CN111130911A - Calculation unloading method based on mobile edge calculation - Google Patents

Calculation unloading method based on mobile edge calculation Download PDF

Info

Publication number
CN111130911A
CN111130911A CN201911420925.0A CN201911420925A CN111130911A CN 111130911 A CN111130911 A CN 111130911A CN 201911420925 A CN201911420925 A CN 201911420925A CN 111130911 A CN111130911 A CN 111130911A
Authority
CN
China
Prior art keywords
computation
calculation
task
representing
user
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.)
Granted
Application number
CN201911420925.0A
Other languages
Chinese (zh)
Other versions
CN111130911B (en
Inventor
冯维
刘浩
许晓荣
姚英彪
许丹
夏晓威
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 CN201911420925.0A priority Critical patent/CN111130911B/en
Publication of CN111130911A publication Critical patent/CN111130911A/en
Application granted granted Critical
Publication of CN111130911B publication Critical patent/CN111130911B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
    • 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]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • 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
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0896—Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities
    • 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)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention discloses a calculation unloading method based on mobile edge calculation, which is characterized by comprising the following steps: s1, establishing a system model to obtain local calculation time delay and calculation unloading time delay of a user task; s2, calculating the average time delay of the task according to the obtained local calculation time delay and the calculation unloading time delay, and establishing a joint optimization model for calculating unloading, bandwidth and calculation resource distribution according to the calculated average time delay; s3, establishing an original optimization problem model based on the joint optimization model; and S4, solving the original optimization problem in the original optimization problem model by adopting an optimization algorithm of joint calculation unloading, bandwidth and calculation resource allocation. The invention aims at a mobile edge network with limited system resources, performs joint optimization on computation unloading, bandwidth and computation resource allocation to minimize the average time delay of a user for completing computation tasks, and provides a JOCBA algorithm to solve the optimization problem.

Description

Calculation unloading method based on mobile edge calculation
Technical Field
The invention relates to the technical field of mobile edges, in particular to a calculation unloading method based on mobile edge calculation.
Background
With the popularity of mobile devices, new applications that are computationally intensive and energy intensive are emerging (e.g., real-time online gaming, virtual reality, etc.). However, mobile devices typically have limited battery capacity and computing power, which can be a bottleneck limiting the development of mobile applications. To solve this problem, researchers have considered offloading the user's computing tasks to a Mobile Edge Computing (MEC) server to save mobile user energy consumption and improve system performance, but there are still many issues that need to be solved. On one hand, the computational offload model of MEC is divided into two types in total: 1) a binary offload model; 2) the model is partially unloaded. The former is easier to implement in practice and is suitable for simple tasks that are not partitionable, whereas the user's tasks are mostly divisible. The latter takes into account the detachability of tasks, allowing tasks to be computed in parallel on the mobile device and the edge server, but making the optimization problem more complex. On the other hand, considering the limitation of system wireless bandwidth resources and edge server computing resources, how to better distribute these limited resources among all users of the service becomes a difficulty. In addition, the computational offloading decision of the user also affects the allocation of system resources, so the joint optimization of the two becomes a problem to be solved.
Disclosure of Invention
The invention aims to provide a calculation unloading method based on mobile edge calculation aiming at the defects of the prior art, and the method carries out joint optimization on calculation unloading, bandwidth and calculation resource allocation aiming at a mobile edge network with limited system resources so as to minimize the average time delay of a user for completing a calculation task, and provides a JOCBA algorithm to solve the optimization problem.
In order to achieve the purpose, the invention adopts the following technical scheme:
a computation offload method based on mobile edge computation comprises the following steps:
s1, establishing a system model to obtain local calculation time delay and calculation unloading time delay of a user task;
s2, calculating the average time delay of the task according to the obtained local calculation time delay and the calculation unloading time delay, and establishing a joint optimization model for calculating unloading, bandwidth and calculation resource distribution according to the calculated average time delay;
s3, establishing an original optimization problem model based on the joint optimization model;
and S4, solving the original optimization problem in the original optimization problem model by adopting an optimization algorithm of joint calculation unloading, bandwidth and calculation resource allocation.
Further, the local computation time delay of the user task in step S1 is:
Figure BDA0002352362670000021
wherein the content of the first and second substances,
Figure BDA0002352362670000022
representing user ukThe task size for the calculation is in bits, and the offload decisions of all users are represented as a set
Figure BDA0002352362670000023
CkIndicating the number of CPU cycles required to complete each bit;
Figure BDA0002352362670000024
representing user ukThe local computing power of.
Further, the calculation of the offload delay by the user in step S1 is as follows:
Figure BDA0002352362670000025
wherein D iskRepresenting user ukThe task size for local calculation is bit;
Figure BDA0002352362670000026
representing user ukThe task upload rate of (1); pkRepresenting user ukThe transmit power of (a); h iskRepresenting user ukChannel gain with the base station; n is0Power of additive white Gaussian noise, set α { α ═1,α2,L,αKDenotes a system bandwidth allocation policy, αk∈[0,1]A normalized scaling factor representing a system bandwidth;
Figure BDA0002352362670000027
representing the download rate after the task is completed; pMIndicating the transmission power of the base station βkRepresenting the output to input ratio of the computational task; allocation policy for mobile edge compute server computing resources represented as a set
Figure BDA0002352362670000028
Figure BDA0002352362670000029
Indicating assignment of Mobile edge computing MEC Server to user ukThe calculation rate of (c).
Further, the step S2 is specifically to execute the local computation task and the computation offload task at the same time, and the user ukThe time required to complete the task is:
Figure BDA00023523626700000210
wherein, tkIndicating the time required to complete the task.
Further, the step S2 of establishing a joint optimization model for computation offload, bandwidth and computation resource allocation is:
Figure BDA0002352362670000031
Figure BDA0002352362670000032
Figure BDA0002352362670000033
Figure BDA0002352362670000034
Figure BDA0002352362670000035
Figure BDA0002352362670000036
wherein the content of the first and second substances,
Figure BDA0002352362670000037
representing the non-negativity of the system resource allocation;
Figure BDA0002352362670000038
a capacity constraint representing the total bandwidth of the system;
Figure BDA0002352362670000039
representing a capacity constraint of a mobile edge computing MEC server computing resource, wherein FMECIndicating the maximum number of CPU cycles per second provided by the edge server;
Figure BDA00023523626700000310
the local computation part representing the task must not be larger than the original task size.
Further, the original optimization problem in step S3 includes a computation offload decision problem, a joint allocation problem of wireless bandwidth and computation resources.
Further, the computation offload solution problem is:
Figure BDA00023523626700000311
Figure BDA00023523626700000312
according to the calculation unloading decision optimization problem, the final unloading decision of the user is obtained as follows:
Dloc*＝{M1,M2，L，MK} (7)
wherein the content of the first and second substances,
Figure BDA00023523626700000313
Figure BDA0002352362670000041
wherein D isloc*Representing a computational offload policy.
Further, the joint allocation problem of the wireless bandwidth and the computing resource is as follows:
Figure BDA0002352362670000042
Figure BDA0002352362670000043
according to the joint allocation optimization problem of the wireless bandwidth and the computing resource, the obtained final joint allocation strategies of the wireless bandwidth and the computing resource are respectively as follows:
Figure BDA0002352362670000044
Figure BDA0002352362670000045
wherein
Figure BDA0002352362670000046
Figure BDA0002352362670000047
Wherein, αk *A joint allocation policy representing wireless bandwidth;
Figure BDA0002352362670000048
representing a joint allocation policy of computing resources.
Further, the step S4 specifically includes:
s41, a group of wireless bandwidth α and a computing resource F of a mobile edge computing MEC server are selected in advance, and computing unloading D of tasks is obtained according to a formula (8)loc(ii) a And setting the maximum iteration number as tmax；
S42, judging whether the iteration index value t is smaller than the maximum iteration time tmax(ii) a If not, go to step S46; if yes, go to step S43;
s43, calculating an unloading strategy D according to the obtained tasklocAnd according to the formula (12) and the formula (13), obtaining a wireless bandwidth strategy α*And computing resource policy F of mobile edge computing MEC server*；
S44, α obtained by calculation*And F*Substituting into the formula (8) to obtain the calculation unloading strategy
Figure BDA0002352362670000051
S45, updating the iteration index value t, and executing the step S42;
s46, outputting a calculation unloading strategy Dloc*Wireless bandwidth policy α*Computing resource strategy F of mobile edge computing MEC server*。
Compared with the prior art, the invention has the following advantages:
1. the invention considers the difference between different user tasks and the limited calculation capacity of the MEC server, aims to minimize the average time delay of users, and establishes a joint optimization partial task unloading, bandwidth and calculation resource distribution problem model. And the self-adaptive matching of the user task unloading decision, the communication resource and the computing resource is realized in a more realistic scene.
2. In the invention, the optimization problem is a multivariable strongly-coupled non-convex nonlinear programming problem, so that the solution is very difficult. Therefore, the strongly coupled joint optimization problem is decomposed into two sub-problems, and an optimization algorithm (JOCBA) for joint calculation unloading, bandwidth and calculation resource allocation is provided for solving, and the algorithm can be quickly converged.
Drawings
FIG. 1 is a flowchart of a method for offloading computation based on moving edge computation according to an embodiment;
FIG. 2 is a schematic diagram of a system model according to a first embodiment;
fig. 3 is a schematic diagram of convergence of the algorithm according to the first embodiment.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.
The invention aims to overcome the defects of the prior art and provides a calculation unloading method based on mobile edge calculation.
Example one
The present embodiment provides a calculation unloading method based on moving edge calculation, as shown in fig. 1, including the steps of:
s1, establishing a system model to obtain local calculation time delay and calculation unloading time delay of a user task;
s2, calculating the average time delay of the task according to the obtained local calculation time delay and the calculation unloading time delay, and establishing a joint optimization model for calculating unloading, bandwidth and calculation resource distribution according to the calculated average time delay;
s3, establishing an original optimization problem model based on the joint optimization model;
and S4, solving the original optimization problem in the original optimization problem model by adopting an optimization algorithm of joint calculation unloading, bandwidth and calculation resource allocation.
The embodiment is suitable for a mobile edge computing network, and the system model is shown in fig. 2. In the network, there are K users and one base station BS, and all nodes have only one antenna. Wherein the BS is collocated with the MEC server to provide the computation uninstalling service for the user, and the computation data of the user can be arbitrarily divided by bit for partial local computation and partial uninstallation.
In step S1, a system model is built to obtain the local computation delay and the computation offload delay of the user task.
And establishing a system model, and deducing the expressions of the local calculation time delay and the calculation unloading time delay of the user task.
The local calculation time delay of the user task is as follows:
Figure BDA0002352362670000061
wherein the content of the first and second substances,
Figure BDA0002352362670000062
representing user ukThe task size for local computation is in bits, and the offload decisions of all users are represented as a set
Figure BDA0002352362670000063
CkIndicating the number of CPU cycles required to complete each bit;
Figure BDA0002352362670000064
representing user ukI.e., the number of CPU cycles available per second.
For user ukTask unload part of, calculating time
Figure BDA0002352362670000071
Can be divided into three parts: upload time of task
Figure BDA0002352362670000072
Calculation time at MEC
Figure BDA0002352362670000073
Download time after completion of task calculation
Figure BDA0002352362670000074
Thus, the computation offload latency for the user is:
Figure BDA0002352362670000075
wherein D iskRepresenting user ukThe task size for local calculation is bit;
Figure BDA0002352362670000076
representing user ukThe task upload rate of (1); pkRepresenting user ukThe transmit power of (a); h iskRepresenting user ukChannel gain with the base station BS; n is0Power of additive white Gaussian noise, set α { α ═1,α2,L,αKDenotes a system bandwidth allocation policy, αk∈[0,1]A normalized scaling factor representing a system bandwidth;
Figure BDA0002352362670000077
representing the download rate after the task is completed; pMRepresenting the transmission power of the base station BS βkRepresenting the output to input ratio of the computational task; allocation policy for mobile edge compute server computing resources represented as a set
Figure BDA0002352362670000078
Figure BDA0002352362670000079
Indicating assignment of Mobile edge computing MEC Server to user ukIs calculated byThe rate.
In step S2, a joint optimization model of computation offload, bandwidth, and computation resource allocation is established according to the obtained local computation delay and the average delay of computation offload delay computation task, and according to the calculated average delay.
In this embodiment, a joint optimization model of computation offload, bandwidth, and computation resource allocation is established based on a complex partial offload model, with the objective function of minimizing the average time delay for a user to complete a computation task.
Specifically, based on a partial unloading model, considering that the tasks are in a parallel mode, the local computing task and the computing unloading task can be executed simultaneously, so that the user ukThe time required to complete the task is:
Figure BDA00023523626700000710
wherein, tkIndicating the time required to complete the task.
In the embodiment, how to make a user offloading policy and reasonably allocate system resources to minimize an average time delay for a user to complete a task is mainly considered under the condition that the system bandwidth and the computing resources of the MEC server are limited.
Therefore, the joint optimization model for computing offload, bandwidth and computing resource allocation is established as follows:
Figure BDA0002352362670000081
Figure BDA0002352362670000082
Figure BDA0002352362670000083
Figure BDA0002352362670000084
Figure BDA0002352362670000085
Figure BDA0002352362670000086
wherein the content of the first and second substances,
Figure BDA0002352362670000087
representing the non-negativity of the system resource allocation;
Figure BDA0002352362670000088
a capacity constraint representing the total bandwidth of the system;
Figure BDA0002352362670000089
representing a capacity constraint of a mobile edge computing MEC server computing resource, wherein FMECIndicating the maximum number of CPU cycles per second provided by the edge server;
Figure BDA00023523626700000810
the local computation part representing the task must not be larger than the original task size.
In step S3, an original optimization problem model based on the joint optimization model is established.
In this embodiment, since the joint optimization problem is a multivariable strongly-coupled non-convex nonlinear programming problem, the original optimization problem is decomposed into the following two sub-problems: (1) solving the optimal user offloading decision (i.e. computing offloading decision problem) given the system bandwidth and computing resource allocation scheme; (2) a joint resource allocation problem for communication resources and computing resources (i.e., a joint allocation problem for wireless bandwidth and computing resources) is solved based on a given offload decision.
The calculation method of the user optimal unloading decision under the given condition of the resource allocation scheme is as follows:
when both the bandwidth scaling factor α and the computational resource allocation policy F are determined, the computational offload decision sub-problem may be rewritten as follows:
Figure BDA00023523626700000811
Figure BDA00023523626700000812
according to the calculation unloading decision optimization problem, the final unloading decision of the user is obtained as follows:
Dloc*＝{M1,M2，L，MK} (7)
wherein the content of the first and second substances,
Figure BDA0002352362670000091
Figure BDA0002352362670000092
wherein D isloc*Representing a computational offload policy.
The calculation method of the joint resource allocation problem of the communication resources and the computing resources is as follows:
for existing offload decisions, the system resource allocation sub-problem can be re-expressed as:
Figure BDA0002352362670000093
Figure BDA0002352362670000094
according to the joint allocation optimization problem of the wireless bandwidth and the computing resource, the obtained final joint allocation strategies of the wireless bandwidth and the computing resource are respectively as follows:
Figure BDA0002352362670000095
Figure BDA0002352362670000096
wherein
Figure BDA0002352362670000101
Figure BDA0002352362670000102
Wherein, αk *A joint allocation policy representing wireless bandwidth;
Figure BDA0002352362670000103
representing a joint allocation policy of computing resources.
In step S4, an optimization algorithm of joint computation offload, bandwidth and computation resource allocation is used to solve the original optimization problem in the original optimization problem model.
In this embodiment, based on the solution of the sub-problem, an optimization algorithm (JOCBA) combining calculation offloading, bandwidth and calculation resource allocation is proposed to solve the original optimization problem.
The method specifically comprises the following steps:
s41, a group of wireless bandwidth α and a computing resource F of a mobile edge computing MEC server are selected in advance, and computing unloading D of tasks is obtained according to a formula (8)loc(ii) a And setting the maximum iteration number as tmaxAnd initializing an iteration index t as 0;
s42, judging whether the iteration index value t is smaller than the maximum iteration time tmax(ii) a If not, go to step S46; if yes, go to step S43;
s43, calculating an unloading strategy D according to the obtained tasklocAnd according to the formula (12) and the formula (13), obtaining a wireless bandwidth strategy α*And computing resource policy F of mobile edge computing MEC server*；
S44, α obtained by calculation*And F*Substituting into the formula (8) to obtain the calculation unloading strategy Dloc*；
S45, updating the iteration index value t, and executing the step S42;
s46, outputting a calculation unloading strategy Dloc*Wireless bandwidth policy α*Computing resource strategy F of mobile edge computing MEC server*。
Specifically, to solve the joint optimization problem in step S3, a set of α and F satisfying constraint conditions is randomly selected, and the corresponding optimal task offloading policy D is solved according to equation (8)loc. Secondly, the problem of resource allocation is solved firstly. When the offloading decision of all users is determined, the MEC server may collect the task offloading amount of the users, and then calculate N according to equations (14) and (15) respectivelykAnd the value of M. Based on the above analysis solution of the resource allocation sub-problem, N is calculatedkSubstituting the value of M into equation (12) and substituting the value of M into equation (13) can solve the optimal resource allocation scheme, i.e. αk *And
Figure BDA0002352362670000104
since the offloading decision of each user is independent, the user can determine his own computational offloading scheme by only collecting the transmitting power information of the MEC server, and specifically, the user solves the above αk *And
Figure BDA0002352362670000111
substituting into formula (8) to obtain the optimal unloading decision Dloc*＝{M1,M2，L，MK}。
And repeating the solving process of the sub-problems until the user calculation unloading decision and the system resource allocation strategy are kept unchanged, which indicates that the algorithm provided by the invention is converged and the system reaches the optimum.
The present embodiment also performs convergence simulation on the proposed JOCBA algorithm, assuming that the total bandwidth B of the system is 1MHz, and the power n of additive white gaussian noise0The number of users, K25, is-100 dBm, evenly distributed in the coverage area of the base station. Suppose eachUser ukLocal computing speed of
Figure BDA0002352362670000112
Uniformly selected in the set {0.1,0.2, L,1.0} GHz, channel gain hkIndependently of each other, and in the range of [ -50, -30 [ -50-]Uniform distribution is obeyed between dBm. Simultaneous task size Dk∈[10,100]KB, D of different userskIndependent of each other and subject to uniform distribution. Number of CPUs required per bit Ck∈[500,4000]C between different userskIndependent of each other and subject to uniform distribution. The transmission power of the user is set to a constant Pk0.1W, the base station transmitting power is PM＝1W，βk0.2. And is provided with three different FMECThe values were compared and were: fMEC＝50GHz，FMEC＝100GHz，FMEC150GHz as shown in fig. 3. As can be seen from the figure, the average delay of the user is as FMECThe value increases and decreases because of FMECThe larger the value of (a), the more computing resources are allocated to each user, and the smaller the average delay naturally. However, regardless of FMECBy what value, the JOCBA algorithm provided by the invention can quickly converge in iterations within 10 times, which meets the requirement of a user on quick response of the system.
Compared with the prior art, the embodiment has the following advantages:
1. the invention considers the difference between different user tasks and the limited calculation capacity of the MEC server, aims to minimize the average time delay of users, and establishes a joint optimization partial task unloading, bandwidth and calculation resource distribution problem model. And the self-adaptive matching of the user task unloading decision, the communication resource and the computing resource is realized in a more realistic scene.
2. In the invention, the optimization problem is a multivariable strongly-coupled non-convex nonlinear programming problem, so that the solution is very difficult. Therefore, the strongly coupled joint optimization problem is decomposed into two sub-problems, and an optimization algorithm (JOCBA) for joint calculation unloading, bandwidth and calculation resource allocation is provided for solving, and the algorithm can be quickly converged.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims (9)

1. A computation offload method based on mobile edge computation is characterized by comprising the following steps:
s1, establishing a system model to obtain local calculation time delay and calculation unloading time delay of a user task;
s2, calculating the average time delay of the task according to the obtained local calculation time delay and the calculation unloading time delay, and establishing a joint optimization model for calculating unloading, bandwidth and calculation resource distribution according to the calculated average time delay;
s3, establishing an original optimization problem model based on the joint optimization model;
and S4, solving the original optimization problem in the original optimization problem model by adopting an optimization algorithm of joint calculation unloading, bandwidth and calculation resource allocation.
2. The method for offloading computation based on mobile edge computation of claim 1, wherein the local computation latency of the user task in step S1 is:
Figure FDA0002352362660000011
wherein the content of the first and second substances,
Figure FDA0002352362660000012
representing user ukThe task size for calculation is in bitOffload decisions for all users are represented as a set
Figure FDA0002352362660000013
CkIndicating the number of CPU cycles required to complete each bit; f. ofk locRepresenting user ukThe local computing power of.
3. The method for offloading computation based on mobile edge computation of claim 2, wherein the computation offloading time delay of the user in step S1 is:
Figure FDA0002352362660000015
wherein D iskRepresenting user ukThe task size for local calculation is bit;
Figure FDA0002352362660000016
representing user ukThe task upload rate of (1); pkRepresenting user ukThe transmit power of (a); h iskRepresenting user ukChannel gain with the base station; n is0Power of additive white Gaussian noise, set α { α ═1,α2,L,αKDenotes a system bandwidth allocation policy, αk∈[0,1]A normalized scaling factor representing a system bandwidth;
Figure FDA0002352362660000017
representing the download rate after the task is completed; pMIndicating the transmission power of the base station βkRepresenting the output to input ratio of the computational task; allocation policy representation of mobile edge computing MEC server computing resources as a set
Figure FDA0002352362660000018
fk MECIndicating assignment of Mobile edge computing MEC Server to user ukThe calculation rate of (c).
4. The method for computation offloading based on mobile edge computation of claim 3, wherein step S2 is specifically implemented by executing a local computation task and a computation offloading task simultaneously, and user u is configured to execute the local computation task and the computation offloading task simultaneouslykThe time required to complete the task is:
Figure FDA0002352362660000021
wherein, tkIndicating the time required to complete the task.
5. The method of claim 4, wherein the step S2 of building a joint optimization model of computation offload, bandwidth allocation and computation resource allocation comprises:
Figure FDA0002352362660000022
Figure FDA0002352362660000023
Figure FDA0002352362660000024
Figure FDA0002352362660000025
Figure FDA0002352362660000026
Figure FDA0002352362660000027
wherein, αk≥0,
Figure FDA0002352362660000028
fk MEC≥0,
Figure FDA0002352362660000029
Representing the non-negativity of the system resource allocation;
Figure FDA00023523626600000210
a capacity constraint representing the total bandwidth of the system;
Figure FDA00023523626600000211
representing a capacity constraint of a mobile edge computing MEC server computing resource, wherein FMECIndicating the maximum number of CPU cycles per second provided by the edge server;
Figure FDA00023523626600000212
the local computation part representing the task must not be larger than the original task size.
6. The method of claim 5, wherein the original optimization problem in step S3 includes a computation offload decision problem, a joint allocation problem of wireless bandwidth and computation resources.
7. The method of claim 6, wherein the computation offload solution is based on the computation offload solution of the moving edge computation:
Figure FDA00023523626600000213
Figure FDA0002352362660000031
according to the calculation unloading decision optimization problem, the final unloading decision of the user is obtained as follows:
Dloc*＝{M1,M2，L，MK} (7)
wherein the content of the first and second substances,
Figure FDA0002352362660000032
Figure FDA0002352362660000033
wherein D isloc*Representing a computational offload policy.
8. The method of claim 7, wherein the joint allocation of wireless bandwidth and computing resources is subject to the following problems:
Figure FDA0002352362660000034
Figure FDA0002352362660000035
according to the joint allocation optimization problem of the wireless bandwidth and the computing resource, the obtained final joint allocation strategies of the wireless bandwidth and the computing resource are respectively as follows:
Figure FDA0002352362660000036
Figure FDA0002352362660000041
wherein
Figure FDA0002352362660000042
Figure FDA0002352362660000043
Wherein, αk *A joint allocation policy representing wireless bandwidth;
Figure FDA0002352362660000044
representing a joint allocation policy of computing resources.
9. The method for computation offloading based on moving edge computation of claim 8, wherein the step S4 specifically includes:
s41, a group of wireless bandwidth α and a computing resource F of a mobile edge computing MEC server are selected in advance, and computing unloading D of tasks is obtained according to a formula (8)loc(ii) a And setting the maximum iteration number as tmax；
S42, judging whether the iteration index value t is smaller than the maximum iteration time tmax(ii) a If not, go to step S46; if yes, go to step S43;
s43, calculating an unloading strategy D according to the obtained tasklocAnd according to the formula (12) and the formula (13), obtaining a wireless bandwidth strategy α*And computing resource policy F of mobile edge computing MEC server*；
S44, α obtained by calculation*And F*Substituting into the formula (8) to obtain the calculation unloading strategy Dloc*；
S45, updating the iteration index value t, and executing the step S42;
s46, outputting a calculation unloading strategy Dloc*Wireless bandwidth policy α*Computing resource strategy F of mobile edge computing MEC server*。
CN201911420925.0A 2019-12-31 2019-12-31 Calculation unloading method based on mobile edge calculation Active CN111130911B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911420925.0A CN111130911B (en) 2019-12-31 2019-12-31 Calculation unloading method based on mobile edge calculation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911420925.0A CN111130911B (en) 2019-12-31 2019-12-31 Calculation unloading method based on mobile edge calculation

Publications (2)

Publication Number Publication Date
CN111130911A true CN111130911A (en) 2020-05-08
CN111130911B CN111130911B (en) 2022-05-31

Family

ID=70507571

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911420925.0A Active CN111130911B (en) 2019-12-31 2019-12-31 Calculation unloading method based on mobile edge calculation

Country Status (1)

Country Link
CN (1) CN111130911B (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111615129A (en) * 2020-05-29 2020-09-01 南京邮电大学 Resource allocation method in NOMA-based multi-user mobile edge computing system
CN111918311A (en) * 2020-08-12 2020-11-10 重庆邮电大学 Vehicle networking task unloading and resource allocation method based on 5G mobile edge computing
CN112512063A (en) * 2020-11-25 2021-03-16 福州大学 Resource allocation method for unmanned aerial vehicle assisted edge computing based on radio frequency energy collection
CN112887128A (en) * 2021-01-12 2021-06-01 西安电子科技大学 Resource allocation joint optimization method, system, storage medium and application
CN113296842A (en) * 2021-05-14 2021-08-24 上海交通大学 Optimized task unloading scheduling method based on correlation in mobile edge calculation
CN113434206A (en) * 2021-05-26 2021-09-24 国网河南省电力公司信息通信公司 Method and device for calculating and unloading smart power grid
CN113556764A (en) * 2021-07-30 2021-10-26 云南大学 Method and system for determining calculation rate based on mobile edge calculation network
CN113904947A (en) * 2021-11-15 2022-01-07 湖南大学无锡智能控制研究院 Vehicle-road cooperative distributed edge computing task unloading and resource allocation method and system

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10037231B1 (en) * 2017-06-07 2018-07-31 Hong Kong Applied Science and Technology Research Institute Company Limited Method and system for jointly determining computational offloading and content prefetching in a cellular communication system
CN109756912A (en) * 2019-03-25 2019-05-14 重庆邮电大学 A kind of multiple base stations united task unloading of multi-user and resource allocation methods
CN109814951A (en) * 2019-01-22 2019-05-28 南京邮电大学 The combined optimization method of task unloading and resource allocation in mobile edge calculations network
CN109922479A (en) * 2019-01-11 2019-06-21 西安电子科技大学 A kind of calculating task discharging method based on Time-delay Prediction
CN110035410A (en) * 2019-03-07 2019-07-19 中南大学 Federated resource distribution and the method and system of unloading are calculated in a kind of vehicle-mounted edge network of software definition

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10037231B1 (en) * 2017-06-07 2018-07-31 Hong Kong Applied Science and Technology Research Institute Company Limited Method and system for jointly determining computational offloading and content prefetching in a cellular communication system
CN109922479A (en) * 2019-01-11 2019-06-21 西安电子科技大学 A kind of calculating task discharging method based on Time-delay Prediction
CN109814951A (en) * 2019-01-22 2019-05-28 南京邮电大学 The combined optimization method of task unloading and resource allocation in mobile edge calculations network
CN110035410A (en) * 2019-03-07 2019-07-19 中南大学 Federated resource distribution and the method and system of unloading are calculated in a kind of vehicle-mounted edge network of software definition
CN109756912A (en) * 2019-03-25 2019-05-14 重庆邮电大学 A kind of multiple base stations united task unloading of multi-user and resource allocation methods

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
JIANBO DU: "Computation Offloading and Resource Allocation in Vehicular Networks Based on Dual-Side Cost Minimization", 《IEEE》 *

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111615129A (en) * 2020-05-29 2020-09-01 南京邮电大学 Resource allocation method in NOMA-based multi-user mobile edge computing system
CN111918311A (en) * 2020-08-12 2020-11-10 重庆邮电大学 Vehicle networking task unloading and resource allocation method based on 5G mobile edge computing
CN111918311B (en) * 2020-08-12 2022-04-12 重庆邮电大学 Vehicle networking task unloading and resource allocation method based on 5G mobile edge computing
CN112512063A (en) * 2020-11-25 2021-03-16 福州大学 Resource allocation method for unmanned aerial vehicle assisted edge computing based on radio frequency energy collection
CN112512063B (en) * 2020-11-25 2022-04-29 福州大学 Resource allocation method for unmanned aerial vehicle assisted edge computing based on radio frequency energy collection
CN112887128A (en) * 2021-01-12 2021-06-01 西安电子科技大学 Resource allocation joint optimization method, system, storage medium and application
CN113296842A (en) * 2021-05-14 2021-08-24 上海交通大学 Optimized task unloading scheduling method based on correlation in mobile edge calculation
CN113296842B (en) * 2021-05-14 2022-05-06 上海交通大学 Optimized task unloading scheduling method based on correlation in mobile edge calculation
CN113434206A (en) * 2021-05-26 2021-09-24 国网河南省电力公司信息通信公司 Method and device for calculating and unloading smart power grid
CN113556764A (en) * 2021-07-30 2021-10-26 云南大学 Method and system for determining calculation rate based on mobile edge calculation network
CN113556764B (en) * 2021-07-30 2022-05-31 云南大学 Method and system for determining calculation rate based on mobile edge calculation network
CN113904947A (en) * 2021-11-15 2022-01-07 湖南大学无锡智能控制研究院 Vehicle-road cooperative distributed edge computing task unloading and resource allocation method and system

Also Published As

Publication number Publication date
CN111130911B (en) 2022-05-31

Similar Documents

Publication Publication Date Title
CN111130911B (en) Calculation unloading method based on mobile edge calculation
CN110035410B (en) Method for joint resource allocation and computational offloading in software-defined vehicle-mounted edge network
CN109814951B (en) Joint optimization method for task unloading and resource allocation in mobile edge computing network
CN110941667B (en) Method and system for calculating and unloading in mobile edge calculation network
CN111405568B (en) Computing unloading and resource allocation method and device based on Q learning
CN111240701B (en) Task unloading optimization method for end-side-cloud collaborative computing
CN111586720B (en) Task unloading and resource allocation combined optimization method in multi-cell scene
CN109951869B (en) Internet of vehicles resource allocation method based on cloud and mist mixed calculation
CN111132191B (en) Method for unloading, caching and resource allocation of joint tasks of mobile edge computing server
CN109151864B (en) Migration decision and resource optimal allocation method for mobile edge computing ultra-dense network
CN112105062B (en) Mobile edge computing network energy consumption minimization strategy method under time-sensitive condition
CN111405569A (en) Calculation unloading and resource allocation method and device based on deep reinforcement learning
CN110096362B (en) Multitask unloading method based on edge server cooperation
CN113268341B (en) Distribution method, device, equipment and storage medium of power grid edge calculation task
CN110489176B (en) Multi-access edge computing task unloading method based on boxing problem
CN110233755B (en) Computing resource and frequency spectrum resource allocation method for fog computing in Internet of things
CN111800812B (en) Design method of user access scheme applied to mobile edge computing network of non-orthogonal multiple access
CN114189892A (en) Cloud-edge collaborative Internet of things system resource allocation method based on block chain and collective reinforcement learning
CN109729543B (en) High-energy-efficiency computing task unloading method based on data compression
CN111132074A (en) Multi-access edge computing unloading and frame time slot resource allocation method in Internet of vehicles environment
CN113867843B (en) Mobile edge computing task unloading method based on deep reinforcement learning
CN110519849B (en) Communication and computing resource joint allocation method for mobile edge computing
CN112988347B (en) Edge computing unloading method and system for reducing energy consumption and cost sum of system
CN111711962A (en) Cooperative scheduling method for subtasks of mobile edge computing system
CN111935677B (en) Internet of vehicles V2I mode task unloading method and system

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