A Computation Offloading Method Based on Task Transmission Control and Improved Greedy Strategy

Abstract

The present invention provides a computing offloading method based on task transmission control and improved greedy strategy, which relates to the technical field of computing offloading, and includes: constructing a system model composed of a communication model and a task offloading model; obtaining the ratio of transmission and execution time of adjacent offloading tasks, determining the data volume of the offloading task executed by the optimal edge server, and generating a sorting sequence; according to the sorting sequence and the data volume, allocating the offloading task to the roadside unit or the service vehicle, and combining the distance between the task vehicle and the edge server to determine the optimal edge server and the task allocation result; the base station sends a resource allocation message to the proximal roadside unit, and then it forwards the message to the task vehicle, and the task vehicle sends the complete task information to the allocated roadside unit or service vehicle after delay waiting, and returns the result after the offloading task is completed. Through the task transmission control mechanism, the present invention alleviates the communication load of the edge server and improves the computing offloading and resource allocation efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of resource allocation and computing offloading, and particularly to a computing offloading method based on task transmission control and improved greedy strategy. Background Art

[0002] With the development of scientific and technological fields such as wireless communication technology and artificial intelligence, in-vehicle applications have become increasingly intelligent and diverse, covering various aspects such as road safety, entertainment services, traffic control, intensive computing services, etc. These applications also pose challenges to the limited computing resources and energy of vehicles. However, with the increasing demand for resources and the higher requirements for latency in in-vehicle applications, due to the low coverage rate of cloud computing service architectures, long data transmission distances, high response delays, unstable communication quality, and the transmission of excessive data will also increase the burden on the cloud core network, in-vehicle cloud computing faces new challenges. To solve the problems existing in the above cloud service technologies, mobile edge computing has become a promising technology by transferring cloud computing resources to the edge close to mobile devices.

[0003] The emergence of in-vehicle edge computing has created conditions for the implementation of more in-vehicle applications, but also faces new challenges. In the scenario of in-vehicle edge computing, vehicles are mobile, and it is necessary to ensure the stability of the communication link between the edge server and the vehicle and the efficiency of data transmission; the resources in in-vehicle edge computing are diverse but also limited, including not only roadside units but also other vehicles with idle resources. To meet the latency requirements of vehicle tasks in the edge environment, reasonable resource allocation and computing offloading methods are required, and the workload of the edge server also needs to be considered; during peak traffic periods or congested hours, task vehicles will generate a large number of computing and communication tasks, and most solutions achieve computing task offloading through vehicle-to-infrastructure (V2I) communication, which may lead to communication overload of the edge server and low computing offloading and resource allocation efficiency.

[0004] Therefore, it is urgent to provide a solution to improve the above problems. Summary of the Invention

[0005] The purpose of the present invention is to provide a computing offloading method based on task transmission control and improved greedy strategy to improve the problems in the above background art.

[0006] A computing offloading method based on task transmission control and improved greedy strategy provided by the present invention adopts the following technical solutions;

[0007] Construct a system model among multiple task vehicles, multiple roadside units, and multiple service vehicles, and construct multiple optional edge servers based on the roadside units and service vehicles. The system model includes a communication model and a task offloading model;

[0008] Obtain the time ratio of the transmission and execution of adjacent offloading tasks, and based on the time ratio, obtain the data volume ratio of a single optimal edge server to execute the offloading tasks, so as to obtain a sorting sequence;

[0009] Allocate the offloading tasks to the roadside unit or the service vehicle based on the order and data volume size of the sorting sequence, respectively obtain the distances between the task vehicle and the edge server corresponding to the start and completion of the offloading task based on the coordinates of the task vehicle, and obtain the optimal edge server corresponding to all tasks, that is, the task allocation result;

[0010] Based on the task allocation result, the base station sends a resource allocation message back to the proximal roadside unit, and the proximal roadside unit then sends the resource allocation message to the task vehicle. After a delay waiting time, the task vehicle sends the complete task information to the allocated roadside unit or service vehicle. When the optimal edge server finishes executing the offloading task, it sends the offloading result back to the task vehicle.

[0011] The beneficial effect of a computing offloading method based on task transmission control and improved greedy strategy provided by the present invention lies in that, based on the task transmission control mechanism, the present invention can effectively relieve the communication load of the edge server and improve the computing offloading and resource allocation efficiency.

[0012] Optionally, when constructing the system model between multiple task vehicles, multiple roadside units, and multiple service vehicles, the specific content inside the system model includes: a base station, a roadside unit, a task vehicle, and a service vehicle. Among them, the task vehicle and the service vehicle travel in a straight line at a fixed speed during the computing offloading. The task vehicle is used to send a task offloading request to the roadside unit, and after transmitting the complete task information to the roadside unit, it receives the result returned by the roadside unit after completing the task. The service vehicle is used to receive the complete task information sent by the task vehicle and send the task execution result to the task vehicle. The base station is used to receive the task offloading request sent by the roadside unit and return a resource allocation message to the roadside unit. The roadside unit is used to receive the task offloading request sent by the task vehicle and return the resource allocation message to the task vehicle.

[0013] Optionally, the mathematical expression of the communication model is:

[0014] ;

[0015] ;

[0016] Among them, represents the bandwidth of the transmission channel between the task vehicle and the edge server , represents the transmission power of the task vehicle , represents the channel gain, Indicates the ambient noise when the service provider acts as the receiver, for the task vehicle and the edge server The Euclidean distance between them, Indicates the path loss component, Indicates the task vehicle and the edge server The uplink data rate of the communication between them;

[0017] The mathematical expression of the task offloading model is:

[0018] ;

[0019] ;

[0020] Where, Indicates the amount of computation of the task of the task vehicle The size of the amount of data of the computing task of the task vehicle Indicates the task vehicle The size of the amount of data of the computing task, Indicates the computing intensity of the task, Indicates the transmission time for the task vehicle to send the offloading task to the edge server.

[0021] Optionally, the process of respectively obtaining the distances corresponding to the task vehicle and the edge server at the start and completion of the offloading task based on the coordinates of the task vehicle includes:

[0022] ;

[0023] ;

[0024] ;

[0025] Where, Indicates the task vehicle The distance from the edge server The distance, Indicates the task vehicle The distance from the roadside unit The distance, Indicates the task vehicle The distance from the service vehicle The distance, 、 Are respectively the coordinates of the task vehicle Relative to the edge server When sending the offloading task, 、 Are respectively the task vehicle in the corresponding coordinate axis direction The driving sub-speed and are respectively expressed as the coordinates of the task vehicle relative to the roadside unit when sending the unloading task. represents the total completion time of the task vehicle executing the unloading task at the roadside unit . represents the total completion time of the task vehicle executing the unloading task at the service vehicle . and are respectively the coordinates of the task vehicle relative to the service vehicle when sending the unloading task. and are respectively the speeds of the service vehicle in the corresponding coordinate axis directions.

[0026] The process of screening the optimal edge server based on the distance and the communication range radius in the optional edge servers includes:

[0027] ;

[0028] ;

[0029] ;

[0030] where , , are respectively the communication range radii of the roadside unit , the service vehicle , and the edge server .

[0031] Optionally, the mathematical expression for the sum of the total completion times of the tasks of the edge server is:

[0032] ;

[0033] where represents the sum of the total completion times of the tasks, represents the number of tasks that the edge server needs to execute, , , respectively represent the execution times of the 1st, 2nd,..., th tasks of the edge server in the execution order. When , Obtain the minimum value. The edge server executes the assigned offloading tasks in ascending order of task execution time to obtain the minimum value of the sum of the total task completion times.

[0034] Optionally, the process of obtaining the data volume ratio of a single optimal edge server to execute offloading tasks includes: based on the uplink data rate of communication and the data volume size of the computing tasks of the task vehicle to obtain the transmission time for the task vehicle to send the offloading task to the edge server, obtain the task offloading time based on the optimal computing volume size and the CPU cycle frequency of the edge server, and obtain the task data volume ratio based on the task transmission time and the task offloading time, where

[0035] The mathematical expression for the transmission time for the task vehicle to send the offloading task to the edge server is:

[0036] ;

[0037] The task offloading time The mathematical expression is:

[0038] ;

[0039] The task data volume ratio The mathematical expression is:

[0040] ;

[0041] Among them, represents the transmission time for the task vehicle to send the offloading task to the edge server, represents the data volume size of the computing tasks of the task vehicle ; represents the task vehicle and the edge server The uplink data rate of communication between them, represents the bandwidth of the transmission channel, represents the task vehicle The transmission power of; represents the channel gain, represents the Gaussian white noise power, represents the task vehicle The data volume size of the previous executed task, represents the edge server The CPU cycle frequency of; represents the computing intensity of the task, represents the task vehicle The data volume size of the previous executed task of the task, Indicates the ratio of the current vehicle task data volume to the previous vehicle task data volume.

[0042] Optionally, the process of the optimal edge server executing the offloading task includes:

[0043] When the offloading task is assigned to the th and the th edge servers, respectively obtain the set of queuing waiting times after the task is offloaded and the set , and obtain the set and the set after removing the equal elements of the set ;

[0044] Based on the set , assign the current optimal edge server to the user to execute the offloading task, and obtain the minimum value of the maximum working time of all optional edge servers of the task vehicle. Among them, when , the improved greedy strategy is expressed as:

[0045] ;

[0046] ;

[0047] Among them, represents the queuing waiting time of the task of the task vehicle at the corresponding edge server, represents the number of edge servers that the task vehicle can select, represents the total completion time of the task vehicle assigned to the edge server to execute the offloading task, represents the th edge server, represents the th edge server.

[0048] Optionally, when the task vehicle executes the offloading task to the optimal edge server, the mathematical expression of the total completion time of the task vehicle is:

[0049] ;

[0050] Among them, represents the set of roadside units, represents the set of service vehicles, represents the total completion time of the task vehicle , represents the task vehicle at the roadside unit The total completion time of the unloading task executed at denotes the task vehicle at the service vehicle The total completion time of the unloading task executed at denotes the local total time. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 denotes a flowchart of a computing offloading method based on task transmission control and improved greedy strategy provided by the present invention;

[0052] Figure 2 denotes a schematic diagram of resource allocation and task offloading process of a computing offloading method based on task transmission control and improved greedy strategy provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Apparently, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention. Unless otherwise defined, the technical terms or scientific terms used herein shall have the ordinary meanings as understood by those of ordinary skill in the art to which the present invention pertains. The terms such as "including" used in the present invention mean that the elements or items appearing before the term cover the elements or items listed after the term and their equivalents, without excluding other elements or items.

[0054] In some embodiments, referring to Figure 1 , which denotes a flowchart of a computing offloading method based on task transmission control and improved greedy strategy provided by the present invention, includes the following steps:

[0055] S1. Construct a system model among multiple task vehicles, multiple roadside units and multiple service vehicles, and construct multiple optional edge servers based on the roadside units and service vehicles. The system model includes a communication model and a task offloading model;

[0056] S2. Obtain the time ratio of the transmission and execution of adjacent offloading tasks, and obtain the data volume ratio of a single optimal edge server to execute the offloading task based on the time ratio to obtain a sorting sequence;

[0057] S3. Allocate the offloading tasks to the roadside units or service vehicles based on the order and data volume size of the sorting sequence, obtain the distances between the task vehicles and the edge servers corresponding to the start and completion of the offloading tasks respectively based on the coordinates of the task vehicles, and obtain the optimal edge server corresponding to all tasks, that is, the task allocation result;

[0058] S4. Based on the task allocation result, the base station sends a resource allocation message back to the proximal roadside unit, and the proximal roadside unit then sends the resource allocation message to the task vehicle. After a delay waiting time, the task vehicle sends the complete task information to the allocated roadside unit or service vehicle. When the optimal edge server finishes executing the offloading task, it sends the offloading result back to the task vehicle.

[0059] In some embodiments, when performing step S1, when constructing the system model among multiple task vehicles, multiple roadside units, and multiple service vehicles, the internal structure of the system model specifically includes: a base station, roadside units, task vehicles, and service vehicles. Among them, task vehicles and service vehicles travel in a straight line at a fixed speed during the computing offloading period. Task vehicles are used to send task offloading requests to roadside units, and after transmitting the complete task information to the roadside units, they receive the results returned by the roadside units after completing the tasks. Service vehicles are used to receive the complete task information sent by task vehicles and send the task execution results to task vehicles. The base station is used to receive the task offloading requests sent by roadside units and return resource allocation messages to roadside units. Roadside units are used to receive the task offloading requests sent by task vehicles and return resource allocation messages to task vehicles.

[0060] Specifically, the base station is responsible for calculating offloading resource allocation. Edge servers are deployed in roadside units. Task vehicles can generate computing tasks. Service vehicles, as mobile edge servers, have idle computing resources and can provide services to other users. Limited by the limited computing, storage performance, and energy consumption of a single vehicle, vehicles need to offload tasks to edge servers through computing offloading to meet the requirements of computing-intensive and latency-sensitive tasks. Roadside units, as fixed edge servers, and service vehicles, as mobile edge servers, both serve as service providers to provide task offloading services for task vehicles at the same time. In the scenario considered by this invention, a single task vehicle has at most one computing task and the computing tasks are independently offloaded, and the service provider executes each offloading task in sequence.

[0061] Further, the task vehicles in the model are represented as a set , where is the number of task vehicles. The roadside units are represented as a set , where is the number of roadside units. The service vehicles are represented as a set , where is the number of service vehicles. The roadside units equipped with fixed edge servers and the service vehicles as mobile edge servers together constitute the edge server set , and the total number of edge servers is .

[0062] Refer toFigure 2 , the task vehicle sends a task offloading request message containing the basic task information to the nearest roadside unit. The roadside unit receives the task offloading request and sends it to the base station responsible for resource allocation within the corresponding area. The base station is generally located in the central area of the edge environment. The roadside unit and the vehicle are often far from the central cloud, but the distance between the base station and the roadside unit, between the roadside unit and the vehicle, and between the task vehicle and the service vehicle is often very close, the propagation delay is very small, and the small data volume of the message also determines that its transmission delay is very short. Therefore, the delay of message transmission in the vehicle-mounted edge computing scenario is very small.

[0063] Furthermore, the base station collects the basic information of all tasks and the working status information of the roadside unit and the service vehicle, determines the execution server for each user task, and then returns the allocation message to the user's proximal roadside unit. The user's proximal roadside unit then returns the corresponding resource allocation message to the task vehicle. After receiving the message, the task vehicle sends the complete task information to the corresponding task execution server. Finally, the roadside unit or the service vehicle receives the offloading task and returns the result to the task vehicle after completion.

[0064] Furthermore, when performing step S1, the mathematical expression of the communication model is:

[0065] ;

[0066] ;

[0067] Among them, represents the bandwidth of the transmission channel between the task vehicle and the edge server , represents the transmission power of the task vehicle , represents the channel gain, represents the ambient noise when the service side is the receiving end, is the Euclidean distance between the task vehicle and the edge server , represents the path loss component, represents the task vehicle and the edge server the uplink data rate of the communication between them.

[0068] The mathematical expression of the task offloading model is:

[0069] ;

[0070] ;

[0071] Among them, Represents the computing workload of the task vehicle of the task, Represents the data volume of the computing task of the task vehicle of the task, Represents the computing intensity of the task, Represents the transmission time for the task vehicle to send the offloading task to the edge server.

[0072] Furthermore, during the period from when the task vehicle sends the offloading task to when the edge server returns the execution result, the moving speed of the task vehicle remains basically constant. Taking the position of the edge server at the start of sending the offloading task as the coordinate origin, with the due east and due north being the positive directions of the X and Y axes respectively. The process of obtaining the coordinates of the task vehicle from the latitude and longitude information of the edge server and the task vehicle, and respectively obtaining the distances corresponding to the task vehicle and the edge server at the start and completion of the offloading task based on the coordinates of the task vehicle includes:

[0073] ;

[0074] ;

[0075] ;

[0076] Among them, Represents the task vehicle distance from the edge server of the distance, Represents the task vehicle distance from the roadside unit of the distance, Represents the task vehicle distance from the service vehicle of the distance, , Are respectively the coordinates of the task vehicle relative to the edge server at the time of sending the offloading task, , Are respectively the driving sub-speeds of the task vehicle in the corresponding coordinate axis directions, , Are respectively represented as the coordinates of the task vehicle relative to the roadside unit at the time of sending the offloading task, Represents the total completion time of the task vehicle executing the offloading task at the roadside unit , Represents the total completion time of the task vehicle executing the offloading task at the service vehicle at the time of sending the offloading task, , are the coordinates of the task vehicle relative to the service vehicle when sending the unloading task, , are respectively the speeds of the service vehicle in the directions of the corresponding coordinate axes;

[0077] The process of screening the optimal edge server from the optional edge servers based on the distance and the communication range radius includes:

[0078] ;

[0079] ;

[0080] ;

[0081] Among them, , , are respectively the communication range radii of the roadside unit , the service vehicle , and the edge server .

[0082] For the task vehicle and its optional edge servers that meet the above conditions, the base station selects one of the optional edge servers with a certain strategy to execute the unloading task of the task vehicle.

[0083] Actually, in the vehicle-mounted edge computing scenario, to allocate resources in the edge environment for the task vehicle, it is necessary to screen out the edge servers available for the user according to the communication range of the edge server. First, for a certain task vehicle, the position when the task vehicle starts to send the unloading task must be within the communication range of the optional edge server; second, since the time from when the task vehicle sends the task unloading request to when it receives the calculation result is not long, simply considering the mobility of the vehicle, before the task vehicle receives the execution result, it cannot drive out of the communication range of the edge server that executes the unloading task, because this will affect the transmission of the execution result.

[0084] Furthermore, analyzing the rationality of step S3, assuming that under the result of a certain resource allocation and computing offloading method, the edge server needs to execute tasks, and the corresponding task vehicles are respectively represented as in the execution order.

[0085] Furthermore, to allocate to the edge server Performing these tasks is optimal for the entire edge environment with the goal of reducing the average user latency. In the case where there is no connection between edge servers in terms of offloading tasks, The sum of the total completion time of all tasks on The minimum value should be taken. The mathematical expression of the sum of the total task completion time of the edge server is:

[0086] ;

[0087] in, represents the total completion time of the task, Represents an edge server The number of tasks that need to be executed, 、 、 Represents edge servers In the order of execution, 1, 2, ..., The execution time of a task, when hour, To obtain the minimum value, the edge server executes the assigned offloading tasks in the order of task execution time from small to large, and obtains the minimum value of the sum of the total task completion time.

[0088] Further, according to ,Since the transmission time of computing tasks is shorter than the execution time of computing tasks and there is an order of magnitude difference, the edge server can be approximated The sum of the total completion time of all uninstallation tasks as follows:

[0089] ;

[0090] in, represents the total completion time of the task, Represents an edge server The number of tasks that need to be executed, 、 、 Represents edge servers In order of execution, 1, 2, ..., The execution time of a task, when hour, To obtain the minimum value, the edge server executes the assigned offloading tasks in the order of task execution time from small to large, and obtains the minimum value of the sum of the total task completion time.

[0091] In fact, the above analysis is for a single edge server. To extend the magnitude relationship of the execution time of tasks to all roadside units and service vehicles in the system, that is, it is necessary to consider the magnitude relationship of the computational workload of all tasks, which can be expressed as follows:

[0092] ;

[0093] Among them, 、 、 represent the computational workload magnitudes of the 1st, 2nd, …, th tasks in the execution order of the edge server , represents the edge server, represents the set of edge servers.

[0094] It can be seen from the above formula that at the beginning stage of executing the resource allocation and computing offloading algorithm, the base station should sort all the offloading tasks of task vehicles in ascending order according to the computational workload , and give priority to allocating tasks with smaller computational workloads. In this way, when the offloading tasks are allocated to different roadside units or service vehicles, it can be ensured that for any edge server, the computational workload magnitudes of the allocated offloading tasks are increasing.

[0095] Furthermore, based on the greedy strategy, the minimum value in the set of the maximum current queuing waiting times of the optional edge servers can be obtained. The specific steps are as follows:

[0096] When the offloading task is allocated to the th and the th edge servers, respectively obtain the queuing waiting time sets and the set after the offloading of the task, and obtain the set after removing the equal elements of the set and the set ;

[0097] Based on the set , allocate the current optimal edge server for the user to execute the offloading task, and obtain the minimum value of the maximum working time of all the optional edge servers of the task vehicle.

[0098] Specifically, if there are optional edge servers that the task vehicle can choose, which are respectively , if the offloading tasks of the task vehicle are respectively allocated to these edge servers for execution, their total completion times are respectively . And before the offloading of this task, the task vehicle The queuing waiting times of the tasks at the corresponding edge servers are respectively: .

[0099] If under the current computing offloading method, the offloading task is assigned to the edge server , while under other possible methods, the offloading task is assigned to any optional edge server except , that is and and .

[0100] Furthermore, considering the case of serial number , when the offloading task is assigned to the edge server , the queuing waiting time of the task vehicle after task offloading at the optional edge server is represented as a set : , while in the case of task assignment to it is the set . After removing the elements with equal values from the two sets, it is the set : and .

[0101] Specifically, if the greedy strategy is adopted, that is, the current optimal edge server is assigned to the user to execute the offloading task, then there is:

[0102] ;

[0103] Furthermore, combining the relationship , it can be obtained that , , that is, it can be ensured that after the task of the task vehicle is offloaded, compared with other possible resource allocation methods, under the same allocation order, the greedy strategy can minimize the maximum value of the working time of all optional edge servers of the task vehicle .

[0104] Actually, when , the improved greedy strategy is expressed as:

[0105] ;

[0106] ;

[0107] Among them, represents the queuing waiting time of the task of the task vehicle at the corresponding edge server, represents the number of edge servers that the task vehicle can choose, represents the total completion time for the task vehicle assigned to the edge server to execute the offloading task. represents the th edge server. represents the th edge server.

[0108] The above is the case of serial number When the situation is essentially the same. Therefore, before and after a single task offloading, adopting a greedy strategy can simultaneously ensure that the maximum value of the current user task completion delay and the working time of all edge servers is minimized.

[0109] The base station calculates the task completion time in the case of task offloading for all roadside units or service vehicles within the communication range of the current locations of all users in sequence, and the roadside unit or service vehicle corresponding to the minimum value obtained is the optimal edge server to which the task vehicle offloading task should be assigned. This can ensure that after each calculation of task offloading, the total completion time of the current task is minimized, the maximum user experience value in terms of delay is maximized, and for all optional roadside units and service vehicles of the current task, it can ensure that after task offloading, the maximum value among the current queuing waiting times of all optional edge servers is minimized, that is, a certain balance in time is achieved among the edge servers.

[0110] Specifically, the mathematical expression of the maximum user experience value is:

[0111] ;

[0112] where the indicator , when the computing task of the user vehicle is assigned to the edge server , , otherwise , represents the number of roadside units, represents the number of service vehicles.

[0113] In some embodiments, when executing step S4, the delay waiting time is the time that the task vehicle needs to wait after receiving the computing offloading and resource allocation message and sending the offloading task to the edge server , and its mathematical expression is:

[0114] ;

[0115] where is the task vehicle After receiving the computing offloading and resource allocation message, send the offloading task to the edge server The time to wait represents the queuing waiting time represents the transmission time for the task vehicle to send the offloading task to the edge server

[0116] After obtaining the allocation result of the task vehicle's offloading task, the base station sends the resource allocation message back to the roadside unit closest to the task vehicle, and the user-proximal roadside unit then sends it to the corresponding task vehicle. According to the message content, after the task vehicle receives the message and after the waiting time it sends the complete task information content to the allocated roadside unit or service vehicle, and after the edge server completes the offloading task and obtains the result, it directly sends it back to the task vehicle

[0117] Furthermore, when the task vehicle executes the offloading task to the optimal edge server, the total completion time of the task vehicle has the following mathematical expression:

[0118] ;

[0119] wherein represents the set of roadside units represents the set of service vehicles represents the task vehicle the total completion time represents the task vehicle at the roadside unit the total completion time of executing the offloading task represents the task vehicle at the service vehicle the total completion time of executing the offloading task represents the local total time

[0120] Although the embodiments of the present invention have been described in detail above, it is obvious to those skilled in the art that various modifications and changes can be made to these embodiments. However, it should be understood that such modifications and changes are all within the scope and spirit of the present invention described in the claims. Moreover, the present invention described herein can have other embodiments and can be implemented or realized in various ways

Claims

1. A computing offloading method based on task transmission control and improved greedy strategy, characterized in that, Including: Construct a system model among multiple task vehicles, multiple roadside units, and multiple service vehicles, and construct multiple optional edge servers based on the roadside units and service vehicles. The system model includes a communication model and a task offloading model; Obtain the transmission time for a task vehicle to send an offloading task to an edge server based on the uplink data rate of communication and the data volume size of the computing task of the task vehicle. Obtain the execution time based on the data volume size of the task vehicle's execution of the previous offloading task and the CPU cycle frequency of the edge server. Obtain the time ratio of the transmission and execution of adjacent offloading tasks, and obtain the task data volume ratio based on the ratio of the data volume size of the computing task of the task vehicle to the data volume size of the task vehicle's execution of the previous offloading task; Sort according to the size of the task data volume ratio to obtain a sorting sequence. Sequentially calculate the transmission time for a task vehicle to send an offloading task to an edge server using the task offloading model according to the order of the sorting sequence. Calculate the distances corresponding to the task vehicle and the edge server at the start and completion of the offloading task based on the transmission time, and screen and obtain the optimal edge server, i.e., the task allocation result, from the optional edge servers based on the distance and the communication range radius of the edge server; Based on the task allocation result, the base station sends a resource allocation message back to the proximal roadside unit, and the proximal roadside unit then sends the resource allocation message to the task vehicle. After a delay waiting time, the task vehicle sends the complete task information to the allocated roadside unit or service vehicle. When the optimal edge server finishes executing the offloading task, it sends the offloading result back to the task vehicle; The process for the optimal edge server to execute the offloading task includes: When the offloading tasks are assigned to the th and the th edge servers, the queuing waiting time sets and the set after task offloading are obtained respectively. After removing the equal elements in the set and the set , the set is obtained; Based on the set Allocate the current optimal edge server for the user to execute the offloading task, and obtain the minimum of the maximum working times of all the optional edge servers of the task vehicle. Among them, when The improved greedy strategy is expressed as: ； ； Among them, represents the queuing waiting time of the task of the task vehicle at the corresponding edge server, represents the number of edge servers that the task vehicle can choose, represents the total completion time for the task vehicle assigned to the edge server to execute the offloading task, represents the th edge server, represents the th edge server.

2. The computational offloading method based on task transmission control and improved greedy strategy according to claim 1, wherein When constructing the system model among multiple task vehicles, multiple roadside units, and multiple service vehicles, the specific components inside the system model include: a base station, roadside units, task vehicles, and service vehicles. Among them, the task vehicles and service vehicles travel in a straight line at a fixed speed during the computing offloading period. The task vehicle is used to send a task offloading request to the roadside unit, and after transmitting the complete task information to the roadside unit, it receives the result returned by the roadside unit after completing the task execution. The service vehicle is used to receive the complete task information sent by the task vehicle and send the task execution result to the task vehicle. The base station is used to receive the task offloading request sent by the roadside unit and return a resource allocation message to the roadside unit. The roadside unit is used to receive the task offloading request sent by the task vehicle and return the resource allocation message to the task vehicle; 3. The computational offloading method based on task transmission control and improved greedy strategy according to claim 1, characterized in that The mathematical expression of the communication model is: ； ； Among them, represents the bandwidth of the transmission channel between the mission vehicle and the edge server ; represents the transmission power of the mission vehicle ; represents the channel gain; represents the ambient noise when the service provider is the receiving end; is the Euclidean distance between the mission vehicle and the edge server ; represents the path loss component; represents the uplink data rate of the communication between the mission vehicle and the edge server ; The mathematical expression of the task offloading model is: ； ； Among them, represents the computational workload of the task of the mission vehicle and represents the data volume of the computational task of the mission vehicle represents the computational intensity of the task, and represents the transmission time for the mission vehicle to send the offloading task to the edge server. It should be noted that there seems to be some incorrect or repeated tags in the original text. The above translation is based on the best understanding and compliance with the requirements.

4. The computational offloading method based on task transmission control and improved greedy strategy according to claim 1, characterized in that The process of obtaining the distances corresponding to the task vehicle and the edge server at the start and completion of the offloading task based on the coordinates of the task vehicle respectively includes: ； ； ； Among them, represents the distance of the mission vehicle from the edge server . represents the distance of the mission vehicle from the roadside unit . represents the distance of the mission vehicle from the service vehicle . , are respectively the coordinates of the mission vehicle relative to the edge server when sending the offloading task. , are respectively the driving sub-speeds of the mission vehicle in the corresponding coordinate axis directions. , are respectively expressed as the coordinates of the mission vehicle relative to the roadside unit when sending the offloading task. represents the total completion time of the mission vehicle executing the offloading task at the roadside unit . represents the total completion time of the mission vehicle executing the offloading task at the service vehicle . , are respectively the coordinates of the mission vehicle relative to the service vehicle when sending the offloading task. , are respectively the speeds of the service vehicle in the corresponding coordinate axis directions; The process of screening and obtaining the optimal edge server from the optional edge servers based on the distance and the communication range radius includes: ； ； ； Among them, , , are the communication range radii of the roadside unit , the service vehicle , and the edge server respectively.

5. A computing offloading method based on task transmission control and improved greedy strategy according to claim 1, characterized in that, The mathematical expression for the sum of the total task completion times of the edge servers is as follows: ； Among them, represents the sum of the total task completion times, represents the edge server the number of tasks that need to be executed, , , respectively represent the edge server in the execution order, the execution times of the 1st, 2nd, …, th tasks. When , obtains the minimum value, and the edge server executes the assigned offloading tasks in ascending order of task execution time to obtain the minimum value of the sum of the total task completion times.

6. A computing offloading method based on task transmission control and improved greedy strategy according to claim 1, characterized in that The mathematical expression for the transmission time of the task vehicle sending the offloading task to the edge server is as follows: ； The execution time has a mathematical expression of: ； The ratio of the task data volumes The mathematical expression thereof is as follows: ； Among them, represents the transmission time for the task vehicle to send the unloading task to the edge server, represents the task vehicle in terms of the data volume size of the computing task, represents the task vehicle and the edge server in terms of the uplink data rate of communication therebetween, represents the bandwidth of the transmission channel, represents the task vehicle in terms of the transmission power, represents the channel gain, represents the Gaussian white noise power, represents the task vehicle in terms of the data volume size of the previous executed task, represents the edge server in terms of the CPU cycle frequency, represents the computing intensity of the task, represents the task vehicle in terms of the data volume size of the previous executed task of the task.

7. A computing offloading method based on task transmission control and improved greedy strategy according to claim 1, characterized in that When the task vehicle executes the unloading task to the optimal edge server, the task vehicle The mathematical expression of the total completion time is as follows: ； Among them, represents the set of roadside units, represents the set of service vehicles, represents the mission vehicle 's total completion time, represents the mission vehicle at the roadside unit the total completion time of the unloading task performed, represents the mission vehicle at the service vehicle the total completion time of the unloading task performed, represents the local total time.

Images