Resource allocation and task unloading proportion optimization method based on vehicle mobile edge calculation

Abstract

The invention relates to the technical field of wireless communication, in particular to a resource allocation and task unloading proportion optimization method based on vehicle mobile edge computing, and the method comprises the steps: constructing a communication and task unloading computing system model between a plurality of base stations and a task vehicle; based on the communication and task unloading calculation system model, on the premise of considering that the task vehicle keeps communication with the base station, determining a system utility function of the task vehicle, the system utility function representing a weighted sum of completion time delay of the task vehicle and equipment energy consumption; setting constraint conditions by taking minimization of the system utility function as a target, and constructing a system utility function model; and based on the system utility function model, solving a frequency allocation optimal solution of the task vehicle, a transmission power optimal solution between the task vehicle and the base station and a task unloading proportion optimal solution. The invention aims to solve the problem of how to reduce the effectiveness of a communication system when a vehicle moves.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technologies, and in particular to a method for optimizing resource allocation and task offloading ratio based on vehicle mobile edge computing. Background Art

[0002] With the increasing requirements for vehicle intelligence, a large number of computing-intensive and delay-sensitive applications continue to emerge, such as intelligent traffic control, autonomous driving, image or video-assisted real-time navigation, etc. These applications always require abundant computing resources to maintain system performance, while the limited computing power of the vehicle itself can hardly meet the needs of these complex applications.

[0003] At present, most of the research on mobile edge computing only considers the unilateral impact of system delay or energy consumption, and does not fully consider the problem of limited communication time between vehicles and base stations. Therefore, an optimization resource allocation and task offloading strategy based on simulated annealing algorithm is needed to effectively reduce the utility of the system by jointly optimizing resource allocation and task offloading ratio. Summary of the invention

[0004] The main purpose of this application is to provide a resource allocation and task offloading ratio optimization method based on vehicle mobile edge computing, aiming to solve the problem of how to reduce the effectiveness of the communication system when the vehicle is moving.

[0005] To achieve the above objectives, the present application provides a method for optimizing resource allocation and task offloading ratio based on vehicle mobile edge computing, the method comprising:

[0006] S1, building a communication and task offloading computing system model between multiple base stations and task vehicles;

[0007] S2, based on the communication and task offloading computing system model, under the premise of considering that the task vehicle maintains communication with the base station, determining the system utility function of the task vehicle, wherein the system utility function is characterized as a weighted sum of the completion delay of the task vehicle and the energy consumption of the equipment;

[0008] S3, setting constraints with the goal of minimizing the system utility function and constructing a system utility function model;

[0009] S4, based on the system utility function model, solving the optimal solution for the frequency allocation of the task vehicle, the optimal solution for the transmission power between the task vehicle and the base station, and the optimal solution for the task offloading ratio.

[0010] Optionally, in S1, the step of constructing the communication and task offloading computing system model includes:

[0011] S11, the mission vehicle is assembled To represent, the set of base stations is represented as Each mission vehicle There is only one task R to be processed n :

[0012]

[0013] Where, d n is the data size of the task, c n To calculate R n The number of CPU cycles required, The maximum tolerable constraint delay for completing the task;

[0014] S12, assuming that the total system bandwidth is B, divide it into S sub-channels, the size of each sub-channel is W = B / S, and the task OT is calculated using the Shannon formula n The transfer rate unloaded to task vehicle n is:

[0015]

[0016] In the formula, p n is the transmission power between task vehicle n and service node j, h nj For the task vehicle n and the service node The channel gain between 2 is the average noise power from the mission vehicle to the service node;

[0017] S13, the service nodes are arranged in parallel. Assuming that the center coordinates of the first service node j are expressed as (r, r), the center coordinates of the remaining service nodes are expressed as (r, r+(j-1)2r). The task vehicle n is randomly distributed within the communication coverage of each service node, and its initial position coordinates are expressed as (x nj ,y nj ), and at a speed v n Travel at a constant speed, and define the right direction as the positive direction. The lateral distance between the task vehicle n from the initial position and the out-of-service node coverage range is defined as s n , its formula is expressed as:

[0018]

[0019] Among them, r represents the communication coverage radius of the service node and the driving direction of the vehicle;

[0020] Calculate the communication delay when the vehicle leaves the coverage area of ​​the service node

[0021]

[0022] S14, to ensure that each service node can successfully perform offload calculations on vehicles within its communication range, the task must meet the following constraints:

[0023]

[0024] In the formula, represents the transmission delay of the mission vehicle unloading to the service node j through the uplink,

[0025] Represents the computing delay consumed by the task vehicle at service node j.

[0026] Optionally, in S2, the expression of the system utility function is:

[0027]

[0028] in,

[0029]

[0030] Where, T n is the total delay that vehicle n needs to spend in completing the task, E n is the total energy consumption of task vehicle n in the process of completing the task, λ t represents the preference factor of system delay, λ e is expressed as the preference factor of system energy consumption, and λ t +λ e =1, is the computing power of the mission vehicle n, κ is the computing energy consumption coefficient related to the vehicle’s own chip, c n is the number of CPU cycles, p n is the transmission power between the mission vehicle n and the service node, d n is the data size of the task, ξ n is the uninstall ratio.

[0031] Optionally, in S3, the expression of the system utility function model is:

[0032] P1:

[0033] st.C1:

[0034] C2:

[0035] C3:

[0036] C4:

[0037] C5:

[0038]

[0039] Where C1 represents the unloading ratio ξ n The value is between 0 and 1; C2 represents the transmission power p of the constrained task vehicle n From 0 to P max C3 represents the computing resources f allocated by service node j to task vehicle n nj Greater than or equal to 0; C4 represents the total computing power allocated to all mission vehicles Less than the computing resources F of the service node j ; C5 means that the task offloading calculation delay is less than the minimum value of the communication holding delay and the maximum tolerable delay.

[0040] Optionally, the steps in S4 specifically include:

[0041] S41, solving the optimal solution of the frequency allocation based on the Lagrange multiplier method;

[0042] S42, solving the optimal solution of the transmission power based on a quasi-convex technology;

[0043] S43, solving the optimal solution of the task offloading ratio based on a simulated annealing algorithm.

[0044] Optionally, the S43 specifically includes:

[0045] S431, initialize the initial maximum temperature to 1200, the number of loop iterations to 50, the initial temperature to 1200 and the termination temperature to 0.1, and randomly assign an initial unloading factor at the starting point of the algorithm iteration

[0046] S432, perturb the initial solution to generate a new solution, that is, satisfy Where e is the random perturbation, L is the step size related to the initial solution and the value range;

[0047] S433, obtain new unloading factor ξ n Then, the task vehicle transmission power is updated based on the Lagrange multiplier method, and the transmission power is solved according to the quasi-convex technology, and substituted into the C5 to determine whether it is satisfied. If it is satisfied, it is a new solution. If it is not satisfied, repeat to step S432;

[0048] S434, calculate the difference of the objective function If ΔC>0, then accept ξ n as the new current solution, otherwise with probability Accept n As the current new solution;

[0049] S435, loop and iterate steps S432-S434 50 times, determine whether the current temperature T reaches the termination temperature, if so, output the current new solution as the optimal solution; otherwise, return to step S42 to continue execution.

[0050] This application has at least the following beneficial effects:

[0051] Most existing vehicle mobile edge computing system designs only consider the goal of reducing system latency or energy consumption, and do not consider the mobility of the vehicle, and cannot meet more comprehensive user needs. The present invention models the system utility function based on the time and energy consumption of the trade-off system, can flexibly adjust the latency and energy consumption factors to better meet real-time needs, and establish a communication delay constraint between the service node and the task vehicle. By jointly optimizing the node's computing frequency allocation, vehicle transmission power and task offloading ratio, the system utility is effectively reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A system model diagram of the present invention in which multiple base stations provide unloading calculations for multi-task vehicles;

[0053] Figure 2 A flowchart of a method for optimizing resource allocation and task offloading ratio in vehicle edge computing of the present invention;

[0054] Figure 3 is a graph showing the relationship between the number of system vehicles and system utility in comparison between the solution proposed in the present invention and the benchmark solution;

[0055] Figure 4 It is a graph showing the relationship between the number of iterations and system utility when comparing the scheme proposed in the present invention with the greedy unloading scheme.

[0056] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0057] In order to better understand the above technical solution, exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0058] First embodiment

[0059] Reference Figure 1, a system model consisting of multiple task vehicles and base stations on a two-way road is considered. The base station in the middle of the road provides offloading computing services to task vehicles with limited computing resources. The task requesting vehicle is represented by a set. The set of base stations is collectively referred to as a service node, denoted as. The task vehicle can perform its task locally and communicate with nearby base stations to offload the task to the MEC server assigned to the base station for auxiliary computing processing. Assuming that each task vehicle has only one task to be requested, it is represented as, where is the data size of the task, is the number of CPU cycles required for calculation, and is the maximum constrained delay that can be tolerated to complete the task.

[0060] Reference Figure 2 A method for optimizing resource allocation and task offloading ratio based on vehicle mobile edge computing includes the following steps:

[0061] S1, building a communication and task offloading computing system model between multiple base stations and task vehicles;

[0062] S2, based on the communication and task offloading computing system model, under the premise of considering that the task vehicle maintains communication with the base station, determining the system utility function of the task vehicle, wherein the system utility function is characterized as a weighted sum of the completion delay of the task vehicle and the energy consumption of the equipment;

[0063] S3, setting constraints with the goal of minimizing the system utility function and constructing a system utility function model;

[0064] S4, based on the system utility function model, solving the optimal solution for the frequency allocation of the task vehicle, the optimal solution for the transmission power between the task vehicle and the base station, and the optimal solution for the task offloading ratio.

[0065] Second embodiment

[0066] Based on the first embodiment, in this embodiment, the communication between the task vehicle and the service node adopts Orthogonal Frequency Division Multiple Access (OFDMA). The total system bandwidth is B, which is divided into S sub-channels, and the size of each sub-channel is W = B / S. According to Shannon's formula, the task OT can be calculated n The transfer rate for offloading to the service vehicle is:

[0067]

[0068] where p n is the transmission power between the mission vehicle n and the service node, h nj For the task vehicle n and the service node The channel gain between 2is the average noise power from the mission vehicle to the service node.

[0069] The service nodes are arranged in parallel. Assuming that the center coordinates of the first service node j are expressed as (r, r), the center coordinates of the remaining service nodes can be expressed as (r, r+(j-1)2r). The task vehicle n is randomly distributed within the communication coverage of each service node, and its initial position coordinates are expressed as (x nj ,y nj ), and at a speed v n The vehicle is driven at a constant speed, and the right direction is defined as the positive direction. The lateral distance between the task vehicle n from the initial position and the area outside the coverage of the service node is defined as s n , its formula is expressed as:

[0070]

[0071] Where r represents the communication coverage radius of the service node and the driving direction of the vehicle. Therefore, the communication delay of the vehicle leaving the coverage of the service node is:

[0072]

[0073] To ensure that each service node can successfully perform offload calculations on vehicles within its communication range, the following constraints must be met:

[0074]

[0075] Third embodiment

[0076] Based on any of the above embodiments, in this embodiment, in order to balance the resources between the task vehicle and the service node, the task vehicle decomposes the task into two parts: local calculation and edge service node offloading calculation for calculation.

[0077] In the local computing model, the latency of some tasks being computed locally in the vehicle is:

[0078]

[0079] At the same time, the energy consumption of the mission vehicle calculated locally is:

[0080]

[0081] in is the computing power of the mission vehicle n, and κ is the computing energy consumption coefficient related to the vehicle’s own chip.

[0082] In the edge service node offloading computing model, the total delay of task vehicle n offloading the task to service node j for edge computing consists of three parts: task transmission delay, task computing delay of the service node, and calculation result return delay. Since the amount of result data after the service node calculation is much smaller than the initial data amount, the return delay of the downlink can be ignored.

[0083] The transmission delay of part of the task offloading to service node j through the uplink is:

[0084]

[0085] At the same time, the computational delay consumed at service node j is:

[0086]

[0087] Among them, f nj It represents the computing resources provided by the service node j assigned to the user vehicle n. The computing resources of the service node j are F j , and can allocate computing resources to multiple mission vehicles at the same time. Therefore, it is necessary to meet the following requirements:

[0088]

[0089] The total delay of offloading edge computing is:

[0090]

[0091] Since the base station is fixed and can be powered on, the energy consumption of the MEC server-assisted computing in the service node is not counted. Therefore, the total energy consumption of the offload edge computing of the task vehicle n is:

[0092]

[0093] where p n is the transmission power of mission vehicle n.

[0094] In order to make full use of system resources and meet the task requirements of low latency and low energy consumption, the user utility is defined as the trade-off between the delay and energy consumption generated by the task vehicle n. The utility function is expressed as:

[0095] C=λ t T n +λ e E n

[0096] Among them, λ t represents the preference factor of system delay, λ e is expressed as the preference factor of system energy consumption, and λ t +λ e =1,0≤λt ≤1,0≤λ e ≤1. When λ e ≤λ t The optimization problem is more concerned with optimizing latency than energy consumption, and vice versa. The optimization of different requirements for latency and energy consumption can be achieved by adjusting the preference factor, T n and E n are the total delay and total energy consumption of the task vehicle n in the process of completing the task, respectively:

[0097]

[0098] The utility function of the system is defined as the sum of the utilities of all mission vehicles in the system, so the system utility function is:

[0099]

[0100] Fourth embodiment

[0101] Based on the first embodiment, in this embodiment, in S3, the utility function problem of all task vehicles in the system can be modeled as:

[0102] P1:

[0103] st.C1:

[0104] C2:

[0105] C3:

[0106] C4:

[0107] C5:

[0108]

[0109] In the formula, C1 represents the size of the unloading ratio between 0 and 1; C2 represents the transmission power of the constrained task vehicle between 0 and P max C3 means that the computing resources allocated by service node j to task vehicle n must not be less than 0; C4 means that the sum of computing resources allocated to all task vehicles must be less than the computing resources of the service node; C5 means that the task offloading computing delay must be less than the smallest of the communication holding delay and the maximum tolerable delay.

[0110] Fifth embodiment

[0111] Based on the first embodiment, in this embodiment, in S4, the specific steps of solving the alternating iterative subproblem based on the simulated annealing algorithm are:

[0112] In view of the non-convex characteristics of the original nonlinear programming problem, the original problem can be decomposed into sub-problems of resource allocation and task offloading ratio optimization. In the case of f nj and p n This paper uses Lagrange multiplier method, quasi-convex optimization technique and bisection method to solve these two sub-problems.

[0113] When optimizing the calculation frequency, the problem P1 can be decomposed into nj The expression is:

[0114]

[0115] st.C1,C3,C4

[0116] This problem is a convex function and can be solved by using the Lagrangian function to find the optimal frequency for:

[0117]

[0118] When power optimization is performed, problem P1 can be decomposed into the following n The expression is:

[0119]

[0120] st.C1,C2

[0121] Since the objective function The second-order derivative of is not always positive, so the above equation is a non-convex optimization problem and can be solved using quasi-convex function technology.

[0122] S41, input initialization parameters, including initial maximum temperature, number of loop iterations and termination temperature, set initial maximum temperature T = T0, generally set to 1200, number of inner loop iterations to 50, temperature attenuation factor α = 0.9, and randomly give an initial solution at the starting point of the algorithm iteration. By using the Lagrange multiplier method and quasi-convex technology, we can obtain f nj , p n And calculate the objective function value The value of

[0123] S42, the internal loop is increased to 50 times, i.e., steps S4.3 to S4.6 are repeated;

[0124] S43, perturb the initial solution to generate a new solution that satisfies the formula Where e is the random perturbation, L is the step size related to the initial solution and the value range;

[0125] S44, a new solution is obtained n After that, update p n ,f nj Substitute into C5 to determine whether it is satisfied. If it is satisfied, it is a new solution. If it is not satisfied, repeat to step S4.3;

[0126] S45, calculate the difference of the objective function If ΔC>0, then accept ξ n as the new current solution, otherwise with probability Accept n As the current new solution;

[0127] S46, whether the inner loop condition is satisfied at the current temperature T, if not, continue from step S4.3, if satisfied, terminate the program and output the solution, which is the optimal solution;

[0128] S47, according to the cooling formula T * = αT to cool down, and then go to step S4.2 until the temperature value is 0.1, and output the global optimal solution C(ξ n ).

[0129] Sixth embodiment

[0130] Based on any of the above embodiments, in this embodiment, the simulation values ​​of the parameters are as follows: the number of base stations is 4; the number of vehicles is 32; the total system bandwidth is B=20MHz; the task data volume d n is 0.8MB; the number of cycles for task calculation is 0.6GHz; the maximum transmission power of the task vehicle is P max =42dBm; energy coefficient κ = 5×10 -27 ; The computing power of the MEC server and the local task vehicle are F j =15GHz and The speed of the vehicle v n The value is within 50-100 km / h and the vehicle is traveling at a constant speed; the average noise power σ 2 =-100dBm, time preference factor λ t and energy consumption preference factor λ e are 0.3 and 0.7 respectively; the channel fading model of the user vehicle is:

[0131] L(dB)=140.7+36.7log 10 d [km]

[0132] In this embodiment, the solution obtained by the above method is compared with the following three conventional solutions:

[0133] 1) Average resource allocation scheme: All computing resources of service nodes are evenly allocated, and the task offloading ratio is optimized based on the simulated annealing algorithm.

[0134] 2) Random offloading scheme: The resource allocation subproblem is optimized, while the task offloading is compared with random splitting.

[0135] 3) Local computing solution: All tasks of all mission vehicles are not offloaded and are all calculated locally.

[0136] like Figure 3 As shown in the figure, compared with the other three benchmark schemes, the proposed algorithm can more reasonably realize the utilization of global resources when the number of vehicles is the same, so the utility value of the system is the lowest.

[0137] like Figure 4 As shown in the figure, the proposed scheme has a faster convergence speed than the greedy unloading scheme, and the utility value of the system is lower, which shows the superiority of the proposed algorithm.

[0138] In addition, it can be understood by those skilled in the art that all or part of the processes in the method for implementing the above embodiment can be completed by instructing the relevant hardware through a computer program. The computer program includes program instructions, and the computer program can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in a computer system to implement the process steps of the embodiment of the above method.

[0139] Therefore, the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the various steps of the resource allocation and task offloading ratio optimization method based on vehicle mobile edge computing as described in the above embodiment.

[0140] The computer-readable storage medium may be a USB flash drive, a mobile hard disk, a read-only memory (ROM), a magnetic disk, or an optical disk, etc., which may store program codes.

[0141] It should be noted that since the storage medium provided in the embodiment of the present application is the storage medium used to implement the method of the embodiment of the present application, based on the method introduced in the embodiment of the present application, the person skilled in the art can understand the specific structure and deformation of the storage medium, so it is not repeated here. All storage media used in the method of the embodiment of the present application belong to the scope of protection of this application.

[0142] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0143] The present application is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, 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 generate 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 flowchart and / or block diagram. 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.

[0144] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate 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 A function specified in one or more boxes.

[0145] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0146] It should be noted that in the claims, any reference signs placed between brackets shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claim. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The present application may be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third etc. does not indicate any order. These words may be interpreted as names.

[0147] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0148] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for optimizing resource allocation and task offloading ratio based on vehicle mobile edge computing, characterized in that: The method comprises the following steps: S1, building a communication and task offloading computing system model between multiple base stations and task vehicles; S2, based on the communication and task offloading computing system model, under the premise of considering that the task vehicle maintains communication with the base station, determining the system utility function of the task vehicle, wherein the system utility function is characterized as a weighted sum of the completion delay of the task vehicle and the energy consumption of the equipment; S3, setting constraints with the goal of minimizing the system utility function and constructing a system utility function model; S4, based on the system utility function model, solving the optimal solution for the frequency allocation of the task vehicle, the optimal solution for the transmission power between the task vehicle and the base station, and the optimal solution for the task offloading ratio.

2. The method according to claim 1, characterized in that In S1, the steps of constructing the communication and task offloading computing system model include: S11, the mission vehicle is assembled To represent, the set of base stations is represented as Each mission vehicle There is only one task R to be processed n : Where, d n is the data size of the task, c n To calculate R n The number of CPU cycles required, The maximum tolerable constraint delay for completing the task; S12, assuming that the total system bandwidth is B, divide it into S sub-channels, the size of each sub-channel is W = B / S, and the task OT is calculated using the Shannon formula n The transfer rate unloaded to task vehicle n is: In the formula, p n is the transmission power between task vehicle n and service node j, h nj is the channel gain between task vehicle n and service node j∈J, σ 2 is the average noise power from the mission vehicle to the service node; S13, the service nodes are arranged in parallel. Assuming that the center coordinates of the first service node j are expressed as (r, r), the center coordinates of the remaining service nodes are expressed as (r, r+(j-1)2r). The task vehicle n is randomly distributed within the communication coverage of each service node, and its initial position coordinates are expressed as (x nj ,y nj ), and at a speed v n Travel at a constant speed, and define the right direction as the positive direction. The lateral distance between the task vehicle n from the initial position and the out-of-service node coverage range is defined as s n , its formula is expressed as: Among them, r represents the communication coverage radius of the service node and the driving direction of the vehicle; Calculate the communication delay when the vehicle leaves the coverage area of ​​the service node S14, to ensure that each service node can successfully perform offload calculations on vehicles within its communication range, the task must meet the following constraints: In the formula, represents the transmission delay of the mission vehicle unloading to the service node j through the uplink, Represents the computing delay consumed by the task vehicle at service node j.

3. The method according to claim 2, characterized in that In S2, the expression of the system utility function is: in, Where, T n is the total delay that the task vehicle n takes to complete the task, E n is the total energy consumption of task vehicle n in the process of completing the task, λ t represents the preference factor of system delay, λ e is expressed as the preference factor of system energy consumption, and λ t +λ e =1, is the computing power of the mission vehicle n, κ is the computing energy consumption coefficient related to the vehicle’s own chip, c n is the number of CPU cycles, p n is the transmission power between the mission vehicle n and the service node, d n is the data size of the task, ξ n is the uninstall ratio.

4. The method according to claim 3, characterized in that In S3, the expression of the system utility function model is: Where C1 represents the unloading ratio ξ n The value is between 0 and 1; C2 represents the transmission power p of the constrained task vehicle n From 0 to P max C3 represents the computing resources f allocated by service node j to task vehicle n nj Greater than or equal to 0; C4 represents the total computing power allocated to all mission vehicles Less than the computing resources F of the service node j ; C5 means that the task offloading calculation delay is less than the minimum value of the communication holding delay and the maximum tolerable delay.

5. The method according to claim 4, characterized in that The steps in S4 specifically include: S41, solving the optimal solution of the frequency allocation based on the Lagrange multiplier method; S42, solving the optimal solution of the transmission power based on a quasi-convex technology; S43, solving the optimal solution of the task offloading ratio based on a simulated annealing algorithm.

6. The method according to claim 5, characterized in that The S43 specifically includes: S431, initialize the initial maximum temperature to 1200, the number of loop iterations to 50, the initial temperature to 1200 and the termination temperature to 0.1, and randomly assign an initial unloading factor at the starting point of the algorithm iteration S432, perturb the initial solution to generate a new solution, that is, satisfy Where e is the random perturbation, L is the step size related to the initial solution and the value range; S433, obtain new unloading factor ξ n Then, the task vehicle transmission power is updated based on the Lagrange multiplier method, and the transmission power is solved according to the quasi-convex technology, and substituted into the C5 to determine whether it is satisfied. If it is satisfied, it is a new solution. If it is not satisfied, repeat to step S432; S434, calculate the difference of the objective function If ΔC>0, then accept ξ n as the new current solution, otherwise with probability Accept n As the current new solution; S435, loop and iterate steps S432-S434 50 times, determine whether the current temperature T reaches the termination temperature, if so, output the current new solution as the optimal solution; otherwise, return to step S42 to continue execution.

Images