CN110099384B - Multi-user multi-MEC task unloading resource scheduling method based on edge-end cooperation - Google Patents

Multi-user multi-MEC task unloading resource scheduling method based on edge-end cooperation Download PDF

Info

Publication number
CN110099384B
CN110099384B CN201910337470.XA CN201910337470A CN110099384B CN 110099384 B CN110099384 B CN 110099384B CN 201910337470 A CN201910337470 A CN 201910337470A CN 110099384 B CN110099384 B CN 110099384B
Authority
CN
China
Prior art keywords
task
mec
time
processing
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.)
Active
Application number
CN201910337470.XA
Other languages
Chinese (zh)
Other versions
CN110099384A (en
Inventor
朱晓荣
吴柳青
朱洪波
唐思宇
朱振宇
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.)
Nanjing University of Posts and Telecommunications
Original Assignee
Nanjing University of Posts and Telecommunications
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 Nanjing University of Posts and Telecommunications filed Critical Nanjing University of Posts and Telecommunications
Priority to CN201910337470.XA priority Critical patent/CN110099384B/en
Publication of CN110099384A publication Critical patent/CN110099384A/en
Priority to PCT/CN2020/085258 priority patent/WO2020216135A1/en
Priority to JP2021562894A priority patent/JP7162385B2/en
Application granted granted Critical
Publication of CN110099384B publication Critical patent/CN110099384B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/02—Resource partitioning among network components, e.g. reuse partitioning
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/16—Central resource management; Negotiation of resources or communication parameters, e.g. negotiating bandwidth or QoS [Quality of Service]
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Quality & Reliability (AREA)
  • Mobile Radio Communication Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a multi-user multi-MEC task unloading resource scheduling method based on edge-end cooperation, which comprises the following steps: step 1, task analysis: analyzing the task completion time delay and determining the task completion benefit; step 2, problem formation: forming an optimization problem by taking the overall efficiency of the maximized task completion as a target; step 3, ensuring a steady state: the stability of the task backlog queue is ensured to simplify the problem; step 4, channel allocation: determining optimal channel allocation by a given task unloading allocation strategy; step 5, task scheduling: determining optimal task scheduling by a given channel resource allocation strategy; step 6, joint optimization: and combining the steps 4 and 5 to obtain optimal channel allocation and task scheduling. The invention fully considers the service diversity, carries out priority division on the tasks, improves the user side income to the maximum extent and realizes the multi-user multi-MEC task unloading resource scheduling.

Description

Multi-user multi-MEC task unloading resource scheduling method based on edge-end cooperation
Technical Field
The invention relates to a multi-user multi-MEC task unloading resource scheduling method based on edge-end cooperation, and belongs to the technical field of network resource allocation.
Background
The Mobile Edge Computing (MEC) technology belongs to a distributed Computing, and puts data processing, application program running and even implementation of some functional services on nodes at the Edge of a network. The mobile edge cloud is composed of one or more edge servers, namely servers equipped with computing storage functions on the traditional base station, and the traditional base station is updated to be the mobile edge computing base station. Because the MEC base station is close to the user terminal, the mobile terminal can be helped to process tasks, the task processing time delay is reduced, and the energy consumption of the mobile terminal is reduced. Different from Cloud Computing (Cloud Computing), edge Computing puts down functions such as data processing and the like from a network center to a network edge node, processes data nearby, does not need to upload a large amount of data to a remote core management platform, and can reduce time for data to come to and go from a Cloud end and network bandwidth cost.
The complete edge calculation unloading process is divided into the following three parts: 1) the mobile terminal sends a task unloading request to an MEC service area (including some necessary information of task calculation amount); 2) unloading the task to an MEC server for processing; 3) the MEC server transmits an offload response (including a task processing result, etc.) to the mobile terminal. Time is divided into frames, where the frames are divided into control subframes and calculation offload subframes. In the control sub-frame, control information is exchanged between the MEC server and the mobile terminal to determine an offload schedule. In the calculation of the offload sub-frame, the workload is first sent to the MEC server, which returns the result to the mobile terminal after completing the processing workload. For an edge cloud consisting of a plurality of MEC servers, whether to offload tasks and to which MEC server the tasks are offloaded for processing is still an open question of how to ensure that all mobile terminals and the edge cloud servers can be allocated and efficiently adapt to changes in energy and user requirements.
The task amount of a user, the computing capacity of an MEC server, the computing capacity of a mobile terminal and the occupation condition of channel resources are comprehensively considered, a multi-user multi-MEC task unloading resource scheduling model is established by taking the task completion benefit maximization as a target, and great practical significance is brought to multi-user multi-MEC task unloading resource scheduling based on edge-end cooperation.
Disclosure of Invention
The invention aims to provide a multi-user multi-MEC task unloading resource scheduling method based on edge-end cooperation so as to realize optimal multi-user multi-MEC task unloading resource scheduling.
In order to achieve the purpose, the invention adopts the technical scheme that:
a multi-user multi-MEC task unloading resource scheduling method based on edge-end cooperation comprises the following steps:
step 1, task analysis: analyzing the task completion time delay and determining the task completion benefit;
step 2, problem formation: forming an optimization problem by taking the overall efficiency of the maximized task completion as a target;
step 3, ensuring a steady state: the stability of the task backlog queue is ensured to simplify the problem;
step 4, channel allocation: determining optimal channel allocation by a given task unloading allocation strategy;
step 5, task scheduling: determining optimal task scheduling by a given channel resource allocation strategy;
Step 6, joint optimization: and combining the steps 4 and 5 to obtain optimal channel allocation and task scheduling.
In the step 1, the benefit is determined by two factors: expected profit values generated by the attributes of the tasks and time delay loss generated in the task processing process; the specific steps of the step 1 are as follows:
step 11, considering that in K times, all users have I tasks that need to be executed locally or unloaded to the MEC for processing at time t, where a set J of the MEC is {1, 2, 3
Figure BDA0002039631700000026
Denotes the ith task, I ═ 1, 2, 3, …, I; describing task m by using task model i Size: d i (t) (bit), i.e. task m i The packet size of (d); task m i The workload to be processed is D i (t)X i (CPU cycles) wherein X i Represents the CPU cycle required for processing 1bit data volume; setting the spectrum bandwidth possessed by MEC j to be B j (t) has N j (t) subcarriers, each subcarrier having a bandwidth of
Figure BDA0002039631700000023
According to the Shannon formula, the task m on the subcarrier n i Has a data transmission rate of
Figure BDA0002039631700000024
Task m i Total data transmission rate of
Figure BDA0002039631700000025
Wherein N is j (t) denotes a subcarrierWave number, pi i，n (t) is a channel allocation indicator when pi i，n When (t) is equal to 1, the subcarrier n is assigned to the task m i Unloading is carried out; when pi i，n When (t) is 0, this indicates that subcarrier n is not allocated to task m i Unloading is carried out; p is a radical of formula i Representing task m i The transmitting power of the terminal; h is n，j Denotes the channel gain of the user subcarrier n, sets the mobility of the user not high during the task unloading, so h n，j ＝127+logd i，j ，d i，j Representing a task m i The distance between the user terminal and MEC j; sigma 2 Is the channel noise power; task m i The transfer delay offloaded to MEC j is
Figure BDA0002039631700000031
Step 12, when the user leaves the task in the local processing, the time delay only comprises the task processing time; consider a mobile user terminal u i Has a CPU processing capacity of
Figure BDA0002039631700000032
The local execution latency is
Figure BDA0002039631700000033
Step 13, the time delay for unloading the task to the MEC server is composed of the following three parts: a. b, when a large number of tasks need to be unloaded to the edge cloud for processing, the loads of the MEC servers are exceeded, and the tasks may need to be queued in each MEC server, namely queuing waiting time, c, and task processing time;
wherein the unloading transmission time is
Figure BDA0002039631700000034
A queue waiting time of
Figure BDA0002039631700000035
Setting the CPU processing capability of MEC j to
Figure BDA0002039631700000036
Task m i Is treated for a time of
Figure BDA0002039631700000037
Thus, task m i The total latency to offload to the execution at MEC server j is
Figure BDA0002039631700000038
Step 14, for any MEC server, the task arrival process is modeled as a Bernoulli process, and the task arrival rate of the MEC server j is set to be lambda j (ii) a The number of tasks waiting in the queue is assumed to be the queue state: q j (t) {0, 1, 2, 3. }, queue Q of MEC j j (t) the update formula is
Q j (t+1)＝Q j (t)-V j (t)+A j (t)
Wherein, V j (t) represents the processing speed of the task at MEC j, i.e. V is processed and completed within a time of length 1 at time t j (t) tasks; a. the j (t) indicates whether the task arrives at time t, A j (t) is an element of {0, 1 }; thus, there is Pr { A j (t)＝1}＝λ j And Pr { A j (t)＝0}＝1-λ j (ii) a Based on the litter's law, considering that the execution delay including the queuing delay and the processing delay is proportional to the average queue length of the task buffer in K time points, the average queue length is expressed as follows:
Figure BDA0002039631700000039
step 15, set u i Representing a task m i At the expected profit, L (T), formulated according to its priority i ) Representing a task m i At time T i Internally completing the paid delay loss;
Figure BDA00020396317000000310
c is a proportionality coefficient and is determined according to the sensitivity of a system to time delay, and the larger C is, the larger the time delay loss caused by the time delay is; rho i In order to lose the tolerance of time delay, when the time delay is less than the tolerance, the time delay does not influence the satisfaction of a user, namely the income of the user is not lost, and when the time delay is more than rho i The delay affects the user satisfaction and correspondingly causes delay loss.
In the step 2, a user benefit value is introduced as an index for measuring the system performance, and an optimization problem is established with the aim of maximizing the user-side task completion total benefit within a period of time; the method comprises the following specific steps:
step 21, task m i The off-load to MEC processing yields the benefit of
Figure BDA0002039631700000041
Figure BDA0002039631700000042
Wherein u is i Representing a task m i At the expected profit, L (T), formulated according to its priority i，j (t)) is task m i Latency penalty incurred by offloading to MEC processing;
step 22, task m i Revenue generated by local execution
Figure BDA0002039631700000043
Step 23, the symbols pi are distributed by joint optimization of the subcarriers i，n (t) task assigner s i，j (t) obtaining an optimization problem with the goal of maximizing the total profit for the user-side task completion over a period of time:
P1：
Figure BDA0002039631700000044
s.t.C1：
Figure BDA0002039631700000045
C2：
Figure BDA0002039631700000046
C3：
Figure BDA0002039631700000047
C4：
Figure BDA0002039631700000048
C5：
Figure BDA0002039631700000049
C6：
Figure BDA00020396317000000410
C7：
Figure BDA00020396317000000411
wherein C1 ensures that a task can only be selected to be processed locally or offloaded to a MEC server for execution; c2 ensuring s i，j (t) is a binary variable; c3 ensuring pi i，n (t) is a binary variable; c4 ensures that one subcarrier can be allocated to only one user at most; c5 ensuring that the base station allocates transmission power to users not exceeding the maximum transmission power, p, of the base station max Is the maximum transmit power of the base station; c6 ensuring that unload transfer energy does not exceed task m i Residual energy of the mobile terminal equipment
Figure BDA00020396317000000412
C7 ensuring that task execution latency satisfies maximum latency requirement
Figure BDA00020396317000000413
Expected benefit u per task in the objective function due to optimization problem P1 i Is fixed and does not change with time t, and time delay lossThe loss function L (-) is a linear function, thus resulting in a simplified optimization problem P2:
P2：
Figure BDA0002039631700000051
s.t.C1：
Figure BDA0002039631700000052
C2：
Figure BDA0002039631700000053
C3：
Figure BDA0002039631700000054
C4：
Figure BDA0002039631700000055
C5：
Figure BDA0002039631700000056
C6：
Figure BDA0002039631700000057
C7：
Figure BDA0002039631700000058
in the step 3, the stability of the task backlog queues of each MEC server is ensured, the problem is simplified to solve the optimal task unloading resource scheduling strategy under the steady-state condition based on the lyapunov theory, and the specific steps are as follows:
step 31, setting the task arrival state of each queue as a bernoulli process, and making Θ (t) ═ Q 1 (t)，Q 2 (t)，...，Q j (t)，...，Q J (t)) represents the queue state, Θ (t) being based on the task arrival rate λ j Evolving at a time slot t e {0, 1, 2. }; defining a quadratic lyapunov function:
Figure BDA0002039631700000059
ω j representing a weight set, wherein different weights can cause different queues to have different positions in a task scheduling strategy, and setting all omega j Are both 1; obviously, the Lyapunov function is non-negative if and only if all Θ's are j (t) is 0, L (Θ (t)) equals 0;
step 32, defining the mean of the differences of the quadratic lyapunov functions at a time as the lyapunov drift function Δ (Θ (t)) in order to predict the changes of the respective queue states:
Figure BDA00020396317000000510
wherein the content of the first and second substances,
Figure BDA00020396317000000511
represents the mean of the differences of the quadratic lyapunov functions;
This drift is the expected change in the lyapunov function over a time instant;
step 33, at each time t, observing the current Θ (t) value and taking control action, according to the consistent Θ (t), greedy minimizing drift plus penalty function expectation:
Figure BDA0002039631700000061
step 34, determining a time delay sensitive parameter v 0 V is provided 0 1, the optimization problem P2 is reduced to:
P3：
Figure BDA0002039631700000062
s.t.C1：
Figure BDA0002039631700000063
C2：
Figure BDA0002039631700000064
C3：
Figure BDA0002039631700000065
C4：
Figure BDA0002039631700000066
C5：
Figure BDA0002039631700000067
C6：
Figure BDA0002039631700000068
in the step 4, under the condition that a task unloading allocation strategy is given, converting the optimization problem P3 into a channel resource allocation problem, and solving optimal channel allocation by using a KKT condition; the method comprises the following specific steps:
step 41, setting given task unloading distribution strategy S' i，j (t), the optimization problem P3 is a question of R i，j (t) convex problem, assuming there are l tasks to offload to MEC processing, i.e. S i，j The number of (t) ═ 1 is l, and the optimization objective function is converted into the following formula:
Figure BDA0002039631700000069
f(R ij (t)，S′ ij (t)) is with respect to R ij (t);
step 42, since f (R) i，j (t)，S′ i，j (t)) is a convex function and all constraints are linear functions, so the optimization problem is a convex optimization problem, from the KKT condition, one can obtain information about R i，j (t) optimal solution
Figure BDA00020396317000000610
Step 43, construct the lagrangian function of the optimization problem as follows:
Figure BDA00020396317000000611
wherein, mu i，j Is the undetermined coefficient of each constraint condition;
If R is i，j (t) and μ i，j The KKT condition is satisfied at any point, yielding:
Figure BDA0002039631700000071
by solving the above equation, the optimal R is obtained i，j (t)：
Figure BDA0002039631700000072
From this, a fixed task offload distribution policy S 'can be derived' i，j (t) optimal solution:
Figure BDA0002039631700000073
in the step 5, a given channel resource allocation strategy is set, and the optimization problem P3 is converted into a 0-1 integer programming problem; the method comprises the following specific steps:
step 51, setting a given channel resource allocation strategy, and converting the optimization problem P3 into a 0-1 integer programming problem as follows:
P4：
Figure BDA0002039631700000074
s.t.C1：
Figure BDA0002039631700000075
C2：
Figure BDA0002039631700000076
step 52, at each time t, the task allocation strategy S (t) is solved by taking the total time delay of all task processing as a target, namely, the optimal MEC server corresponding to each task is solved, and each task is unloaded to the optimal MEC j * Server processed time delay
Figure BDA0002039631700000077
Step 53, calculating the time delay T of the task left in the local processing i (t) time delay to offload to MEC processing
Figure BDA0002039631700000078
And (T) i (t) + δ) where δ is the delay tolerance, if
Figure BDA0002039631700000079
The task is at MEC j * And processing, otherwise, locally processing, and updating the task allocation strategy S (t).
The specific steps of the step 6 are as follows:
and 61, obtaining the optimal channel resource allocation under the fixed task unloading allocation according to the step 4.
And step 62, obtaining the optimal task unloading distribution strategy under the fixed channel according to the step 5.
And step 63, repeating the steps 61 and 62 until the optimal channel allocation and task scheduling strategy is obtained.
Has the beneficial effects that: the invention aims to comprehensively consider the task amount of a user, the computing capacity of an MEC server, the computing capacity of a mobile terminal and the occupation condition of channel resources, aim at maximizing task completion benefits, distribute computing resources and channel resources, establish a multi-user multi-MEC computing unloading frame, ensure the stability of an MEC server task extrusion queue and solve under KKT conditions by utilizing the Lyapunov theory, and realize the optimal multi-user multi-MEC task unloading resource scheduling. The invention fully considers the service diversity, carries out priority division on the tasks, improves the user side income to the maximum extent and realizes the multi-user multi-MEC task unloading resource scheduling.
Drawings
FIG. 1 is a diagram of multi-user multi-MEC task offload resource scheduling based on edge-end coordination;
fig. 2 is a schematic diagram of a multi-user multi-MEC computing offloading scenario based on edge-end collaboration.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
The invention comprehensively considers the task amount of the user, the computing capacity of the MEC server, the computing capacity of the mobile terminal and the channel resource occupation condition, aims at maximizing task completion benefits, distributes computing resources and channel resources, establishes a multi-user multi-MEC computing unloading frame, ensures the stability of the MEC server task extrusion queue and the KKT condition by utilizing the Lyapunov theory to solve, and realizes the optimal multi-user multi-MEC task unloading resource scheduling.
The invention discloses a multi-user multi-MEC task unloading resource scheduling method based on edge-end cooperation, which comprises the following steps:
step 1, task analysis: for a mobile terminal, different benefits can be generated when different tasks are completed, and the benefits are mainly determined by two factors: expected profit value generated by the attribute (such as priority) of the task, and delay loss generated in the task processing process; the method comprises the following specific steps:
step 11, as shown in fig. 2, it is considered that, at time t, all users have I tasks that need to be executed locally or unloaded to the MEC for processing at time K, where a set J of the MEC is {1, 2, 3
Figure BDA0002039631700000081
Denotes the ith task, I ═ 1, 2, 3, …, I; describing task m by using task model i Size: d i (t) (bit), i.e. task m i The packet size of (d); task m i The workload to be processed is D i (t)X i (CPU cycles) wherein X i Represents the CPU cycle required for processing 1bit data volume; setting the spectrum bandwidth possessed by MEC j to be B j (t) has N j (t) subcarriers, each subcarrier having a bandwidth of
Figure BDA0002039631700000082
According to the Shannon formula, the task m on the subcarrier n i Has a data transmission rate of
Figure BDA0002039631700000083
Task m i Total data transmission rate of
Figure BDA0002039631700000091
Wherein N is j (t) represents the number of subcarriers,. pi i，n (t) is a channel allocation indicator when pi i，n When (t) is 1, it means that subcarrier n is allocated to task m i Unloading is carried out; when pi i，n When (t) is 0, it indicates that subcarrier n is not allocated to task m i Unloading is carried out; p is a radical of i Representing a task m i The transmitting power of the terminal; h is n，j Denotes the channel gain of the user subcarrier n, sets the mobility of the user not high during the task unloading, so h n，j ＝127+logd i，j ，d i，j Representing a task m i The distance between the user terminal and MEC j; sigma 2 Is the channel noise power; task m i The transfer delay offloaded to MEC j is
Figure BDA0002039631700000092
Step 12, when the user leaves the task in the local processing, the time delay only comprises the task processing time; consider a mobile user terminal u i Has a CPU processing capacity of
Figure BDA0002039631700000093
The local execution latency is
Figure BDA0002039631700000094
Step 13, the time delay for unloading the task to the MEC server is composed of the following three parts: a. b, when a large number of tasks need to be unloaded to the edge cloud for processing, the loads of the MEC servers are exceeded, and the tasks may need to be queued in each MEC server, namely queuing waiting time, c, and task processing time;
wherein the unloading transmission time is
Figure BDA0002039631700000095
A queue waiting time of
Figure BDA0002039631700000096
Setting the CPU processing capability of MEC j to
Figure BDA0002039631700000097
Task m i Is treated for a time of
Figure BDA0002039631700000098
Thus, task m i The total latency to offload to the execution at MEC server j is
Figure BDA0002039631700000099
Step 14, for any MEC server, the task arrival process is modeled as a Bernoulli process, and the task arrival rate of the MEC server j is set to be lambda j (ii) a The number of tasks waiting in the queue is assumed to be the queue state: q j (t) {0, 1, 2, 3. }, queue Q of MEC j j (t) the update formula is
Q j (t+1)＝Q j (t)-V j (t)+A j (t)
Wherein, V j (t) represents the processing speed of the task at MEC j, i.e. at time tThe process in a time of length 1 completes V j (t) tasks; a. the j (t) indicates whether the task arrives at time t, A j (t) is an element of {0, 1 }; thus, there is Pr { A j (t)＝1}＝λ j And Pr { A j (t)＝0}＝1-λ j (ii) a Based on the litter's law, considering that the execution delay including the queuing delay and the processing delay is proportional to the average queue length of the task buffer in K time points, the average queue length is expressed as follows:
Figure BDA0002039631700000101
step 15, set u i Representing a task m i At the expected profit, L (T), formulated according to its priority i ) Representing a task m i At time T i Internally completing the paid delay loss;
Figure BDA0002039631700000102
wherein, C is a proportionality coefficient, which is determined according to the sensitivity of the system to time delay, the larger C is, the larger the time delay loss caused by time delay is, and C is 1 in the invention; rho i In order to lose the tolerance of time delay, when the time delay is less than the tolerance, the time delay does not influence the satisfaction of a user, namely the income of the user is not lost, and when the time delay is more than rho i The delay affects the user satisfaction and correspondingly causes delay loss.
Step 2, problem formation: introducing a user benefit value as an index for measuring the system performance, and establishing an optimization problem by taking the total benefit of user-side task completion in a period of maximization as a target; the method comprises the following specific steps:
step 21, task m i The yield of the offloading to MEC processing is
Figure BDA0002039631700000103
Figure BDA0002039631700000104
Wherein u is i Representing a task m i At the expected profit, L (T), formulated according to its priority i，j (t)) is task m i Latency penalty incurred by offloading to MEC processing;
step 22, task m i Revenue generated by local execution
Figure BDA0002039631700000105
Step 23, the symbols pi are distributed by joint optimization of the subcarriers i，n (t) task assigner s i，j (t) obtaining an optimization problem with the goal of maximizing the total profit for the user-side task completion over a period of time:
P1：
Figure BDA0002039631700000106
s.t.C1：
Figure BDA0002039631700000107
C2：
Figure BDA0002039631700000108
C3：
Figure BDA0002039631700000109
C4：
Figure BDA00020396317000001010
C5：
Figure BDA00020396317000001011
C6：
Figure BDA00020396317000001012
C7：
Figure BDA0002039631700000111
wherein C1 ensures that a task can only be selected to be processed locally or offloaded to a MEC server for execution; c2 ensuring s i，j (t) is a binary variable; c3 ensuring pi i，n (t) is a binary variable; c4 ensures that one subcarrier can be allocated to only one user at most; c5 ensuring that the base station allocates transmission power to users not exceeding the maximum transmission power, p, of the base station max Is the maximum transmit power of the base station; c6 ensuring that unload transfer energy does not exceed task m i Residual energy of the mobile terminal equipment
Figure BDA0002039631700000112
C7 ensuring that task execution latency satisfies maximum latency requirement
Figure BDA0002039631700000113
Expected benefit u per task in the objective function due to optimization problem P1 i Is fixed and does not vary with time t, the delay loss function L (-) is a linear function, thus resulting in a simplified optimization problem P2:
P2：
Figure BDA0002039631700000114
s.t.C1：
Figure BDA0002039631700000115
C2：
Figure BDA0002039631700000116
C3：
Figure BDA0002039631700000117
C4：
Figure BDA0002039631700000118
C5：
Figure BDA0002039631700000119
C6：
Figure BDA00020396317000001110
C7：
Figure BDA00020396317000001111
step 3, ensuring a steady state: the stability of the task backlog queues of each MEC server is guaranteed, and the problem is simplified into the optimal task unloading resource scheduling strategy under the steady-state condition based on the Lyapunov theory; the method comprises the following specific steps:
step 31, setting the task arrival state of each queue as a bernoulli process, and making Θ (t) ═ Q 1 (t)，Q 2 (t)，...，Q j (t)，...，Q J (t)) represents the queue state, Θ (t) being based on the task arrival rate λ j Evolving at a time slot t e {0, 1, 2. }; defining a quadratic lyapunov function:
Figure BDA00020396317000001112
ω j representing a weight set, wherein different weights can cause different queues to have different positions in a task scheduling strategy, and setting all omega j Are both 1; obviously, the Lyapunov function is non-negative if and only if all Θ's are j (t) is 0, L (Θ (t)) equals 0;
step 32, defining the mean of the differences of the quadratic lyapunov functions at a time as the lyapunov drift function Δ (Θ (t)) in order to predict the changes of the respective queue states:
Figure BDA0002039631700000121
Wherein, the first and the second end of the pipe are connected with each other,
Figure BDA0002039631700000122
represents the mean of the differences of the quadratic lyapunov functions;
this drift is the expected change in the lyapunov function over a time instant;
step 33, at each time t, observing the current Θ (t) value and taking control action, according to the consistent Θ (t), greedy minimizing drift plus penalty function expectation:
Figure BDA0002039631700000123
step 34, determining a time delay sensitive parameter v 0 V is provided 0 1, the optimization problem P2 is reduced to:
P3：
Figure BDA0002039631700000124
s.t.C1：
Figure BDA0002039631700000125
C2：
Figure BDA0002039631700000126
C3：
Figure BDA0002039631700000127
C4：
Figure BDA0002039631700000128
C5：
Figure BDA0002039631700000129
C6：
Figure BDA00020396317000001210
step 4, channel allocation: under the condition that a task unloading distribution strategy is given, converting an optimization problem P3 into a channel resource distribution problem, and solving optimal channel distribution by utilizing a KKT condition; the method comprises the following specific steps:
step 41, setting given task unloading distribution strategy S' i，j (t), the optimization problem P3 is a question of R i，j (t) convex problem, assuming there are l tasks to offload to MEC processing, i.e. S i，j The number of (t) ═ 1 is l, and the optimization objective function is converted into the following formula:
Figure BDA00020396317000001211
f(R ij (t)，S′ ij (t)) is with respect to R ij (t);
step 42, since f (R) i，j (t)，S′ i，j (t)) is a convex function and all constraints are linear functions, so the optimization problem is a convex optimization problem, from the KKT condition, one can obtain information about R i，j (t) optimal solution
Figure BDA0002039631700000131
Step 43, construct the lagrangian function of the optimization problem as follows:
Figure BDA0002039631700000132
Wherein, mu i，j Is the undetermined coefficient of each constraint condition;
if R is i，j (t) and μ i，j The KKT condition is satisfied at any point, yielding:
Figure BDA0002039631700000133
by solving the above equation, the optimal R is obtained i，j (t)：
Figure BDA0002039631700000134
From this, a fixed task offload distribution policy S 'can be derived' i，j (t) optimal solution:
Figure BDA0002039631700000135
step 5, task scheduling: given a channel resource allocation strategy, the optimization problem P3 can be converted into a 0-1 integer programming problem; the method comprises the following specific steps:
step 51, setting a given channel resource allocation strategy, and converting the optimization problem P3 into a 0-1 integer programming problem as follows:
P4：
Figure BDA0002039631700000136
s.t.C1：
Figure BDA0002039631700000137
C2：
Figure BDA0002039631700000138
step 52, at each time t, the task allocation strategy S (t) is solved by taking the total time delay of all task processing as a target, namely, the optimal MEC server corresponding to each task is solved, and each task is unloaded to the optimal MEC j * Server processed time delay
Figure BDA0002039631700000139
Step 53, calculating the time delay T of the task left in the local processing i (t) time delay to offload to MEC processing
Figure BDA0002039631700000142
And (T) i (t) + δ) where δ is the delay tolerance, if
Figure BDA0002039631700000141
TaskThen at MEC j * And processing, otherwise, locally processing, and updating the task allocation strategy S (t).
Step 6, joint optimization: alternately iterating step 4 and step 5 until the user benefit in a period of time is maximized, specifically comprising the following steps:
step 61, obtaining the optimal channel resource allocation under the fixed task unloading allocation according to step 4;
Step 62, obtaining an optimal task unloading allocation strategy under the fixed channel according to the step 5;
and step 63, repeating the steps 61 and 62 until the optimal channel allocation and task scheduling strategy is obtained.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.

Claims (2)

1. A multi-user multi-MEC task unloading resource scheduling method based on edge-end cooperation is characterized in that: the method comprises the following steps:
step 1, task analysis: analyzing the task completion time delay and determining the task completion benefit;
in the step 1, the benefit is determined by two factors: expected profit values generated by the attributes of the tasks and time delay loss generated in the task processing process; the specific steps of the step 1 are as follows:
step 11, considering that all users have I tasks to be executed locally or unloaded to the MEC for processing at time t within K times, where a set J of the MEC is {1,2,3, …, J …, J }, and a set of the tasks is
Figure FDA0003670705650000011
Denotes the ith task, I ═ 1,2,3, …, I; describing task m by using task model i Size: d i (t), i.e. task m i The packet size of (d); task m i The workload to be processed is D i (t)X i Wherein X is i To representCPU cycle required for processing 1bit data volume; setting the spectrum bandwidth possessed by MEC j to be B j (t) has N j (t) subcarriers, each subcarrier having a bandwidth of
Figure FDA0003670705650000012
According to the Shannon formula, the task m on the subcarrier n i Has a data transmission rate of
Figure FDA0003670705650000013
Task m i Total data transmission rate of
Figure FDA0003670705650000014
Wherein N is j (t) represents the number of subcarriers,. pi i,n (t) is a channel allocation indicator when pi i,n When (t) is 1, it means that subcarrier n is allocated to task m i Unloading is carried out; when pi i,n When (t) is 0, it indicates that subcarrier n is not allocated to task m i Unloading is carried out; p is a radical of i Representing a task m i The transmitting power of the terminal; h is n,j Denotes the channel gain of the user subcarrier n, sets the mobility of the user not high during the task unloading, so h n,j ＝127+logd i,j ，d i,j Representing a task m i The distance between the user terminal and MEC j; sigma 2 Is the channel noise power; task m i The transfer delay offloaded to MEC j is
Figure FDA0003670705650000015
Step 12, when the user leaves the task in the local processing, the time delay only comprises the task processing time; consider a mobile user terminal u i Has a CPU processing capacity of f i l Then the local execution delay is
Figure FDA0003670705650000016
Step 13, the time delay for unloading the task to the MEC server is composed of the following three parts: a. b, when a large number of tasks need to be unloaded to the edge cloud for processing, the loads of the MEC servers are exceeded, and the tasks may need to be queued at each MEC server, namely queuing waiting time, c, task processing time;
Wherein the unloading transfer time is
Figure FDA0003670705650000021
The queue waiting time is
Figure FDA0003670705650000022
Setting the CPU processing capability of MEC j to
Figure FDA0003670705650000023
Task m i Is treated for a time of
Figure FDA0003670705650000024
Thus, task m i The total latency to offload to the execution at MEC server j is
Figure FDA0003670705650000025
Step 14, for any MEC server, the task arrival process is modeled as a Bernoulli process, and the task arrival rate of the MEC server j is set to be lambda j (ii) a The number of tasks waiting in the queue is assumed to be the queue state: q j (t) {0,1,2,3, … }, queue Q of MEC j j (t) the update formula is
Q j (t+1)＝Q j (t)-V j (t)+A j (t)
Wherein, V j (t) represents the processing speed of the task at MEC j, i.e. V is processed and completed within a time of length 1 at time t j (t)A task; a. the j (t) indicates whether the task arrives at time t, A j (t) is an element of {0,1 }; thus, there is Pr { A j (t)＝1}＝λ j And Pr { A j (t)＝0}＝1-λ j (ii) a Based on the litter's law, considering that the execution delay including the queuing delay and the processing delay is proportional to the average queue length of the task buffer in K time points, the average queue length is expressed as follows:
Figure FDA0003670705650000026
step 15, set u i Representing a task m i At the expected profit, L (T), formulated according to its priority i ) Representing a task m i At time T i Internally completing the paid delay loss;
Figure FDA0003670705650000027
c is a proportionality coefficient and is determined according to the sensitivity of a system to time delay, and the larger C is, the larger the time delay loss caused by the time delay is; rho i In order to lose the tolerance of time delay, when the time delay is less than the tolerance, the time delay does not influence the satisfaction of a user, namely the income of the user is not lost, and when the time delay is more than rho i The time delay affects the satisfaction degree of the user, and correspondingly time delay loss is generated;
step 2, problem formation: forming an optimization problem by taking the overall efficiency of the maximized task completion as a target;
in the step 2, a user benefit value is introduced as an index for measuring the system performance, and an optimization problem is established with the aim of maximizing the user-side task completion total benefit within a period of time; the method comprises the following specific steps:
step 21, task m i The yield of the offloading to MEC processing is
Figure FDA0003670705650000031
Figure FDA0003670705650000032
Wherein u is i Representing a task m i At the expected profit, L (T), formulated according to its priority i,j (t)) is task m i Latency penalty incurred by offloading to MEC processing;
step 22, task m i Revenue generated by local execution
Figure FDA0003670705650000033
Step 23, the symbols pi are distributed by joint optimization of the subcarriers i,n (t) task assigner s i,j (t) obtaining an optimization problem with the goal of maximizing the total profit for the user-side task completion over a period of time:
P1
Figure FDA0003670705650000034
Figure FDA0003670705650000035
Figure FDA0003670705650000036
Figure FDA0003670705650000037
Figure FDA0003670705650000038
Figure FDA0003670705650000039
Figure FDA00036707056500000310
Figure FDA00036707056500000311
wherein C1 ensures that a task can only be selected to be processed locally or offloaded to a MEC server for execution; c2 ensuring s i,j (t) is a binary variable; c3 ensures pi i,n (t) is a binary variable; c4 ensures that one subcarrier can be allocated to only one user at most; c5 ensuring that the base station allocates transmission power to users not exceeding the maximum transmission power, p, of the base station max Is the maximum transmit power of the base station; c6 ensuring that unload transfer energy does not exceed task m i Residual energy of the mobile terminal equipment
Figure FDA00036707056500000312
C7 ensuring that task execution latency satisfies maximum latency requirement
Figure FDA00036707056500000313
Expected benefit u per task in the objective function due to optimization problem P1 i Is fixed and does not vary with time t, the delay loss function L (-) is a linear function, thus resulting in a simplified optimization problem P2:
P2
Figure FDA00036707056500000314
Figure FDA00036707056500000315
Figure FDA00036707056500000316
Figure FDA0003670705650000041
Figure FDA0003670705650000042
Figure FDA0003670705650000043
Figure FDA0003670705650000044
Figure FDA0003670705650000045
step 3, ensuring a steady state: the stability of the task backlog queue is ensured to simplify the problem;
in the step 3, the stability of the task backlog queues of each MEC server is ensured, the problem is simplified to solve the optimal task unloading resource scheduling strategy under the steady-state condition based on the lyapunov theory, and the specific steps are as follows:
step 31, setting the task arrival state of each queue as a bernoulli process, and making Θ (t) ═ Q 1 (t),Q 2 (t),…,Q j (t),…,Q J (t)) represents the queue state, Θ (t) being based on the task arrival rate λ j Evolving at a time slot t e {0,1, 2. }; defining a quadratic lyapunov function:
Figure FDA0003670705650000046
ω j Representing a weight set, wherein different weights can cause different queues to have different positions in a task scheduling strategy, and setting all omega j Are both 1; obviously, the Lyapunov function is non-negative if and only if all Θ's are j (t) is 0, L (Θ (t)) equals 0;
step 32, defining the mean of the differences of the quadratic lyapunov functions at a time as the lyapunov drift function Δ (Θ (t)) in order to predict the changes of the respective queue states:
Figure FDA0003670705650000047
wherein, the first and the second end of the pipe are connected with each other,
Figure FDA0003670705650000048
represents the mean of the differences of the quadratic lyapunov functions;
this drift is the expected change in the lyapunov function over a time instant;
step 33, at each time t, observing the current Θ (t) value and taking control action, according to the consistent Θ (t), greedy minimizing drift plus penalty function expectation:
Figure FDA0003670705650000049
step 34, determining a time delay sensitive parameter v 0 V is provided 0 1, the optimization problem P2 is reduced to:
P3
Figure FDA00036707056500000410
Figure FDA0003670705650000051
Figure FDA0003670705650000052
Figure FDA0003670705650000053
Figure FDA0003670705650000054
Figure FDA0003670705650000055
Figure FDA0003670705650000056
step 4, channel allocation: determining optimal channel allocation by a given task unloading allocation strategy;
in the step 4, under the condition that a task unloading allocation strategy is given, converting the optimization problem P3 into a channel resource allocation problem, and solving optimal channel allocation by using a KKT condition; the method comprises the following specific steps:
Step 41, setting given task unloading distribution strategy S' i,j (t), the optimization problem P3 is a question of R i,j (t) convex problem, assuming there are l tasks to offload to MEC processing, i.e. S i,j The number of (t) ═ 1 is l, and the optimization objective function is converted into the following formula:
Figure FDA0003670705650000057
f(R ij (t),S′ ij (t)) is with respect to R ij (t);
step 42, since f (R) i,j (t),S′ i,j (t)) is a convex function and all constraints are linear functions, so the optimization problem is a convex optimization problem, from the KKT condition, one can obtain information about R i,j (t) optimal solution
Figure FDA0003670705650000058
Step 43, construct the lagrangian function of the optimization problem as follows:
Figure FDA0003670705650000059
wherein, mu i,j Is the undetermined coefficient of each constraint condition;
if R is i,j (t) and μ i,j The KKT condition is satisfied at any point, yielding:
Figure FDA00036707056500000510
by solving the above equation, the optimal R is obtained i,j (t)：
Figure FDA0003670705650000061
From this, a fixed task offload distribution policy S 'can be derived' i , j (t) optimal solution:
Figure FDA0003670705650000062
step 5, task scheduling: determining optimal task scheduling by a given channel resource allocation strategy;
in the step 5, a given channel resource allocation strategy is set, and the optimization problem P3 is converted into a 0-1 integer programming problem; the method comprises the following specific steps:
step 51, setting a given channel resource allocation strategy, and converting the optimization problem P3 into a 0-1 integer programming problem as follows:
P4
Figure FDA0003670705650000063
Figure FDA0003670705650000064
Figure FDA0003670705650000065
step 52, at each time t, the task allocation strategy S (t) is solved by taking the total time delay of all task processing as a target, namely, the optimal MEC server corresponding to each task is solved, and each task is unloaded to the optimal MEC j * Server processed time delay
Figure FDA0003670705650000066
Step 53, calculating the time delay T of the task left in the local processing i (t) time delay to offload to MEC processing
Figure FDA0003670705650000067
And (T) i (t) + δ) where δ is the delay tolerance, if
Figure FDA0003670705650000068
The task is at MEC j * Processing, otherwise, processing locally, and updating a task allocation strategy S (t);
step 6, joint optimization: and combining the steps 4 and 5 to obtain optimal channel allocation and task scheduling.
2. The multi-user multi-MEC task offload resource scheduling method based on edge-to-end coordination according to claim 1, characterized in that: the specific steps of the step 6 are as follows:
step 61, obtaining the optimal channel resource allocation under the fixed task unloading allocation according to step 4;
step 62, obtaining an optimal task unloading distribution strategy under the fixed channel according to step 5;
and step 63, repeating the steps 61 and 62 until the optimal channel allocation and task scheduling strategy is obtained.
CN201910337470.XA 2019-04-25 2019-04-25 Multi-user multi-MEC task unloading resource scheduling method based on edge-end cooperation Active CN110099384B (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CN201910337470.XA CN110099384B (en) 2019-04-25 2019-04-25 Multi-user multi-MEC task unloading resource scheduling method based on edge-end cooperation
PCT/CN2020/085258 WO2020216135A1 (en) 2019-04-25 2020-04-17 Multi-user multi-mec task unloading resource scheduling method based on edge-end collaboration
JP2021562894A JP7162385B2 (en) 2019-04-25 2020-04-17 Multi-User Multi-MEC Task Unload Resource Scheduling Method Based on Edge-Terminal Collaboration

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910337470.XA CN110099384B (en) 2019-04-25 2019-04-25 Multi-user multi-MEC task unloading resource scheduling method based on edge-end cooperation

Publications (2)

Publication Number Publication Date
CN110099384A CN110099384A (en) 2019-08-06
CN110099384B true CN110099384B (en) 2022-07-29

Family

ID=67445680

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910337470.XA Active CN110099384B (en) 2019-04-25 2019-04-25 Multi-user multi-MEC task unloading resource scheduling method based on edge-end cooperation

Country Status (3)

Country Link
JP (1) JP7162385B2 (en)
CN (1) CN110099384B (en)
WO (1) WO2020216135A1 (en)

Families Citing this family (80)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110099384B (en) * 2019-04-25 2022-07-29 南京邮电大学 Multi-user multi-MEC task unloading resource scheduling method based on edge-end cooperation
CN112399388B (en) 2019-08-13 2024-06-14 中兴通讯股份有限公司 Method, device and system for realizing edge calculation
CN110489176B (en) * 2019-08-27 2022-05-10 湘潭大学 Multi-access edge computing task unloading method based on boxing problem
CN110557287B (en) * 2019-09-10 2020-12-25 北京邮电大学 Resource allocation method and device based on Lyapunov optimization
CN110493360B (en) * 2019-09-26 2022-08-19 重庆大学 Mobile edge computing unloading method for reducing system energy consumption under multiple servers
CN110972147B (en) * 2019-10-28 2021-05-25 华东师范大学 Data-driven edge response time optimization method
CN110647403B (en) * 2019-10-31 2023-04-11 桂林电子科技大学 Cloud computing resource allocation method in multi-user MEC system
CN110856183B (en) * 2019-11-18 2021-04-16 南京航空航天大学 Edge server deployment method based on heterogeneous load complementation and application
CN112835773B (en) * 2019-11-25 2023-08-11 合肥欣奕华智能机器股份有限公司 Method and device for evaluating task running instantaneity in operating system
CN110958625B (en) * 2019-12-05 2021-06-04 湖南师范大学 Real-time multi-modal language analysis system and method based on mobile edge intelligence
CN111372268B (en) * 2019-12-24 2022-07-08 中国石油大学(华东) Method for optimizing joint wireless resource management and task unloading in ultra-dense network
CN111107566B (en) * 2019-12-25 2023-09-26 国网冀北电力有限公司唐山供电公司 Unloading method based on collaborative content caching in electric power Internet of things scene
CN110764919A (en) * 2019-12-26 2020-02-07 帷幄匠心科技（杭州）有限公司 Edge computing architecture based on multiple ARM processors
CN110928691B (en) * 2019-12-26 2021-07-09 广东工业大学 Traffic data-oriented edge collaborative computing unloading method
CN111258677B (en) * 2020-01-16 2023-12-15 北京兴汉网际股份有限公司 Task unloading method for heterogeneous network edge computing
CN111401744B (en) * 2020-03-17 2023-04-28 重庆邮电大学 Dynamic task unloading method in uncertainty environment in mobile edge calculation
CN111511028B (en) * 2020-04-13 2023-04-07 北京邮电大学 Multi-user resource allocation method, device, system and storage medium
CN111538595B (en) * 2020-04-23 2023-05-02 电子科技大学中山学院 MEC server task scheduling method based on task cutting
CN112468547B (en) * 2020-11-13 2023-04-07 广州中国科学院沈阳自动化研究所分所 Regional-based industrial edge computing task cloud collaborative unloading method
CN112512056B (en) * 2020-11-14 2022-10-18 北京工业大学 Multi-objective optimization calculation unloading method in mobile edge calculation network
CN112559171B (en) * 2020-12-02 2022-02-08 河海大学 Multi-user task unloading method based on delayed acceptance in mobile edge computing environment
CN112835708B (en) * 2020-12-14 2023-11-10 浙大城市学院 High-quality task allocation and service deployment method for mobile group perception system in edge computing environment
CN112738822A (en) * 2020-12-25 2021-04-30 中国石油大学(华东) NOMA-based security offload and resource allocation method in mobile edge computing environment
CN112596910B (en) * 2020-12-28 2024-02-20 广东电网有限责任公司电力调度控制中心 Cloud computing resource scheduling method in multi-user MEC system
CN112752302A (en) * 2021-01-05 2021-05-04 全球能源互联网研究院有限公司 Power service time delay optimization method and system based on edge calculation
CN112860337B (en) * 2021-01-28 2022-11-29 山东师范大学 Method and system for unloading dependent tasks in multi-access edge computing
CN112988346B (en) * 2021-02-07 2024-02-23 中国联合网络通信集团有限公司 Task processing method, device, equipment and storage medium
CN113452956B (en) * 2021-02-26 2023-03-28 深圳供电局有限公司 Intelligent distribution method and system for power transmission line inspection tasks
CN112996056A (en) * 2021-03-02 2021-06-18 国网江苏省电力有限公司信息通信分公司 Method and device for unloading time delay optimized computing task under cloud edge cooperation
CN113114733B (en) * 2021-03-24 2022-07-08 重庆邮电大学 Distributed task unloading and computing resource management method based on energy collection
CN113032120B (en) * 2021-03-26 2023-06-30 重庆大学 Industrial field big data task cooperative scheduling method based on edge calculation
CN113179296B (en) * 2021-04-08 2022-10-25 中国科学院计算技术研究所 Task unloading method for vehicle-mounted edge computing system
CN113179482B (en) * 2021-04-21 2023-08-22 中南大学 Data transmission resource system and method based on mobile edge calculation
CN113687924B (en) * 2021-05-11 2023-10-20 武汉理工大学 Intelligent dynamic task computing and unloading method based on edge computing system
CN113296941B (en) * 2021-05-12 2023-10-24 广州中国科学院沈阳自动化研究所分所 Cache task scheduling method and device based on polygonal edge calculation
CN113395679B (en) * 2021-05-25 2022-08-05 安徽大学 Resource and task allocation optimization system of unmanned aerial vehicle edge server
CN113434206B (en) * 2021-05-26 2023-03-21 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
CN113377447B (en) * 2021-05-28 2023-03-21 四川大学 Multi-user computing unloading method based on Lyapunov optimization
CN113326126B (en) * 2021-05-28 2024-04-05 湘潭大学 Task processing method, task scheduling method, device and computer equipment
CN113377531B (en) * 2021-06-04 2022-08-26 重庆邮电大学 Mobile edge computing distributed service deployment method based on wireless energy drive
CN113377533B (en) * 2021-06-07 2023-11-21 重庆邮电大学 Dynamic computing unloading and server deployment method in unmanned aerial vehicle assisted mobile edge computing
CN113364626B (en) * 2021-06-08 2022-09-30 国网江苏省电力有限公司 Service placement and bandwidth allocation method for video analysis application facing edge environment
CN113360248B (en) * 2021-06-15 2022-03-15 燕山大学 Game optimization method based on information visible MEC system task scheduling strategy
CN113411394B (en) * 2021-06-16 2022-09-09 广东电网有限责任公司电力调度控制中心 Method and device for processing unloading task in ultra-dense network scene
CN113452625B (en) * 2021-06-28 2022-04-15 重庆大学 Deep reinforcement learning-based unloading scheduling and resource allocation method
CN113423115B (en) * 2021-07-01 2023-07-25 兰州理工大学 Energy collaboration and task unloading optimization method based on edge calculation
CN113590279B (en) * 2021-07-08 2024-05-14 华中科技大学 Task scheduling and resource allocation method for multi-core edge computing server
CN113507712B (en) * 2021-07-09 2023-07-28 中国传媒大学 Resource allocation and calculation task unloading method based on alternate direction multiplier
CN113535393B (en) * 2021-07-12 2023-04-04 湖南工学院 Computing resource allocation method for unloading DAG task in heterogeneous edge computing
CN113347277B (en) * 2021-07-15 2022-10-18 湘潭大学 Unloading distribution method based on task segmentation in edge calculation
CN113727371B (en) * 2021-08-06 2023-04-18 北京科技大学 IRS (inter-Range instrumentation) assisted MEC (Multi-media communication) network wireless and computing resource allocation method and device
CN113590232B (en) * 2021-08-23 2023-04-25 南京信息工程大学 Relay edge network task unloading method based on digital twinning
CN113709883B (en) * 2021-08-30 2023-12-05 北京邮电大学 Dynamic resource allocation method and device under multi-unmanned aerial vehicle auxiliary industrial scene
CN113778682B (en) * 2021-09-13 2023-04-18 电子科技大学 Resource allocation method for MEC system
CN113934534B (en) * 2021-09-27 2022-12-06 苏州大学 Method and system for computing and unloading multi-user sequence tasks under heterogeneous edge environment
CN113905347B (en) * 2021-09-29 2024-04-16 华北电力大学 Cloud edge end cooperation method for air-ground integrated power Internet of things
CN114051266B (en) * 2021-11-08 2024-01-12 首都师范大学 Wireless body area network task unloading method based on mobile cloud-edge calculation
CN114051205B (en) * 2021-11-08 2022-09-13 南京大学 Edge optimization method based on reinforcement learning dynamic multi-user wireless communication scene
CN114301907B (en) * 2021-11-18 2023-03-14 北京邮电大学 Service processing method, system and device in cloud computing network and electronic equipment
CN114138373B (en) * 2021-12-07 2023-10-24 吉林大学 Edge computing task unloading method based on reinforcement learning
CN114143355B (en) * 2021-12-08 2022-08-30 华北电力大学 Low-delay safety cloud side end cooperation method for power internet of things
CN114172809B (en) * 2021-12-13 2023-10-03 重庆邮电大学 Video computing cloud edge collaborative task scheduling method based on target tracking
CN114189521B (en) * 2021-12-15 2024-01-26 福州大学 Method for collaborative computing offloading in F-RAN architecture
CN114448991B (en) * 2021-12-28 2022-10-21 西安电子科技大学 Multi-edge server selection method, system, medium, device and terminal
CN114423021B (en) * 2022-01-20 2023-05-26 重庆邮电大学 Service-aware resource optimization method
CN114461299B (en) * 2022-01-26 2023-06-06 中国联合网络通信集团有限公司 Unloading decision determining method and device, electronic equipment and storage medium
CN114650567B (en) * 2022-03-17 2024-04-23 江苏科技大学 Unmanned aerial vehicle auxiliary V2I network task unloading method
CN114693141B (en) * 2022-04-06 2022-10-25 长春理工大学 Transformer substation inspection method based on end edge cooperation
CN114866548B (en) * 2022-04-26 2022-12-23 中南大学 Task unloading method based on mobile fog calculation
CN114844900B (en) * 2022-05-05 2022-12-13 中南大学 Edge cloud resource cooperation method based on uncertain demand
CN114968404B (en) * 2022-05-24 2023-11-17 武汉大学 Distributed unloading method for computing tasks of location privacy protection
CN115022331B (en) * 2022-05-30 2024-05-14 中国电信股份有限公司 Edge computing resource allocation method and device, storage medium and electronic equipment
CN115022893A (en) * 2022-05-31 2022-09-06 福州大学 Resource allocation method for minimizing total computation time in multi-task edge computing system
CN115277789B (en) * 2022-08-26 2024-03-26 中国长江三峡集团有限公司 Safety protection system and method for cascade hydropower station
CN115665805A (en) * 2022-12-06 2023-01-31 北京交通大学 Point cloud analysis task-oriented edge computing resource scheduling method and system
CN116320354B (en) * 2023-01-16 2023-09-29 浙江大学 360-degree virtual reality video user access control system and control method
CN115858048B (en) * 2023-03-03 2023-04-25 成都信息工程大学 Hybrid critical task oriented dynamic arrival edge unloading method
CN116366661A (en) * 2023-06-02 2023-06-30 江西师范大学 Collaborative edge user allocation method based on blockchain and auction theory
CN116980982B (en) * 2023-09-20 2024-01-23 山东高速信息集团有限公司 Task processing method and system under flexible networking architecture
CN117492856A (en) * 2023-10-17 2024-02-02 南昌大学 Low-delay edge computing and unloading method for credit assessment in financial Internet of things

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107995660A (en) * 2017-12-18 2018-05-04 重庆邮电大学 Support Joint Task scheduling and the resource allocation methods of D2D- Edge Servers unloading
CN109582448A (en) * 2018-10-17 2019-04-05 中国电子科技集团公司第二十八研究所 A kind of edge calculations method for scheduling task towards criticality and timeliness
CN109656703A (en) * 2018-12-19 2019-04-19 重庆邮电大学 A kind of mobile edge calculations auxiliary vehicle task discharging method

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7085837B2 (en) * 2001-12-04 2006-08-01 International Business Machines Corporation Dynamic resource allocation using known future benefits
US10440096B2 (en) * 2016-12-28 2019-10-08 Intel IP Corporation Application computation offloading for mobile edge computing
CN109067842B (en) * 2018-07-06 2020-06-26 电子科技大学 Calculation task unloading method facing Internet of vehicles
CN110099384B (en) * 2019-04-25 2022-07-29 南京邮电大学 Multi-user multi-MEC task unloading resource scheduling method based on edge-end cooperation

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107995660A (en) * 2017-12-18 2018-05-04 重庆邮电大学 Support Joint Task scheduling and the resource allocation methods of D2D- Edge Servers unloading
CN109582448A (en) * 2018-10-17 2019-04-05 中国电子科技集团公司第二十八研究所 A kind of edge calculations method for scheduling task towards criticality and timeliness
CN109656703A (en) * 2018-12-19 2019-04-19 重庆邮电大学 A kind of mobile edge calculations auxiliary vehicle task discharging method

Also Published As

Publication number Publication date
JP2022522057A (en) 2022-04-13
WO2020216135A1 (en) 2020-10-29
CN110099384A (en) 2019-08-06
JP7162385B2 (en) 2022-10-28

Similar Documents

Publication Publication Date Title
CN110099384B (en) Multi-user multi-MEC task unloading resource scheduling method based on edge-end cooperation
Mao et al. Energy efficiency and delay tradeoff for wireless powered mobile-edge computing systems with multi-access schemes
Hussein et al. Efficient task offloading for IoT-based applications in fog computing using ant colony optimization
CN110113195B (en) Method for joint unloading judgment and resource allocation in mobile edge computing system
CN108809695B (en) Distributed uplink unloading strategy facing mobile edge calculation
CN110234127B (en) SDN-based fog network task unloading method
Chen et al. Multiuser computation offloading and resource allocation for cloud–edge heterogeneous network
CN113810233B (en) Distributed computation unloading method based on computation network cooperation in random network
CN113286329B (en) Communication and computing resource joint optimization method based on mobile edge computing
CN112235836A (en) Industrial edge network system architecture and resource scheduling method
CN113535393B (en) Computing resource allocation method for unloading DAG task in heterogeneous edge computing
Zhang et al. Theoretical analysis on edge computation offloading policies for IoT devices
CN112512065B (en) Method for unloading and migrating under mobile awareness in small cell network supporting MEC
CN110809291B (en) Double-layer load balancing method of mobile edge computing system based on energy acquisition equipment
CN113641417B (en) Vehicle security task unloading method based on branch-and-bound method
CN109743217B (en) Self-adaptive resource adjusting method based on SVRA algorithm
CN115629873A (en) System and method for controlling unloading of vehicle-road cloud cooperative tasks and stability of task queue
CN110187964A (en) The deadline minimizes mist computation migration method in scenes of internet of things
CN112860429A (en) Cost-efficiency optimization system and method for task unloading in mobile edge computing system
CN115033359A (en) Internet of things agent multi-task scheduling method and system based on time delay control
CN113821346B (en) Edge computing unloading and resource management method based on deep reinforcement learning
CN115865930A (en) MEC dynamic adjustment method based on 5G power Internet of things
CN116366576A (en) Method, device, equipment and medium for scheduling computing power network resources
CN111611069B (en) Multi-type task migration method among multiple data centers
CN113190342A (en) Method and system architecture for multi-application fine-grained unloading of cloud-edge cooperative network

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