coverage function and the importance of data transmission delay based on historical data.

By comparing the comprehensive allocation evaluation coefficients of each allocation scenario, the allocation scenario with the highest comprehensive allocation evaluation coefficient is obtained, and the deployment location of the edge node is selected based on this allocation scenario.

The resource status initialization module is used by each edge node to initialize the local resource scheduling ledger, record the initial resource status, and send the task information to the edge node layer after the IoT device generates a task.

Each edge node initializes its local resource scheduling ledger and records the initial resource status as follows:

When an edge node starts up, it obtains the initial configuration information of computing resources, storage resources, and network resources by calling the underlying hardware detection interface of the system.

Read the configuration parameters of the edge node's operating system and related services to determine the amount of resources allocated to the system's own operation;

Based on the obtained hardware configuration information and system resource usage, the initial available resource status is calculated.

The calculated initial resource status information is written into the local resource scheduling ledger using a data structure that combines key-value pairs and lists.

At the same time, a unique identifier (UUID) is assigned to each resource record to facilitate subsequent querying, updating and management of resource status.

After an IoT device generates a task, the process of sending the task information to the edge node layer is as follows:

During operation, IoT devices generate tasks based on their own functions and user instructions. The task information includes several key fields:

Task Identifier: Generate a unique identifier for each task to distinguish different tasks in the system;

Task type: Define the nature of the task, such as data acquisition task, data processing task, equipment control task, etc.;

Data requirements: Specify the data source, data format, and data volume required for the task; for example, a data acquisition task needs to specify the type of sensor and sampling frequency; a data processing task needs to provide the storage location and format requirements of the data to be processed.

Computational requirements: Quantify the computational resource requirements of the task, which include the CPU computational resources that the task is expected to require;

The estimated CPU computing resources required for a task are calculated using a task complexity assessment model. The process is as follows: Assuming the task complexity is T and the data volume is D, the formula for calculating the estimated CPU computing resources required for the task is as follows:

C_{_j} = α_T + β_D;

Where α and β are weighting coefficients adjusted according to the actual situation, and _Cj is the CPU computing resources expected to be required for the task;

It's important to note that data processing is a crucial step in the various tasks generated by IoT devices, and the CPU is the core component for data processing. In the initial stages of task allocation or in certain specific scenarios, simplifying the model is necessary to make the problem easier to handle and analyze. Focusing only on CPU computing resource requirements, without considering other resource requirements, can transform complex multi-resource-constrained problems into single-resource-constrained problems, significantly reducing the complexity of task allocation algorithms and facilitating the rapid finding of feasible allocation solutions. For example, in some simple IoT monitoring systems, the main task is to perform simple calculations and analyses on sensor data. In this case, CPU computing resources may be the primary limiting factor; ignoring other resource requirements can simplify the calculation process and improve the efficiency of task allocation.

Real-time requirement: Set the time limit for task completion, denoted by T _deadline,j ;

The generated task information is encapsulated according to a specific communication protocol;

During the encapsulation process, communication control information such as the source device address (the network address of the IoT device itself) and the target edge node address (determined according to the pairing relationship between the device and the edge node or the load balancing strategy) are added to form a complete task data packet.

IoT devices send encapsulated task data packets to the edge node layer via wireless network. During the transmission process, a retransmission mechanism is used to ensure reliable transmission of task information. If no acknowledgment response is received from the edge node within a specified time, the task data packet is retransmitted. The number of retransmissions can be dynamically adjusted according to network conditions. At the same time, the transmission time of the task information is recorded for subsequent transmission delay calculation and task timeliness judgment.

The intelligent task allocation module is used to calculate the optimal task allocation scheme based on the intelligent task allocation algorithm, combined with its own resource status and the resource information of other nodes, after the edge node receives a task, and then allocate the task to the edge node for processing according to the optimal task allocation scheme.

After receiving the task data packet sent by the IoT device, the edge node first parses the task information according to the encapsulation protocol and extracts key information;

Then, query the local resource scheduling ledger to check its current computing resource status and obtain the computing capabilities of the edge node;

If its own resources cannot meet the task requirements, it will enter the resource information exchange stage; if its own resources meet the task requirements, it will participate in the subsequent optimal solution calculation as a potential task execution node.

Edge nodes interact with other edge nodes through the blockchain network. Each edge node broadcasts its current resource status information (obtained from the local resource scheduling ledger) to other nodes, while also receiving resource status information from other nodes.

It should be noted that, in order to ensure the timeliness and accuracy of information, an information update cycle can be set to periodically exchange and update resource information;

To minimize task completion time, meet task real-time requirements, and satisfy CPU computing resource requirements, a task allocation model is constructed. There are k tasks in total, and x _{_ij} is a decision variable. When task j is allocated to edge node i, x _{_ij} = 1; otherwise, x_{_ij} = 0.

For each task j assigned to edge node i, the task execution time is calculated, and the optimal task allocation scheme is determined based on the task execution time. The process of calculating the task execution time is as follows:

Calculate the data transmission time based on the data volume of the task and the network bandwidth between the edge node and the IoT device;

The task computation time is determined based on the estimated CPU computing resources required for the task and the computing capabilities of the edge nodes.

The task execution time is obtained by adding the data transmission time to the task computation time. The formula for calculating the task execution time is as follows:

In the formula, t is the task execution time, m is the number of edge nodes, k is the number of tasks, x _{_ij} is the decision variable, C_{_j} is the CPU computing resources expected to be required for the task, c_{_i} is the computing power of the edge node, b_{_i} is the network bandwidth between the edge node and the IoT device, and D _{_j} is the data volume of the task.

Furthermore, the following constraints must be met: the task execution time must be less than or equal to the set task completion time limit, and the resource usage of each edge node cannot exceed its available resources.

The constraint expressions are as follows:

In the formula, t _{_ij} is the task execution time, x_{_ij} is the decision variable, m is the number of edge nodes, k is the number of tasks, T_{_deadlinej} is the set time limit for task completion, C_{_j} is the CPU computing resources expected to be required for the task, c_{_i} is the computing power of the edge node, b _{_i} is the network bandwidth between the edge node and the IoT device, and D_{_j} is the data volume of the task.

Under the conditions of satisfying time and resource constraints, the optimal task allocation scheme is solved by minimizing the task execution time;

Based on the calculated optimal task allocation scheme, the edge nodes will assign tasks to the corresponding edge nodes for processing.

It should be noted that the edge node that assigns the task sends a task assignment instruction to the edge node to which the task is assigned, which includes detailed information about the task. After receiving the instruction, the edge node to which the task is assigned obtains the necessary data from the IoT device (if needed) and begins to execute the task. At the same time, each edge node updates its local resource scheduling ledger in real time to record the resource usage during the task execution process.

The dynamic resource adjustment module is used to monitor the resource load of edge nodes in real time. When resource load imbalance occurs, resources are redistributed according to the dynamic resource adaptive adjustment strategy to ensure stable system operation.

Real-time monitoring of edge node resource load, obtaining the total CPU computing resources and CPU computing resource usage of edge nodes, and calculating the percentage of available CPU resources. The calculation formula is as follows:

In the formula, P _storage is the percentage of available CPU resources, S _total is the total CPU computing resources of the edge node, and S _used is the CPU computing resource usage.

By comparing the percentage of available CPU resources of each edge node, if the percentage of available CPU resources of a node exceeds 1.5 times the average system load, and at least one node has a load below 50% of the average load, then the computing resource load is determined to be unbalanced.

Resource reallocation is achieved using a priority-based preemptive scheduling algorithm. When a high-priority task requests resources, if the current node does not have sufficient resources, the low-priority task is temporarily interrupted, and CPU core computing resources are reallocated to ensure that the high-priority task is executed first.

The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims (6)

1. The block chain edge computing resource collaborative scheduling system of the Internet of things equipment is characterized by comprising an equipment edge networking module, a resource state initializing module, an intelligent task allocation module and a dynamic resource adjusting module, wherein the modules are connected with one another:

The equipment edge networking module is used for deploying edge nodes in an equipment dense area of the Internet of things, and the edge node deployment process comprises the following steps:

Randomly distributing m edge nodes to the junction of the grid units, calculating the comprehensive distribution evaluation coefficient of each distribution condition, comparing the comprehensive distribution evaluation coefficients of each distribution condition to obtain the distribution condition with the maximum comprehensive distribution evaluation coefficient, and selecting the position of the distributed edge nodes according to the distribution condition;

The comprehensive allocation evaluation coefficient acquisition process comprises the steps of constructing an evaluation index system according to a coverage function and data transmission delay to obtain a comprehensive allocation evaluation coefficient, wherein the specific calculation formula is as follows: In the formula (I), in the formula (II), Is the comprehensive allocation evaluation coefficient, C is the coverage function,Is the delay of the data transmission and,Is the weight coefficient of the coverage function,Is a data transmission delay weight coefficient;

the coverage function and the data transmission delay acquisition process are as follows, the effective coverage radius of the edge node is set as r, a circle is drawn by taking the candidate position as a circle center, r as a radius, a coverage area is constructed, and a coverage function C is introduced to ensure that all the Internet of things equipment can be effectively covered: Wherein, C is a coverage function, S is the area of the intensive area of the equipment of the Internet of things, Calculating the total number of the Internet of things devices according to the number of the Internet of things devices in each grid unit, and calculating the data transmission delay by combining the signal transmission speed, wherein the specific calculation formula is as follows: In the formula (I), in the formula (II), Is the delay of the data transmission and,The linear distance between each piece of Internet of things equipment and the nearest edge node is the linear distance, the signal transmission speed is v, and N is the total number of pieces of Internet of things equipment;

the edge node is accessed to a blockchain network, and the Internet of things equipment is connected with the edge node through a wireless network to carry out data transmission and task submission;

the resource state initialization module is used for initializing a local resource scheduling account book by each edge node, recording the initial resource state, and sending task information to the edge node layer after the Internet of things equipment generates tasks;

the intelligent task allocation module is used for calculating an optimal task allocation scheme according to the self resource state and the resource information of other nodes according to an intelligent task allocation algorithm after the edge node receives the task, and allocating the task to the edge node for processing according to the optimal task allocation scheme;

and the dynamic resource adjusting module is used for monitoring the resource load condition of the edge node in real time, and when the resource load imbalance occurs, the dynamic resource self-adaptive adjusting strategy is used for carrying out resource reallocation so as to realize the stable operation of the system.

2. The blockchain edge computing resource collaborative scheduling system of the internet of things device according to claim 1, wherein the process of acquiring the intensive area of the internet of things device is as follows:

boundary definition is carried out on the equipment-intensive area of the Internet of things through a geographic information system technology, the equipment-intensive area of the Internet of things is abstracted into a two-dimensional plane graph, and the area S of the equipment-intensive area of the Internet of things is obtained;

dividing the area into n grid cells, and counting the number of the Internet of things devices in each grid cell Area average device distribution densityThe calculation formula is as follows:

;

in the formula, For the area-averaged device distribution density,The number of the Internet of things devices in each grid unit is n, the number of the grid units is n, and S is the dense area of the Internet of things devices;

Comparing and analyzing the average equipment distribution density of the equipment-intensive area of the Internet of things with a preset threshold value, and if the average equipment distribution density of the equipment-intensive area of the Internet of things is greater than or equal to the preset distribution density threshold value, deploying edge nodes in the equipment-intensive area of the Internet of things;

If the average equipment distribution density of the equipment-intensive area of the Internet of things is smaller than a preset distribution density threshold, edge nodes do not need to be deployed in the equipment-intensive area of the Internet of things.

3. The cooperative scheduling system for computing resources by a blockchain edge of an internet of things device according to claim 2, wherein after an edge node receives a task, the process of computing an optimal task allocation scheme by combining a self resource state and resource information of other nodes according to an intelligent task allocation algorithm is as follows:

After the edge node receives the task data packet sent by the Internet of things equipment, analyzing the task information according to an encapsulation protocol, and extracting key information;

Inquiring a local resource scheduling account book, and checking the current computing resource state of the local resource scheduling account book to obtain the computing capacity of the edge node;

if the self resources can not meet the task demands, entering a resource information interaction stage;

if the self resources meet the task demands, the self resources are used as a task execution node to participate in subsequent calculation of the optimal scheme;

The edge nodes interact with other edge nodes through a block chain network, and each edge node broadcasts current resource state information to other nodes and receives the resource state information of other nodes;

Constructing a task allocation model which meets constraint conditions that the task execution time is less than or equal to the set task completion time limit, the resource usage of each edge node cannot exceed the available resource, setting k tasks in total, For decision variables, when task j is assigned to edge node i,OtherwiseWherein the constraint expression is as follows:;

in the formula, It is the time of execution of the task,Is a decision variable, m is the number of edge nodes, k is the number of tasks,It is the time period for completion of the task that is set,Is the CPU computing resource required for task prediction,Is the computational power of the edge node and,Is the network bandwidth between the edge node and the internet of things device,Is the amount of data that is to be processed,Is the upper limit of the total amount of CPU computing resources of the ith edge node;

under the condition that the time limit constraint and the resource constraint are met, calculating task execution time for the situation that each task j is distributed to the edge node i, and solving an optimal task distribution scheme according to the task execution time.

4. The blockchain edge computing resource collaborative scheduling system of the internet of things device according to claim 3, wherein the process of obtaining the computing task execution time is as follows:

Calculating data transmission time according to the data volume of the task and the network bandwidth between the edge node and the Internet of things equipment;

acquiring task computing time according to CPU computing resources and edge node computing capacity required by task prediction;

adding the data transmission time and the task calculation time to obtain task execution time, wherein the task execution time calculation formula is as follows:

;

Where t is the task execution time, m is the number of edge nodes, k is the number of tasks, Is a decision variable that is used to determine the quality of the product,Is the CPU computing resource required for task prediction,Is the computational power of the edge node and,Is the network bandwidth between the edge node and the internet of things device,Is the amount of data for the task.

5. The cooperative scheduling system for blockchain edge computing resources of an internet of things device according to claim 4, wherein the process of monitoring the resource load condition of the edge node in real time is as follows:

the method comprises the steps of monitoring the resource load condition of an edge node in real time, obtaining the total amount of CPU computing resources of the edge node and the consumption of the CPU computing resources, and computing the percentage of available CPU resources, wherein the computing formula is as follows:

;

in the formula, Is the percentage of available CPU resources,Is the total amount of CPU computing resources of the edge node,Is the CPU computing resource usage.

6. The cooperative scheduling system for blockchain edge computing resources of an internet of things device according to claim 5, wherein when resource load imbalance occurs, a resource reallocation process is performed according to a dynamic resource adaptive adjustment policy as follows:

Comparing the available CPU resource percentages of the edge nodes, and judging that the load of the computing resources is unbalanced if the available CPU resource percentage of a certain node exceeds 1.5 times of the average load of the system and at least one node load is lower than 50% of the average load;

And when the high-priority task requests resources, if the current node resources are insufficient, temporarily interrupting the low-priority task, and reallocating CPU core computing resources to ensure the priority execution of the high-priority task.