Decision-making method and system for MEC task offloading and migration based on particle swarm

Abstract

The present invention belongs to the field of mobile communication technology and provides a decision-making method and system for MEC task offloading and migration based on particle swarms. The method includes: based on a system network model, constructing a system cost model according to the relationship between the connection time between a mobile device and an MEC server and the performance of the MEC server; constructing a system delay benefit according to the system cost model, the total number of tasks, and the computing rate of the mobile device; constructing a system energy consumption benefit according to the system energy consumption cost model, the total number of tasks, and the energy consumed by a single bit task calculated by the mobile device; constructing a system benefit according to the system delay benefit and the system energy consumption benefit; constructing a planning model with the maximum system benefit as the objective function, combining time constraints, energy consumption constraints, and task quantity constraints; solving the planning model using a particle swarm algorithm to obtain the task offloading decision, task offloading amount, and / or migration amount when the system benefit is maximized, thereby completing the offloading and / or migration of the current task.

Description

Technical Field

[0001] The present invention belongs to the field of mobile communication technology, and in particular relates to a decision method and system for MEC task offloading and migration based on particle swarm. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] With the rapid development of 5G, mobile users are increasingly demanding application processing speed and service quality, widening the gap between limited computing resources and complex computing requirements. Emerging applications, such as virtual reality and real-time online gaming, often require large computational loads and very low latency. Processing these applications in user equipment (UE), which has limited computing resources and battery capacity, is challenging.

[0004] Multi-access Edge Computing (MEC) is an effective solution to these problems. In an MEC system, MEC servers with sufficient computing and communication resources are deployed at the edge of the network, closer to the user end (UE), providing them with complex computing resources. Users can offload intensive computing tasks to MEC servers for processing and return results, eliminating the need for processing in distant cloud computing centers. This significantly reduces the burden on the core network.

[0005] However, deploying MEC hosts in multi-access networks presents new challenges. Different MEC servers have varying computing and transmission capabilities. Therefore, selecting a suitable MEC host to meet user computing requests is a challenge for MEC service providers. Compared to traditional cloud server resources, MEC servers have limited resources.

[0006] In MEC networks, service continuity and accuracy face significant challenges due to user mobility. Uncertain connection times between mobile devices and MEC servers significantly impact system performance. User movement trajectories are difficult to accurately predict, complicating offloading decisions. Current research focuses on single- or multi-user service scenarios for a single MEC server. However, considering only a single server is impractical in real-world networks, especially in 5G and IoT environments. When a user leaves the current MEC server, their computing tasks must be migrated to another MEC server. Task migration can lead to increased computational and communication latency. User mobility has not been fully explored in existing work. Summary of the Invention

[0007] In order to solve the technical problem of how to optimally offload and migrate tasks in high-speed motion scenarios in the above-mentioned background technology, the present invention provides a decision-making method and system for MEC task offloading and migration based on particle swarms. The method comprehensively considers the energy consumption, task execution time, connection time, bandwidth resource limitations, and different computing capabilities of edge computing layer nodes to maximize system benefits while strictly meeting the task response time constraints. The present invention can provide services for users to select the best MEC server and maximize the use of computing resources and communication resources in the MEC server.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions:

[0009] A first aspect of the present invention provides a decision-making method for MEC task offloading and migration based on particle swarm.

[0010] The decision-making method for MEC task offloading and migration based on particle swarm includes:

[0011] Build a system network model that includes several mobile devices and several MEC servers;

[0012] Based on the system network model, a system cost model is constructed according to the relationship between the connection time between mobile devices and MEC servers and the performance of MEC servers;

[0013] Build system latency benefits based on the system cost model, the total number of tasks, and the computing rate of mobile devices;

[0014] Calculate the energy consumed by a single bit task based on the system energy cost model, the total number of tasks, and the number of mobile devices, and build the system energy efficiency;

[0015] Build system benefits based on system latency benefits and system energy consumption benefits;

[0016] Taking the maximum system benefit as the objective function, the planning model is constructed by combining time constraints, energy consumption constraints and task volume constraints;

[0017] The particle swarm algorithm is used to solve the planning model to obtain the task offloading decision, task offloading amount and / or migration amount when the system benefit is maximized, so as to complete the offloading and / or migration of the current task.

[0018] Furthermore, the system cost model includes the computational cost of the mobile device processing part of the task locally, as well as the transmission cost, migration cost and computational cost of the MEC server corresponding to the transmission task, and the idle cost generated by the mobile device occupying the MEC server;

[0019] Furthermore, the transmission cost is:

[0020]

[0021] Furthermore, the migration cost is:

[0022]

[0023] Furthermore, the computational cost is:

[0024]

[0025] Furthermore, the idle cost is:

[0026]

[0027] in, Represents mobile devices and MEC i The transmission time between Represents mobile devices and MEC j The transmission time between Represents mobile devices and MEC i The connection time between i represents the amount of tasks offloaded from mobile devices to MEC, and α represents the data transmission cost per unit time; MEC i The task calculation time in MEC j The task calculation time in represents the task computation time in the mobile device; β represents the task computation cost per unit time; Indicates that the task is from MEC i Migrate to MEC j time, γ represents the data migration cost per unit time; represents the idle time, and δ represents the idle cost per unit time.

[0028] Furthermore, the energy consumption cost model is the energy consumption generated by the mobile device in computing tasks, the energy consumption generated by task transmission and calculation result return, MEC i Energy consumption generated by server computing tasks and MEC j The cumulative energy consumption generated by server computing tasks;

[0029] Furthermore, the energy consumption of the mobile device during the computing task is:

[0030]

[0031] Furthermore, the energy consumption generated by the task transmission and calculation result return is:

[0032]

[0033] Furthermore, the MEC i The energy consumption generated by the server computing task is:

[0034]

[0035] Furthermore, the MEC j The energy consumption generated by the server computing task is:

[0036]

[0037] Among them, W represents the total amount of tasks, x i represents the amount of tasks offloaded from mobile devices to MEC, P Cl represents the energy consumption of a mobile device computing a single-bit task, P Rl represents the energy consumption of a mobile device transmitting a single bit; P Ci MEC i Calculate the energy consumption of a single-bit task, P Cj MEC j Calculate the energy consumption of a single-bit task; MEC i The task calculation time in MEC j The task computation time in ρx i represents the size of the calculation result, and ε represents the energy consumption cost per joule.

[0038] Furthermore, the system delay benefit is:

[0039]

[0040] Among them, β represents the task calculation cost per unit time, W represents the total number of tasks, V Cl represents the computing rate of the mobile device, represents the transmission cost, represents the computation cost, represents the migration cost, represents idle cost;

[0041] Furthermore, the energy efficiency of the system is:

[0042]

[0043] Where ε represents the energy cost per joule, P Cl represents the energy consumption of a mobile device computing a single-bit task, Indicates the energy consumption cost of the MEC system.

[0044] Furthermore, the objective function is:

[0045]

[0046] Among them, η i represents the weight of system delay benefit, η j represents the weight of the system energy efficiency, represents the system delay benefit, Indicates the energy efficiency of the system.

[0047] Furthermore, the time constraint is:

[0048]

[0049] Among them, D m Indicates the maximum duration allowed to complete the task. Represents mobile devices and MEC i The connection time between Represents task x i Transfer from mobile device to MEC i time, Represents task x i In MEC i The calculation time in ; Indicates that the connection time is insufficient to complete task x i Transmission to MEC i In this scenario, the delay of task processing, Indicates that the connection time is sufficient to upload task x i However, it is not enough to complete the task processing delay in computing scenarios. Indicates the task processing latency when the connection time is sufficient to upload and complete the calculation.

[0050] Furthermore, the energy consumption constraint is:

[0051]

[0052] Among them, E i,j (x i ) represents the energy consumed by the mobile device, Indicates the energy consumption threshold of the mobile device.

[0053] Furthermore, the task load constraint is

[0054] 0 <x i <W,i∈N;

[0055] Among them, x i represents the amount of tasks offloaded from mobile devices to MEC, and W represents the total amount of tasks.

[0056] Furthermore, the process of using the particle swarm algorithm to solve the planning model to obtain the task offloading decision, task offloading amount and / or migration amount when the system benefit is maximized, thereby completing the offloading and / or migration of the current task includes:

[0057] Consider the amount of task offloaded from mobile devices x i Located between 0-W, a single particle will be defined as a one-dimensional vector, and the value of each particle represents a randomly distributed task offloading amount;

[0058] During each iteration, the individual fitness value of each particle is calculated The current fitness value is compared with the historical optimal fitness value P of the particle individual. best And the global optimal fitness value G of the entire population best Compare and update P best and G best ;

[0059] Update the velocity and position of each particle until the iteration is completed;

[0060] Compare and evaluate the maximum system benefits that can be achieved by each MEC server combination, complete the task offloading and migration decision that maximizes the system benefits, and obtain the joint optimal task offloading decision x i Initial uninstallation of MEC i and Migrate MEC j .

[0061] The second aspect of the present invention provides a decision system for MEC task offloading and migration based on particle swarm.

[0062] The decision-making system for MEC task offloading and migration based on particle swarm includes:

[0063] A model building module is configured to: build a system network model including a plurality of mobile devices and a plurality of MEC servers;

[0064] A system cost model building module is configured to: build a system cost model based on the system network model and the relationship between the connection time between the mobile device and the MEC server and the performance of the MEC server;

[0065] A latency benefit building module is configured to: build a system latency benefit based on a system cost model, a total amount of tasks, and a computing rate of a mobile device;

[0066] An energy consumption benefit building module is configured to: calculate the energy consumed by a single bit task according to a system energy consumption cost model, the total number of tasks, and mobile devices, and build a system energy consumption benefit;

[0067] A system benefit building module is configured to: build system benefits based on system delay benefits and system energy consumption benefits;

[0068] A planning model building module is configured to: take maximum system benefit as the objective function, combine time constraints, energy consumption constraints and task volume constraints, and build a planning model;

[0069] The solution and execution module is configured to: use the particle swarm algorithm to solve the planning model, obtain the task offloading decision, task offloading amount and / or migration amount when the system benefit is maximized, and thus complete the offloading and / or migration of the current task.

[0070] A third aspect of the present invention provides a computer-readable storage medium.

[0071] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the decision-making method for MEC task offloading and migration based on particle swarm as described in the first aspect above.

[0072] A fourth aspect of the present invention provides a computer device.

[0073] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the particle swarm-based MEC task offloading and migration decision method as described in the first aspect above are implemented.

[0074] Compared with the prior art, the present invention has the following beneficial effects:

[0075] This paper proposes a decision-making method and system for MEC task offloading and migration based on particle swarm. Taking into account the relationship between the connection time of mobile user devices and MEC servers, a MEC task processing model (Transmission Calculation Migration Idle and Energy, TCMIE) is established in high-speed mobile scenarios. By defining the system benefit function, a task offloading and migration scheme based on optimal system benefit is designed. While optimizing latency and energy consumption, it fully utilizes the advantages of edge computing and ensures global optimality, strong robustness, and fast convergence.

[0076] The present invention comprehensively considers communication resources, computing resources, energy consumption, connection time and task volume, selects the best MEC server to provide services for user devices, and provides optimal task offloading and migration tasks.

[0077] The method of the present invention has low computational complexity, high accuracy, and higher system benefits than typical similar task offloading and migration solutions. The system benefits of the solution proposed by the present invention are 12% higher than those of several existing solutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0079] Figure 1 This is a diagram of the task offloading and migration scenario in multi-access edge computing in the present invention.

[0080] Figure 2 This is a flow chart of the particle swarm decision-making method in the present invention.

[0081] Figure 3 This is a comparison chart of system benefit simulation results between the present invention and the existing solution under the same task volume.

[0082] Figure 4 This is a comparison chart of system benefit simulation results between the present invention and the existing solution at the same connection time.

[0083] Figure 5 The figure compares the delay performance simulation results of the present invention and the existing solution in a multi-user scenario.

[0084] Figure 6 This is a comparison chart of system benefit simulation results between the present invention and the existing solution in a multi-user scenario;

[0085] Figure 7 Flowchart of the decision-making method for MEC task offloading and migration based on particle swarm in the present invention. DETAILED DESCRIPTION

[0086] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0087] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0088] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0089] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the methods and systems according to the various embodiments of the present disclosure. It should be noted that each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code can include one or more executable instructions for implementing the logical functions specified in the various embodiments. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, or they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the flowchart and / or block diagram, and the combination of the boxes in the flowchart and / or block diagram, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or can be implemented using a combination of dedicated hardware and computer instructions.

[0090] Explanation of terms:

[0091] High-speed mobility scenarios refer to scenarios where users are moving at high speeds (such as high-speed railways and subways) and a large amount of user data demand is concentrated due to the dense user population. This scenario is characterized by both high-speed user movement and high density.

[0092] Example 1

[0093] like Figure 7 As shown, this embodiment provides a decision method for MEC task offloading and migration based on particle swarms. This embodiment uses the method applied to the server as an example. It can be understood that the method can also be applied to terminals, and can also be applied to systems including terminals, servers, and implemented through the interaction between terminals and servers. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, mainly referring to edge servers, which have limited resources compared to cloud servers but are located at the edge layer of the network, close to users. The terminal can be a smartphone, tablet computer, laptop computer, desktop computer, smart watch, etc., but is not limited to this. The terminal and the server can be directly or indirectly connected via wired or wireless communication, which is not limited in this application. In this embodiment, the method includes the following steps:

[0094] Build a system network model that includes several mobile devices and several MEC servers;

[0095] Based on the system network model, a system cost model is constructed according to the relationship between the connection time between mobile devices and MEC servers and the performance of MEC servers;

[0096] Build system latency benefits based on the system cost model, the total number of tasks, and the computing rate of mobile devices;

[0097] Calculate the energy consumed by a single bit task based on the system energy cost model, the total number of tasks, and the number of mobile devices, and build the system energy efficiency;

[0098] Build system benefits based on system latency benefits and system energy consumption benefits;

[0099] Taking the maximum system benefit as the objective function, the planning model is constructed by combining time constraints, energy consumption constraints and task volume constraints;

[0100] The particle swarm algorithm is used to solve the planning model to obtain the task offloading decision, task offloading amount and / or migration amount when the system benefit is maximized, so as to complete the offloading and / or migration of the current task.

[0101] The particle swarm-based MEC task offloading and migration decision-making method of the present invention can select the best MEC server for users to provide services and maximize the use of computing resources and storage resources in the MEC server. In order to more clearly describe the purpose, technical solutions and advantages of the present invention, the technical solutions of the present invention will be described in detail and in full with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them.

[0102] like Figure 1 As shown in Figure 1, assume that there are N MEC servers serving mobile users. In high-speed mobile scenarios, mobile users have computationally intensive tasks W that need to be processed. Due to their limited computing power, the tasks W that need to be processed can be partially split and processed simultaneously on the mobile device and offloaded to the MEC server for processing. In high-speed mobile scenarios, mobile users are very likely to switch between base stations, which may lead to the switching of serving MEC servers. This is because if the user is in close contact with the MEC server (MEC i ) is too short. When the user leaves the MEC i When the communication range is reached, the unfinished tasks will be sent from the MEC i Transfer to the next available MEC server (MEC j ) to continue processing. This process is called "migration". If the mobile user is connected to the MEC i If the connection time is too long, i After forwarding the task results to the user, the user remains connected to the MEC i , the process is "idle".

[0103] The particle swarm-based MEC task offloading and migration method of this embodiment specifically includes the following steps:

[0104] Step 1: Build a system cost model based on the relationship between the connection time between the mobile device UE and the MEC server and the MEC system performance. The system cost model includes the computational cost of the mobile device processing some tasks locally, as well as the transmission cost corresponding to the transmission task, the migration cost, the computational cost of the MEC server, and the idle cost incurred by the mobile device occupying the MEC server.

[0105] Transmission costs:

[0106] Calculate the cost:

[0107] Migration costs:

[0108] Idle cost:

[0109] The sum of the above four costs is the communication and computation costs. Represents mobile devices and MEC i The transmission time between Represents mobile devices and MEC j The transmission time between Represents mobile devices and MEC i The connection time between i represents the amount of tasks offloaded from mobile devices to MEC, and α represents the data transmission cost per unit time; MEC i The task calculation time in MEC j The task calculation time in represents the task computation time in the mobile device; β represents the task computation cost per unit time of the local device; Indicates that the task is from MEC i Migrate to MEC j Time, γ local device data migration cost per unit time; represents the idle time, and δ is the idle cost per unit time of the local device.

[0110] In the task processing model of this embodiment, only the energy consumed by mobile user devices and the energy consumed by MEC computing are considered, while the energy consumed by MEC server transmission is ignored. Therefore, the energy consumption generated by the three processes of local computing, task transmission, and MEC server computing will be taken into account.

[0111] The energy consumption of mobile devices during computing tasks is expressed as:

[0112]

[0113] The energy consumption generated by task transmission and calculation result return is expressed as:

[0114]

[0115] MEC i The energy consumption generated by the computing task is expressed as:

[0116]

[0117] MEC j The energy consumption generated by the computing task is expressed as:

[0118]

[0119] The total energy cost can be expressed as:

[0120]

[0121] Among them, P Cl represents the energy consumption of a mobile device computing a single-bit task, P Rl represents the energy consumption of a mobile device transmitting a single bit; P Ci Indicates ME Ci Calculate the energy consumption of a single-bit task, P Cj MEC j Calculate the energy consumption of a single-bit task; ρx i represents the size of the calculation result, and ε represents the energy consumption cost per joule.

[0122] Step 2: In the model proposed in this embodiment, the optimization objectives of the MEC system mainly include energy consumption and task execution delay. The system benefit is defined as the benefit improved by the mobile device and MEC server collaboratively processing tasks compared with the mobile UE independently processing tasks. The system benefit includes two parts: delay benefit and energy consumption benefit.

[0123] The system delay benefit can be expressed as:

[0124]

[0125] The energy efficiency of the system can be expressed as:

[0126]

[0127] The system benefit function can be expressed as:

[0128]

[0129] Among them, W represents the total amount of tasks, V Cl represents the computing rate of the mobile device, P Clrepresents the energy consumed by the mobile device to calculate a single bit task; η i represents the weight of delay benefit, η j Represents the weight of energy efficiency.

[0130] Step 3: The optimization problem of system benefits involves the joint optimization task offloading decision x i Initial uninstallation of MEC i and Migrate MEC j ;The task offloading decision problem can be expressed as:

[0131]

[0132] Constraint 1:

[0133]

[0134] Constraint 2:

[0135]

[0136] Constraint 3:

[0137] 0 <x i <W,i∈N;

[0138] Among them, D m Indicates the maximum duration allowed for completing the task, E i,j (x i ) represents the energy consumed by the mobile device, Indicates the energy consumption threshold of the mobile device; Represents task x i Transfer from mobile device to MEC i time, Represents task x i In MEC i The calculation time in ; Indicates that the connection time is insufficient to complete task x i Transmission to MEC i In this scenario, the delay of task processing, Indicates that the connection time is sufficient to upload task x i However, it is not enough to complete the task processing delay in computing scenarios. Indicates the task processing latency when the connection time is sufficient to upload and complete the calculation.

[0139] Specifically expressed as:

[0140]

[0141] Specifically expressed as:

[0142]

[0143] Specifically expressed as:

[0144]

[0145] in, Indicates UE and MEC i The uplink transmission rate between i,j MEC i and MEC j The transmission rate between Indicates UE and MEC j The transmission rate of the downlink between HO Indicates the time when the mobile user switches to MEC. Indicates the task calculation time of the mobile device; VC i MEC i The calculation rate.

[0146] The above constraint 1 ensures that the system delay cannot be greater than the maximum duration of task completion; constraint 2 ensures that the energy consumption of mobile users does not exceed the energy consumption threshold of mobile devices; constraint 3 ensures that the amount of tasks offloaded to MEC by users does not exceed the total task amount.

[0147] Step 4: The problem considered in Step 3 is formulated as a mixed integer nonlinear programming problem, which jointly optimizes the task offloading decision x i Initial uninstallation of MEC i and Migrate MEC j Due to the combinatorial nature of the problem, it is difficult and impractical to find the optimal solution. We introduce the particle swarm optimization (PSO) algorithm to solve it. The specific process is shown in Figure 2 As shown. The mobile device has multiple MECs at the initial location i Task offloading can be performed because when task migration occurs on mobile devices, MEC i The remaining tasks need to be migrated to MEC j , MEC j There are also multiple options to choose from; first determine the initial uninstall MEC selected by the user i and Migrate MEC j possible combinations; for each combination, the particle swarm algorithm is introduced to obtain the maximum system benefit that this MEC server combination can achieve and the corresponding task offloading amount x i .

[0148] The specific steps of the particle swarm algorithm include: considering the possible task offloading amount x of the mobile device: iBetween 0-W, then a single particle will be defined as a one-dimensional vector, and the value of each particle represents a randomly distributed task offloading amount; during each iteration, the individual fitness value of each particle is calculated The current fitness value is compared with the historical optimal fitness value P of the particle individual. best And the global optimal fitness value G of the entire population best Compare and update P best and G best Then update the speed and position of each particle until the iteration is complete. Finally, compare and evaluate the maximum system benefit that can be achieved by each MEC server combination, complete the task offloading and migration decision that maximizes the system benefit, and obtain the joint optimized task offloading decision x i Initial uninstallation of MEC i and Migrate MEC j .

[0149] like Figure 3 and Figure 4 As shown in the figure, the system benefit comparison of the solution proposed in this invention (Optimization of Communication Calculation and Energy, O-CCE) in a single-user scenario with the traditional delay optimization solution and energy consumption optimization solution is shown. The O-CCE solution proposed in this invention has the best system benefit, the energy consumption optimization solution ranks second, and the delay optimization solution has the worst. The system benefit can be improved by up to 12% compared with the energy consumption optimization solution and 20%-30% compared with the delay optimization solution. Figure 5 As shown in the figure, in the multi-user scenario, the average system delay of the optimal delay solution is the smallest, followed by the O-CCE solution proposed in the present invention, and the energy consumption optimal solution is the largest. In order to achieve the minimum system delay in the optimal delay solution, mobile devices and MEC servers need to process tasks in parallel. Although the delay is reduced, it sacrifices the system benefit and does not fully utilize the computing advantages of MEC. In order to pursue the minimum system energy consumption, the optimal energy consumption solution will select the MEC server with the minimum computing energy consumption for task offloading. Although the energy consumption of the system is reduced, the system delay is also increased. As shown in the figure, the average system delay of the optimal delay solution is the smallest. Figure 6 As shown, in the multi-user scenario, the proposed solution has the best system benefit, followed by the optimal energy consumption solution, and the optimal latency solution has the worst. Compared with the optimal energy consumption solution, the average system benefit increased by 5%, and compared with the optimal latency solution, the average system benefit increased by 28%.

[0150] Example 2

[0151] This embodiment provides a decision system for MEC task offloading and migration based on particle swarm.

[0152] The decision-making system for MEC task offloading and migration based on particle swarm includes:

[0153] A model building module is configured to: build a system network model including a plurality of mobile devices and a plurality of MEC servers;

[0154] A system cost model building module is configured to: build a system cost model based on the system network model and the relationship between the connection time between the mobile device and the MEC server and the performance of the MEC server;

[0155] A latency benefit building module is configured to: build a system latency benefit based on a system cost model, a total amount of tasks, and a computing rate of a mobile device;

[0156] An energy consumption benefit building module is configured to: calculate the energy consumed by a single bit task according to a system energy consumption cost model, the total number of tasks, and mobile devices, and build a system energy consumption benefit;

[0157] A system benefit building module is configured to: build system benefits based on system delay benefits and system energy consumption benefits;

[0158] A planning model building module is configured to: take maximum system benefit as the objective function, combine time constraints, energy consumption constraints and task volume constraints, and build a planning model;

[0159] The solution and execution module is configured to: use the particle swarm algorithm to solve the planning model, obtain the task offloading decision, task offloading amount and / or migration amount when the system benefit is maximized, and thus complete the offloading and / or migration of the current task.

[0160] It should be noted that the aforementioned model building module, system cost model building module, latency benefit building module, energy consumption benefit building module, system benefit building module, planning model building module, and solution and execution module implement the same examples and application scenarios as those implemented in the steps of Example 1, but are not limited to the contents disclosed in Example 1. It should be noted that the aforementioned modules, as part of the system, can be executed in a computer system, such as a set of computer-executable instructions.

[0161] Example 3

[0162] This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the steps in the decision-making method for MEC task offloading and migration based on particle swarm as described in the first embodiment above are implemented.

[0163] Example 4

[0164] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the particle swarm-based MEC task offloading and migration decision-making method as described in the first embodiment are implemented.

[0165] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage and optical storage, etc.) containing computer-usable program code.

[0166] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0167] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0168] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0169] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware through a computer program. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0170] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. The decision-making method of MEC task offloading and migration based on particle swarm is characterized by: include: Build a system network model that includes several mobile devices and several MEC servers; Based on the system network model, a system cost model is constructed according to the relationship between the connection time between the mobile device and the MEC server and the performance of the MEC server. The system cost model includes the idle cost caused by the mobile device occupying the MEC server. The idle cost is: ， in, Indicates that the mobile device and The connection time between represents the amount of tasks offloaded from mobile devices to MEC, represents the idle time, δ represents the idle cost per unit time; Based on the system cost model, the total number of tasks, and the computing rate of the mobile device, the system delay benefit is constructed; the system delay benefit is: Among them, β represents the task calculation cost per unit time, W represents the total number of tasks, represents the computing rate of the mobile device, represents the transmission cost, represents the computation cost, represents the migration cost, represents the idle cost, j Represents the migration MEC j Subscript in j ; The energy consumed by a single bit task is calculated based on the system energy cost model, the total number of tasks, and the number of mobile devices, and the system energy efficiency is constructed. The system energy efficiency is: in, represents the energy cost per joule, represents the energy consumption of a mobile device computing a single-bit task, Represents the energy consumption cost of the MEC system; Build system benefits based on system latency benefits and system energy consumption benefits; Taking the maximum system benefit as the objective function, a planning model is constructed by combining time constraints, energy consumption constraints, and task volume constraints; the time constraints are: in, Indicates the maximum duration allowed to complete the task. Indicates that the mobile device and The connection time between Indicates a task Transfer from mobile device to time, Indicates a task exist The calculation time in ; Indicates that the connection time is insufficient to complete the task Transfer to In this scenario, the delay of task processing, Indicates that the connection time is sufficient to upload the task However, it is not enough to complete the task processing delay in computing scenarios. Indicates the task processing delay when the connection time is sufficient to upload and complete the calculation; The energy consumption constraint is: < ， ； in, represents the energy consumed by the mobile device, represents the energy consumption threshold of the mobile device, N represents the number of MECs used as the initial offloading MEC; The task load constraint is W， ； in, represents the amount of tasks offloaded from mobile devices to MEC, and W represents the total amount of tasks; The particle swarm algorithm is used to solve the planning model to obtain the task offloading decision, task offloading amount and / or migration amount when the system benefit is maximized, so as to complete the offloading and / or migration of the current task.

2. The method for offloading and migrating MEC tasks based on particle swarm according to claim 1, characterized in that: The system cost model includes the computational cost of the mobile device processing part of the task locally, the transmission cost corresponding to the transmission task, the migration cost and the computational cost of the MEC server, and the idle cost generated by the mobile device occupying the MEC server; The transmission cost is: The migration cost is: The computational cost is: in, Indicates that the mobile device and The transmission time between Indicates that the mobile device and The transmission time between Indicates that the mobile device and The connection time between represents the amount of tasks offloaded from mobile devices to MEC, and α represents the data transmission cost per unit time; express The task calculation time in express The task calculation time in represents the task computation time in the mobile device; β represents the task computation cost per unit time; Indicates that the task Migrate to time, γ represents the data migration cost per unit time.

3. The method for offloading and migrating MEC tasks based on particle swarm according to claim 1, characterized in that: The energy consumption cost model is the energy consumption generated by the mobile device in computing tasks, task transmission and calculation result transmission, The energy consumption generated by the server computing tasks and The cumulative energy consumption generated by server computing tasks; The energy consumption of the mobile device during the computing task is: The energy consumption generated by the task transmission and calculation result return is: described The energy consumption generated by the server computing task is: described The energy consumption generated by the server computing task is: Among them, W represents the total amount of tasks, represents the amount of tasks offloaded from mobile devices to MEC, represents the energy consumption of a mobile device computing a single-bit task, represents the energy consumption of a mobile device transmitting a single bit; express Calculate the energy consumption of a single-bit task, express Calculate the energy consumption of a single-bit task; express The task calculation time in express The task calculation time in ; Indicates the size of the calculation result, Indicates the cost per joule of energy consumed.

4. The method for offloading and migrating MEC tasks based on particle swarm according to claim 1, characterized in that: The objective function is: in, represents the weight of system delay benefit, represents the weight of the system energy efficiency, represents the system delay benefit, Indicates the energy efficiency of the system.

5. The method for offloading and migrating MEC tasks based on particle swarm according to claim 1, characterized in that: The process of using the particle swarm algorithm to solve the planning model to obtain the task offloading decision, task offloading amount and / or migration amount when the system benefit is maximized, thereby completing the offloading and / or migration of the current task includes: Considering task offloading on mobile devices Located between 0-W, a single particle will be defined as a one-dimensional vector, and the value of each particle represents a randomly distributed task offloading amount; During each iteration, the individual fitness value of each particle is calculated , compare the current fitness value with the historical optimal fitness value of the particle individual And the global optimal fitness value of the entire population Compare and update and ; Update the velocity and position of each particle until the iteration is completed; Compare and evaluate the maximum system benefits that can be achieved by each MEC server combination, complete the task offloading and migration decisions that maximize the system benefits, and obtain the joint optimal task offloading decision , Initial Uninstall and migration .

6. The decision system for MEC task offloading and migration based on particle swarm is characterized by: include: The model construction module is configured to: construct a system network model including a plurality of mobile devices and a plurality of MEC servers; the system cost model includes the idle cost generated by the mobile devices occupying the MEC servers; the idle cost is: ， in, Indicates that the mobile device and The connection time between represents the amount of tasks offloaded from mobile devices to MEC, represents the idle time, δ represents the idle cost per unit time; A system cost model building module is configured to: build a system cost model based on the system network model and the relationship between the connection time between the mobile device and the MEC server and the performance of the MEC server; The delay benefit building module is configured to build a system delay benefit based on the system cost model, the total number of tasks, and the computing rate of the mobile device; the system delay benefit is: Among them, β represents the task calculation cost per unit time, W represents the total number of tasks, represents the computing rate of the mobile device, represents the transmission cost, represents the computation cost, represents the migration cost, represents the idle cost, j Represents the migration MEC j Subscript in j ; The energy consumption benefit building module is configured to calculate the energy consumed by a single bit task based on the system energy consumption cost model, the total number of tasks, and the mobile device, and build the system energy consumption benefit; the system energy consumption benefit is: in, represents the energy cost per joule, represents the energy consumption of a mobile device computing a single-bit task, Represents the energy consumption cost of the MEC system; A system benefit building module is configured to: build system benefits based on system delay benefits and system energy consumption benefits; The planning model building module is configured to build a planning model with the maximum system benefit as the objective function, combined with time constraints, energy consumption constraints and task volume constraints; the time constraints are: in, Indicates the maximum duration allowed to complete the task. Indicates that the mobile device and The connection time between Indicates a task Transfer from mobile device to time, Indicates a task exist The calculation time in ; Indicates that the connection time is insufficient to complete the task Transfer to In this scenario, the delay of task processing, Indicates that the connection time is sufficient to upload the task However, it is not enough to complete the task processing delay in computing scenarios. Indicates the task processing delay when the connection time is sufficient to upload and complete the calculation; The energy consumption constraint is: < ， ； in, represents the energy consumed by the mobile device, represents the energy consumption threshold of the mobile device, N represents the number of MECs used as the initial offloading MEC; The task load constraint is W， ； in, represents the amount of tasks offloaded from mobile devices to MEC, and W represents the total amount of tasks; The solution and execution module is configured to: use the particle swarm algorithm to solve the planning model, obtain the task offloading decision, task offloading amount and / or migration amount when the system benefit is maximized, and thus complete the offloading and / or migration of the current task.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps in the decision-making method for MEC task offloading and migration based on particle swarm are implemented as described in any one of claims 1 to 5.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps in the decision-making method for MEC task offloading and migration based on particle swarm are implemented as described in any one of claims 1 to 5.

Images