Optimization method for edge computing unloading decision and resource allocation of Internet of Vehicles

Through improved genetic algorithms, optimize the task offload ratio, channel bandwidth and edge server resource allocation in edge computing in Internet of Vehicles, the problem of difficult to minimize the completion time of on-board tasks is solved, and more efficient task offload decisions and resource allocation are achieved, reducing local The risk of optimal solution.

CN119031388APending Publication Date: 2024-11-26CHANGCHUN UNIV OF SCI & TECH
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411163721.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-23
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

The prior art has failed to effectively optimize the synergistic effects of real-time task offload decisions, communication resources and computing resources in the edge computing of Internet of Vehicles, making it difficult to minimize the completion time of on-vehicle tasks.

Method used

Using improved genetic algorithms, by establishing a communication model and task offload model between vehicles and edge servers, the task offload ratio, channel bandwidth and allocation of edge server computing resources are optimized, and combined with adaptive crossover and mutation operators, the completion time of on-board tasks is reduced.

Benefits of technology

It effectively reduces the completion time of on-board tasks, improves the algorithm's ability to explore in the search space, avoids the risk of falling into local optimal solutions, and significantly improves the optimization effect of task unloading decisions and resource allocation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119031388A_ABST
    Figure CN119031388A_ABST
Patent Text Reader

Abstract

The invention relates to an optimization method for edge computing unloading decision and resource allocation of the Internet of Vehicles, and belongs to the technical field of edge computing of the Internet of Vehicles. According to the method, under the scene that the task generated by the vehicle can be partially uploaded to the edge server to be processed, a communication model of the vehicle and the edge server and a task unloading model of the vehicle are established, and the task unloading proportion, the channel bandwidth and the resources allocated by the edge server for processing the unloading task are combined, so that the total completion time of the system is minimized as the target; the method comprises the following steps: establishing a system total overhead optimization function model, solving the optimization function model by using an improved genetic algorithm under the condition that wireless transmission resources and computing resources are limited, and obtaining optimal solutions of a task unloading proportion, a channel bandwidth and resource allocation, thereby effectively reducing the vehicle-mounted task completion time and improving the system performance. And meanwhile, the method is excellent in the aspect of reserving the optimal solution set, so that the exploration capability of the algorithm in a wide search space is improved, and the risk that the algorithm falls into local optimum and cannot jump out is remarkably reduced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of Internet of Vehicles edge computing, and in particular relates to an optimization method for Internet of Vehicles edge computing offloading decision and resource allocation. Background Art

[0002] In recent years, with the growth of car ownership, the intelligence of cars has been continuously improved, such as autonomous driving, image-assisted navigation, and autonomous driving. However, such intelligent services often require high computing power, which puts a huge burden on the local computing resources on the vehicle. The capabilities required to process future vehicle data will easily exhaust the vehicle's on-board resources, and most vehicles cannot perform these tasks locally. In order to meet these challenges, Vehicular Edge Computing (VEC) solutions have emerged. By offloading tasks to nearby edge nodes, such as roadside units (RSUs) or base stations connected to VEC servers, edge computing services can be provided to vehicles. Therefore, it is a problem worth studying to combine limited computing resources with efficient resource allocation to facilitate offloading decisions for complex tasks.

[0003] At present, most of the research on edge computing in the Internet of Vehicles only optimizes the offloading decision, and few studies have studied the synergistic effects of real-time task offloading decision, communication resource allocation and computing resource allocation. Therefore, it is necessary to design an optimization method that combines offloading decision and resource allocation to solve the above problems. Summary of the invention

[0004] The present invention aims to achieve the optimal configuration of task offloading ratio, channel bandwidth and edge server computing resources through a joint optimization strategy based on an improved genetic algorithm. In the context of limited wireless transmission resources and edge computing server processing resources, the present invention is committed to solving the optimization problem of how to minimize the completion time of vehicle-mounted tasks. The improved genetic algorithm adopted improves the rationality of individual crossover mutation probability through improved adaptive mutation and crossover operators, which not only retains excellent individuals, but also enhances the exploration ability of low-fitness individuals, preventing population evolution from falling into a local optimal solution.

[0005] The technical solution adopted by the present invention is as follows:

[0006] A method for optimizing offloading decision and resource allocation of edge computing in Internet of Vehicles includes the following steps:

[0007] Step 1: Establish a communication model between the vehicle and the edge server and calculate the uplink task data rate of each task of the vehicle;

[0008] Step 2: Establish a task offloading model for the vehicle, offload the tasks to the base station according to the task offloading ratio, calculate the local processing time, uplink time and edge server processing time of each task according to the task offloading ratio, task size and uplink task data rate, and calculate the total processing time of the corresponding vehicle offloading task according to the uplink time and edge server processing time of the offloaded task;

[0009] Step 3: Combine the task offloading ratio, channel bandwidth and the resources allocated by the edge server to process the offloading tasks, and establish an edge server system total overhead optimization function model with the goal of minimizing the total system completion time, where the total system completion time is the maximum value of the total completion time of each task, and the total completion time of each task is the maximum value of the local processing time of the task and the total processing time of the corresponding vehicle offloading task;

[0010] Step 4: Use the improved genetic algorithm to solve the edge server system total overhead optimization function model. The solution process specifically includes the following steps:

[0011] Step 4.1: Perform binary matrix column encoding on the three variables of task offloading ratio, channel bandwidth, and resources allocated by the edge server to process offloaded tasks. Each matrix obtained represents a chromosome i, and a population set consisting of I chromosomes is initialized;

[0012] Step 4.2: Calculate the fitness value of each chromosome according to the fitness function;

[0013] Step 4.3: Determine whether the population has reached the genetic termination condition. If so, output the optimal solution; if not, randomly pair the chromosomes in the population according to the selection strategy to select the chromosomes that will produce the next generation;

[0014] Step 4.4: Perform a crossover operation on the two paired chromosomes according to the improved adaptive crossover probability. The calculation formula of the improved adaptive crossover probability is:

[0015]

[0016] in, is the maximum crossover probability set, is the minimum crossover probability set, f max is the chromosome with the highest fitness value in the current population, f′ is the maximum fitness value of the two chromosomes to be crossed, and f avg is the average fitness value of the current population;

[0017] Step 4.5: Perform mutation operation on the new chromosome obtained after the crossover operation according to the improved adaptive mutation probability. The calculation formula of the improved adaptive mutation probability is:

[0018]

[0019] in, is the maximum mutation probability set, is the set minimum mutation probability, and f is the fitness value of the chromosome to be mutated;

[0020] Step 4.6: After generating a new population, return to step 4.2.

[0021] The present invention takes the urban road traffic network as the application scenario and establishes an optimization method for joint task offloading decision, communication resources and edge server computing resources. The method adopts an improved genetic optimization algorithm to solve the problem of task offloading and resource allocation. Beneficial effects of the present invention:

[0022] Considering a scenario with a base station and multiple one-way straight vehicles, the tasks generated by the vehicles can be partially uploaded to the edge server for processing. In this scenario, the present invention provides an optimization method for offloading decision and resource allocation of edge computing in the Internet of Vehicles based on an improved genetic algorithm. The method proposes a communication model and a partial offloading model, and designs an optimization problem of minimizing task processing time. When wireless transmission resources and edge server computing resources are limited, the improved genetic algorithm is used to jointly solve the task offloading ratio, channel bandwidth, and allocation of edge server computing resources, effectively reducing the completion time of the on-board tasks; the improved genetic algorithm performs well in retaining the optimal solution set by optimizing the crossover and mutation process, which not only improves the algorithm's exploration ability in a wide search space, but also significantly reduces the risk of the algorithm falling into a local optimum and being unable to jump out. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is the scenario of offloading edge computing tasks in the Internet of Vehicles to which the present invention is applied;

[0024] Figure 2 A flowchart of the method for optimizing offloading decision and resource allocation of edge computing in Internet of Vehicles according to an embodiment of the present invention;

[0025] Figure 3 is a coding diagram;

[0026] Figure 4 It is a schematic diagram of the cross operation process;

[0027] Figure 5 Schematic diagram of the mutation operation process;

[0028] Figure 6 Flowchart of the improved genetic algorithm. DETAILED DESCRIPTION

[0029] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] The purpose of the present invention is to provide an optimization method for offloading decision and resource allocation of edge computing in Internet of Vehicles, which uses an improved genetic algorithm to solve the optimization problem of minimizing the user task completion time and obtain the optimal task offloading strategy and resource allocation solution.

[0031] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0032] The embodiment of the present invention provides an optimization method for offloading decision and resource allocation of edge computing in Internet of Vehicles. The method is applied to task offloading scenarios such as Figure 1 As shown, it includes a vehicle, an edge server (VEC server), wherein the edge node (RSU) connects multiple vehicles. Figure 1 In the ,end-user vehicle can choose to offload its tasks to the cloud server or the VEC server.

[0033] like Figure 2 As shown, the optimization method for offloading decision and resource allocation of edge computing in Internet of Vehicles based on the improved genetic algorithm in this embodiment mainly includes the following steps:

[0034] Step 1: Establish a communication model between the vehicle and the edge server and calculate the uplink task data rate of each task of the vehicle;

[0035] Step 2: Establish a task offloading model for the vehicle, offload the tasks to the base station according to the task offloading ratio, calculate the local processing time, uplink time and edge server processing time of each task according to the task offloading ratio, task size and uplink task data rate, and calculate the total processing time of the corresponding vehicle offloading task according to the uplink time and edge server processing time of the offloaded task;

[0036] Step 3: Combine the task offloading ratio, channel bandwidth and the resources allocated by the edge server to process the offloading tasks, and establish an edge server system total overhead optimization function model with the goal of minimizing the total system completion time, where the total system completion time is the maximum value of the total completion time of each task, and the total completion time of each task is the maximum value of the local processing time of the task and the total processing time of the corresponding vehicle offloading task;

[0037] Step 4: Use the improved genetic algorithm to solve the total overhead optimization function model of the edge server system to obtain the optimal solution for task offloading ratio, channel bandwidth and edge server computing resource allocation.

[0038] Specifically, in step 1, a communication model between the vehicle and the edge server is established. The key performance indicators of task processing time are task uplink time and task execution time. The task communication time from the vehicle to the server is affected by the communication model, while the task execution time refers to the time to process the task on the server, which is affected by the offloading model.

[0039] Assume that each car has a computing task A m ={f m , D m , L m}, where m∈(1, 2, 3...M), f m represents the local computing resources of the vehicle, D m Indicates the size of the task to be processed, M indicates the total number of vehicle users, and when describing the number of CPU cycles required to process a task with a size of 1 Byte, L is defined m is the time cost corresponding to the task, which is the number of CPU cycles consumed to process a single byte task. The vehicle uploads the unloading task to the edge server through frequency division multiple access, and the user vehicles will not cause co-frequency interference with each other during the calculation and unloading process. The total bandwidth of the edge server system is B max , assigned to task A m The bandwidth is B m Therefore, Task A m The uplink task data rate is:

[0040] R m =B m log2(1+SNR m )

[0041] Among them, SNR is task A m The signal-to-noise ratio of B is in the range of [5, 25] db in this embodiment; max is the total bandwidth of the edge server system, B m Assigned to Task A m bandwidth.

[0042] In step 2, a vehicle task offloading model is established. Task A m Unload to the base station according to the task offloading ratio, and the remaining ratio (1-P m ) is handled locally, so task A m Local processing time Uplink time Can be expressed as:

[0043]

[0044]

[0045] Assume that the computing resources of the edge server are F max , and task A m After uploading some tasks to the edge server, the edge server will assign F m The time it takes for edge servers to process offloaded tasks It can be expressed as:

[0046]

[0047] After the task processing is completed, the edge server sends the processed data back to the vehicle, usually ignoring the impact of the task return delay. Therefore, the total processing time of the vehicle unloading task is It can be expressed as:

[0048]

[0049] In step 3, the total cost optimization function model of the edge server system is established. Since the task offloading and local task processing are performed simultaneously, task A m The total completion time is expressed as:

[0050]

[0051] Then the total completion time T of the system can be expressed as:

[0052] T=Max(T1,T2,...T m ...T M )

[0053] The goal of the present invention is to optimize the task A m The task offloading ratio P m , channel bandwidth B m and the resources allocated by the edge server F m To minimize the total completion time of the system, in summary, the established edge server system delay optimization function, that is, the edge server system total overhead optimization function model is expressed as:

[0054]

[0055]

[0056]

[0057]

[0058] C4: 0 ≤ P m ≤1

[0059] Among them, F max Compute resources for edge server systems, B max is the total bandwidth of the edge server system. In the formula, C1 indicates that the processing time after the user unloads is less than the time for the task to be processed locally; C2 indicates that the total computing resources obtained by each vehicle user is limited to the total computing resources that the edge server can provide, which means that the allocation of resources is limited to ensure that it does not exceed the maximum computing capacity of the server; C3 indicates that the total bandwidth resources allocated to each user is less than the total bandwidth of the edge server system; C4 represents that the user has the right to choose not to unload, partially unload or completely unload the task. This choice is affected by multiple factors, including the current number of users, the computing resource capacity of the vehicle, the computing resource capacity of the edge server, and the available channel bandwidth. These factors work together to affect the user's unloading decision to optimize task processing and improve efficiency.

[0060] Genetic Algorithm, or GA for short, is an optimization technique that simulates the biological evolution process. It generates a set of candidate solutions (individuals) randomly, and continuously iterates to improve these solutions through operations such as selection, crossover, and mutation until the optimal solution to the problem is approached. In this process, individuals with high fitness are more likely to be selected to reproduce, while mutation ensures the diversity and evolutionary potential of the population. Genetic algorithms have been proven to be an effective tool for solving various complex and difficult optimization problems.

[0061] (1) Population initialization is the process of creating an initial set of candidate solutions for the algorithm. These candidate solutions are called individuals, and the entire set is called a population. The size of the population and the encoding method of the individuals have a significant impact on the algorithm's search ability. Population initialization usually follows the following steps:

[0062] Determine the population size: The size of the population directly affects the coverage of the search space. A larger population may provide a wider search range, but the computational cost is also higher.

[0063] Randomly generate individuals: According to the scale and complexity of the problem, a certain number of individuals are randomly generated. Each individual represents a potential solution to the problem.

[0064] Ensure diversity: During the initialization phase, emphasis should be placed on maintaining the diversity of individuals within the population, in order to prevent the algorithm from falling into a local optimal solution too early, thereby ensuring a wide coverage of the search space.

[0065] (2) Encoding is a key step in converting the solution to a form that can be operated by the genetic algorithm. In genetic algorithms, encoding specifically refers to converting the representation of the solution into a chromosome structure. This string of chromosomes is the representation of the individual in the algorithm. The choice of the specific encoding method should be determined based on the unique attributes of the problem.

[0066] The process of solving the edge server system total cost optimization function model using the improved genetic algorithm specifically includes the following steps:

[0067] Step 4.1: Offload the task ratio P m , channel bandwidth B m and the resources F allocated by the edge server to handle the offloaded tasks m The three variables are encoded by binary matrix columns, and each matrix obtained represents a chromosome i, and the population set consisting of I chromosomes is initialized. This step designs the initialization population and encoding, and the specific implementation is as follows:

[0068] Assume there are M on-board tasks, each on-board task A m Both variables need to be optimized m 、F m and B m , and these variables are all decimal variables, so we need to change P m 、F m and B m Binary matrix column encoding is performed. Therefore, a matrix K with 15M rows and 3M columns i It can be represented as a chromosome i, i∈{1, 2, …, I}, and the encoding method is shown in Figure 3 .

[0069]

[0070] Among them, the accuracy when encoding 15M rows is:

[0071]

[0072] Among them, U max and U min Represent three variables P respectively m , F m , B m The maximum and minimum values ​​of .

[0073] The initial population consists of I chromosomes

[0074] Step 4.2: Design a fitness function and calculate the fitness value of each chromosome based on the fitness function.

[0075] Considering that the goal of genetic algorithm optimization is to minimize the optimization function, the inverse of the total completion time T of the system in step 3 is used as the fitness function to evaluate the quality of the individual. The formula of the fitness function is as follows:

[0076]

[0077] Among them, fitness is the fitness function.

[0078] Step 4.3: Determine whether the population has reached the genetic termination condition. If so, output the optimal solution; if not, randomly pair the chromosomes in the population according to the selection strategy to select the chromosomes that will produce the next generation.

[0079] In genetic algorithms, individual selection operations involve identifying and retaining high-quality individuals from the current population, which will serve as parents to give birth to offspring. Generally, individuals with high fitness values ​​have a higher probability of being selected to ensure that their genes can continue in the next round. A typical means of implementing this strategy is to use the roulette method. This method first evaluates the fitness value f(i) of each individual, and then determines the respective selection probability P(i) based on these values. Finally, the individual is selected based on the size of the random number generated from 0 to 1 and the selection probability P(i). The calculation formulas for each individual's fitness value f(i) and selection probability P(i) are as follows:

[0080]

[0081]

[0082] Among them, T i is the total system delay of individual i, f(i) and f(j) are the fitness values ​​of individuals i and j respectively, and I is the total number of individuals in the population.

[0083] The genetic termination condition in this step can be that the difference between the highest fitness value in the population and the average fitness value of the population is less than the set limit value (for example, 0.001), or the algorithm has completed the preset maximum number of iterations, and the genetic algorithm will stop running. This setting is intended to ensure that the algorithm stops when it reaches the desired fitness level or reaches the maximum iteration limit, thereby effectively utilizing computing resources.

[0084] Step 4.4: Perform a crossover operation on the two paired chromosomes according to the improved adaptive crossover probability.

[0085] Step 4.5: Perform mutation operation on the new chromosome obtained after the crossover operation according to the improved adaptive mutation probability.

[0086] In the implementation of the genetic algorithm, a single-point crossover strategy is used to reorganize the chromosome sequence. The specific steps are as follows: a crossover position is randomly determined in the two paired chromosomes A and B, denoted as e. Next, a probability random number for the crossover position e is generated, and the probability random number is compared with the crossover probability P c If the probability random number is not greater than the crossover probability P c , then at the crossover point e, the corresponding parts of chromosome A and chromosome B will be exchanged to generate two new chromosomes. The crossover operation is shown in Figure 4 .

[0087] The mutation process involves flipping a specific gene in the chromosome from 0 to 1, which is based on a random number and the mutation probability P. m The comparison is determined by taking chromosome A as an example. For mutation operation, see Figure 5 .

[0088] The traditional population crossover probability P c and mutation probability P m The calculation formula is as follows:

[0089]

[0090]

[0091] Among them, k1, k2, k3, k4 are constants between [0, 1], f max is the highest fitness value of individuals in the current population, f′ is the maximum fitness value of the two individuals to be crossed, and f avg is the average fitness value of the current population, and f is the fitness value of the individual to be mutated.

[0092] The present invention improves the crossover probability and mutation probability of the population. The improved mutation operator improves the rationality of the individual crossover mutation probability, which not only improves the algorithm's exploration ability in a wide search space, but also significantly reduces the risk of the algorithm falling into a local optimum and being unable to jump out. The calculation formulas for the improved adaptive crossover probability and mutation probability are as follows:

[0093]

[0094]

[0095] in, They represent the maximum crossover probability, minimum crossover probability, maximum mutation probability and minimum mutation probability respectively; f max is the chromosome with the highest fitness value in the current population; f avgis the average fitness value of the current population; f′ is the maximum fitness value of the two chromosomes to be crossed; f is the fitness value of the chromosome to be mutated.

[0096] Step 4.6: Generate a new population, then return to step 4.2 and recalculate the fitness value of each chromosome in the new population until the optimal solution for task offloading ratio, channel bandwidth, and edge server computing resource allocation is obtained.

[0097] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0098] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for optimizing offloading decision and resource allocation of edge computing in Internet of Vehicles, characterized in that: The following steps are involved: Step 1: Establish a communication model between the vehicle and the edge server and calculate the uplink task data rate of each task of the vehicle; Step 2: Establish a task offloading model for the vehicle, offload the tasks to the base station according to the task offloading ratio, calculate the local processing time, uplink time and edge server processing time of each task according to the task offloading ratio, task size and uplink task data rate, and calculate the total processing time of the corresponding vehicle offloading task according to the uplink time and edge server processing time of the offloaded task; Step 3: Combine the task offloading ratio, channel bandwidth and the resources allocated by the edge server to process the offloading tasks, and establish an edge server system total overhead optimization function model with the goal of minimizing the total system completion time, where the total system completion time is the maximum value of the total completion time of each task, and the total completion time of each task is the maximum value of the local processing time of the task and the total processing time of the corresponding vehicle offloading task; Step 4: Use the improved genetic algorithm to solve the edge server system total overhead optimization function model. The solution process specifically includes the following steps: Step 4.1: Perform binary matrix column encoding on the three variables of task offloading ratio, channel bandwidth, and resources allocated by the edge server to process offloaded tasks. Each matrix obtained represents a chromosome i, and a population set consisting of I chromosomes is initialized; Step 4.2: Calculate the fitness value of each chromosome according to the fitness function; Step 4.3: Determine whether the population has reached the genetic termination condition. If so, output the optimal solution; if not, randomly pair the chromosomes in the population according to the selection strategy to select the chromosomes that will produce the next generation; Step 4.4: Perform a crossover operation on the two paired chromosomes according to the improved adaptive crossover probability. The calculation formula of the improved adaptive crossover probability is: in, is the maximum crossover probability set, is the minimum crossover probability set, f max is the chromosome with the highest fitness value in the current population, f′ is the maximum fitness value of the two chromosomes to be crossed, and f avg is the average fitness value of the current population; Step 4.5: Perform mutation operation on the new chromosome obtained after the crossover operation according to the improved adaptive mutation probability. The calculation formula of the improved adaptive mutation probability is: in, is the maximum mutation probability set, is the set minimum mutation probability, and f is the fitness value of the chromosome to be mutated; Step 4.6: After generating a new population, return to step 4.

2.

2. The optimization method for offloading decision and resource allocation of edge computing in Internet of Vehicles according to claim 1 is characterized in that: In step 1, assume that each vehicle has a computation task A m ={f m , D m , L m }, where m∈(1, 2, 3...M), f m represents the local computing resources of the vehicle, D m Indicates the size of the task to be processed, M indicates the total number of vehicle users, and when describing the number of CPU cycles required to process a task with a size of 1 Byte, L is defined m is the time cost corresponding to the task, which is the number of CPU cycles consumed to process a single byte task; the vehicle uploads the unloading task to the edge server through frequency division multiple access, task A m The uplink task data rate is: R m =B m log2(1+SNR m ) Among them, SNR is task A m The signal-to-noise ratio, B max is the total bandwidth of the edge server system, B m Assigned to Task A m bandwidth.

3. The optimization method for offloading decision and resource allocation of edge computing in Internet of Vehicles according to claim 2 is characterized in that: The value range of the signal-to-noise ratio (SNR) is [5, 25] db.

4. The optimization method for offloading decision and resource allocation of edge computing in Internet of Vehicles according to claim 2 or 3, characterized in that: In step 2, Task A m By task offloading ratio P m Unload to the base station, the remaining ratio (1-P m ) is left to be processed locally, and the calculation results in task A m Uplink time for: Assume that the computing resources of the edge server are F max , and task A m After uploading some tasks to the edge server, the edge server will assign F m computing resources to process these tasks, and the time it takes for edge servers to process offloaded tasks for: Total processing time for vehicle unloading tasks for:

5. The optimization method for offloading decision and resource allocation of edge computing in Internet of Vehicles according to claim 2 or 3, characterized in that: The formula of the total cost optimization function model of the edge server system is established as follows: C1: C2: C3: C4: 0≤P m ≤1 Among them, T m For Task A m The total completion time of the system, T is the total completion time of the system, F max Compute resources for edge server systems, B max is the total bandwidth of the edge server system.

6. The optimization method for offloading decision and resource allocation of edge computing in Internet of Vehicles according to claim 5 is characterized in that: The formula of the fitness function is: Among them, fitness is the fitness function.

7. The optimization method for offloading decision and resource allocation of edge computing in Internet of Vehicles according to claim 2 or 3, characterized in that: The size of the matrix obtained after encoding in step 4.1 is 15M×3M.

8. The optimization method for offloading decision and resource allocation of edge computing in Internet of Vehicles according to claim 2 or 3, characterized in that: The selection strategy in step 4.3 adopts the roulette algorithm.

9. The optimization method for offloading decision and resource allocation of edge computing in Internet of Vehicles according to claim 1 or 2, characterized in that: In step 4.4, the chromosome sequence is recombined using a single-point crossover strategy. The specific steps include: A crossover position is randomly determined in the two paired chromosomes, denoted as e; Generate a random number with the probability of the crossover position e; The probability random number of the crossover position e is combined with the improved adaptive crossover probability P c Compare, if the probability random number is less than or equal to the improved adaptive crossover probability P c , then at the crossover position e, the corresponding parts of the two chromosomes are exchanged to generate two new chromosomes.

10. The optimization method for offloading decision and resource allocation of edge computing in Internet of Vehicles according to claim 2 or 3, characterized in that: The genetic termination condition is that the difference between the highest fitness value in the population and the average fitness value of the population is less than the set limit value, or the genetic termination condition is that the current number of iterations of the algorithm reaches the preset maximum number of iterations.

Citation Information

Cited By

  • Cooperative perception and calculation unloading joint optimization method in Internet of Vehicles

    CN120769236A