A resource management method for edge computing system based on blockchain

Abstract

The present invention discloses a resource management method for an edge computing system based on blockchain, comprising a network controller sensing current terminal device task information and current system wireless environment information; the network controller sensing computing resource information of an edge server and a cloud server; uploading relevant information from a corresponding device to the network controller via a wireless connection; inputting current task information and current system wireless environment information into a trained optimization model deployed in the network controller, calculating terminal device transmission power control based on the current state, terminal device task offloading decision, and computing resource allocation to be used in the task calculation process and blockchain consensus process, and transmitting the calculations to each computing entity for execution; the present invention utilizes task offloading and computing resource allocation algorithms to jointly optimize task offloading and resource allocation with extremely low time complexity, thereby improving the execution efficiency of the system and user experience.

Description

Technical Field

[0001] The present invention relates to the field of blockchain and edge computing technology, and in particular to a blockchain-based edge computing system resource management method. Background Art

[0002] With the increasing maturity of the Internet of Things and mobile Internet technologies, computing-intensive applications such as face recognition, image processing, and autonomous driving have rapidly emerged. Traditional cloud computing not only occupies a large amount of bandwidth resources, but also can no longer meet the low latency requirements of various current applications. Therefore, Mobile Edge Computing (MEC) came into being.

[0003] Mobile edge computing can use wireless access networks to provide the required services and cloud computing functions to various mobile terminal users nearby. Users can offload part or all of the computing tasks to edge computing servers for processing according to their own needs, thereby providing users with ultra-low latency solutions. Compared with the traditional cloud computing-based IoT architecture, edge computing solves the problems of extended communication time and large transmission traffic generated by cloud computing services. However, as data access in a distributed environment, edge computing also makes the hardware and software specifications of edge devices complicated, and the computing and storage capabilities of terminal devices are limited, making it difficult to load high-computing security algorithms, and network security issues need to be urgently addressed.

[0004] Blockchain has a series of characteristics such as distributed data storage, point-to-point transmission, asymmetric encryption, consensus mechanism, smart contracts, etc. It is a new distributed infrastructure that facilitates data storage and traceability, can ensure the consistency of data in the entire network in a distributed environment, and provide protection for data security in the network. Blockchain technology can be combined with edge computing to establish a secure and reliable network environment to provide users with higher quality services.

[0005] The existing task offloading decision and resource allocation solutions ignore the computational latency of the joint optimization task and the consensus latency of the blockchain, and fail to organically combine blockchain with edge computing. At the same time, the existing solutions fail to dynamically allocate computing resources to the blockchain consensus computing process for edge servers. The above shortcomings greatly limit the performance of task offloading decision and resource allocation. Summary of the invention

[0006] 1. Technical issues to be resolved

[0007] In order to overcome the shortcomings of the above-mentioned prior art, the present invention is aimed at the tasks of the blockchain system based on mobile edge computing, comprehensively considers factors such as the amount of task data in the system structure and the network transmission rate to determine the task offloading decision and the resource allocation strategy, inherits the present invention to the network controller deployment module and the resource configuration module of the edge server layer, and combines the network status perception and collection module to realize the dynamic optimization of the task offloading decision and the resource allocation strategy, and accelerates the neural network training process and the optimization process on the basis of practicality, thereby improving the user experience.

[0008] (II) Technical solution

[0009] In order to solve the above technical problems, the present invention proposes a resource management method for edge computing system based on blockchain, comprising the following steps:

[0010] S1: The network controller senses the current terminal device task information and the current wireless environment information of the system; at the same time, the network controller senses the computing resource information of the edge server and the cloud server; and uploads the relevant information from the corresponding device to the network controller through the wireless connection;

[0011] S2: Input the current task information and the current wireless environment information of the system described in step S1 into the trained optimization model deployed in the network controller, and calculate the terminal device transmission power control based on the current state, the terminal device task offloading decision, and the task calculation process and the allocation of computing resources to be used in the blockchain consensus process;

[0012] S3: extracting the task offloading decision information of the terminal device obtained in step S2, if the information indicates that the task is executed locally, the network controller sends a corresponding task offloading decision instruction to the terminal device via a wireless connection, and the terminal device directly executes the task locally;

[0013] S4: Extract the task offloading decision information of the terminal device obtained in step S2. If the information indicates that the task is executed on the edge server or cloud server, in addition to the task calculation process, the task must also be authenticated through the consensus process of the blockchain system deployed on the edge server; the network controller sends the corresponding task offloading decision instruction and transmission power control instruction to the terminal device through a wireless connection, and the terminal device transmits the task to the edge server according to the corresponding transmission power, and joins the blockchain to execute the consensus process; if the task is executed on the edge server, the result is sent back to the terminal device after the task calculation process and the blockchain consensus process are all completed; if the task is executed on the cloud server, it is necessary to further transmit the task from the edge server to the cloud server through a wired link to process the task. If the task is executed on the edge server, the network controller sends a computing resource allocation instruction to the edge server through a wireless connection; if the task is executed on the cloud server, the network controller sends a computing resource allocation instruction to the cloud server through a wired link; the above instructions are used to control the calculation process of the parallel execution of tasks and the consensus process of the blockchain;

[0014] Furthermore, the step S1 is specifically as follows:

[0015] The characteristics of the network architecture of this method are as follows: represents the edge server (base station) set, One of the base stations exist In the example, the i-th edge server (base station) covers N i Terminal devices, represents the terminal equipment set under the i-th base station, One of the terminal devices In addition, a network controller is placed at the edge layer to control the operation of the entire method.

[0016] The algorithm is deployed on a network controller to work. The network controller perceives the current terminal device task information and the current wireless environment information of the system. The task characteristics are composed of the task calculation amount and data amount. Let c ij Indicates the computational amount of the task generated by each terminal device, Order ij Indicates the amount of data generated by each terminal device. The current wireless environment characteristics are composed of the channel gain from the terminal device to the base station. Let g ij represents the channel gain from the terminal device ij to its corresponding base station, At the same time, the network controller perceives the computing resource information of the edge server and the cloud server, and F care the upper bounds of computing resources of the i-th edge server and cloud server respectively.

[0017] During the working process, the network controller will upload the above information from the corresponding device to the network controller through wireless connection.

[0018] Furthermore, the step S2 is specifically as follows:

[0019] The current task data volume s and the current channel gain g between the terminal device and the base station obtained in step S1 are input into the trained optimization model deployed on the network controller to calculate the resource management instructions.

[0020] First, a fast numerical method is designed to obtain the transmit power control command of the terminal device under the current state, and the transmit power control command is separated from other optimization variables. Among them, p ij is the transmission power allocated to the terminal device ij. It can be found that the total delay of the optimization target of this method decreases monotonically with the increase of the transmission power. Therefore, the transmission power control instruction can be obtained by increasing the transmission power so that the energy consumption of the terminal device reaches the energy consumption constraint.

[0021] The feature of task scheduling in this method is that, regarding the task offloading decision of the terminal device, let α ij =1 means that task ij is offloaded to the edge server for execution, otherwise, α ij =0; similarly, β ij =1 means that task ij is offloaded to the cloud server for execution, otherwise, β ij =0; obviously, 1-α ij -β ij =1 means that task ij is executed locally, At the same time, regarding the allocation of computing resources used in the task computing process and the blockchain consensus process, let and are the computing resources allocated by the ith edge server to the blockchain consensus process and the computing process of task ij, respectively. make The computing resources allocated by the cloud server to the computing process of task ij, The above optimization variables are designed based on the deep reinforcement learning technology to design an optimization algorithm and establish a Markov decision process, where the state space of this method is set to b = {s, g} and the action space is set to a = {α, β, f block ,f e ,f c}. Finally, the optimization result is obtained according to the algorithm output.

[0022] Furthermore, the step S3 is specifically as follows:

[0023] Extract the offloading decision information obtained in step S2. If the information indicates that the task is executed locally, the network controller sends the task offloading decision instructions α and β to the terminal device via a wireless connection, executes the instructions in the manner described in this step, and directly transmits the result back to the user.

[0024] Furthermore, the step S4 is specifically as follows:

[0025] Extract the offloading decision information obtained in step S2. If the information indicates that the task is executed on the edge server or cloud server, the network controller sends the task offloading decision instruction and the transmission power control instructions α, β and p to the terminal device through the wireless connection, and sends the computing resource allocation instruction f to the edge server through the wireless connection. block and f e , or send computing resource allocation instructions to the cloud server through a wired link f c , and execute the instructions as described in this step. After the task calculation process and the blockchain consensus process are all completed, the results are returned to the user. After the corresponding instructions are executed, the total delay is obtained, that is, the sum of the larger of the task calculation delay and the blockchain consensus delay and the task transmission delay. The formula is:

[0026]

[0027] in, is the total processing delay of task ij, is the transmission delay of task ij (if task ij is executed on the terminal device, the value is 0), is the computational delay of task ij, is the blockchain consensus delay of the i-th base station.

[0028] (III) Beneficial effects

[0029] Compared with the prior art, the present invention adopts the above technical solution and has the following beneficial effects:

[0030] 1. When making task offloading decisions and resource allocation, the task computing delay and blockchain consensus delay are jointly optimized through analysis of actual scenarios, making the technology more practical;

[0031] 2. When making task offloading decisions and resource allocation, a Markov decision process is constructed. Based on deep reinforcement learning technology, complex optimization problems are solved through neural networks, which is more practical;

[0032] 3. By analyzing and optimizing the target structure, the original problem is decomposed into two sub-problems for solution, which greatly reduces the learning difficulty of deep reinforcement learning and greatly improves the calculation speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 is a schematic diagram of a system model of an embodiment;

[0034] Figure 2 It is a schematic diagram of the training process of task offloading decision and resource allocation;

[0035] Figure 3 It is a schematic diagram of the application process of task offloading decision and resource allocation;

[0036] Figure 4 It is a schematic diagram comparing the rewards of this scheme with other schemes as the wired transmission rate increases;

[0037] Figure 5 This is a schematic diagram comparing the rewards of this scheme with other schemes as the amount of task data increases;

[0038] Figure 6 This is a diagram comparing the rewards of this scheme with other schemes as the bandwidth increases. DETAILED DESCRIPTION

[0039] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0040] The present invention proposes a task scheduling method combining an edge computing system with a blockchain system, and the embodiment includes the following steps:

[0041] Step 1: The network controller senses the current terminal device task information and the current wireless environment information of the system; at the same time, the network controller senses the computing resource information of the edge server and the cloud server; and uploads the relevant information from the corresponding device to the network controller through a wireless connection;

[0042] Step 2: Input the current task information and the current wireless environment information of the system described in step 1 into the trained optimization model deployed in the network controller, and calculate the terminal device transmission power control based on the current state, the terminal device task offloading decision, and the task calculation process and the allocation of computing resources to be used in the blockchain consensus process;

[0043] Step 3: extract the task offloading decision information of the terminal device obtained in step 2. If the information indicates that the task is executed locally, the network controller sends a corresponding task offloading decision instruction to the terminal device via a wireless connection, and the terminal device directly executes the task locally.

[0044] Step 4: Extract the task offloading decision information of the terminal device obtained in step 2. If the information indicates that the task is executed on the edge server or cloud server, in addition to the task calculation process, the consensus process must be executed in parallel through the blockchain deployed on the edge server to increase security and credibility; the network controller sends the corresponding task offloading decision instructions and transmission power control instructions to the terminal device through a wireless connection. The terminal device transmits the task to the edge server according to the corresponding transmission power and joins the blockchain to execute the consensus process; if the task is executed on the edge server, the result is sent back to the terminal device after the task calculation process and the blockchain consensus process are all completed; if the task is executed on the cloud server, it is necessary to further transmit the task from the edge server to the cloud server through a wired link to process the task. If the task is executed on the edge server, the network controller sends a computing resource allocation instruction to the edge server through a wireless connection; if the task is executed on the cloud server, the network controller sends a computing resource allocation instruction to the cloud server through a wired link; the above instructions are used to control the parallel execution of the task calculation process and the blockchain consensus process;

[0045] Furthermore, the step 1 includes:

[0046] The characteristics of the network architecture of this method are as follows: represents the edge server (base station) set, One of the base stations exist In the example, the i-th edge server (base station) covers N i Terminal devices, represents the terminal equipment set under the i-th base station, One of the terminal devices In addition, a network controller is placed at the edge layer to control the operation of the entire method.

[0047] The algorithm is deployed on a network controller to work. The network controller perceives the current terminal device task information and the current wireless environment information of the system. The task characteristics are composed of the task calculation amount and data amount. Let c ij Indicates the computational amount of the task generated by each terminal device, Order ij Indicates the amount of data generated by each terminal device. The current wireless environment characteristics are composed of the channel gain from the terminal device to the base station. Let g ij represents the channel gain from the terminal device ij to its corresponding base station, At the same time, the network controller perceives the computing resource information of the edge server and the cloud server, and Fc are the upper bounds of computing resources of the i-th edge server and cloud server respectively.

[0048] During the working process, the network controller will upload the above information from the corresponding device to the network controller through wireless connection.

[0049] Furthermore, the step 2 includes:

[0050] The current task data volume s obtained in step 1 and the current channel gain g between the terminal device and the base station are input into the trained optimization model deployed on the network controller to calculate the resource management instructions.

[0051] First, a fast numerical method is designed to obtain the transmit power control command of the terminal device in the current state, and the transmit power control command is separated from other optimization variables. Among them, p ij is the transmission power allocated to the terminal device ij. The transmission delay from the terminal device to the base station is:

[0052]

[0053]

[0054]

[0055] in, is the wireless transmission rate from the terminal device to the base station, is the transmission energy consumption of the wireless transmission process from the terminal device to the base station, W is the bandwidth, is the noise power, and χ is the inter-channel interference. It can be found that since the transmission power decreases monotonically with the increase of the transmit power, the total delay of the optimization target of this method decreases monotonically with the increase of the transmit power. Therefore, the transmit power control instruction can be obtained by increasing the transmit power so that the energy consumption of the terminal device reaches the energy consumption constraint.

[0056] The feature of task scheduling in this method is that, regarding the task offloading decision of the terminal device, let α ij =1 means that task ij is offloaded to the edge server for execution, otherwise, α ij =0; similarly, β ij =1 means that task ij is offloaded to the cloud server for execution, otherwise, β ij =0; obviously, 1-α ij -β ij =1 means that task ij is executed locally, At the same time, regarding the allocation of computing resources used in the task computing process and the blockchain consensus process, let and are the computing resources allocated by the ith edge server to the blockchain consensus process and the computing process of task ij, respectively. make The computing resources allocated by the cloud server to the computing process of task ij, The above optimization variables are designed based on the deep reinforcement learning technology to design an optimization algorithm and establish a Markov decision process, where the state space of this method is set to b = {s, g} and the action space is set to a = {α, β, f block ,f e ,f c}. Finally, the optimization result is obtained according to the algorithm output.

[0057] Furthermore, the step three includes:

[0058] Extract the offloading decision information obtained in step 2. If the information indicates that the task is executed locally, the network controller sends the task offloading decision instructions α and β to the terminal device via a wireless connection, executes the instructions as described in this step, and directly transmits the results back to the user.

[0059] During the local execution process, the calculation delay is:

[0060]

[0061] in, is the computing resource of the terminal device. The computing energy consumption of local execution is:

[0062]

[0063] Furthermore, the step 4 includes:

[0064] Extract the offloading decision information obtained in step 2. If the information indicates that the task is executed on the edge server or cloud server, the network controller sends the task offloading decision instruction and the transmission power control instructions α, β and p to the terminal device through the wireless connection, and sends the computing resource allocation instruction f to the edge server through the wireless connection. block and f e , or send computing resource allocation instructions to the cloud server through a wired link f c , and execute the instructions as described in this step. After the task calculation process and blockchain consensus process are completed, the results are returned to the user.

[0065] Among them, during the execution process of the edge server, the calculation delay is:

[0066]

[0067] During the execution of the cloud server, the calculation delay is:

[0068]

[0069] At the same time, when the cloud server is executed, the sum of the wired transmission delays from the terminal device to the base station and from the base station to the cloud server is:

[0070]

[0071] in, It is the wired transmission delay per unit data volume from the base station to the cloud server.

[0072] At the same time, based on the PBFT protocol, the blockchain consensus delay is calculated Find the total delay of task processing:

[0073]

[0074] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Therefore, the present invention will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention shall be included in the scope of protection of the claims of the present invention.

Claims

1. A resource management method for edge computing system based on blockchain, characterized in that: The following steps are involved: S1: The network controller senses the current terminal device task information and the current wireless environment information of the system; at the same time, the network controller senses the computing resource information of the edge server and the cloud server; and uploads the relevant information from the corresponding device to the network controller through the wireless connection; S2: Input the current terminal device task information and the current wireless environment information of the system described in step S1 into the trained optimization model deployed in the network controller, and calculate the terminal device transmission power control based on the current state, the terminal device task offloading decision, and the task calculation process and the allocation of computing resources to be used in the consensus process of the blockchain; Specifically, the current task data volume and the current channel gain between the terminal device and the base station obtained in step S1 are input into the trained optimization model deployed on the network controller to calculate the resource management instructions; Firstly, a fast numerical method is designed to obtain the transmit power control command of the terminal device in the current state, and the transmit power control command is separated from other optimization variables. Among them, p ij The transmission power allocated to the terminal device ij; the total delay of the optimization target of this method decreases monotonically with the increase of the transmission power, so the transmission power control instruction is obtained by increasing the transmission power so that the energy consumption of the terminal device reaches the energy consumption constraint; The feature of task scheduling in this method is that, regarding the task offloading decision of the terminal device, let α ij =1 means that task ij is offloaded to the edge server for execution, otherwise, α ij =0; similarly, β ij =1 means that task ij is offloaded to the cloud server for execution, otherwise, β ij =0; obviously, 1-α ij -β ij =1 means that task ij is executed locally, At the same time, regarding the allocation of computing resources used in the task computing process and the blockchain consensus process, let and are the computing resources allocated by the ith edge server to the blockchain consensus process and the computing process of task ij, respectively. make The computing resources allocated by the cloud server to the computing process of task ij, The above optimization variables are designed based on the deep reinforcement learning technology to design an optimization algorithm and establish a Markov decision process, where the state space of this method is set to b = {s, g} and the action space is set to a = {α, β, f block ,f e ,f c }; Finally, the optimization result is obtained according to the algorithm output; S3: extracting the task offloading decision information of the terminal device obtained in step S2, if the information indicates that the task is executed locally, the network controller sends a corresponding task offloading decision instruction to the terminal device via a wireless connection, and the terminal device directly executes the task locally; S4: Extract the task offloading decision information of the terminal device obtained in step S2. If the information indicates that the task is executed on the edge server or cloud server, in addition to the task calculation process, the task must also be authenticated through the consensus process of the blockchain system deployed on the edge server; the network controller sends the corresponding task offloading decision instruction and transmission power control instruction to the terminal device through a wireless connection, and the terminal device transmits the task to the edge server according to the corresponding transmission power, and joins the blockchain to execute the consensus process; if the task is executed on the edge server, the result is returned to the terminal device after the task calculation process and the blockchain consensus process are all completed; if the task is executed on the cloud server, it is necessary to further transmit the task from the edge server to the cloud server through a wired link to process the task; if the task is executed on the edge server, the network controller sends a computing resource allocation instruction to the edge server through a wireless connection; if the task is executed on the cloud server, the network controller sends a computing resource allocation instruction to the cloud server through a wired link; the above instructions are used to control the calculation process of the parallel execution of the task and the consensus process of the blockchain.

2. According to a blockchain-based edge computing system resource management method according to claim 1, it is characterized in that: The step S1 includes: the network architecture of the method is characterized by: represents the edge server set, One of the base stations exist In the example, the i-th edge server covers N i Terminal devices, represents the terminal equipment set under the i-th base station, One of the terminal devices In addition, a network controller is placed at the edge layer to control the operation of the entire method; The algorithm is deployed on a network controller to work. The network controller perceives the current terminal device task information and the current wireless environment information of the system. The task characteristics are composed of the task calculation amount and data amount. Let c ij Indicates the computational amount of the task generated by each terminal device, Order ij Indicates the amount of data generated by each terminal device. The current wireless environment characteristics are composed of the channel gain from the terminal device to the base station. Let g ij represents the channel gain from the terminal device ij to its corresponding base station, At the same time, the network controller perceives the computing resource information of the edge server and the cloud server, and F c are the upper bounds of the computing resources of the i-th edge server and cloud server respectively; during the working process, the network controller will upload the above information from the corresponding device to the network controller through wireless connection.

3. According to a blockchain-based edge computing system resource management method according to claim 1, it is characterized in that: The step S3 includes: extracting the offloading decision information obtained in step S2. If the information indicates that the task is executed locally, the network controller sends the task offloading decision instructions α and β to the terminal device via a wireless connection, executes the instructions in the manner described in step S3 of claim 1, and directly transmits the result back to the user.

4. According to a blockchain-based edge computing system resource management method according to claim 1, it is characterized in that: The step S4 includes: extracting the offloading decision information obtained in step S2, if the information indicates that the task is executed on the edge server or the cloud server, the network controller sends the task offloading decision instruction and the transmission power control instructions α, β and p to the terminal device through the wireless connection, and sends the computing resource allocation instruction f to the edge server through the wireless connection. block and f e , or send computing resource allocation instructions to the cloud server through a wired link f c , and execute the instructions in the manner described in step S4 of claim 1, and after the task calculation process and the blockchain consensus process are all completed, the results are returned to the user; after the corresponding instructions are executed, the total delay is obtained, that is, the sum of the larger of the task calculation delay and the blockchain consensus delay and the task transmission delay, and the formula is: in, is the total processing delay of task ij, is the transmission delay of task ij, is the computational delay of task ij, is the blockchain consensus delay of the i-th base station.

Images