CN116405569A - Task unloading matching method and system based on vehicle and edge computing server - Google Patents

Task unloading matching method and system based on vehicle and edge computing server Download PDF

Info

Publication number
CN116405569A
CN116405569A CN202211612847.6A CN202211612847A CN116405569A CN 116405569 A CN116405569 A CN 116405569A CN 202211612847 A CN202211612847 A CN 202211612847A CN 116405569 A CN116405569 A CN 116405569A
Authority
CN
China
Prior art keywords
vehicle
server
vehicles
steps
task
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
CN202211612847.6A
Other languages
Chinese (zh)
Inventor
金敏晨
冯维
姜显扬
许晓荣
朱芳
夏晓威
吴端坡
胡楚哲
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.)
Hangzhou Dianzi University
Original Assignee
Hangzhou Dianzi 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 Hangzhou Dianzi University filed Critical Hangzhou Dianzi University
Priority to CN202211612847.6A priority Critical patent/CN116405569A/en
Publication of CN116405569A publication Critical patent/CN116405569A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5061—Partitioning or combining of resources
    • G06F9/5072—Grid computing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00—Indexing scheme relating to G06F9/00
    • G06F2209/50—Indexing scheme relating to G06F9/50
    • G06F2209/502—Proximity
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00—Indexing scheme relating to G06F9/00
    • G06F2209/50—Indexing scheme relating to G06F9/50
    • G06F2209/509—Offload
    • 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)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Medical Informatics (AREA)
  • Mathematical Physics (AREA)
  • Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention belongs to the technical field of information and communication engineering, and particularly relates to a task unloading matching method and system based on a vehicle and an edge computing server. The method comprises the following steps: s1, an initialization stage: the central control center MBS collects basic information of vehicles and roadside units RSU; s2, calculating time delay and energy consumption between different vehicles and roadside units (RSUs) by considering random movement of the vehicles and various uploading rate calculation methods; s3, establishing an optimization model taking joint optimization task unloading time delay and energy consumption as targets and taking task unloading decision and random movement of a vehicle as constraints; and S4, solving the model established in the step S3 by adopting a matching algorithm based on cost minimization, and obtaining the final matching condition of the vehicle server. The invention has the characteristics of being capable of jointly optimizing time delay and energy consumption and considering mobility and transmission reliability of vehicles.

Description

Task unloading matching method and system based on vehicle and edge computing server
Technical Field
The invention belongs to the technical field of information and communication engineering, and particularly relates to a task unloading matching method and system based on a vehicle and an edge computing server.
Background
With the rapid development of vehicle networks, how to process massive data of computationally intensive tasks in time is urgent. Because of the limited computing and battery resources of the vehicle, the vehicle cannot meet all computing demands. Traditional solutions are to upload tasks to the cloud to aid in computation, but due to the huge delay, it is impractical to require all vehicles to interact directly with the cloud. Transmitting large amounts of data to the cloud will put tremendous strain on the network bandwidth. Meanwhile, the security of user privacy data is difficult to ensure by traditional cloud computing. Mobile edge computing (MEC, mobile edge computing) is a new promising solution to the vehicle resource constraint challenges that can deploy edge computing infrastructure in roadside units (RSUs) and offload computing tasks from the vehicle to edge servers. The Mobile Edge Computing (MEC) enables computing resources to be closer to the Internet of vehicles, and has the characteristics of low time delay, large bandwidth, high reliability and the like.
Task offloading is a very important problem in moving edge computing. The prior art investigated the impact of vehicle mobility on computational offloading. However, these works mostly only consider constant movement of the vehicle speed, which is not a true vehicle running mode. In addition, the mobility of the vehicle also affects the wireless channel, requiring consideration of various uplink transmission rates. However, most efforts only assume that the uplink transmission rate is constant within the coverage of one roadside unit RSU. Previous studies only consider time delays or energy consumption in vehicle edge calculations and do not jointly optimize both for better performance. Therefore, how to combine the above matters is a urgent problem to be solved.
For example, a task offloading method based on mobile edge calculation in the internet of vehicles described in chinese patent document with application number of cn202210242936.X, the method includes: constructing a multi-edge server joint unloading model; acquiring an unloading task of a vehicle, and constructing an unloading base station selection vector according to the unloading task of the vehicle; calculating the load and energy consumption for task unloading by adopting a resource allocation method based on the equivalent maximum tolerant delay according to the unloading base station selection vector; taking energy consumption as an adaptive function, and adopting a genetic algorithm to carry out iterative optimization on an unloading task of the vehicle to obtain an unloading base station selection scheme; optimizing a task unloading strategy by using reinforcement learning according to an unloading base station selection scheme to obtain a task unloading ratio and unloading power; according to the selected unloading base station, the formulated unloading ratio and the unloading power, task unloading is completed, while the total energy consumption expenditure of the system is effectively reduced, and the effectiveness of the task unloading and resource allocation of the internet of vehicles is realized, the method has the disadvantage that the problem of jointly optimizing the random movement and various uploading rates of the vehicles to obtain better performance is not considered because only the time delay or the energy consumption in the calculation of the edges of the vehicles is still considered.
Disclosure of Invention
The invention provides a task unloading matching method and a task unloading matching system based on a vehicle and an edge computing server, which can jointly optimize time delay and energy consumption and consider the mobility and transmission reliability of the vehicle, and solve the problem that the method has certain limitation because the prior task unloading method in the mobile edge computing only considers that the speed of the vehicle is constant and the uplink transmission rate is constant in the coverage area of a roadside unit RSU and does not consider the random movement and various uploading rates of the vehicle.
In order to achieve the aim of the invention, the invention adopts the following technical scheme:
the task unloading matching method based on the vehicle and the edge computing server comprises the following steps:
s1, an initialization stage:
the central control center MBS collects basic information of vehicles and roadside units RSU;
s2, calculating time delay and energy consumption between different vehicles and roadside units (RSUs) by considering random movement of the vehicles and various uploading rate calculation methods;
s3, establishing an optimization model taking joint optimization task unloading time delay and energy consumption as targets and taking task unloading decision and random movement of a vehicle as constraints;
and S4, solving the model established in the step S3 by adopting a matching algorithm based on cost minimization, and obtaining the final matching condition of the vehicle server.
Preferably, in step S1, the basic information of the vehicle and the roadside unit RSU includes a task data size of the vehicle, a calculation capability of the vehicle, an upper and lower limit of a moving speed and an acceleration, and a calculation capability of the roadside unit RSU.
Preferably, step S2 includes the steps of:
s21, set in the VEC server layer, deploy
Figure 83674DEST_PATH_IMAGE001
RSU of different coverage areas, the set of RSUs is recorded as
Figure 833324DEST_PATH_IMAGE002
Figure 864601DEST_PATH_IMAGE003
，
Figure 66913DEST_PATH_IMAGE004
Representing the braiding of different RSUs within the coverage area;
Figure DEST_PATH_IMAGE005
is expressed as the coverage radius of (2)
Figure 804187DEST_PATH_IMAGE006
；
Figure 532977DEST_PATH_IMAGE005
For each ofThe computing resources provided by the vehicle are noted as
Figure 173037DEST_PATH_IMAGE007
The method comprises the steps of carrying out a first treatment on the surface of the Network layer of vehicle
Figure 361180DEST_PATH_IMAGE008
Vehicle composition, representing a collection of vehicles
Figure 790893DEST_PATH_IMAGE009
The method comprises the steps of carrying out a first treatment on the surface of the Binary group for vehicle task
Figure 16600DEST_PATH_IMAGE010
To represent;
Figure 624299DEST_PATH_IMAGE011
the data size representing the task is represented by a size of the data,
Figure 801202DEST_PATH_IMAGE012
representing the required CPU cycles;
s22, setting
Figure 273422DEST_PATH_IMAGE013
For the length of time slot, the first
Figure 852171DEST_PATH_IMAGE014
The time slots are semi-closed intervals; dividing the coverage area of each into the same length by adopting a discretization method
Figure 630771DEST_PATH_IMAGE015
Is defined between cells of (a); obtaining
Figure 796435DEST_PATH_IMAGE005
The right boundary index of coverage of (c) is
Figure 302372DEST_PATH_IMAGE016
The method comprises the steps of carrying out a first treatment on the surface of the Position index of road
Figure 234162DEST_PATH_IMAGE017
The method comprises the steps of carrying out a first treatment on the surface of the Vehicle with a vehicle body having a vehicle body support
Figure 449243DEST_PATH_IMAGE018
First, the
Figure 600739DEST_PATH_IMAGE014
Position of each time slot
Figure 411831DEST_PATH_IMAGE019
Tracking through indexes corresponding to the positions of the time slots; vehicle with a vehicle body having a vehicle body support
Figure 699593DEST_PATH_IMAGE018
Is the position of (2)
Figure 234043DEST_PATH_IMAGE019
Sum speed of
Figure 420305DEST_PATH_IMAGE020
Every time according to the following formula
Figure 438683DEST_PATH_IMAGE013
Updating once per second:
Figure 925159DEST_PATH_IMAGE021
(1)
Figure 341097DEST_PATH_IMAGE022
(2)
wherein,,
Figure 296546DEST_PATH_IMAGE023
and
Figure 964287DEST_PATH_IMAGE024
minimum and maximum speeds of vehicle movement, respectively;
Figure 226641DEST_PATH_IMAGE025
for vehicles
Figure 442509DEST_PATH_IMAGE018
The amount of speed change of the first time slot,
Figure 790314DEST_PATH_IMAGE026
for vehicles
Figure 261746DEST_PATH_IMAGE018
First, the
Figure 880072DEST_PATH_IMAGE014
Acceleration of each time slot, subject to truncation
Figure 434550DEST_PATH_IMAGE027
Standard gaussian distribution on;
Figure 892819DEST_PATH_IMAGE028
and
Figure 26998DEST_PATH_IMAGE029
respectively vehicles
Figure 30988DEST_PATH_IMAGE018
A maximum deceleration value and a maximum acceleration value of (a);
s23, if the vehicle
Figure 303837DEST_PATH_IMAGE018
Selecting to offload to VEC server, vehicle
Figure 485289DEST_PATH_IMAGE018
Is composed of four parts: latency of waiting
Figure 939271DEST_PATH_IMAGE030
Transmission delay time
Figure 906090DEST_PATH_IMAGE031
Calculating time delay
Figure 271212DEST_PATH_IMAGE032
Time delay of switching
Figure 644687DEST_PATH_IMAGE033
The method comprises the steps of carrying out a first treatment on the surface of the If the vehicle is
Figure 261613DEST_PATH_IMAGE018
Selecting a locally calculated vehicle
Figure 535469DEST_PATH_IMAGE018
Is the calculated time delay
Figure 835607DEST_PATH_IMAGE034
；
Latency of waiting
Figure 945645DEST_PATH_IMAGE030
For vehicles
Figure 349951DEST_PATH_IMAGE018
The time required to reach the unloading:
Figure 917460DEST_PATH_IMAGE035
(3)
the transmission signal-to-noise ratio calculating method comprises the following steps:
Figure 703014DEST_PATH_IMAGE036
(4)
wherein,,
Figure 346354DEST_PATH_IMAGE037
for the transmission power of the vehicle,
Figure 668531DEST_PATH_IMAGE038
is that
Figure 730028DEST_PATH_IMAGE018
To the point of
Figure 670171DEST_PATH_IMAGE005
The distance between the two plates is set to be equal,
Figure 177638DEST_PATH_IMAGE039
in order to be a path loss index,
Figure 940058DEST_PATH_IMAGE040
for the reference channel gain at the reference distance,
Figure 715116DEST_PATH_IMAGE041
is additive white noise power;
s24, uplink transmission rate
Figure 387012DEST_PATH_IMAGE042
Different transmission models and coding rate calculations are selected by signal-to-noise ratio:
Figure 614731DEST_PATH_IMAGE043
(5)
wherein,,
Figure 852946DEST_PATH_IMAGE044
for the coding rate of the code-rate,
Figure 639767DEST_PATH_IMAGE045
is the uplink bandwidth;
vehicle with a vehicle body having a vehicle body support
Figure 859396DEST_PATH_IMAGE018
Is the first of (2)
Figure 653040DEST_PATH_IMAGE014
Data volume transmitted in each time slot
Figure 233626DEST_PATH_IMAGE046
Is approximately at
Figure 248856DEST_PATH_IMAGE047
The average uplink transmission rate over the time slot interval
Figure 514752DEST_PATH_IMAGE013
：
Figure 952949DEST_PATH_IMAGE048
(6)
Transmission delay time
Figure 578971DEST_PATH_IMAGE031
For vehicles
Figure 478400DEST_PATH_IMAGE018
Uploading tasks to a server
Figure 118460DEST_PATH_IMAGE005
The time required satisfies the following formula:
Figure 73647DEST_PATH_IMAGE049
(7)
Figure 4825DEST_PATH_IMAGE050
(8)
wherein,,
Figure 542117DEST_PATH_IMAGE051
for vehicles
Figure 540028DEST_PATH_IMAGE018
Just get into
Figure 549223DEST_PATH_IMAGE005
Time slots required for coverage:
Figure 798938DEST_PATH_IMAGE052
(9)
the transmission delay must be such that the transmission is completed within the coverage of the server, i.e. the transmission delay cannot exceed the maximum transmission delay that can be uploaded at the corresponding server
Figure 440004DEST_PATH_IMAGE053
：
Figure 110282DEST_PATH_IMAGE054
(10)
Maximum transmission delay
Figure 649848DEST_PATH_IMAGE053
Is the time required from the vehicle entering the server coverage to leaving the server coverage:
Figure 624626DEST_PATH_IMAGE055
(11)
s25, calculating time delay as time required by processing tasks; if the vehicle is
Figure 556416DEST_PATH_IMAGE018
At the local computing task, the time delay is calculated
Figure 771497DEST_PATH_IMAGE032
The method comprises the following steps:
Figure 250889DEST_PATH_IMAGE056
(12)
wherein,,
Figure 202927DEST_PATH_IMAGE057
the frequency is calculated for the local area of the vehicle,
Figure 366055DEST_PATH_IMAGE012
representing the CPU cycles required for the computing task;
if the vehicle is
Figure 204567DEST_PATH_IMAGE018
Selective offloading to a server
Figure 703130DEST_PATH_IMAGE005
Calculating, then calculating the time delay
Figure 222973DEST_PATH_IMAGE032
The method comprises the following steps:
Figure 709449DEST_PATH_IMAGE058
(13)
wherein,,
Figure 892431DEST_PATH_IMAGE059
is a server
Figure 80836DEST_PATH_IMAGE005
Is used for calculating the frequency of the calculation.
S26, at the server
Figure 14157DEST_PATH_IMAGE005
When a vehicle moves out of the coverage range of a server during processing tasks, switching is needed, and a calculation result needs to be transmitted from a current server to a server in a range to which the position of the vehicle belongs after calculation is completed and then transmitted to the vehicle; the data quantity of the calculation result is smaller, and the feedback delay is ignored; obtaining the switching time delay
Figure 102942DEST_PATH_IMAGE033
The method comprises the following steps:
Figure 955361DEST_PATH_IMAGE060
(14)
wherein the method comprises the steps of
Figure 178531DEST_PATH_IMAGE061
To unload the server index of the range to which the vehicle position belongs after completion,
Figure 338380DEST_PATH_IMAGE062
indexing the transmitted server;
Figure 720819DEST_PATH_IMAGE063
the time required for completing one-time switching for two adjacent servers;
Figure 822768DEST_PATH_IMAGE061
the value of (2) is found by the following inequality:
Figure 552476DEST_PATH_IMAGE064
(15)
wherein,,
Figure 421075DEST_PATH_IMAGE065
representing a vehicle
Figure 736650DEST_PATH_IMAGE018
From entering the road to the task at the server
Figure 556969DEST_PATH_IMAGE005
Calculating the number of time slots required by completion;
the vehicle
Figure 957995DEST_PATH_IMAGE018
Selecting total time delay to offload
Figure 895864DEST_PATH_IMAGE066
The method comprises the following steps:
Figure 813748DEST_PATH_IMAGE067
(16)
s27, if the vehicle selects to calculate locally, the energy consumption of the vehicle
Figure 319816DEST_PATH_IMAGE068
To calculate the energy consumption:
Figure 863929DEST_PATH_IMAGE069
(17)
wherein,,
Figure 169271DEST_PATH_IMAGE070
for a vehicleThe energy consumption cost factor of the vehicle is calculated,
Figure 990596DEST_PATH_IMAGE071
for the energy coefficients specified in the vehicle CPU model,
Figure 792199DEST_PATH_IMAGE057
for vehicles
Figure 644181DEST_PATH_IMAGE018
Is a local calculation frequency of (2);
if the vehicle is
Figure 799219DEST_PATH_IMAGE018
Selecting to unload to
Figure 865264DEST_PATH_IMAGE005
Calculating the energy consumption of the vehicle
Figure 667129DEST_PATH_IMAGE068
The energy consumption required for transmission:
Figure 795622DEST_PATH_IMAGE072
(18)
wherein,,
Figure 878984DEST_PATH_IMAGE073
is uplink transmission power;
s28, finally obtaining the vehicle
Figure 157126DEST_PATH_IMAGE018
Is of (2)
Figure 34952DEST_PATH_IMAGE066
And energy consumption
Figure 916320DEST_PATH_IMAGE068
The following expression is present:
Figure 39259DEST_PATH_IMAGE074
（19）。
preferably, step S3 includes the steps of:
s31, setting the optimization variable definition of task unloading as
Figure 611055DEST_PATH_IMAGE075
The method comprises the steps of carrying out a first treatment on the surface of the Wherein,,
Figure 164177DEST_PATH_IMAGE076
representing a vehicle
Figure 985371DEST_PATH_IMAGE018
Offloading tasks to
Figure 20323DEST_PATH_IMAGE005
Otherwise, the device can be used to determine whether the current,
Figure 682511DEST_PATH_IMAGE077
；
s32, converting the task unloading problem into a matching problem between the vehicle and the RSU; setting a vehicle to select only one server, and selecting one server at most simultaneously
Figure 902140DEST_PATH_IMAGE078
A server; the total time delay and the total energy consumption are jointly optimized, and the cost function of task calculation is obtained as follows:
Figure 505903DEST_PATH_IMAGE079
the method comprises the steps of carrying out a first treatment on the surface of the Wherein the method comprises the steps of
Figure 734759DEST_PATH_IMAGE080
Representing time delay
Figure 297459DEST_PATH_IMAGE066
The weight of the vehicle is occupied;
s33, modeling the total optimization problem is finally obtained as follows:
Figure 579666DEST_PATH_IMAGE081
Figure 860606DEST_PATH_IMAGE082
(20)。
preferably, step S4 includes the steps of:
s41, adopting an algorithm based on cost minimization, firstly calculating the vehicle
Figure 893153DEST_PATH_IMAGE018
Selective offloading to a server
Figure 341319DEST_PATH_IMAGE005
Cost of (2)
Figure DEST_PATH_IMAGE083
If the constraint condition in step S33 is not satisfied, then
Figure 637171DEST_PATH_IMAGE083
Is infinite;
s42, calculating the priority of each server by the following formula:
Figure 952877DEST_PATH_IMAGE084
(21)
wherein,,
Figure 257957DEST_PATH_IMAGE085
=1 indicates a vehicle
Figure 795248DEST_PATH_IMAGE018
May be offloaded to the server, if the constraints are not met,
Figure 416329DEST_PATH_IMAGE086
the method comprises the steps of carrying out a first treatment on the surface of the Pressing on server
Figure 937440DEST_PATH_IMAGE087
And (5) ascending order arrangement is carried out to obtain a server priority list.
Preferably, the step S4 further includes the steps of:
s43, setting cost values which are unloaded to different servers according to selection for each vehicle, and arranging the cost values in ascending order to be used as a preference list of the vehicles;
s44, each vehicle selects the first server in the preference list, sends a matching request, and temporarily divides the vehicles into vehicle sets of the corresponding servers
Figure 780632DEST_PATH_IMAGE088
Neutralizing;
processing the matching request according to the order of the server priority list; for each server, if the received matching request exceeds
Figure 985479DEST_PATH_IMAGE078
The requests are arranged in an ascending order according to the cost value, and the front is arranged
Figure 967342DEST_PATH_IMAGE078
The vehicles corresponding to the matching requests are reserved; for each of the remaining vehicles, continuing to send a matching request to the next server in the preference list until a server is found that does not exceed the capacity limit, and classifying the corresponding vehicle into
Figure 631541DEST_PATH_IMAGE088
Neutralizing; when all of the servers have been processed,
Figure 173031DEST_PATH_IMAGE088
the final vehicle server match.
The invention also provides a task unloading matching system based on the vehicle and the edge computing server, which comprises the following steps:
the initialization stage module is used for enabling the centralized control center MBS to collect basic information of vehicles and roadside units RSU;
the time delay and energy consumption calculation module is used for calculating time delay and energy consumption between different vehicles and roadside units (RSUs) by taking into consideration random movement of the vehicles and various uploading rate calculation methods;
the model building module is used for building an optimization model which aims at jointly optimizing task unloading time delay and energy consumption and aims at task unloading decision and random movement of the vehicle as constraints;
and the model solving module is used for solving the established model by adopting a matching algorithm based on cost minimization to obtain the final matching condition of the vehicle server.
Preferably, the model building module specifically comprises:
setting the optimization variables of task offloading to be defined as
Figure 684915DEST_PATH_IMAGE075
The method comprises the steps of carrying out a first treatment on the surface of the Wherein,,
Figure 24629DEST_PATH_IMAGE076
representing a vehicle
Figure 739906DEST_PATH_IMAGE018
Offloading tasks to
Figure 65845DEST_PATH_IMAGE005
Otherwise, the device can be used to determine whether the current,
Figure 353607DEST_PATH_IMAGE077
；
converting the task unloading problem into a matching problem between the vehicle and the RSU; setting a vehicle to select only one server, and selecting one server at most simultaneously
Figure 221813DEST_PATH_IMAGE078
A server; the total time delay and the total energy consumption are jointly optimized, and the cost function of task calculation is obtained as follows:
Figure 657342DEST_PATH_IMAGE079
the method comprises the steps of carrying out a first treatment on the surface of the Wherein the method comprises the steps of
Figure 147492DEST_PATH_IMAGE080
Representing time delay
Figure 165126DEST_PATH_IMAGE066
The weight of the occupied；
The final overall optimization problem is modeled as:
Figure 440119DEST_PATH_IMAGE089
Figure 793303DEST_PATH_IMAGE082
(20)。
compared with the prior art, the invention has the beneficial effects that: (1) According to the method, the edge computing server is deployed on the roadside unit, and the computing task of the vehicle can be selected to be computed locally or uploaded to the server for computing; under the condition that the vehicle randomly runs, from the matching angle, optimizing an unloading scheme between the vehicle and a server to jointly optimize the time delay and the energy consumption of the system; (2) The invention is different from the previous research that the uplink transmission rate is assumed to be constant, and the invention selects different transmission models and encoding rates according to the signal to noise ratio to calculate the uplink transmission rate, thereby ensuring the reliability of transmission; (3) The matching algorithm based on the cost minimization can effectively reduce time delay and energy consumption.
Drawings
FIG. 1 is a flow chart of a method for matching task offloading based on a vehicle and an edge computing server in accordance with the present invention;
FIG. 2 is a model architecture diagram of a vehicle and edge computing server based task offload matching system of the present invention;
fig. 3 is a cost simulation diagram of different unloading algorithms for different numbers of vehicles according to an embodiment of the present invention.
Detailed Description
In order to more clearly illustrate the embodiments of the present invention, specific embodiments of the present invention will be described below with reference to the accompanying drawings. It is evident that the drawings in the following description are only examples of the invention, from which other drawings and other embodiments can be obtained by a person skilled in the art without inventive effort.
Examples:
as shown in fig. 1, the present invention provides a task offload matching method based on a vehicle and an edge computing server, comprising the steps of:
s1, an initialization stage:
the central control center MBS collects basic information of vehicles and roadside units RSU;
s2, calculating time delay and energy consumption between different vehicles and roadside units (RSUs) by considering random movement of the vehicles and various uploading rate calculation methods;
s3, establishing an optimization model taking joint optimization task unloading time delay and energy consumption as targets and taking task unloading decision and random movement of a vehicle as constraints;
and S4, solving the model established in the step S3 by adopting a matching algorithm based on cost minimization, and obtaining the final matching condition of the vehicle server.
The present invention contemplates a three-tier network consisting of a VEC server tier, a vehicle network tier, and a centralized control tier as shown in FIG. 1. The centralized control center gathers the required data and solves the offloading policies. In step S1, the basic information of the vehicle and the roadside unit RSU includes the task data size of the vehicle, the computing power of the vehicle, the upper and lower limits of the moving speed and the acceleration, and the computing power of the roadside unit RSU.
Step S2 further comprises the steps of:
s21, set in the VEC server layer, deploy
Figure 461045DEST_PATH_IMAGE090
RSU of different coverage areas, the set of RSUs is recorded as
Figure 785716DEST_PATH_IMAGE002
Figure 405178DEST_PATH_IMAGE003
，
Figure 628349DEST_PATH_IMAGE004
Representing the braiding of different RSUs within the coverage area;
Figure 286732DEST_PATH_IMAGE005
is expressed as the coverage radius of (2)
Figure 747801DEST_PATH_IMAGE006
；
Figure 269656DEST_PATH_IMAGE005
The computing resources provided for each vehicle are noted as
Figure 229390DEST_PATH_IMAGE007
The method comprises the steps of carrying out a first treatment on the surface of the Network layer of vehicle
Figure 442197DEST_PATH_IMAGE008
Vehicle composition, representing a collection of vehicles
Figure 774083DEST_PATH_IMAGE009
The method comprises the steps of carrying out a first treatment on the surface of the Binary group for vehicle task
Figure 968304DEST_PATH_IMAGE010
To represent;
Figure 166067DEST_PATH_IMAGE011
the data size (in bits) representing the task represents the required CPU cycles;
for simplicity, local computing may be considered to be offloaded on one server, noted as
Figure 670648DEST_PATH_IMAGE091
. The tasks of the vehicle can be selected to be calculated locally or can be offloaded to a server for calculation.
The vehicle in the invention adopts a random moving mode. The invention adopts a time slot base to study the system, and specifically comprises the following steps:
s22, setting
Figure 558838DEST_PATH_IMAGE013
For the length of time slot, the first
Figure 159846DEST_PATH_IMAGE014
The time slots are semi-closed intervals
Figure 579326DEST_PATH_IMAGE092
The method comprises the steps of carrying out a first treatment on the surface of the Each is discretized by adopting a discretization method
Figure 320886DEST_PATH_IMAGE005
The coverage area is divided into the same length
Figure 93277DEST_PATH_IMAGE015
As shown in fig. 2; obtaining
Figure 504666DEST_PATH_IMAGE005
The right boundary index of coverage of (c) is
Figure 801656DEST_PATH_IMAGE016
The method comprises the steps of carrying out a first treatment on the surface of the Position index of road
Figure 645109DEST_PATH_IMAGE017
The method comprises the steps of carrying out a first treatment on the surface of the Vehicle with a vehicle body having a vehicle body support
Figure 852099DEST_PATH_IMAGE018
First, the
Figure 621341DEST_PATH_IMAGE014
Position of each time slot
Figure 921740DEST_PATH_IMAGE019
Tracking through indexes corresponding to the positions of the time slots; vehicle with a vehicle body having a vehicle body support
Figure 818151DEST_PATH_IMAGE018
Is the position of (2)
Figure 128916DEST_PATH_IMAGE019
Sum speed of
Figure 85371DEST_PATH_IMAGE020
Every time according to the following formula
Figure 858417DEST_PATH_IMAGE013
Updating once per second:
Figure 73366DEST_PATH_IMAGE021
(1)
Figure 143697DEST_PATH_IMAGE022
(2)
wherein,,
Figure 271053DEST_PATH_IMAGE023
and
Figure 29931DEST_PATH_IMAGE024
minimum and maximum speeds of vehicle movement, respectively;
Figure 815615DEST_PATH_IMAGE025
for vehicles
Figure 523808DEST_PATH_IMAGE018
First, the
Figure 9016DEST_PATH_IMAGE014
The amount of speed change of the time slot,
Figure 821901DEST_PATH_IMAGE026
for vehicles
Figure 660544DEST_PATH_IMAGE018
First, the
Figure 738090DEST_PATH_IMAGE014
Acceleration of each time slot, subject to truncation
Figure 895665DEST_PATH_IMAGE027
Standard gaussian distribution on;
Figure 176604DEST_PATH_IMAGE028
and
Figure 333785DEST_PATH_IMAGE029
respectively vehicles
Figure 282149DEST_PATH_IMAGE018
A maximum deceleration value and a maximum acceleration value of (a);
s23, if the vehicle
Figure 342116DEST_PATH_IMAGE018
Selecting to offload to VEC server, vehicle
Figure 359619DEST_PATH_IMAGE018
Is composed of four parts: latency of waiting
Figure 540065DEST_PATH_IMAGE030
Transmission delay time
Figure 93668DEST_PATH_IMAGE031
Calculating time delay
Figure 91580DEST_PATH_IMAGE032
Time delay of switching
Figure 347112DEST_PATH_IMAGE033
The method comprises the steps of carrying out a first treatment on the surface of the If the vehicle is
Figure 73192DEST_PATH_IMAGE018
Selecting a locally calculated vehicle
Figure 730570DEST_PATH_IMAGE018
Is the calculated time delay
Figure 899383DEST_PATH_IMAGE034
；
Latency of waiting
Figure 392943DEST_PATH_IMAGE030
For vehicles
Figure 384033DEST_PATH_IMAGE018
Reach to be unloaded
Figure 614026DEST_PATH_IMAGE005
The time required is:
Figure 45751DEST_PATH_IMAGE035
(3)
transmission signal to noise ratio
Figure 197247DEST_PATH_IMAGE093
The calculation method comprises the following steps:
Figure 726448DEST_PATH_IMAGE036
(4)
wherein,,
Figure 109150DEST_PATH_IMAGE037
for the transmission power of the vehicle,
Figure 229553DEST_PATH_IMAGE038
is that
Figure 133924DEST_PATH_IMAGE018
To the point of
Figure 954899DEST_PATH_IMAGE005
The distance between the two plates is set to be equal,
Figure 238113DEST_PATH_IMAGE039
in order to be a path loss index,
Figure 919630DEST_PATH_IMAGE040
for the reference channel gain at the reference distance,
Figure 875079DEST_PATH_IMAGE041
is additive white noise power;
s24, uplink transmission rate
Figure 277241DEST_PATH_IMAGE042
By signal-to-noise ratio selectionSelecting different transmission models and encoding rate calculation:
Figure 274016DEST_PATH_IMAGE043
(5)
wherein,,
Figure 749603DEST_PATH_IMAGE044
for the coding rate of the code-rate,
Figure 831829DEST_PATH_IMAGE045
is the uplink bandwidth;
vehicle with a vehicle body having a vehicle body support
Figure 726098DEST_PATH_IMAGE018
Is the first of (2)
Figure 187166DEST_PATH_IMAGE014
Data volume transmitted in each time slot
Figure 69540DEST_PATH_IMAGE046
Is approximately at
Figure 248849DEST_PATH_IMAGE047
The average uplink transmission rate over the time slot interval
Figure 899140DEST_PATH_IMAGE013
：
Figure 198404DEST_PATH_IMAGE048
(6)
Transmission delay time
Figure 533570DEST_PATH_IMAGE031
For vehicles
Figure 419749DEST_PATH_IMAGE018
Uploading tasks to a server
Figure 623197DEST_PATH_IMAGE005
The time required is full ofThe following formula is used:
Figure 793278DEST_PATH_IMAGE049
(7)
Figure 515990DEST_PATH_IMAGE050
(8)
wherein,,
Figure 325683DEST_PATH_IMAGE051
for vehicles
Figure 99866DEST_PATH_IMAGE018
Just get into
Figure 921192DEST_PATH_IMAGE005
Time slots required for coverage:
Figure 519532DEST_PATH_IMAGE052
(9)
the transmission delay must be such that the transmission is completed within the coverage of the server, i.e. the transmission delay cannot exceed the maximum transmission delay that can be uploaded at the corresponding server
Figure 55337DEST_PATH_IMAGE053
：
Figure 475954DEST_PATH_IMAGE054
(10)
Maximum transmission delay
Figure 276420DEST_PATH_IMAGE053
Is the time required from the vehicle entering the server coverage to leaving the server coverage:
Figure 812705DEST_PATH_IMAGE055
(11)
s25, calculating time delay
Figure 472357DEST_PATH_IMAGE032
The time required for processing a task; if the vehicle is
Figure 618036DEST_PATH_IMAGE018
At the local computing task, the time delay is calculated
Figure 882796DEST_PATH_IMAGE032
The method comprises the following steps:
Figure 259157DEST_PATH_IMAGE056
(12)
wherein,,
Figure 389793DEST_PATH_IMAGE057
for vehicles
Figure 89895DEST_PATH_IMAGE018
Is used to calculate the frequency of the local calculation of (a),
Figure 631998DEST_PATH_IMAGE012
representing the CPU cycles required for the computing task;
if the vehicle chooses to offload to the server
Figure 8621DEST_PATH_IMAGE005
Calculating, namely calculating time delay as follows:
Figure 377286DEST_PATH_IMAGE058
(13)
wherein,,
Figure 295126DEST_PATH_IMAGE059
is a server
Figure 252587DEST_PATH_IMAGE005
Is used for calculating the frequency of the calculation.
S26, at the server
Figure 613161DEST_PATH_IMAGE005
When a vehicle moves out of the coverage range of a server during processing tasks, switching is needed, and a calculation result needs to be transmitted from a current server to a server in a range to which the position of the vehicle belongs after calculation is completed and then transmitted to the vehicle; the data quantity of the calculation result is smaller, and the feedback delay is ignored; obtaining the switching time delay
Figure 157537DEST_PATH_IMAGE033
The method comprises the following steps:
Figure 386393DEST_PATH_IMAGE060
(14)
wherein the method comprises the steps of
Figure 214672DEST_PATH_IMAGE061
To unload the server index of the range to which the vehicle position belongs after completion,
Figure 493950DEST_PATH_IMAGE062
indexing the transmitted server;
Figure 774890DEST_PATH_IMAGE063
the time required for completing one-time switching for two adjacent servers;
Figure 541858DEST_PATH_IMAGE061
the value of (2) is found by the following inequality:
Figure 975375DEST_PATH_IMAGE064
(15)
wherein,,
Figure 677752DEST_PATH_IMAGE065
representing a vehicle
Figure 695256DEST_PATH_IMAGE018
From entering the road to the task at the server
Figure 770309DEST_PATH_IMAGE005
Calculating the number of time slots required by completion;
the vehicle
Figure 573180DEST_PATH_IMAGE018
Selecting to unload to
Figure 695725DEST_PATH_IMAGE005
Is not less than a threshold
Figure 685678DEST_PATH_IMAGE066
The method comprises the following steps:
Figure 295913DEST_PATH_IMAGE067
(16)
s27, if the vehicle selects to calculate locally, the energy consumption of the vehicle
Figure 264875DEST_PATH_IMAGE068
To calculate the energy consumption:
Figure 246738DEST_PATH_IMAGE069
(17)
wherein,,
Figure 409472DEST_PATH_IMAGE070
as a factor of the energy consumption cost of the vehicle,
Figure 384250DEST_PATH_IMAGE071
for the energy coefficients specified in the vehicle CPU model,
Figure 958451DEST_PATH_IMAGE057
for vehicles
Figure 268472DEST_PATH_IMAGE018
Is a local meter of (2)Calculating the frequency;
if the vehicle is
Figure 810181DEST_PATH_IMAGE018
Selectively offloading to calculation, vehicle energy consumption
Figure 245708DEST_PATH_IMAGE068
The energy consumption required for transmission:
Figure 674416DEST_PATH_IMAGE072
(18)
wherein,,
Figure 185032DEST_PATH_IMAGE073
is uplink transmission power;
s28, finally obtaining the vehicle
Figure 387605DEST_PATH_IMAGE018
Is of (2)
Figure 517235DEST_PATH_IMAGE066
And the energy consumption has the following expression:
Figure 252979DEST_PATH_IMAGE074
（19）。
step S3 includes the steps of:
s31, setting the optimization variable definition of task unloading as
Figure 433031DEST_PATH_IMAGE075
The method comprises the steps of carrying out a first treatment on the surface of the Wherein,,
Figure 372168DEST_PATH_IMAGE076
representing a vehicle
Figure 554757DEST_PATH_IMAGE018
Offloading tasks to
Figure 318576DEST_PATH_IMAGE005
Otherwise, the device can be used to determine whether the current,
Figure 780781DEST_PATH_IMAGE077
；
s32, converting the task unloading problem into a matching problem between the vehicle and the RSU; setting a vehicle to select only one server, and selecting one server at most simultaneously
Figure 128586DEST_PATH_IMAGE078
A server; the invention aims to jointly optimize the total time delay and the total energy consumption, and the cost function of task calculation is obtained as follows:
Figure 353680DEST_PATH_IMAGE079
the method comprises the steps of carrying out a first treatment on the surface of the Wherein the method comprises the steps of
Figure 470541DEST_PATH_IMAGE080
Representing time delay
Figure 103648DEST_PATH_IMAGE066
The weight of the vehicle is occupied;
s33, modeling the total optimization problem is finally obtained as follows:
Figure 33689DEST_PATH_IMAGE081
Figure 167867DEST_PATH_IMAGE082
(20)。
step S4 includes the steps of:
s41, in order to solve the problem of unloading and matching of multiple vehicles and multiple servers, the invention adopts an algorithm based on cost minimization, and firstly calculates the vehicles
Figure 217862DEST_PATH_IMAGE018
Selective offloading to a server
Figure 300831DEST_PATH_IMAGE005
Cost of (2)
Figure 357649DEST_PATH_IMAGE083
If the constraint condition in step S33 is not satisfied, then
Figure 374147DEST_PATH_IMAGE083
Is infinite;
s42, calculating the priority of each server by the following formula:
Figure 91698DEST_PATH_IMAGE084
(21)
wherein,,
Figure 535449DEST_PATH_IMAGE085
=1 indicates a vehicle
Figure 79563DEST_PATH_IMAGE018
May be offloaded to the server, if the constraints are not met,
Figure 438432DEST_PATH_IMAGE086
the method comprises the steps of carrying out a first treatment on the surface of the Pressing on server
Figure 384391DEST_PATH_IMAGE087
And (5) ascending order arrangement is carried out to obtain a server priority list.
S43, setting cost values which are unloaded to different servers according to selection for each vehicle, and arranging the cost values in ascending order to be used as a preference list of the vehicles;
s44, each vehicle selects the first server in the preference list, sends a matching request, and temporarily divides the vehicles into vehicle sets of the corresponding servers
Figure 795781DEST_PATH_IMAGE088
Neutralizing;
processing the matching request according to the order of the server priority list; for each server, if the received matching request exceeds
Figure 922131DEST_PATH_IMAGE078
And, toThe requests are arranged in ascending order according to the cost value, and the requests are arranged before
Figure 201803DEST_PATH_IMAGE078
The vehicles corresponding to the matching requests are reserved; for each of the remaining vehicles, continuing to send a matching request to the next server in the preference list until a server is found that does not exceed the capacity limit, and classifying the corresponding vehicle into
Figure 877635DEST_PATH_IMAGE088
Neutralizing; when all of the servers have been processed,
Figure 410991DEST_PATH_IMAGE088
the final vehicle server match.
The invention also provides a task unloading matching system based on the vehicle and the edge computing server, which comprises the following steps:
the initialization stage module is used for enabling the centralized control center MBS to collect basic information of vehicles and roadside units RSU;
the time delay and energy consumption calculation module is used for calculating time delay and energy consumption between different vehicles and roadside units (RSUs) by taking into consideration random movement of the vehicles and various uploading rate calculation methods;
the model building module is used for building an optimization model which aims at jointly optimizing task unloading time delay and energy consumption and aims at task unloading decision and random movement of the vehicle as constraints;
and the model solving module is used for solving the established model by adopting a matching algorithm based on cost minimization to obtain the final matching condition of the vehicle server.
The model building module specifically comprises the following steps:
setting the optimization variables of task offloading to be defined as
Figure 195276DEST_PATH_IMAGE075
The method comprises the steps of carrying out a first treatment on the surface of the Wherein,,
Figure 888426DEST_PATH_IMAGE076
representing a vehicle
Figure 638338DEST_PATH_IMAGE018
Offloading tasks to
Figure 516164DEST_PATH_IMAGE005
Otherwise, the device can be used to determine whether the current,
Figure 397533DEST_PATH_IMAGE077
；
converting the task unloading problem into a matching problem between the vehicle and the RSU; setting a vehicle to select only one server, and selecting one server at most simultaneously
Figure 851298DEST_PATH_IMAGE078
A server; the total time delay and the total energy consumption are jointly optimized, and the cost function of task calculation is obtained as follows:
Figure 891935DEST_PATH_IMAGE079
the method comprises the steps of carrying out a first treatment on the surface of the Wherein the method comprises the steps of
Figure 816028DEST_PATH_IMAGE080
Representing time delay
Figure 279633DEST_PATH_IMAGE066
The weight of the vehicle is occupied;
the final overall optimization problem is modeled as:
Figure 829432DEST_PATH_IMAGE089
Figure 803204DEST_PATH_IMAGE082
(20)。
as shown in fig. 3, a cost simulation diagram under different unloading algorithms under different vehicle numbers is shown. As can be seen from fig. 3, under any number of vehicles, the matching algorithm based on the cost minimization of the method is much smaller than that of all local calculation, and the effect of reducing time delay and energy consumption is remarkable.
The invention designs a cooperation scheme between a plurality of road-side units (RSUs) and a plurality of vehicles. An edge computing server is deployed on the roadside unit, and the computing task of the vehicle can be selected to be calculated locally or uploaded to the server for calculation. Under the condition that the vehicle randomly runs, from the matching angle, an unloading scheme between the vehicle and the server is optimized so as to jointly optimize the time delay and the energy consumption of the system.
According to the method, the edge computing server is deployed on the roadside unit, and the computing task of the vehicle can be selected to be computed locally or uploaded to the server for computing; under the condition that the vehicle randomly runs, from the matching angle, optimizing an unloading scheme between the vehicle and a server to jointly optimize the time delay and the energy consumption of the system; the invention is different from the previous research that the uplink transmission rate is assumed to be constant, and the invention selects different transmission models and encoding rates according to the signal to noise ratio to calculate the uplink transmission rate, thereby ensuring the reliability of transmission; the matching algorithm based on the cost minimization can effectively reduce time delay and energy consumption.
The foregoing is only illustrative of the preferred embodiments and principles of the present invention, and changes in specific embodiments will occur to those skilled in the art upon consideration of the teachings provided herein, and such changes are intended to be included within the scope of the invention as defined by the claims.

Claims (8)

1. The task unloading matching method based on the vehicle and the edge computing server is characterized by comprising the following steps of:
s1, an initialization stage:
the central control center MBS collects basic information of vehicles and roadside units RSU;
s2, calculating time delay and energy consumption between different vehicles and roadside units (RSUs) by considering random movement of the vehicles and various uploading rate calculation methods;
s3, establishing an optimization model taking joint optimization task unloading time delay and energy consumption as targets and taking task unloading decision and random movement of a vehicle as constraints;
and S4, solving the model established in the step S3 by adopting a matching algorithm based on cost minimization, and obtaining the final matching condition of the vehicle server.
2. The task offload matching method based on the vehicle and the edge calculation server according to claim 1, wherein in step S1, the basic information of the vehicle and the roadside unit RSU includes a task data size of the vehicle, a calculation capability of the vehicle, upper and lower limits of a moving speed and acceleration, and a calculation capability of the roadside unit RSU.
3. The vehicle and edge computing server-based task offload matching method according to claim 1, wherein step S2 includes the steps of:
s21, set in the VEC server layer, deploy
Figure QLYQS_2
RSU of different coverage areas, the set of RSUs is recorded as
Figure QLYQS_6
Figure QLYQS_9
，
Figure QLYQS_3
Representing the braiding of different RSUs within the coverage area;
Figure QLYQS_7
is expressed as the coverage radius of (2)
Figure QLYQS_11
；
Figure QLYQS_13
The computing resources provided for each vehicle are noted as
Figure QLYQS_1
The method comprises the steps of carrying out a first treatment on the surface of the Network layer of vehicle
Figure QLYQS_5
Vehicle composition, representing a collection of vehicles
Figure QLYQS_10
The method comprises the steps of carrying out a first treatment on the surface of the Binary group for vehicle task
Figure QLYQS_12
To represent;
Figure QLYQS_4
the data size representing the task is represented by a size of the data,
Figure QLYQS_8
representing the required CPU cycles;
s22, setting
Figure QLYQS_17
For the length of time slot, the first
Figure QLYQS_19
The time slots are semi-closed intervals
Figure QLYQS_23
The method comprises the steps of carrying out a first treatment on the surface of the Each is discretized by adopting a discretization method
Figure QLYQS_15
The coverage area is divided into the same length
Figure QLYQS_18
Is defined between cells of (a); obtaining
Figure QLYQS_22
The right boundary index of coverage of (c) is
Figure QLYQS_26
The method comprises the steps of carrying out a first treatment on the surface of the Position index of road
Figure QLYQS_14
The method comprises the steps of carrying out a first treatment on the surface of the Vehicle with a vehicle body having a vehicle body support
Figure QLYQS_20
First, the
Figure QLYQS_24
Position of each time slot
Figure QLYQS_28
Tracking through indexes corresponding to the positions of the time slots; vehicle with a vehicle body having a vehicle body support
Figure QLYQS_16
Is the position of (2)
Figure QLYQS_21
Sum speed of
Figure QLYQS_25
Every time according to the following formula
Figure QLYQS_27
Updating once per second:
Figure QLYQS_29
(1)
Figure QLYQS_30
(2)
wherein,,
Figure QLYQS_34
and
Figure QLYQS_37
minimum and maximum speeds of vehicle movement, respectively;
Figure QLYQS_39
for vehicles
Figure QLYQS_33
First, the
Figure QLYQS_35
The amount of speed change of the time slot,
Figure QLYQS_38
for vehicles
Figure QLYQS_41
First, the
Figure QLYQS_31
Acceleration of each time slot, subject to truncation
Figure QLYQS_36
Standard gaussian distribution on;
Figure QLYQS_40
and
Figure QLYQS_42
respectively vehicles
Figure QLYQS_32
A maximum deceleration value and a maximum acceleration value of (a);
s23, if the vehicle
Figure QLYQS_45
Selecting to offload to VEC server, vehicle
Figure QLYQS_46
Is composed of four parts: latency of waiting
Figure QLYQS_48
Transmission delay time
Figure QLYQS_44
Calculating time delay
Figure QLYQS_47
Time delay of switching
Figure QLYQS_49
The method comprises the steps of carrying out a first treatment on the surface of the If the vehicle selection is calculated locally, the vehicle
Figure QLYQS_50
Is the calculated time delay
Figure QLYQS_43
；
Latency of waiting
Figure QLYQS_51
For vehicles
Figure QLYQS_52
Reach to be unloaded
Figure QLYQS_53
The time required is:
Figure QLYQS_54
(3)
transmission signal to noise ratio
Figure QLYQS_55
The calculation method comprises the following steps:
Figure QLYQS_56
(4)
wherein,,
Figure QLYQS_57
for the transmission power of the vehicle,
Figure QLYQS_58
is that
Figure QLYQS_59
To the point of
Figure QLYQS_60
The distance between the two plates is set to be equal,
Figure QLYQS_61
in order to be a path loss index,
Figure QLYQS_62
for the reference channel gain at the reference distance,
Figure QLYQS_63
is additive white noise power;
s24, uplink transmission rate
Figure QLYQS_64
Different transmission models and coding rate calculations are selected by signal-to-noise ratio:
Figure QLYQS_65
(5)
wherein,,
Figure QLYQS_66
for the coding rate of the code-rate,
Figure QLYQS_67
is the uplink bandwidth;
vehicle with a vehicle body having a vehicle body support
Figure QLYQS_68
Data volume transmitted in the first time slot of (a)
Figure QLYQS_69
Is approximately at
Figure QLYQS_70
The average uplink transmission rate over the time slot interval
Figure QLYQS_71
：
Figure QLYQS_72
(6)
Transmission delay time
Figure QLYQS_73
For vehicles
Figure QLYQS_74
Uploading tasks to a server
Figure QLYQS_75
The time required satisfies the following formula:
Figure QLYQS_76
(7)
Figure QLYQS_77
(8)
wherein,,
Figure QLYQS_78
for vehicles
Figure QLYQS_79
Just get into
Figure QLYQS_80
Time slots required for coverage:
Figure QLYQS_81
(9)
the transmission delay must be such that the transmission is completed within the coverage of the server, i.e. the transmission delay cannot exceed the maximum transmission delay that can be uploaded at the corresponding server
Figure QLYQS_82
：
Figure QLYQS_83
(10)
Maximum transmission delay
Figure QLYQS_84
Is the time required from the vehicle entering the server coverage to leaving the server coverage:
Figure QLYQS_85
(11)
s25, calculating time delay as time required by processing tasks; if the vehicle is
Figure QLYQS_86
At the local computing task, the time delay is calculated
Figure QLYQS_87
The method comprises the following steps:
Figure QLYQS_88
(12)
wherein,,
Figure QLYQS_89
for vehicles
Figure QLYQS_90
Is used to calculate the frequency of the local calculation of (a),
Figure QLYQS_91
representing the CPU cycles required for the computing task;
if the vehicle is
Figure QLYQS_92
Selective offloading to a server
Figure QLYQS_93
Calculating, then calculating the time delay
Figure QLYQS_94
The method comprises the following steps:
Figure QLYQS_95
(13)
wherein,,
Figure QLYQS_96
is a server
Figure QLYQS_97
Is a calculated frequency of (2);
s26, when the server processes the task and the vehicle moves out of the coverage area of the server, switching is needed, and the calculation result is transmitted from the current server to the server in the range of the calculated vehicle position and then transmitted to the vehicle; the data quantity of the calculation result is smaller, and the feedback delay is ignored; obtaining the switching time delay
Figure QLYQS_98
The method comprises the following steps:
Figure QLYQS_99
(14)
wherein the method comprises the steps of
Figure QLYQS_100
To unload the server index of the range to which the vehicle position belongs after completion,
Figure QLYQS_101
indexing the transmitted server;
Figure QLYQS_102
the time required for completing one-time switching for two adjacent servers;
Figure QLYQS_103
the value of (2) is found by the following inequality:
Figure QLYQS_104
(15)
wherein,,
Figure QLYQS_105
representing a vehicle
Figure QLYQS_106
From entering the road to the task at the server
Figure QLYQS_107
Calculating the number of time slots required by completion;
the vehicle
Figure QLYQS_108
Selecting to unload to
Figure QLYQS_109
Is not less than a threshold
Figure QLYQS_110
The method comprises the following steps:
Figure QLYQS_111
(16)
s27, if the vehicle selects to calculate locally, the energy consumption of the vehicle
Figure QLYQS_112
To calculate the energy consumption:
Figure QLYQS_113
(17)
wherein,,
Figure QLYQS_114
as a factor of the energy consumption cost of the vehicle,
Figure QLYQS_115
for the energy coefficients specified in the vehicle CPU model,
Figure QLYQS_116
for vehicles
Figure QLYQS_117
Is a local calculation frequency of (2);
if the vehicle is
Figure QLYQS_118
Selecting to unload to
Figure QLYQS_119
Calculating the energy consumption of the vehicle
Figure QLYQS_120
The energy consumption required for transmission:
Figure QLYQS_121
(18)
wherein,,
Figure QLYQS_122
is uplink transmission power;
s28, finally obtaining the vehicle
Figure QLYQS_123
Is of (2)
Figure QLYQS_124
And energy consumption
Figure QLYQS_125
The following expression is present:
Figure QLYQS_126
（19）。
4. the vehicle and edge computing server-based task offload matching method of claim 3, wherein step S3 comprises the steps of:
s31, setting the optimization variable definition of task unloading as
Figure QLYQS_127
The method comprises the steps of carrying out a first treatment on the surface of the Wherein,,
Figure QLYQS_128
representing a vehicle
Figure QLYQS_129
Offloading tasks to
Figure QLYQS_130
Otherwise, the device can be used to determine whether the current,
Figure QLYQS_131
；
s32, converting the task unloading problem into a matching problem between the vehicle and the RSU; setting a vehicle to select only one server, and selecting one server at most simultaneously
Figure QLYQS_132
A server; the total time delay and the total energy consumption are jointly optimized, and the cost function of task calculation is obtained as follows:
Figure QLYQS_133
the method comprises the steps of carrying out a first treatment on the surface of the Wherein the method comprises the steps of
Figure QLYQS_134
Representing time delay
Figure QLYQS_135
The weight of the vehicle is occupied;
s33, modeling the total optimization problem is finally obtained as follows:
Figure QLYQS_136
Figure QLYQS_137
(20)。
5. the vehicle and edge computing server-based task offload matching method of claim 4, wherein step S4 comprises the steps of:
s41, adopting an algorithm based on cost minimization, firstly calculating the vehicle
Figure QLYQS_138
Selective offloading to a server
Figure QLYQS_139
Cost of (2)
Figure QLYQS_140
If the constraint condition in step S33 is not satisfied, then
Figure QLYQS_141
Is infinite;
s42, calculating the priority of each server by the following formula:
Figure QLYQS_142
(21)
wherein,,
Figure QLYQS_143
=1 indicates a vehicle
Figure QLYQS_144
May be offloaded to the server, if the constraints are not met,
Figure QLYQS_145
the method comprises the steps of carrying out a first treatment on the surface of the Pressing on server
Figure QLYQS_146
And (5) ascending order arrangement is carried out to obtain a server priority list.
6. The vehicle and edge computing server-based task offload matching method of claim 5, wherein step S4 further comprises the steps of:
s43, setting cost values which are unloaded to different servers according to selection for each vehicle, and arranging the cost values in ascending order to be used as a preference list of the vehicles;
s44, each vehicle selects the first server in the preference list, sends a matching request, and temporarily divides the vehicles into vehicle sets of the corresponding servers
Figure QLYQS_147
Neutralizing;
processing the matching request according to the order of the server priority list; for each server, if the received matching request exceeds
Figure QLYQS_148
The requests are arranged in an ascending order according to the cost value, and the front is arranged
Figure QLYQS_149
The vehicles corresponding to the matching requests are reserved; for each of the remaining vehicles, continuing to send a matching request to the next server in the preference list until a server is found that does not exceed the capacity limit, and classifying the corresponding vehicle into
Figure QLYQS_150
Neutralizing; when all of the servers have been processed,
Figure QLYQS_151
the final vehicle server match.
7. A vehicle and edge computing server based task offload matching system for implementing the vehicle and edge computing server based task offload matching method of any of claims 1-6, wherein the vehicle and edge computing server based task offload matching system comprises:
the initialization stage module is used for enabling the centralized control center MBS to collect basic information of vehicles and roadside units RSU;
the time delay and energy consumption calculation module is used for calculating time delay and energy consumption between different vehicles and roadside units (RSUs) by taking into consideration random movement of the vehicles and various uploading rate calculation methods;
the model building module is used for building an optimization model which aims at jointly optimizing task unloading time delay and energy consumption and aims at task unloading decision and random movement of the vehicle as constraints;
and the model solving module is used for solving the established model by adopting a matching algorithm based on cost minimization to obtain the final matching condition of the vehicle server.
8. The vehicle and edge computing server based task offload matching system of claim 7, wherein the model building module is specifically as follows:
setting the optimization variables of task offloading to be defined as
Figure QLYQS_152
The method comprises the steps of carrying out a first treatment on the surface of the Wherein,,
Figure QLYQS_153
representing a vehicle
Figure QLYQS_154
Offloading tasks to
Figure QLYQS_155
Otherwise, the device can be used to determine whether the current,
Figure QLYQS_156
；
converting the task unloading problem into a matching problem between the vehicle and the RSU; setting a vehicle to select only one server, and selecting one server at most simultaneously
Figure QLYQS_157
A server; the total time delay and the total energy consumption are jointly optimized, and the cost function of task calculation is obtained as follows:
Figure QLYQS_158
the method comprises the steps of carrying out a first treatment on the surface of the Wherein the method comprises the steps of
Figure QLYQS_159
Representing time delay
Figure QLYQS_160
The weight of the vehicle is occupied;
the final overall optimization problem is modeled as:
Figure QLYQS_161
Figure QLYQS_162
(20)。
CN202211612847.6A 2022-12-15 2022-12-15 Task unloading matching method and system based on vehicle and edge computing server Pending CN116405569A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211612847.6A CN116405569A (en) 2022-12-15 2022-12-15 Task unloading matching method and system based on vehicle and edge computing server

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211612847.6A CN116405569A (en) 2022-12-15 2022-12-15 Task unloading matching method and system based on vehicle and edge computing server

Publications (1)

Publication Number Publication Date
CN116405569A true CN116405569A (en) 2023-07-07

Family

ID=87012904

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211612847.6A Pending CN116405569A (en) 2022-12-15 2022-12-15 Task unloading matching method and system based on vehicle and edge computing server

Country Status (1)

Country Link
CN (1) CN116405569A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117648182A (en) * 2023-11-28 2024-03-05 南京审计大学 Method for processing safety key calculation task by mobile audit equipment
CN117648172A (en) * 2024-01-26 2024-03-05 南京邮电大学 Vehicle-mounted edge calculation scheduling optimization method and system

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117648182A (en) * 2023-11-28 2024-03-05 南京审计大学 Method for processing safety key calculation task by mobile audit equipment
CN117648172A (en) * 2024-01-26 2024-03-05 南京邮电大学 Vehicle-mounted edge calculation scheduling optimization method and system
CN117648172B (en) * 2024-01-26 2024-05-24 南京邮电大学 Vehicle-mounted edge calculation scheduling optimization method and system

Similar Documents

Publication Publication Date Title
CN111786839B (en) Calculation unloading method and system for energy efficiency optimization in vehicle-mounted edge calculation network
CN109068391B (en) Internet of vehicles communication optimization algorithm based on edge calculation and Actor-Critic algorithm
CN109951869B (en) Internet of vehicles resource allocation method based on cloud and mist mixed calculation
CN116405569A (en) Task unloading matching method and system based on vehicle and edge computing server
CN111010684B (en) Internet of vehicles resource allocation method based on MEC cache service
CN110650457B (en) Joint optimization method for task unloading calculation cost and time delay in Internet of vehicles
CN113904947B (en) Vehicle-road cooperative distributed edge computing task unloading and resource allocation method and system
CN111339554B (en) User data privacy protection method based on mobile edge calculation
CN113286329B (en) Communication and computing resource joint optimization method based on mobile edge computing
CN114189892A (en) Cloud-edge collaborative Internet of things system resource allocation method based on block chain and collective reinforcement learning
CN111475274A (en) Cloud collaborative multi-task scheduling method and device
CN112084025A (en) Improved particle swarm algorithm-based fog calculation task unloading time delay optimization method
CN112188627B (en) Dynamic resource allocation strategy based on state prediction
CN115209426B (en) Dynamic deployment method for digital twin servers in edge car networking
CN112860429A (en) Cost-efficiency optimization system and method for task unloading in mobile edge computing system
CN114928611B (en) IEEE802.11p protocol-based energy-saving calculation unloading optimization method for Internet of vehicles
CN116566838A (en) Internet of vehicles task unloading and content caching method with cooperative blockchain and edge calculation
CN115297171A (en) Edge calculation unloading method and system for cellular Internet of vehicles hierarchical decision
CN116916386A (en) Large model auxiliary edge task unloading method considering user competition and load
CN115987375A (en) Power equipment association and resource optimization method, system and medium for converged network
CN113709249B (en) Safe balanced unloading method and system for driving assisting service
CN112929412B (en) Method, device and storage medium for joint bandwidth allocation and data stream unloading based on MEC single-vehicle single-cell
CN116634396A (en) Graph attention-based vehicle networking multi-agent edge computing content caching decision method
CN116744367A (en) Unloading method based on double-layer unloading mechanism and multi-agent algorithm under Internet of vehicles
CN114928893B (en) Architecture based on intelligent reflecting surface and task unloading method

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