CN114173300A - Multi-user task unloading method and system - Google Patents

Multi-user task unloading method and system Download PDF

Info

Publication number
CN114173300A
CN114173300A CN202111364676.5A CN202111364676A CN114173300A CN 114173300 A CN114173300 A CN 114173300A CN 202111364676 A CN202111364676 A CN 202111364676A CN 114173300 A CN114173300 A CN 114173300A
Authority
CN
China
Prior art keywords
task
vehicle terminal
ecs
unloading
vehicle
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111364676.5A
Other languages
Chinese (zh)
Inventor
武贵路
李仲亮
陆波
丁文杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ictehi Technology Development Jiangsu Co ltd
Jiangnan University
Original Assignee
Ictehi Technology Development Jiangsu Co ltd
Jiangnan University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ictehi Technology Development Jiangsu Co ltd, Jiangnan University filed Critical Ictehi Technology Development Jiangsu Co ltd
Priority to CN202111364676.5A priority Critical patent/CN114173300A/en
Publication of CN114173300A publication Critical patent/CN114173300A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/22—Traffic simulation tools or models
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/06—Testing, supervising or monitoring using simulated traffic
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/50—Allocation or scheduling criteria for wireless resources
    • H04W72/53—Allocation or scheduling criteria for wireless resources based on regulatory allocation policies
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention discloses a multi-user task unloading method and a multi-user task unloading system, which comprise the following steps: s1: the vehicle terminal sends a task unloading request to the ECS to which the vehicle terminal belongs; s2: the ECS solves a multi-user task unloading strategy with minimum task overhead as an optimal multi-user unloading strategy according to the established task unloading overhead model, wherein the task unloading overhead model meets the requirements of task delay and task energy consumption; s3: and the ECS sends the optimal multi-user unloading strategy to a vehicle terminal, and the vehicle terminal unloads corresponding tasks according to the optimal multi-user unloading strategy. The invention reduces the task unloading cost, enhances the calculation capability of the Internet of vehicles and reduces the data transmission time delay.

Description

Multi-user task unloading method and system
Technical Field
The invention relates to the field of vehicle networking communication, in particular to a multi-user task unloading method and system.
Background
With the continuous development of information communication technology in the field of mobile communication, great convenience is brought to the ground life of people. The internet of vehicles is used as the most promising development prospect in the internet of things, and has attracted wide attention of all social circles. The internet of things is also considered to be a third revolutionary development wave following the development of the information industry field of computers and the internet. The information service of the internet of vehicles in the future has the characteristics of real-time performance, large scale, mass data access, heterogeneous network fusion and the like. In the context of high-speed mobile internet-of-vehicles communications, the processing and transmission of large amounts of data is a major challenge for researchers today. The traditional technical means can not effectively ensure the characteristics of ultrahigh reliability and low time delay during the information transmission of the Internet of vehicles and meet the application service requirements of the Internet of vehicles due to the limitation of computing resources and energy of the vehicle terminals.
In the face of mass data in a complex traffic network environment, the appearance of a Cloud Computing (CC) technology can effectively solve the problem of lack of Computing resources of a vehicle terminal. However, the way of uploading data to the cloud for centralized processing increases the time delay of information transmission in the internet of vehicles. Time delay is one of the most important metrics in the internet of vehicles. To solve this problem, the CC Server is considered to be pulled to the Edge of the car networking network near the vehicle terminal, and as an Edge Computing Server (ECS), it is called Mobile Edge Computing (MEC) technology. The MEC technology can be applied to the Internet of vehicles to solve the problems of lack of computing resources of vehicle terminals, high task delay requirements and the like. However, in this process, the vehicle terminal in the network needs to offload its own task to the ECS according to the designed task offloading policy to complete the calculation processing, so as to reduce the data link transmission delay and processing delay, and the existing task offloading policies all have the problems of complicated calculation and excessive task overhead, which results in that the offloading policy cannot effectively improve the performance of the communication network.
Disclosure of Invention
The invention aims to provide a multi-user task unloading method and a multi-user task unloading system, which can reduce the task unloading cost, enhance the calculation capacity of the Internet of vehicles and reduce the data transmission time delay.
In order to solve the technical problem, the invention provides a multi-user task unloading method, which comprises the following steps:
s1: the vehicle terminal sends a task unloading request to the ECS to which the vehicle terminal belongs;
s2: the ECS solves a multi-user task unloading strategy with minimum task overhead as an optimal multi-user unloading strategy according to the established task unloading overhead model, wherein the task unloading overhead model meets the requirements of task delay and task energy consumption;
s3: and the ECS sends the optimal multi-user unloading strategy to a vehicle terminal, and the vehicle terminal unloads corresponding tasks according to the optimal multi-user unloading strategy.
As a further improvement of the present invention, a vehicle network system is constructed before step S1, the vehicle network system including M ECSs and N vehicle terminals, the vehicle terminals performing local calculation or ECS calculation when discriminating task processing according to their own calculation capability.
As a further improvement of the present invention, the task offload overhead model established in step S2 is:
Figure BDA0003360178720000021
wherein the content of the first and second substances,
Figure BDA0003360178720000022
represents the overhead of the vehicle terminal i task in performing local calculations,
Figure BDA0003360178720000023
representing the cost of the i task of the vehicle terminal in the ECS calculation, aiIndicating the task offloading strategy for the ith vehicle terminal, ai0 means that the vehicle terminal computing task is executed locally.
As a further improvement of the present invention, in the task offloading overhead model, according to the time cost and energy consumption of the task when the task is executed locally, the overhead of the vehicle terminal i when the task executes local computation is evaluated, which is expressed as:
Figure BDA0003360178720000024
the overhead of the vehicle terminal i task in the ECS calculation is expressed as:
Figure BDA0003360178720000031
wherein the content of the first and second substances,
Figure BDA0003360178720000032
and
Figure BDA0003360178720000033
respectively representing a time delay regulation factor and an energy consumption regulation factor,
Figure BDA0003360178720000034
indicating the time required for the vehicle terminal i task to perform the local calculation,
Figure BDA0003360178720000035
representing the energy consumption required by the vehicle terminal i when the task executes local calculation;
Figure BDA0003360178720000036
indicating the time required by the vehicle terminal i task in the ECS calculation,
Figure BDA0003360178720000037
representing the energy consumption required by the vehicle terminal task in the ECS calculation.
As a further improvement of the invention, the time required for the task of the vehicle terminal i to execute the local calculation is as follows:
Figure BDA0003360178720000038
wherein the content of the first and second substances,
Figure BDA0003360178720000039
indicating the calculation performance of the i-th vehicle terminal, niIndicating the number of CPU cycles required to complete the task;
the energy consumption required by the vehicle terminal i when the task executes the local calculation is as follows:
Figure BDA00033601787200000310
wherein, deltaiIn order to calculate the energy consumption factor locally,
Figure BDA00033601787200000311
energy consumption when local hardware circuits work.
As a further improvement of the invention, the time required for the vehicle terminal i task in the ECS calculation is:
Figure BDA00033601787200000312
wherein the content of the first and second substances,
Figure BDA00033601787200000313
the time is calculated for the ECS to perform the task,
Figure BDA00033601787200000314
time to ECS for task transfer;
Figure BDA00033601787200000315
Figure BDA00033601787200000316
Figure BDA00033601787200000317
represents the computational performance of the jth ECS; r isi(a) Representing the data transmission rate, m, of the task a performed between the vehicle terminal i and the uplink of the ECSiRepresenting the size of the data volume;
the energy consumption required by the vehicle terminal i task in the ECS calculation is as follows:
Figure BDA0003360178720000041
wherein, betaiRepresenting the vehicle terminal i task ECS calculating the energy consumption factor,
Figure BDA0003360178720000042
indicating that the vehicle terminal maintains the energy consumption required for normal communication link transmission, there is a relationship that satisfies
Figure BDA0003360178720000043
A threshold value for maintaining normal communication of the link;
Figure BDA0003360178720000044
strategy for representing N vehicle terminals to execute task unloading
Figure BDA0003360178720000045
Uplink reliable data transmission rate with ECS:
Figure BDA0003360178720000046
where W is the channel bandwidth, piRepresenting the i-th vehicle terminal transmission power, gi,jIndicating the channel gain when the ith vehicle terminal offloads the task to the jth ECS,
Figure BDA0003360178720000047
in order to communicate the interference term(s),
Figure BDA0003360178720000048
representing channel noiseAnd (4) power.
As a further improvement of the present invention, when the multi-user task offloading policy with the minimum task overhead is solved in step S2, the vehicle terminal performs interference calculation and task offloading policy update;
the interference calculation comprises the following steps: for the vehicle terminal i, it selects strategy a at time ti(t) processing its own task, if the total received power of all ECSs is known, the vehicle terminal i obtains the received power from the jth ECS
Figure BDA0003360178720000049
If the vehicle terminal i does not establish connection with the jth ECS, the interference caused by accessing the ECS is
Figure BDA00033601787200000410
Is equal in value to the received power of the accessed ECS, expressed as
Figure BDA00033601787200000411
The task offloading policy updating includes: after the vehicle terminal i completes the channel transmission interference calculation, strategy updating needs to be executed according to the current channel state, and the interference caused when the vehicle terminal i establishes communication connection with each ECS is calculated, namely the interference is
Figure BDA0003360178720000051
The vehicle terminal i performs the update according to the following decision:
Figure BDA0003360178720000052
the vehicle terminal i decides whether to execute the set U after decision updatingiPerforming task unloading decision updating;
if the vehicle terminal i changes the existing task unloading strategy, the cost of the vehicle terminal i cannot be reduced, namely ai(t)∈UiThen the vehicle terminal i follows a without changing the task unloading strategyi(t+1)＝ai(t)；Otherwise, the vehicle terminal i selects the strategy for executing task unloading with the minimum expenditure by changing the task unloading strategy of the vehicle terminal i, namely ai(t+1)∈Ui。
As a further improvement of the present invention, the process of solving the multi-user task offloading policy with the minimum task overhead in step S2 includes the following steps:
task unloading strategy executed by N vehicle terminals in vehicle networking established based on M ECSs
Figure BDA0003360178720000053
And then, establishing an optimization problem by taking the minimum energy consumption of the network system as a target:
min Ksys
Figure BDA0003360178720000054
wherein, R is a data transmission rate threshold of a communication link;
solving the energy consumption optimization problem of the network system by adopting a game theory, reducing the calculation complexity by adopting a distributed task unloading algorithm, and making a task unloading decision of a vehicle terminal i into aiThe other vehicle terminal unloading strategy is expressed as
Figure BDA0003360178720000055
In order to minimize the total overhead of the system and the overhead of each vehicle terminal, when the overhead of a vehicle terminal i is minimized, an optimization problem is established:
Figure BDA0003360178720000056
Figure BDA0003360178720000057
for the vehicle terminal i, obtaining an optimal task unloading strategy by adopting a game theory
Figure BDA0003360178720000058
When the vehicle terminal i can not change the unloading decision
Figure BDA0003360178720000061
To reduce self-overhead
Figure BDA0003360178720000062
Temporal task offload policy
Figure BDA0003360178720000063
Meets the Nash balance result achieved when a plurality of vehicle terminals play games together
Figure BDA0003360178720000064
As a further improvement of the invention, a vehicle networking simulation platform integrating traffic simulation software SUMO and network simulation software OMNeT + + is built, and verification of an optimal multi-user task unloading strategy is realized based on a vehicle networking communication simulation framework Venns so as to evaluate the performance of the vehicle networking simulation platform.
The invention also provides a system for unloading tasks by using the multi-user task unloading method.
The invention has the beneficial effects that: the method comprises the steps of establishing a task unloading model facing a multi-user edge computing network, enabling the established task unloading model to achieve minimization of system overhead through two performances of time delay and energy consumption, and selectively solving and determining a task unloading strategy by adopting a game theory; the invention combines the edge computing technology, the game theory technology and the Internet of vehicles, enhances the computing capability of the Internet of vehicles, reduces the data transmission delay, is easy to realize, has standard process and convenient operation, and solves the problem of vehicle resource limitation in the Internet of vehicles;
furthermore, the method can adopt a vehicle communication simulation framework of Veins to build an experimental simulation platform by combining SUMO and OMNeT + + simulation software, so as to realize simulation and verification of the proposed task unloading strategy and evaluate the performance of the algorithm.
Drawings
FIG. 1 is a schematic structural diagram of an embodiment of the present invention;
FIG. 2 is a schematic diagram of the Internet of vehicles communication network architecture of the present invention;
fig. 3 is a central urban map of the sanyang square of the wuxi city in Jiangsu province (five-pointed star is a deployment base station) according to an embodiment of the present invention;
fig. 4 is a distribution diagram of RSU deployment in OMNeT + +, according to an embodiment of the present invention.
Detailed Description
The present invention is further described below in conjunction with the following figures and specific examples so that those skilled in the art may better understand the present invention and practice it, but the examples are not intended to limit the present invention.
Referring to fig. 1, the present invention provides a multi-user task offloading method, including the following steps:
s1: the vehicle terminal sends a task unloading request to the ECS to which the vehicle terminal belongs;
s2: the ECS solves a multi-user task unloading strategy with minimum task overhead as an optimal multi-user unloading strategy according to the established task unloading overhead model, wherein the task unloading overhead model meets the requirements of task delay and task energy consumption;
s3: and the ECS sends the optimal multi-user unloading strategy to a vehicle terminal, and the vehicle terminal unloads corresponding tasks according to the optimal multi-user unloading strategy.
As shown in fig. 2, before implementing the present invention, an ECS-based car networking system is constructed, which includes an ECS, a roadside infrastructure, and vehicle terminals, and a plurality of the vehicle terminals, the vehicle terminals and the roadside infrastructure, and the vehicle terminals and the ECS may be connected by transmission links. Considering a multi-vehicle terminal multi-service task unloading structure in an ECS scene, roadside infrastructures (base stations and roadside units) are randomly deployed at roadsides and jointly constructed into an internet-of-vehicles system. ECS can be deployed near vehicle terminals, roadside infrastructure. When any vehicle terminal needs to process the task of the vehicle terminal, the vehicle terminal selects to process the task through the computing resource of the vehicle terminal or the computing resource of the ECS, and the network system architecture model comprises network elements such as the vehicle terminal, roadside infrastructures (BS and roadside units), the ECS, communication links and the like. The data transmission link comprises a vehicle terminal and vehicle terminal link, a vehicle terminal and roadside infrastructure, a vehicle terminal and an ECS. The network element transmission mode in the vehicle networking can be divided into various types, including but not limited to direct communication and indirect communication between a vehicle terminal and elements in the network.
Considering that the application scene of the internet of vehicles comprises N vehicle terminals and M ECSs, each vehicle terminal has task data to be processed. During task processing, the scene of the internet of vehicles maintains an approximately static scene, that is, some network parameters such as the number of vehicle terminals and channel states in the network are kept unchanged, that is, the considered scene is a quasi-static scene, and the vehicle terminals participate in a communication process and a calculation process respectively in the task unloading process, wherein the quasi-static scene is that the communication channel conditions of the vehicle terminals are unchanged in the task unloading time period of the vehicle terminals, and after the system completes all calculation tasks, the number and positions of the vehicle terminals in the system may be changed. a isiRepresenting a task unloading strategy of the ith vehicle terminal; a isi0 denotes that the vehicle terminal calculation task is executed locally, aiJ indicates that the vehicle calculation task is performed at the jth ECS; n vehicles have respective unloading strategies, so that system task unloading strategy distribution is obtained
Figure BDA0003360178720000081
Establishing a task unloading overhead model, and solving a multi-user task unloading strategy with the minimum task overhead as a specific process of an optimal multi-user unloading strategy:
for a specific task, two dimensional characteristics of the task are considered, namely the data size miAnd the number n of CPU cycles required to complete the taski. And respectively analyzing from two indexes of time and energy consumption.
For local computing executing tasks, adopt
Figure BDA0003360178720000082
Indicating the computational performance (e.g., number of CPU revolutions per second) of the ith local computation, and the computational performance of different vehicle terminals is different. Thus, this task is performedThe time required for local calculation is
Figure BDA0003360178720000083
The energy consumption for the task to execute local calculation is
Figure BDA0003360178720000084
Wherein, deltaiIn order to calculate the energy consumption factor locally,
Figure BDA0003360178720000085
energy consumption when local hardware circuits work.
For the task unloading to the ECS execution, two indexes of the task transmission delay and the energy consumption are also considered. The time required for the ECS to execute the tasks is as follows:
Figure BDA0003360178720000086
wherein the content of the first and second substances,
Figure BDA0003360178720000087
indicating the computational performance (e.g., CPU revolutions per second) of the jth ECS, and different ECSs have different computational performance.
In addition, the transmission time during the task unloading to the ECS is:
Figure BDA0003360178720000091
in the case where the ECS is assumed to have sufficient computational resources to immediately execute the offload task, the sum of the time spent offloading the task to the ECS is:
Figure BDA0003360178720000092
when the vehicle terminal unloads the task, the vehicle terminal establishes communication connection with the ECS, and the energy consumption generated in the process is as follows:
Figure BDA0003360178720000093
wherein the content of the first and second substances,
Figure BDA0003360178720000094
indicating that the vehicle terminal maintains the energy consumption required for normal communication link transmission, there is a relationship that satisfies
Figure BDA0003360178720000095
Wherein the content of the first and second substances,
Figure BDA0003360178720000096
a threshold for maintaining a communication link for normal communication completion.
The overhead can be evaluated in terms of time spent and energy consumed when the task is executed locally, and is expressed as:
Figure BDA0003360178720000097
by using
Figure BDA0003360178720000098
And
Figure BDA0003360178720000099
respectively representing the time delay regulation factor and the energy consumption regulation factor, and can be correspondingly set according to different network scenes
Figure BDA00033601787200000910
And
Figure BDA00033601787200000911
the value is obtained. In view of the generalized application scenario,
Figure BDA00033601787200000912
and
Figure BDA00033601787200000913
take on a value of [0,1]In the meantime. Of particular note, when
Figure BDA00033601787200000914
When the time is needed, the processing task is represented as time delay sensitivity data; when in use
Figure BDA00033601787200000915
In time, the energy required for processing the task is insufficient and cannot be completed.
Similarly, the total overhead of a task when executing an ECS is represented as:
Figure BDA00033601787200000916
the task processing is executed in different modes aiming at different vehicle terminals, and the total cost of the vehicle terminals under different task execution conditions can be calculated and is represented as follows:
Figure BDA00033601787200000917
task unloading strategy executed by N vehicle terminals in vehicle networking established based on M ECSs
Figure BDA00033601787200000918
And then, establishing an optimization problem by taking the minimum energy consumption of the network system as a target:
min Ksys
Figure BDA0003360178720000101
and solving the energy consumption optimization problem of the network system by adopting a game theory, and providing a distributed task unloading algorithm to reduce the computational complexity. The task uninstalling decision for the vehicle terminal i is aiOther vehicle terminal offload policies may be expressed as
Figure BDA0003360178720000102
To minimize the system overhead, the overhead per vehicle terminal may be minimized. When the cost of the vehicle terminal i is minimized, an optimization problem is established:
Figure BDA0003360178720000103
Figure BDA0003360178720000104
for the vehicle terminal i, obtaining an optimal task unloading strategy by adopting a game theory
Figure BDA0003360178720000105
When the vehicle terminal i can not change the unloading decision
Figure BDA0003360178720000106
To reduce self-overhead
Figure BDA0003360178720000107
Temporal task offload policy
Figure BDA0003360178720000108
Meets the Nash balance result achieved when a plurality of vehicle terminals play games together
Figure BDA0003360178720000109
The game theory is adopted to realize the establishment of the mutually satisfied unloading strategies of all vehicle terminals, and the huge calculation overhead caused by centralized task unloading scheduling is overcome.
Further, for a given task offloading strategy, different vehicle terminals i can be connected to the same ECSj (for example, using CDMA technology), and will be subjected to white noise during signal transmission
Figure BDA00033601787200001012
Influence, and interference between different vehicle terminal signalsAnd (4) disturbing.
The uplink reliable data transmission rate of the vehicle terminal and the ECS is as follows:
Figure BDA00033601787200001010
where W is the channel bandwidth, piRepresenting the i-th vehicle terminal transmission power, gi,jIndicating the channel gain when the ith vehicle terminal offloads the task to the jth ECS.
Figure BDA00033601787200001011
Is a communication interference item.
The method comprises the steps that a vehicle terminal sends a task unloading request to an ECS, the ECS completes interference calculation of a link transmission channel and task unloading strategy updating according to channel conditions of all vehicle terminals in a network system within a set time period, namely when the task unloading strategy is adopted, operations needing to be completed by the vehicle terminal comprise the interference calculation and the task unloading strategy updating.
In terms of interference calculation, for the vehicle terminal i, it selects the strategy a at time ti(t) processing the own task. If the total received power of all ECSs is known, the vehicle terminal i obtains the received power from the jth ECS
Figure BDA0003360178720000111
If the vehicle terminal i does not establish connection with the jth ECS, the interference caused by accessing the ECS is
Figure BDA0003360178720000112
Equal in value to the received power of the accessed ECS, expressed as:
Figure BDA0003360178720000113
in the aspect of task unloading strategy updating, after the vehicle terminal i completes channel transmission interference calculation, strategy updating needs to be executed according to the current channel state so as to reduce self calculation overhead. For vehiclesThe terminal i calculates the interference caused when establishing a communication connection with each ECS, i.e. the interference
Figure BDA0003360178720000114
The vehicle terminal i performs the update according to the following decision:
Figure BDA0003360178720000115
the vehicle terminal i can decide whether to execute the decision updated set UiAnd performing task unloading decision updating. For UiMiddle element
Figure BDA0003360178720000116
And not necessarily uniquely. It is worth noting that if the vehicle terminal i changes the existing task unloading strategy, the cost of the vehicle terminal i cannot be reduced, namely ai(t)∈UiThen the vehicle terminal i follows a without changing the task unloading strategyi(t+1)＝ai(t); otherwise, the vehicle terminal i selects the strategy for executing task unloading with the minimum expenditure by changing the task unloading strategy of the vehicle terminal i, namely ai(t+1)∈Ui. And the ECS sends the task unloading strategy to a vehicle terminal, and the vehicle terminal unloads the corresponding task according to the task unloading strategy.
The invention also provides a system for unloading tasks by using the multi-user task unloading method. Specifically, the system includes a vehicle terminal and the ECS thereof;
the vehicle terminal sends a task unloading request to the ECS to which the vehicle terminal belongs;
the ECS solves a multi-user task unloading strategy with minimum task overhead as an optimal multi-user unloading strategy according to the established task unloading overhead model, wherein the task unloading overhead model meets the requirements of task delay and task energy consumption;
and the ECS sends the optimal multi-user unloading strategy to a vehicle terminal, and the vehicle terminal unloads corresponding tasks according to the optimal multi-user unloading strategy.
Examples
As shown in fig. 3 and 4, taking "city map of the city of sanyang square center in wuxi city, jiangsu province" as an example, a vehicle networking simulation framework Veins simulation platform is built for verification, and a SUMO model and an OMNeT + + tool means are fused to verify the proposed task unloading strategy.
Acquiring a map of a central urban area of a three-positive square in Wuxi city, Jiangsu province from an OpenMapStreet official network by a vehicle-mounted terminal, and exporting to obtain a map.
Using the JOSM software to optimize the acquired map file, deleting some extraneous elements in the map, such as subway lines, airlines, etc., will generate a map that completely covers the selected area to avoid the map being too lengthy.
And starting the SUMO model, and configuring the SUMO model by combining the road network file and the traffic file.
Setting simulation network scene parameters and simulating a real working scene;
6 RSU map areas are deployed in each intersection section, and 120 vehicles move along a preset driving track. In the communication environment parameters, the channel bandwidth is 20MHz, and the data size of the vehicle terminal is 20 KB.
Step three, starting the network simulator OMNeT + +, establishing a communication link between the vehicle terminal and the roadside infrastructure, and connecting the communication link with the traffic infrastructure model SUMO;
a vehicle networking simulation framework Veins simulation platform provides a Traci interface, and instantiates each vehicle terminal in the SUMO into an independent network node; it provides a SUMO-launchd. py script to connect SUMO with OMNeT + + and ensure that the connection is successful.
Step four, when the vehicle terminal executes task unloading, the vehicle terminal communicates with the roadside infrastructure, and OMNeT + + provides a corresponding message file;
the vehicle terminal needs to send corresponding information to the roadside infrastructure, and the information content comprises the identity of the vehicle terminal, the position of the vehicle terminal, the transmitting power of the vehicle terminal, a current unloading strategy, the calculation performance of the local vehicle terminal, the size of task data, the number of CPU revolutions required by task execution, the energy consumption coefficient when the task is locally executed and the energy consumption coefficient when the task is executed by an ECS;
fifthly, the roadside infrastructure executes task unloading strategy scheduling;
in order to conveniently realize the algorithm, the vehicle terminal sends an unloading strategy updating request, and the roadside infrastructure sends out an update permission response which is completed in the roadside infrastructure.
In order to solve the problem of vehicle resource limitation in the Internet of vehicles, the multi-user task unloading and simulation platform construction method establishes a task unloading model facing a multi-user edge computing network, and the established task unloading model mainly considers two performances of time delay and energy consumption so as to minimize system overhead. The method comprises the steps of solving and determining a task unloading strategy by adopting a game theory, establishing an experiment simulation platform by adopting a Veins vehicle communication simulation framework and combining SUMO and OMNeT + + simulation software, and realizing simulation and verification of the provided task unloading strategy so as to evaluate algorithm performance. The invention builds the algorithm for verifying the virtual task unloading simulation verification platform, can effectively reduce the calculation complexity of the vehicle terminal and the system calculation overhead, and builds the multi-user task unloading simulation test platform at the same time, thereby improving the accuracy of the test result.
The above-mentioned embodiments are merely preferred embodiments for fully illustrating the present invention, and the scope of the present invention is not limited thereto. The equivalent substitution or change made by the technical personnel in the technical field on the basis of the invention is all within the protection scope of the invention. The protection scope of the invention is subject to the claims.

Claims (10)

1. A multi-user task offloading method is characterized in that: the method comprises the following steps:
s1: the vehicle terminal sends a task unloading request to the ECS to which the vehicle terminal belongs;
s2: the ECS solves a multi-user task unloading strategy with minimum task overhead as an optimal multi-user unloading strategy according to the established task unloading overhead model, wherein the task unloading overhead model meets the requirements of task delay and task energy consumption;
s3: and the ECS sends the optimal multi-user unloading strategy to a vehicle terminal, and the vehicle terminal unloads corresponding tasks according to the optimal multi-user unloading strategy.
2. The method of claim 1, wherein: before step S1, a vehicle network system is constructed, which includes M ECSs and N vehicle terminals, and when the vehicle terminals determine task processing according to their own computing capabilities, local computation or ECS computation is performed.
3. The method of claim 1, wherein: the task offloading overhead model established in step S2:
Figure FDA0003360178710000011
wherein the content of the first and second substances,
Figure FDA0003360178710000012
represents the overhead of the vehicle terminal i task in performing local calculations,
Figure FDA0003360178710000013
representing the cost of the i task of the vehicle terminal in the ECS calculation, aiIndicating the task offloading strategy for the ith vehicle terminal, ai0 means that the vehicle terminal computing task is executed locally.
4. A method of multi-user task offloading as recited in claim 3, wherein: in the task unloading overhead model, according to the time cost and energy consumption of the task when the task is executed locally, the overhead of the vehicle terminal i when the task executes local calculation is evaluated, and the evaluation is represented as:
Figure FDA0003360178710000014
the overhead of the vehicle terminal i task in the ECS calculation is expressed as:
Figure FDA0003360178710000021
wherein the content of the first and second substances,
Figure FDA0003360178710000022
and
Figure FDA0003360178710000023
respectively representing a time delay regulation factor and an energy consumption regulation factor,
Figure FDA0003360178710000024
indicating the time required for the vehicle terminal i task to perform the local calculation,
Figure FDA00033601787100000213
representing the energy consumption required by the vehicle terminal i when the task executes local calculation;
Figure FDA00033601787100000214
indicating the time required by the vehicle terminal i task in the ECS calculation,
Figure FDA00033601787100000215
representing the energy consumption required by the vehicle terminal task in the ECS calculation.
5. The method of claim 4, wherein: the time required for the vehicle terminal i task to perform local calculation is:
Figure FDA0003360178710000025
wherein f isi localIndicating the calculation performance of the i-th vehicle terminal, niIndicating the number of CPU cycles required to complete the task; the energy consumption required when the vehicle terminal i task executes local calculation is：
Figure FDA0003360178710000026
Wherein, deltaiIn order to calculate the energy consumption factor locally,
Figure FDA0003360178710000027
energy consumption when local hardware circuits work.
6. The method of claim 4, wherein: the time required for the vehicle terminal i task in the ECS calculation is:
Figure FDA0003360178710000028
wherein the content of the first and second substances,
Figure FDA0003360178710000029
the time is calculated for the ECS to perform the task,
Figure FDA00033601787100000210
time to ECS for task transfer;
Figure FDA00033601787100000211
Figure FDA00033601787100000212
Figure FDA0003360178710000031
represents the computational performance of the jth ECS; r isi(a) Representing the data transmission rate when performing task a between the vehicle terminal i and the uplink of the ECSRate, miRepresenting the size of the data volume;
the energy consumption required by the vehicle terminal i task in the ECS calculation is as follows:
Figure FDA0003360178710000032
wherein, betaiRepresenting the vehicle terminal i task ECS calculating the energy consumption factor,
Figure FDA0003360178710000033
indicating that the vehicle terminal maintains the energy consumption required for normal communication link transmission, there is a relationship that satisfies
Figure FDA0003360178710000034
Figure FDA0003360178710000035
A threshold value for maintaining normal communication of the link;
Figure FDA0003360178710000036
strategy for representing N vehicle terminals to execute task unloading
Figure FDA0003360178710000037
Uplink reliable data transmission rate with ECS:
Figure FDA0003360178710000038
where W is the channel bandwidth, piRepresenting the i-th vehicle terminal transmission power, gi,jIndicating the channel gain when the ith vehicle terminal offloads the task to the jth ECS,
Figure FDA0003360178710000039
in order to communicate the interference term(s),
Figure FDA00033601787100000310
representing the channel noise power.
7. The method of claim 6, wherein: when the multi-user task unloading strategy with the minimum task overhead is solved in the step S2, the vehicle terminal performs interference calculation and task unloading strategy updating;
the interference calculation comprises the following steps: for the vehicle terminal i, it selects strategy a at time ti(t) processing its own task, if the total received power of all ECSs is known, the vehicle terminal i obtains the received power from the jth ECS
Figure FDA00033601787100000311
If the vehicle terminal i does not establish connection with the jth ECS, the interference caused by accessing the ECS is
Figure FDA00033601787100000312
Is equal in value to the received power of the accessed ECS, expressed as
Figure FDA0003360178710000041
The task offloading policy updating includes: after the vehicle terminal i completes the channel transmission interference calculation, strategy updating needs to be executed according to the current channel state, and the interference caused when the vehicle terminal i establishes communication connection with each ECS is calculated, namely the interference is
Figure FDA0003360178710000047
The vehicle terminal i performs the update according to the following decision:
Figure FDA0003360178710000042
Figure FDA0003360178710000043
the vehicle terminal i decides whether to execute the set U after decision updatingiPerforming task unloading decision updating;
if the vehicle terminal i changes the existing task unloading strategy, the cost of the vehicle terminal i cannot be reduced, namely ai(t)∈UiThen the vehicle terminal i follows a without changing the task unloading strategyi(t+1)＝ai(t); otherwise, the vehicle terminal i selects the strategy for executing task unloading with the minimum expenditure by changing the task unloading strategy of the vehicle terminal i, namely ai(t+1)∈Ui。
8. A method of multi-user task offloading as recited in claim 3, wherein: the process of solving the multi-user task offloading policy with the minimum task cost in step S2 includes the following steps:
task unloading strategy executed by N vehicle terminals in vehicle networking established based on M ECSs
Figure FDA0003360178710000046
And then, establishing an optimization problem by taking the minimum energy consumption of the network system as a target:
min Ksys
Figure FDA0003360178710000044
wherein, R is a data transmission rate threshold of a communication link;
solving the energy consumption optimization problem of the network system by adopting a game theory, reducing the calculation complexity by adopting a distributed task unloading algorithm, and making a task unloading decision of a vehicle terminal i into aiThe other vehicle terminal unloading strategy is expressed as
Figure FDA0003360178710000045
In order to minimize the total system overhead, minimize the overhead per vehicle terminal, when the overhead of the vehicle terminal i is minimized,establishing an optimization problem:
Figure FDA0003360178710000051
Figure FDA0003360178710000052
for the vehicle terminal i, obtaining an optimal task unloading strategy by adopting a game theory
Figure FDA0003360178710000053
When the vehicle terminal i can not change the unloading decision
Figure FDA0003360178710000054
To reduce self-overhead
Figure FDA0003360178710000055
Temporal task offload policy
Figure FDA0003360178710000056
Meets the Nash balance result achieved when a plurality of vehicle terminals play games together
Figure FDA0003360178710000057
9. A method of multi-user task offloading according to any of claims 1-8, characterized by: and (3) building a vehicle networking simulation platform integrating traffic simulation software SUMO and network simulation software OMNeT + +, and realizing verification of an optimal multi-user task unloading strategy based on a vehicle networking communication simulation framework Venns so as to evaluate the performance of the vehicle networking simulation platform.
10. A system for offloading tasks using the multi-user task offloading method of any of claims 1-9.
CN202111364676.5A 2021-11-17 2021-11-17 Multi-user task unloading method and system Pending CN114173300A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111364676.5A CN114173300A (en) 2021-11-17 2021-11-17 Multi-user task unloading method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111364676.5A CN114173300A (en) 2021-11-17 2021-11-17 Multi-user task unloading method and system

Publications (1)

Publication Number Publication Date
CN114173300A true CN114173300A (en) 2022-03-11

Family

ID=80479493

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111364676.5A Pending CN114173300A (en) 2021-11-17 2021-11-17 Multi-user task unloading method and system

Country Status (1)

Country Link
CN (1) CN114173300A (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109413615A (en) * 2018-09-14 2019-03-01 重庆邮电大学 The energy delay compromise proposal of Energy-aware unloading under car networking based on MEC
CN109814951A (en) * 2019-01-22 2019-05-28 南京邮电大学 The combined optimization method of task unloading and resource allocation in mobile edge calculations network
CN111615060A (en) * 2020-04-29 2020-09-01 西安理工大学 Regional multicast method based on bus track and positioning information
CN112367640A (en) * 2020-11-09 2021-02-12 中科怡海高新技术发展江苏股份公司 V2V mode multitask unloading method and system based on mobile edge calculation

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109413615A (en) * 2018-09-14 2019-03-01 重庆邮电大学 The energy delay compromise proposal of Energy-aware unloading under car networking based on MEC
CN109814951A (en) * 2019-01-22 2019-05-28 南京邮电大学 The combined optimization method of task unloading and resource allocation in mobile edge calculations network
CN111615060A (en) * 2020-04-29 2020-09-01 西安理工大学 Regional multicast method based on bus track and positioning information
CN112367640A (en) * 2020-11-09 2021-02-12 中科怡海高新技术发展江苏股份公司 V2V mode multitask unloading method and system based on mobile edge calculation

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
X. CHEN等: "Efficient Multi-User Computation Offloading for Mobile-Edge Cloud Computing", 《IEEE/ACM TRANSACTIONS ON NETWORKING》 *

Similar Documents

Publication Publication Date Title
Xu et al. Game theory for distributed IoV task offloading with fuzzy neural network in edge computing
CN111132175B (en) Cooperative computing unloading and resource allocation method and application
Yang et al. A parallel intelligence-driven resource scheduling scheme for digital twins-based intelligent vehicular systems
CN109947574B (en) Fog network-based vehicle big data calculation unloading method
CN114189892A (en) Cloud-edge collaborative Internet of things system resource allocation method based on block chain and collective reinforcement learning
CN113590232B (en) Relay edge network task unloading method based on digital twinning
CN113687875B (en) Method and device for unloading vehicle tasks in Internet of vehicles
CN112115505A (en) New energy automobile charging station charging data transmission method based on mobile edge calculation and block chain technology
CN113709249B (en) Safe balanced unloading method and system for driving assisting service
CN115209426A (en) Dynamic deployment method of digital twin servers in edge internet of vehicles
CN115297171A (en) Edge calculation unloading method and system for cellular Internet of vehicles hierarchical decision
CN116321298A (en) Multi-objective joint optimization task unloading strategy based on deep reinforcement learning in Internet of vehicles
Li et al. DNN partition and offloading strategy with improved particle swarm genetic algorithm in VEC
Ren et al. Handoff-aware distributed computing in high altitude platform station (HAPS)–assisted vehicular networks
Song et al. Everyone-centric heterogeneous multi-server computation offloading in ITS with pervasive AI
Laroui et al. Virtual mobile edge computing based on IoT devices resources in smart cities
Hu et al. Digital twins-based multi-agent deep reinforcement learning for UAV-assisted vehicle edge computing
Xu et al. Perception synergy optimization with deep reinforcement learning for cooperative perception in C-V2V scenarios
Wang et al. Joint offloading decision and resource allocation in vehicular edge computing networks
CN117221951A (en) Task unloading method based on deep reinforcement learning in vehicle-mounted edge environment
CN114173300A (en) Multi-user task unloading method and system
He et al. A UAV emergency network user allocation method for load balancing
CN112203309B (en) Joint task unloading and caching method based on server cooperation
Alghamdi et al. Optimized Contextual Data Offloading in Mobile Edge Computing
Liu et al. Collaborative scheduling of computing tasks for edge computing

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20220311

RJ01 Rejection of invention patent application after publication