CN110012039B - ADMM-based task allocation and power control method in Internet of vehicles - Google Patents

ADMM-based task allocation and power control method in Internet of vehicles Download PDF

Info

Publication number
CN110012039B
CN110012039B CN201810006519.9A CN201810006519A CN110012039B CN 110012039 B CN110012039 B CN 110012039B CN 201810006519 A CN201810006519 A CN 201810006519A CN 110012039 B CN110012039 B CN 110012039B
Authority
CN
China
Prior art keywords
roadside unit
task
user equipment
calculation
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.)
Active
Application number
CN201810006519.9A
Other languages
Chinese (zh)
Other versions
CN110012039A (en
Inventor
周振宇
刘朋矩
许晨
冯俊豪
唐良瑞
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.)
North China Electric Power University
Original Assignee
North China Electric Power 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 North China Electric Power University filed Critical North China Electric Power University
Priority to CN201810006519.9A priority Critical patent/CN110012039B/en
Publication of CN110012039A publication Critical patent/CN110012039A/en
Application granted granted Critical
Publication of CN110012039B publication Critical patent/CN110012039B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46Multiprogramming arrangements
    • G06F9/50Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/01Protocols
    • H04L67/12Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/50Network services
    • H04L67/60Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
    • H04L67/62Establishing a time schedule for servicing the requests
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/50Network services
    • H04L67/60Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
    • H04L67/63Routing a service request depending on the request content or context
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W52/00Power management, e.g. TPC [Transmission Power Control], power saving or power classes
    • H04W52/02Power saving arrangements
    • H04W52/0209Power saving arrangements in terminal devices
    • YGENERAL 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
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00Energy efficient computing, e.g. low power processors, power management or thermal management
    • YGENERAL 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
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00Reducing energy consumption in communication networks
    • Y02D30/70Reducing energy consumption in communication networks in wireless communication networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention relates to a mobile edge calculation scheme in a car networking scene, which optimizes the problems of calculation task allocation and transmission power control of user equipment in a car on the premise of meeting the time delay requirement. The energy loss of the equipment under the weighting of the calculation task allocation rate is taken as an objective function, a data transmission model of the user equipment and the edge calculation node is obtained by using a queuing theory method, and the optimization problem is solved through the iteration of nonlinear fractional optimization and an alternative direction multiplier method. In each round of circulation, the outer layer circulation solves the problem of nonlinear fractional programming, the inner layer circulation updates the initial value and the variable until the iteration result meets the set threshold, the task quantity distribution ratio of each user equipment is determined, and the minimized energy consumption is obtained. The technical scheme provided by the invention can effectively reduce the energy consumption of the user equipment, meet the requirement of time delay and improve the computing capacity of the whole network.

Description

ADMM-based task allocation and power control method in Internet of vehicles
Technical Field
The invention relates to a mobile edge computing scheme in the field of wireless communication, in particular to a task allocation and power control method based on ADMM in the Internet of vehicles.
Background
As a typical application of the internet of things in the field of transportation, the internet of vehicles enables ubiquitous information sharing among vehicles with little or no human intervention, which is important for implementing future intelligent transportation systems. On one hand, the Internet of vehicles can stimulate the rapid development of a series of application programs with strict timeliness requirements in the fields of road safety, travel assistance, automatic driving and the like; on the other hand, rich multimedia internet of things applications such as augmented reality, streaming video, and online games are rapidly developing, resulting in extremely large workload data to be cached and processed, which requires a large amount of computing, communication, and storage resources. In a traditional cloud computing model, the position of a cloud server is far away from a demand side, and the capacity of a return path and a backbone network is limited, so that unpredictable delay is caused, and the reliable service quality and experience quality of the internet of things cannot be guaranteed.
And as a rapid task processing method in the internet of things, the vehicle edge computing VEC expands the computing mode from a remote central distribution framework to a distributed edge server. In the internet of vehicles, computing, communication, and storage resources are distributed close to users and are scattered at the edge of the network. The internet of vehicles may be considered as a beneficial complement to traditional cloud computing. Handling lower computational requirements and severely time-limited tasks at the network edge can eliminate excessive network crossing points, which not only reduces computational response time, but also alleviates the signal congestion problem for backhaul links with limited capabilities. Further, the internet of vehicles transfers the work load with excessive energy consumption to the VEC node with higher computing capacity and continuous energy supply, so that the endurance time of user equipment in the vehicle, such as smart phones and wearable equipment with limited battery capacity, is greatly prolonged. With an appropriate task allocation strategy, the energy consumption of local computation is reduced at the cost of increased energy consumption for data transmission, as well as delays caused by data transmission, workload processing on edge servers, and trans-regional.
Therefore, it is an important issue to implement computation task allocation and power control in VEC scenarios. First, it is difficult to determine an optimal task allocation ratio under different delay constraints due to rapid changes in channel conditions and network topology caused by rapid movement of vehicles. While the vehicle may also leave the service area of the roadside unit during data transmission or task processing. Secondly, task allocation variables of different problems are coupled with each other due to the limited computing capacity of the VEC nodes, and from the perspective of energy efficiency, the task allocation ratio must be optimized jointly with power control. Finally, because the work tasks on the user equipment and the VEC nodes vary randomly, a certain optimal utilization scheme of the computing and communication resources cannot be obtained.
Disclosure of Invention
In order to solve the defects in the prior art, the invention aims to provide a task allocation and power control method based on an ADMM in the internet of vehicles. By using the algorithm of the invention, the tasks to be calculated and the transmission power can be reasonably distributed on the premise of ensuring the time delay limit, and the energy consumption of the user mobile equipment is effectively reduced.
A task allocation and power control method based on an ADMM in the Internet of vehicles comprises the following steps:
1) determining whether data transmission can be completed before the vehicle leaves the roadside unit service area;
2) the energy loss is obtained by optimizing the iterative process of the nonlinear fractional optimization and the alternative direction multiplier method
Figure GDA0002685470600000021
The minimum calculation task distribution ratio and transmission power;
3) the server of the roadside unit calculates the distributed tasks and sends the calculated result to the mobile equipment in the vehicle through the roadside unit under the control of the central controller;
slave user equipment
Figure GDA0002685470600000022
The distribution ratio of the task amount to the roadside unit m obeys the average arrival rate of
Figure GDA0002685470600000023
In the process of determining whether data transmission can be completed before the vehicle leaves the service area of the roadside unit m, the poisson process of (1) further includes:
1) when the vehicle enters the service range of the roadside unit m, the vehicle speed is determined
Figure GDA0002685470600000024
And distance of vehicle from edge of roadside unit
Figure GDA0002685470600000025
Determining maximum tolerated time
Figure GDA0002685470600000026
When data transmission time
Figure GDA0002685470600000027
If the value is less than the predetermined value, data transmission is possible;
2) at data transmission time
Figure GDA0002685470600000028
After the requirement is met, if the execution time of the whole edge calculation process is met
Figure GDA0002685470600000029
Not more than the time when the vehicle leaves the road section
Figure GDA00026854706000000210
According to a certain task distribution ratio
Figure GDA00026854706000000211
Sending the calculation task to a roadside unit;
the above determining process, which distributes the ratio, delay and energy consumption according to the task amount, is composed of a local calculating process and a data transmission and edge calculating process, and further includes:
1) the distributed tasks are firstly transferred to the in-vehicle transponder from the in-vehicle mobile equipment, then the in-vehicle transponder sends the tasks to the roadside unit by using the maximum transmission power, and the whole process is two-hop transmission; the signal-to-noise ratio of two hops is respectively expressed as:
Figure GDA00026854706000000212
Figure GDA00026854706000000213
wherein,
Figure GDA00026854706000000214
and
Figure GDA00026854706000000215
representing the transmission power of the mobile device and the transponder, respectively
Figure GDA00026854706000000216
And
Figure GDA00026854706000000217
representing the channel gain from the mobile device to the repeater and from the repeater to the roadside unit, by N0And (3) representing the unilateral power spectral density of Gaussian white noise, and obtaining a two-hop total signal-to-noise ratio:
Figure GDA0002685470600000031
and then for the transmitted size is
Figure GDA0002685470600000032
When the channel bandwidth is
Figure GDA0002685470600000033
Time of flight, transmission time
Figure GDA0002685470600000034
Obtained by the following formula:
Figure GDA0002685470600000035
2) for locally computed tasks, the time is computed locally
Figure GDA0002685470600000036
Demand for computing resources by a task to be computed
Figure GDA0002685470600000037
Local computing capability of mobile device
Figure GDA0002685470600000038
Occupation ratio of task to be calculated to CPU resource
Figure GDA0002685470600000039
Average arrival rate of
Figure GDA00026854706000000310
And duty ratio distribution
Figure GDA00026854706000000311
And (3) deriving:
Figure GDA00026854706000000312
3) for the tasks calculated by the server distributed to the roadside units, the total arrival rate of the task amount from different mobile devices waiting to be calculated by the server
Figure GDA00026854706000000313
The roadside unit m has c identical servers, each having a computing power of
Figure GDA00026854706000000314
On the basis of an M/M/c queue model and an Erlang formula, obtaining the average processing time of a calculation task in a roadside unit M:
Figure GDA00026854706000000315
wherein
Figure GDA00026854706000000316
Figure GDA00026854706000000317
Due to the limited processing capacity of the roadside unit m, the tasks to be calculated are waiting in the necessary queue, then processed by the roadside unit m and the results are sent to the user equipment
Figure GDA00026854706000000318
The average waiting time at the roadside unit m for each calculation result is therefore:
Figure GDA00026854706000000319
wherein,
Figure GDA00026854706000000320
for the transmission processing speed of the roadside unit m, since the data length of the calculation result is much smaller than the calculation task, the calculation result is transmitted from the roadside unit m to the user equipment
Figure GDA00026854706000000321
The time delay of (2) can be ignored; when preparing to send the calculation result, if the vehicle is
Figure GDA00026854706000000322
Having moved out of the coverage of the roadside unit m, the calculation will first be sent to the central controller and then forwarded to the vehicle
Figure GDA0002685470600000041
The roadside unit m' is located; transmission delay in this process
Figure GDA0002685470600000042
Average latency at the central controller
Figure GDA0002685470600000043
And waiting time at roadside unit m
Figure GDA0002685470600000044
Can be considered asIs constant, so the latency across the zone can be expressed as:
Figure GDA0002685470600000045
4) execution time of whole moving edge calculation process
Figure GDA0002685470600000046
Is provided with
Figure GDA0002685470600000047
User equipment
Figure GDA0002685470600000048
Should include locally calculated energy consumption and energy consumption of transmitted data; definition of
Figure GDA0002685470600000049
For local computation power, which depends on the inherent characteristics of the CPU and the complexity of the workload, can be considered constant during task execution;
get user equipment by
Figure GDA00026854706000000410
Local computing energy consumption of (1):
Figure GDA00026854706000000411
obtaining user equipment by the following formula
Figure GDA00026854706000000412
Energy loss of transmitting data to the in-vehicle transponder:
Figure GDA00026854706000000413
obtaining user equipment by the following formula
Figure GDA00026854706000000414
Total energy loss of (c):
Figure GDA00026854706000000415
the energy consumption optimization scheme is an ADMM-based calculation task allocation and power control scheme, and aims to minimize m within the service range of a roadside unit mkDefining an optimized set of variables for the overall energy consumption of a vehicle
Figure GDA00026854706000000416
Wherein
Figure GDA00026854706000000417
The optimization problem is:
P1:
Figure GDA00026854706000000418
s.t.
Figure GDA00026854706000000419
Figure GDA00026854706000000420
Figure GDA00026854706000000421
Figure GDA00026854706000000422
Figure GDA00026854706000000423
Figure GDA00026854706000000424
C1and C2To limit the arrival rate of the workload
Figure GDA00026854706000000425
And
Figure GDA00026854706000000426
cannot exceed user equipment respectively
Figure GDA00026854706000000427
And the processing rate of roadside units m, C3Ensuring that the transmission power does not exceed the maximum transmission power, C, of the user equipment4And C5Delay limits for the data transmission and task calculation processes, respectively, C6Assigning ratios to tasks
Figure GDA0002685470600000051
The boundary limit of (2);
in P1, because of different user equipments
Figure GDA0002685470600000052
Are coupled, so the optimization objective is inseparable; in order to solve the problem, the method further comprises the following steps:
1) introducing a local copy of the optimal resource allocation strategy; using a new set of variables to represent locally optimized variables, defining
Figure GDA0002685470600000053
And
Figure GDA0002685470600000054
respectively as
Figure GDA0002685470600000055
And
Figure GDA0002685470600000056
the set of local optimization variables is defined as
Figure GDA0002685470600000057
Wherein
Figure GDA0002685470600000058
The suboptimal problem of P1 can be expressed as:
P2:
Figure GDA0002685470600000059
s.t.
Figure GDA00026854706000000510
Figure GDA00026854706000000511
Figure GDA00026854706000000512
Figure GDA00026854706000000513
Figure GDA00026854706000000514
Figure GDA00026854706000000515
2) p2 target function by introducing local variables
Figure GDA00026854706000000516
Can be separated, and the objective function is decomposed into mKCan beSub-problems that are solved in parallel, these decentralized joint optimization problems can be expressed as:
P3:
Figure GDA00026854706000000517
s.t.
Figure GDA00026854706000000518
the objective function P3 remains a non-convex problem, defining the numerator and denominator of P3 as:
Figure GDA00026854706000000519
Figure GDA0002685470600000061
and define
Figure GDA0002685470600000062
As the optimal objective function value for P3:
Figure GDA0002685470600000063
wherein
Figure GDA0002685470600000064
And
Figure GDA0002685470600000065
respectively representing the optimal local computation task distribution ratio and the power control strategy;
3) obtaining an optimal target value according to a nonlinear fractional optimization problem
Figure GDA0002685470600000066
The sufficient requirements of (A) are: if and only if equation
Figure GDA0002685470600000067
It holds that the optimal local optimization variables are obtained by solving the following problem
Figure GDA0002685470600000068
And
Figure GDA0002685470600000069
P3:
Figure GDA00026854706000000610
s.t.
Figure GDA00026854706000000611
4) defining a set of local variables for each user equipment
Figure GDA00026854706000000612
And defines the function:
Figure GDA00026854706000000613
thus, the convex optimization problem with P2 can be expressed as:
P5:
Figure GDA00026854706000000614
s.t.
Figure GDA00026854706000000615
5) defining an optimal set of variables associated with P5
Figure GDA00026854706000000616
During each iteration of the iterative algorithm of step 2) of claim 1, the following problem is solved:
P6:
Figure GDA00026854706000000617
s.t.
Figure GDA0002685470600000071
wherein the optimal solution
Figure GDA0002685470600000072
Obtained in a previous iteration when limiting the condition
Figure GDA0002685470600000073
When satisfied with
Figure GDA0002685470600000074
Is a set of optimal solutions to the sought optimization problem P1;
for an iterative process, a set of lagrange multipliers corresponding to equation P6 is defined
Figure GDA0002685470600000075
Defining the normal ρ to adjust the convergence speed, the augmented lagrange formula of P6 can be expressed as:
Figure GDA0002685470600000076
the iteration process comprises two layers of loops, wherein an outer loop is a nonlinear fractional optimization problem, and n is used for indicating the iteration times; the inner loop is the update of the original variable and the dual variable, and t is used for indicating the iteration number,
further comprising:
1) distribution ratio to work task
Figure GDA0002685470600000077
Transmission power
Figure GDA0002685470600000078
And an optimal solution
Figure GDA0002685470600000079
Initializing and setting a termination condition;
2) updating optimized variable sets
Figure GDA00026854706000000710
Given the optimal solution for the nth outer loop
Figure GDA00026854706000000711
Further obtaining the transmission power of each user equipment
Figure GDA00026854706000000712
Local variables
Figure GDA00026854706000000713
And
Figure GDA00026854706000000714
can be decomposed into m that can be resolved in parallelKSub-questions; calculating user equipment according to
Figure GDA00026854706000000715
Optimal task distribution ratio obtained at t-th inner loop
Figure GDA00026854706000000716
And transmission power
Figure GDA00026854706000000717
Figure GDA00026854706000000718
3) Updating
Figure GDA00026854706000000719
Obtaining the global optimal task distribution ratio at the t +1 th inner circulation according to the following formula
Figure GDA00026854706000000720
Figure GDA00026854706000000721
Obtaining the Lagrange multiplier at the t +1 th internal circulation according to the following formula
Figure GDA00026854706000000722
Figure GDA00026854706000000723
4) Updating an optimal solution
Figure GDA0002685470600000081
In the iteration process of the initial variable and the dual variable of the ADMM, when t tends to be infinitesimal, the convergence conditions of the objective function, the residual convergence and the dual variable convergence are met; obtained when the inner loop of the nth iteration is terminated
Figure GDA0002685470600000082
And
Figure GDA0002685470600000083
the optimal solution for the (n + 1) th iteration
Figure GDA0002685470600000084
Obtained according to the following formula:
Figure GDA0002685470600000085
5) the cycle is terminated; when the nth outer layer cycle is satisfied
Figure GDA0002685470600000086
Then, the optimal task distribution ratio is obtained by the following formula
Figure GDA0002685470600000087
Optimum transmission power
Figure GDA0002685470600000088
And an optimal solution
Figure GDA0002685470600000089
Figure GDA00026854706000000810
Figure GDA00026854706000000811
Figure GDA00026854706000000812
Compared with the closest prior art, the technical scheme provided by the invention has the beneficial effects that:
the invention introduces how to realize a vehicle networking edge calculation method with higher energy efficiency, solves the energy consumption minimization problem through an alternating direction multiplier method and nonlinear fractional optimization, and considers energy consumption including local calculation and data transmission and delay caused by local calculation, data transmission, waiting time at VEC nodes and roadside units and cross-regions. Under the constraint condition of VEC node computing power, a target function in a fractional form and a coupled optimization variable are provided, and an NP problem is formed.
In order to better build a multitask and multi-server computing mode, a queuing theory is introduced. Dynamic transmission models at the user equipment and VEC nodes are derived taking into account queue heterogeneity. And assuming that the workload generated by each user equipment follows a poisson distribution and the service time of any one user equipment and VEC node follows an exponential distribution, the task transmission models at the user equipment and VEC node can be treated as an M/1 queue and an M/c queue, respectively.
Drawings
FIG. 1 is a diagram of a network of vehicles edge computing system provided by the present invention;
FIG. 2 is a graph of normalized energy consumption versus assignment rate for tasks at different powers provided by the present invention;
FIG. 3 is a graph of normalized energy consumption versus transmission power for different task allocation ratios provided by the present invention;
FIG. 4 is a graph of energy consumption versus number of user devices in various algorithms provided by the present invention;
FIG. 5 is a graph of normalized energy consumption versus roadside unit service radius variation at different powers as provided by the present invention;
FIG. 6 is a graph of algorithm convergence versus iteration number provided by the present invention;
fig. 7 is a graph of normalized energy consumption versus task allocation ratio for different numbers of ues according to the present invention.
Detailed Description
The following describes embodiments of the present invention in further detail with reference to the accompanying drawings.
The following description and the drawings sufficiently illustrate specific embodiments of the invention to enable those skilled in the art to practice them. Other embodiments may incorporate structural, logical, electrical, process, and other changes. The examples merely typify possible variations. Individual components and functions are optional unless explicitly required, and the sequence of operations may vary. Portions and features of some embodiments may be included in or substituted for those of others. The scope of embodiments of the invention encompasses the full ambit of the claims, as well as all available equivalents of the claims. Embodiments of the invention may be referred to herein, individually or collectively, by the term "invention" merely for convenience and without intending to voluntarily limit the scope of this application to any single invention or inventive concept if more than one is in fact disclosed.
Example one
The invention simulates the scene of multi-task and multi-server under the scene of Internet of vehicles, and considering the higher moving speed of vehicles, the task to be calculated can not be completely transmitted in the service range of a roadside unit, and the problem of cross-region can also exist when the calculation result is received. By judging whether the tasks can be distributed to the roadside units for calculation and coordinating the distribution ratio and the transmission power of the tasks, the energy consumption of the user equipment can be reduced under the condition of ensuring the time delay requirement. And determining the roadside unit to which the vehicle belongs by regulating and controlling the central controller, and completing the return of the calculation result. It is also desirable to take into account the latency caused by the limited computing and storage capabilities of the wayside units due to the multitasking by multiple users. The system model diagram is shown in fig. 1, and the whole process includes the following contents:
1. and judging whether the data transmission can be completed in the service range of the roadside unit or not. User equipment
Figure GDA0002685470600000091
The task distribution ratio allocated to the roadside unit m obeys the average arrival ratio of
Figure GDA0002685470600000092
Poisson process of vehicle speed
Figure GDA0002685470600000093
Movement at a distance from the edge of the roadside unit
Figure GDA0002685470600000094
Time of day, data transmission time
Figure GDA0002685470600000095
Need not be greater than
Figure GDA0002685470600000096
And the total execution time of the whole edge calculation process
Figure GDA0002685470600000097
Not more than the time when the vehicle leaves the road section
Figure GDA0002685470600000098
Satisfy the aboveOn condition, at task distribution ratio
Figure GDA0002685470600000101
And sending the calculation task to the roadside unit.
2. The time delay in the execution of the arithmetic execution is mainly composed of transmission time, waiting time and calculation time.
1) The calculation task is transmitted from the in-vehicle user equipment to the roadside unit for two-hop transmission. Wherein the signal to noise ratio from the in-vehicle mobile device to the in-vehicle transponder is
Figure GDA0002685470600000102
The signal-to-noise ratio from the in-vehicle transponder up to the roadside unit is
Figure GDA0002685470600000103
So that the total signal-to-noise ratio is
Figure GDA0002685470600000104
The size of the transmitted data packet is
Figure GDA0002685470600000105
Channel bandwidth of
Figure GDA0002685470600000106
Time of flight, transmission time
Figure GDA0002685470600000107
Obtained by the following formula:
Figure GDA0002685470600000108
2) in a ratio of
Figure GDA0002685470600000109
Is locally computed, with a demand for computing resources of
Figure GDA00026854706000001010
Occupancy of CPU resources
Figure GDA00026854706000001011
Local computing capability of user equipment
Figure GDA00026854706000001012
Then the local computation time is obtained:
Figure GDA00026854706000001013
3) in a ratio of
Figure GDA00026854706000001014
Is assigned to roadside unit computation. The roadside unit is equipped with c equivalent computing capabilities of
Figure GDA00026854706000001015
The server of (1). Since there are multiple user equipments transmitting calculation tasks in the service range of a roadside unit m, the total arrival rate is
Figure GDA00026854706000001016
On the basis of an M/M/c queue model and an Erlang formula, obtaining the average calculation time of a task to be calculated in a roadside unit M:
Figure GDA00026854706000001017
wherein
Figure GDA00026854706000001018
4) Due to the limited processing power of the roadside units m, the tasks to be computed must wait in a queue. The transmission processing speed of the roadside unit m is
Figure GDA00026854706000001019
The average waiting time at the roadside unit m for each calculation result is:
Figure GDA00026854706000001020
5) when preparing to send the calculation result, if the vehicle is
Figure GDA0002685470600000111
Having moved out of service of the roadside unit m, the calculation will first be sent to the central controller and then forwarded to the vehicle
Figure GDA0002685470600000112
The roadside unit m' is located. Transmission delay in this process
Figure GDA0002685470600000113
Average latency at the central controller
Figure GDA0002685470600000114
And waiting time at roadside unit m
Figure GDA0002685470600000115
Can be regarded as a constant, since the data length of the calculation result is far less than the calculation task, the calculation result is from the roadside unit m to the user equipment
Figure GDA0002685470600000116
The delay of (2) can be neglected. The latency across a region can therefore be expressed as:
Figure GDA0002685470600000117
6) full execution time of moving edge computation process
Figure GDA0002685470600000118
Can be expressed as:
Figure GDA0002685470600000119
3. in the calculation process, the energy consumption of the user equipment mainly comprises the energy consumption of local calculation and the energy consumption of data transmission.
1) Calculating power locally
Figure GDA00026854706000001110
Given the inherent nature of the CPU and the complexity of the workload, which can be considered as constants during task computation, the energy consumption of the local computation is then:
Figure GDA00026854706000001111
2) user equipment
Figure GDA00026854706000001112
Has a data transmission power of
Figure GDA00026854706000001113
The energy loss of the data sent by the user equipment to the in-vehicle transponder is as follows:
Figure GDA00026854706000001114
3) during the implementation of the edge calculation, the total energy loss of the user equipment is:
Figure GDA00026854706000001115
example II,
The optimization algorithm of the invention is divided into two layers of iteration processes, the outer layer of iteration process solves the problem of nonlinear fractional optimization, and the inner layer of iteration process updates the variables. The goal is to minimize m within the service range of the roadside unit mkOverall energy consumption of the vehicle. The problem is expressed as:
Figure GDA0002685470600000121
s.t.
Figure GDA0002685470600000122
Figure GDA0002685470600000123
Figure GDA0002685470600000124
Figure GDA0002685470600000125
Figure GDA0002685470600000126
Figure GDA0002685470600000127
due to different user equipment
Figure GDA0002685470600000128
Are coupled, so the optimization objectives are not separable. To solve this problem, a local copy of the optimal resource allocation policy is introduced and local optimization variables are defined, making the objective function separable:
Figure GDA0002685470600000129
s.t.
Figure GDA00026854706000001210
Figure GDA00026854706000001211
Figure GDA00026854706000001212
Figure GDA00026854706000001213
Figure GDA00026854706000001214
Figure GDA00026854706000001215
the objective function can thus be decomposed into mKA sub-problem that can be solved in parallel, this problem is a non-convex problem. By further mathematical transformation and defining the value of the objective function as
Figure GDA00026854706000001216
The problem can be translated into a convex optimization problem and can then be optimized in an iterative process. Adding limiting conditions
Figure GDA00026854706000001217
At each iteration thereafter, the following problem
Figure GDA0002685470600000131
Is solved as follows:
when the condition is limited
Figure GDA0002685470600000132
When satisfied, the result is the optimal solution of the optimization problem. By passingSolving the following augmented Lagrange problem to obtain the variables updated by each inner layer iteration and the optimal solution of the outer layer cycle process:
Figure GDA0002685470600000133
after initializing variables, in the iteration process of dual variables, the optimal solution of the outer loop can be obtained after the conditions of target function convergence, residual convergence and dual variable convergence are met, and the optimal solution of the target function is obtained after the set loop termination condition is met
Figure GDA0002685470600000134
And optimal task allocation ratio
Figure GDA0002685470600000135
And optimum transmission power
Figure GDA0002685470600000136
For the present invention, we performed a number of simulation experiments. As shown in fig. 2, with
Figure GDA0002685470600000137
I.e. more tasks are allocated to the edge compute nodes for computation, the energy consumption decreases first and then increases. This is due to the fact that
Figure GDA0002685470600000138
Smaller, data transmissions consume less energy than the local computation, and then consume more energy than the local computation with the data transmissions. Fig. 3 shows that the transmission rate increases with increasing transmission power, whereas the energy consumed for data transmission increases faster than the transmission rate, thus showing a monotonic increase in transmission energy loss. Figure 4 reflects the impact of the number of user equipments on the energy consumption under three different optimization schemes. FIG. 5 shows that as the roadside unit coverage increases, the energy loss gradually decreases and tends to increaseAnd (4) stabilizing. In the process shown in fig. 6, it is shown that the iteration of the algorithm can be rapidly converged in 6-7 iterations, that is, the optimal solution can be rapidly obtained. Fig. 7 is a relationship between energy consumption and task allocation ratio for different numbers of ues. The results are consistent with the conclusions of fig. 2 and 4.
Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art can make modifications and equivalents to the embodiments of the present invention without departing from the spirit and scope of the present invention, which is set forth in the claims of the present application.

Claims (1)

1. A task allocation and power control method based on an ADMM in the Internet of vehicles is characterized by comprising the following steps:
1) determining whether data transmission can be completed before the vehicle leaves the roadside unit service area;
2) the energy loss is obtained by optimizing the iterative process of the nonlinear fractional optimization and the alternative direction multiplier method
Figure FDA0002810302750000011
The minimum calculation task distribution ratio and transmission power;
3) the server of the roadside unit calculates the distributed tasks and sends the calculated result to the mobile equipment in the vehicle through the roadside unit under the control of the central controller;
slave user equipment
Figure FDA0002810302750000012
The distribution ratio of the task amount to the roadside unit m obeys the average arrival rate of
Figure FDA0002810302750000013
Further, in the process of determining whether the data transmission can be completed before the vehicle leaves the service area of the roadside unit m, the process of poissonThe method comprises the following steps:
1.1 vehicle speed when vehicle enters the service area of roadside unit m
Figure FDA0002810302750000014
And distance of vehicle from edge of roadside unit
Figure FDA0002810302750000015
Determining maximum tolerated time
Figure FDA0002810302750000016
When data transmission time
Figure FDA0002810302750000017
When the value is less than the value, data transmission is carried out;
1.2 at data transfer time
Figure FDA0002810302750000018
After the requirement is met, if the execution time of the whole edge calculation process is met
Figure FDA0002810302750000019
Not more than the time when the vehicle leaves the road section
Figure FDA00028103027500000110
According to a certain task distribution ratio
Figure FDA00028103027500000111
Sending the calculation task to a roadside unit;
distributing the ratio according to the task amount, wherein the time delay and the energy consumption are formed by a local calculation process and a data transmission and edge calculation process, and the method further comprises the following steps:
1.21 the distributed task is firstly transferred to the in-vehicle transponder from the in-vehicle mobile device, then the in-vehicle transponder sends the task to the roadside unit by the maximum transmission power, and the whole process is two-hop transmission; the signal-to-noise ratio of two hops is respectively expressed as:
Figure FDA00028103027500000112
Figure FDA00028103027500000113
wherein,
Figure FDA00028103027500000114
and
Figure FDA00028103027500000115
representing the transmission power of the mobile device and the transponder, respectively
Figure FDA00028103027500000116
And
Figure FDA00028103027500000117
representing the channel gain from the mobile device to the repeater and from the repeater to the roadside unit, by N0And (3) representing the unilateral power spectral density of Gaussian white noise, and obtaining a two-hop total signal-to-noise ratio:
Figure FDA00028103027500000118
and then for the transmitted size is
Figure FDA00028103027500000119
When the channel bandwidth is
Figure FDA00028103027500000120
Time of flight, transmission time
Figure FDA00028103027500000121
Obtained by the following formula:
Figure FDA00028103027500000123
1.22 for locally calculated tasks, calculate time locally
Figure FDA00028103027500000122
Demand for computing resources by a task to be computed
Figure FDA0002810302750000021
Local computing capability of mobile device
Figure FDA0002810302750000022
Occupation ratio of task to be calculated to CPU resource
Figure FDA0002810302750000023
Average arrival rate of
Figure FDA0002810302750000024
And duty ratio distribution
Figure FDA0002810302750000025
And (3) deriving:
Figure FDA0002810302750000026
1.23 for the server assigned to the roadside units to perform the calculated tasks, there is a total arrival rate of the amount of tasks from the different mobile devices waiting to be calculated by the server
Figure FDA0002810302750000027
The roadside unit m has c identical servers, each having a computing power of
Figure FDA0002810302750000028
On the basis of an M/M/c queue model and an Erlang formula, obtaining the average processing time of a calculation task in a roadside unit M:
Figure FDA0002810302750000029
wherein
Figure FDA00028103027500000210
Figure FDA00028103027500000211
Due to the limited processing capacity of the roadside unit m, the tasks to be calculated are waiting in the necessary queue, then processed by the roadside unit m and the results are sent to the user equipment
Figure FDA00028103027500000212
The average waiting time at the roadside unit m for each calculation result is therefore:
Figure FDA00028103027500000213
wherein,
Figure FDA00028103027500000214
for the transmission processing speed of the roadside unit m, since the data length of the calculation result is much smaller than the calculation task, the calculation result is transmitted from the roadside unit m to the user equipment
Figure FDA00028103027500000215
The delay of (2) is ignored; when preparing to send the calculation result, if the vehicle is
Figure FDA00028103027500000216
Having moved out of the coverage of the roadside unit m, the calculation will first be sent to the central controller and then forwarded to the vehicle
Figure FDA00028103027500000217
The roadside unit m' is located; transmission delay in this process
Figure FDA00028103027500000218
Average latency at the central controller
Figure FDA00028103027500000219
And waiting time at roadside unit m
Figure FDA00028103027500000220
Considered constant, the time delay across the region is thus expressed as:
Figure FDA00028103027500000221
1.24 execution time for the entire moving edge calculation procedure
Figure FDA00028103027500000222
Is provided with
Figure FDA00028103027500000223
User equipment
Figure FDA00028103027500000224
Should include locally calculated energy consumption and energy consumption of transmitted data; definition of
Figure FDA00028103027500000225
For local calculation of power, dependent on the inherent characteristics of the CPU and the workloadComplexity, treated as a constant during task execution;
get user equipment by
Figure FDA0002810302750000031
Local computing energy consumption of (1):
Figure FDA0002810302750000032
obtaining user equipment by the following formula
Figure FDA0002810302750000033
Energy loss of transmitting data to the in-vehicle transponder:
Figure FDA0002810302750000034
obtaining user equipment by the following formula
Figure FDA0002810302750000035
Total energy loss of (c):
Figure FDA0002810302750000036
the energy consumption optimization scheme is an ADMM-based calculation task allocation and power control scheme, and aims to minimize m within the service range of a roadside unit mkDefining an optimized set of variables for the overall energy consumption of a vehicle
Figure FDA0002810302750000037
Wherein
Figure FDA0002810302750000038
The optimization problem is:
Figure FDA0002810302750000039
s.t.
Figure FDA00028103027500000310
Figure FDA00028103027500000311
Figure FDA00028103027500000312
Figure FDA00028103027500000313
Figure FDA00028103027500000314
Figure FDA00028103027500000315
C1and C2To limit the arrival rate of the workload
Figure FDA00028103027500000316
And
Figure FDA00028103027500000317
cannot exceed user equipment respectively
Figure FDA00028103027500000318
And the processing rate of roadside units m, C3Ensuring that the transmission power does not exceed the maximum transmission power, C, of the user equipment4And C5Delay limits for the data transmission and task calculation processes, respectively, C6Assigning ratios to tasks
Figure FDA00028103027500000319
The boundary limit of (2);
in P1, because of different user equipments
Figure FDA00028103027500000320
Are coupled, so the optimization objective is inseparable; in order to solve the problem, the method further comprises the following steps:
2.1 introducing a local copy of the optimal resource allocation strategy; using a new set of variables to represent locally optimized variables, defining
Figure FDA00028103027500000321
And
Figure FDA00028103027500000322
respectively as
Figure FDA00028103027500000323
And
Figure FDA00028103027500000324
the set of local optimization variables is defined as
Figure FDA00028103027500000325
Wherein
Figure FDA00028103027500000326
The suboptimal problem of P1 is expressed as:
Figure FDA00028103027500000327
s.t.
Figure FDA0002810302750000041
Figure FDA0002810302750000042
Figure FDA0002810302750000043
Figure FDA0002810302750000044
Figure FDA0002810302750000045
Figure FDA0002810302750000046
2.2P 2 target function by introducing local variables
Figure FDA0002810302750000047
Can be separated and decomposed into objective functionsmKThese decentralized joint optimization problems are expressed as:
Figure FDA0002810302750000048
Figure FDA0002810302750000049
the objective function P3 remains a non-convex problem, defining the numerator and denominator of P3 as:
Figure FDA00028103027500000410
Figure FDA00028103027500000411
and define
Figure FDA00028103027500000412
As the optimal objective function value for P3:
Figure FDA00028103027500000413
wherein
Figure FDA00028103027500000414
And
Figure FDA00028103027500000415
respectively representing the optimal local computation task distribution ratio and the power control strategy;
2.3 obtaining the optimal target value according to the nonlinear fractional optimization problem
Figure FDA00028103027500000416
The sufficient requirements of (A) are: if and only if equation
Figure FDA00028103027500000417
It holds that the optimal local optimization variables are obtained by solving the following problem
Figure FDA00028103027500000418
And
Figure FDA00028103027500000419
Figure FDA0002810302750000051
Figure FDA0002810302750000052
2.4 defining a set of local variables for each user Equipment
Figure FDA0002810302750000053
And defines the function:
Figure FDA0002810302750000054
thus, the convex optimization problem with P2 is expressed as:
Figure FDA0002810302750000055
Figure FDA0002810302750000056
2.5 defining the optimal set of variables associated with P5
Figure FDA0002810302750000057
Step 2) in each iteration process of the iterative algorithm, the following problems are solved:
Figure FDA0002810302750000058
Figure FDA0002810302750000059
wherein the optimal solution
Figure FDA00028103027500000510
Obtained in a previous iteration when limiting the condition
Figure FDA00028103027500000511
When satisfied with
Figure FDA00028103027500000512
Is a set of optimal solutions to the sought optimization problem P1;
for an iterative process, define the set of lagrange multipliers μ corresponding to equation P6m={μ1,...,μmk,...,μmKDefining a normal ρ to adjust the convergence speed, the augmented lagrange formula of P6 is expressed as:
Figure FDA00028103027500000513
the iteration process comprises two layers of loops, wherein an outer loop is a nonlinear fractional optimization problem, and n is used for indicating the iteration times; the inner loop is the update of the original variable and the dual variable, and t is used for indicating the iteration number,
further comprising:
2.51 assignment ratio to work task
Figure FDA00028103027500000514
Transmission power
Figure FDA00028103027500000515
And an optimal solution
Figure FDA00028103027500000516
Initializing, setting termination conditions;
2.52 updating optimized variable sets
Figure FDA00028103027500000517
Given the optimal solution for the nth outer loop
Figure FDA00028103027500000518
Further obtaining the transmission power of each user equipment
Figure FDA00028103027500000519
Local variables
Figure FDA00028103027500000520
And
Figure FDA00028103027500000521
is decomposed into m that can be resolved in parallelKSub-questions; calculating user equipment according to
Figure FDA00028103027500000522
Optimal task distribution ratio obtained at t-th inner loop
Figure FDA0002810302750000061
And transmission power
Figure FDA0002810302750000062
Figure FDA0002810302750000063
2.53 update
Figure FDA0002810302750000064
Obtaining the global optimal task distribution ratio at the t +1 th inner circulation according to the following formula
Figure FDA0002810302750000065
Figure FDA0002810302750000066
Obtaining the Lagrange multiplier at the t +1 th internal circulation according to the following formula
Figure FDA0002810302750000067
Figure FDA0002810302750000068
2.54 update the optimal solution
Figure FDA0002810302750000069
In the iteration process of the initial variable and the dual variable of the ADMM, when t tends to be infinitesimal, the convergence conditions of the objective function, the residual convergence and the dual variable convergence are met; obtained when the inner loop of the nth iteration is terminated
Figure FDA00028103027500000610
And
Figure FDA00028103027500000611
the optimal solution for the (n + 1) th iteration
Figure FDA00028103027500000612
Obtained according to the following formula:
Figure FDA00028103027500000613
2.55 the cycle is terminated; when the nth outer layer cycle is satisfied
Figure FDA00028103027500000614
Then, the optimal task is obtained by the following formulaDistribution ratio
Figure FDA00028103027500000615
Optimum transmission power
Figure FDA00028103027500000616
And an optimal solution
Figure FDA00028103027500000617
Figure FDA00028103027500000618
Figure FDA00028103027500000619
Figure FDA00028103027500000620
CN201810006519.9A 2018-01-04 2018-01-04 ADMM-based task allocation and power control method in Internet of vehicles Active CN110012039B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810006519.9A CN110012039B (en) 2018-01-04 2018-01-04 ADMM-based task allocation and power control method in Internet of vehicles

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810006519.9A CN110012039B (en) 2018-01-04 2018-01-04 ADMM-based task allocation and power control method in Internet of vehicles

Publications (2)

Publication Number Publication Date
CN110012039A CN110012039A (en) 2019-07-12
CN110012039B true CN110012039B (en) 2021-01-05

Family

ID=67164159

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810006519.9A Active CN110012039B (en) 2018-01-04 2018-01-04 ADMM-based task allocation and power control method in Internet of vehicles

Country Status (1)

Country Link
CN (1) CN110012039B (en)

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110717300B (en) * 2019-09-27 2022-10-21 云南电网有限责任公司 Edge calculation task allocation method for real-time online monitoring service of power internet of things
CN110856045B (en) * 2019-09-30 2021-12-07 咪咕视讯科技有限公司 Video processing method, electronic device, and storage medium
CN110650457B (en) * 2019-10-14 2020-07-14 河海大学 Joint optimization method for task unloading calculation cost and time delay in Internet of vehicles
CN110798858B (en) * 2019-11-07 2023-04-25 华北电力大学(保定) Distributed task unloading method based on cost efficiency
CN111090522B (en) * 2019-12-13 2023-07-28 南京邮电大学 Scheduling system and decision method for service deployment and migration in mobile edge computing environment
CN111475301B (en) * 2020-04-09 2021-06-11 清华大学 Satellite resource allocation method and device and electronic equipment
CN111585615B (en) * 2020-04-17 2021-06-15 华北电力大学(保定) Direct current energy supply method
CN111935205B (en) * 2020-06-19 2022-08-26 东南大学 Distributed resource allocation method based on alternating direction multiplier method in fog computing network
CN112218351B (en) * 2020-10-27 2022-07-12 中国联合网络通信集团有限公司 Data transmission method, device and system
CN113590708B (en) * 2021-06-17 2024-02-20 贝壳找房(北京)科技有限公司 Adaptive delay consuming method, program product and storage medium
CN114264220B (en) * 2021-12-23 2022-11-22 湖南大学 Method for accurately sensing and detecting relative displacement of mobile equipment
CN115941479A (en) * 2022-12-26 2023-04-07 重庆邮电大学 Cooperative task unloading method based on alternative direction multiplier method in intelligent video monitoring system

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105501216A (en) * 2016-01-25 2016-04-20 合肥工业大学 Internet of vehicles based hierarchical energy management control method for hybrid vehicle
CN106304290A (en) * 2016-08-12 2017-01-04 辛建芳 Internet of Things cooperative node Poewr control method based on N strategy
CN106502098A (en) * 2016-11-19 2017-03-15 合肥工业大学 A kind of optimum speed closed loop fast prediction control method and device based on car networking
CN106936892A (en) * 2017-01-09 2017-07-07 北京邮电大学 A kind of self-organizing cloud multi-to-multi computation migration method and system
CN106972898A (en) * 2017-03-15 2017-07-21 北京大学 Car networking data transmission scheduling method based on channel estimating

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2015065920A1 (en) * 2013-10-28 2015-05-07 Huawei Technologies Co., Ltd System and method for joint power allocation and routing for software defined networks

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105501216A (en) * 2016-01-25 2016-04-20 合肥工业大学 Internet of vehicles based hierarchical energy management control method for hybrid vehicle
CN106304290A (en) * 2016-08-12 2017-01-04 辛建芳 Internet of Things cooperative node Poewr control method based on N strategy
CN106502098A (en) * 2016-11-19 2017-03-15 合肥工业大学 A kind of optimum speed closed loop fast prediction control method and device based on car networking
CN106936892A (en) * 2017-01-09 2017-07-07 北京邮电大学 A kind of self-organizing cloud multi-to-multi computation migration method and system
CN106972898A (en) * 2017-03-15 2017-07-21 北京大学 Car networking data transmission scheduling method based on channel estimating

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Energy Efficient Optimization for Computation;Zheng Chang, Zhenyu Zhou, Tapani Ristaniemi, and Zhisheng Niu;《2017 IEEE Global Communications Conference》;20171208;第1-5页 *
Mobile Edge Computing Potential in;Tarik Taleb, Sunny Dutta, Adlen Ksentini等;《Enabling Mobile and Wireless Technologies for Smart Cities》;20170331;第38-43页 *

Also Published As

Publication number Publication date
CN110012039A (en) 2019-07-12

Similar Documents

Publication Publication Date Title
CN110012039B (en) ADMM-based task allocation and power control method in Internet of vehicles
Luo et al. Adaptive video streaming with edge caching and video transcoding over software-defined mobile networks: A deep reinforcement learning approach
Peng et al. Deep reinforcement learning based resource management for multi-access edge computing in vehicular networks
CN109951821B (en) Task unloading scheme for minimizing vehicle energy consumption based on mobile edge calculation
Yang et al. A parallel intelligence-driven resource scheduling scheme for digital twins-based intelligent vehicular systems
CN109669768B (en) Resource allocation and task scheduling method for edge cloud combined architecture
CN107766135B (en) Task allocation method based on particle swarm optimization and simulated annealing optimization in moving cloud
Wang et al. Regional intelligent resource allocation in mobile edge computing based vehicular network
CN109905470A (en) A kind of expense optimization method for scheduling task based on Border Gateway system
CN113286329B (en) Communication and computing resource joint optimization method based on mobile edge computing
CN114138373B (en) Edge computing task unloading method based on reinforcement learning
CN113254188B (en) Scheduling optimization method and device, electronic equipment and storage medium
Zhang et al. User scheduling and task offloading in multi-tier computing 6G vehicular network
CN114205353B (en) Calculation unloading method based on hybrid action space reinforcement learning algorithm
Zheng et al. Digital twin empowered heterogeneous network selection in vehicular networks with knowledge transfer
CN115297171B (en) Edge computing and unloading method and system for hierarchical decision of cellular Internet of vehicles
Shu et al. Joint offloading strategy based on quantum particle swarm optimization for MEC-enabled vehicular networks
CN113709698A (en) Internet of vehicles fog computing cooperation task unloading control method based on task block
CN116261119A (en) Intelligent collaborative task calculation and on-demand resource allocation method in vehicle-mounted environment
Wang et al. Joint optimization of resource allocation and computation offloading based on game coalition in C-V2X
Li et al. Federated multiagent actor–critic learning task offloading in intelligent logistics
Xiao et al. GTTC: A low-expenditure IoT multi-task coordinated distributed computing framework with fog computing
Zhang et al. Cybertwin-driven multi-intelligent reflecting surfaces aided vehicular edge computing leveraged by deep reinforcement learning
CN116744367A (en) Unloading method based on double-layer unloading mechanism and multi-agent algorithm under Internet of vehicles
Chen et al. A Hybrid Deep Reinforcement Learning Approach for Jointly Optimizing Offloading and Resource Management in Vehicular Networks

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
GR01 Patent grant
GR01 Patent grant