CN109857546A - The mobile edge calculations discharging method of multiserver and device based on Lyapunov optimization - Google Patents
The mobile edge calculations discharging method of multiserver and device based on Lyapunov optimization Download PDFInfo
- Publication number
- CN109857546A CN109857546A CN201910004742.4A CN201910004742A CN109857546A CN 109857546 A CN109857546 A CN 109857546A CN 201910004742 A CN201910004742 A CN 201910004742A CN 109857546 A CN109857546 A CN 109857546A
- Authority
- CN
- China
- Prior art keywords
- task
- mobile device
- server
- formula
- follows
- 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
Links
Classifications
-
- 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
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The present invention provides a kind of mobile edge calculations discharging method of multiserver based on Lyapunov optimization and devices, method therein includes: to construct task buffer queue first for the task of the generation of each mobile device, and need unloading but not processed task creation virtual task buffer queue for each mobile device；And establish power consumption of terminal model, Bandwidth Model and QoE model；Then in each preset time piece, the task buffer queuing message of mobile device, the bandwidth situation of virtual cache queuing message and server are collected；Further according to the information of collection, Liapunov drift plus penalty are constructed；Again based on Liapunov drift plus penalty, the calculating unloading decision in current time piece is solved.The present invention is realized to be calculated based on multiserver MEC and be unloaded, and the deadline of task is effectively reduced and reduces the technical effect of energy consumption of mobile equipment.
Description
Technical field
The present invention relates to mobile edge calculations technical fields, and in particular to a kind of multiserver based on Lyapunov optimization
Mobile edge calculations discharging method and device.
Background technique
In 5G wireless system, super-intensive edge device, including small-sized honeycomb base station, wireless access point, notes will be disposed
The computing capability of this computer, tablet computer and smart phone, every kind of equipment is all suitable with computer.Moreover, at each moment
There is large number of equipment idle.If can the available huge calculating in collection network edge and storage resource, this will be sufficient to achieve generally
Existing mobile computing.In brief, the main target of the wireless system from 1G to 4G is to pursue higher and higher radio speed,
Transition to support the flow from centered on voice to centered on multimedia.As wireless network speed is close to cable network
Speed, the mission of 5G become increasingly complex, to support the explosive growth of Internet of Things and internet.Functionally, 5G system
It will support to communicate, calculate, control and content delivery.In application aspect, the various new opplication kimonos based on 5G, which are done honest work, to be risen, example
Such as real-time game on line, virtual reality and ultra high-definition video flowing, this needs unprecedented high access speed and low latency.In mistake
It goes in 10 years, the different visions of Next Generation Internet are also taken off, including Internet of Things, and tactile internet (prolongs with Millisecond
Late) and social networks.
It is predicted according to Cisco, arrives the year two thousand twenty, internet will increase about 50,000,000,000 internet of things equipment (such as sensor, wearable
Equipment), wherein most resource is communicated and is stored, and must rely on cloud or edge device, to enhance its energy for calculating
Power.Currently, in the prior art, what is generallyd use is traditional mobile cloud computing (Mobile Cloud Computing, MCC) mould
Formula, i.e., one of infrastructure, platform, software (or application) needed for being obtained by mobile network with on-demand, easy extension way
The delivery and use pattern of kind IT resource or (information) service.
Present invention applicant is in implementing the present invention, it may, have found that method in the prior art at least has following technology
Problem:
Only rely on the Millisecond delay needed for cloud computing technology is insufficient for new application in 5G and services.In addition, eventually
Data exchange between end subscriber and long-distance cloud will will lead to data tsunami, and backhaul network is made to paralyse.
Method in the prior art known to as a result, is limited there are computing resource and leads to the technology that task postpones and energy consumption is high
Problem.
Summary of the invention
In view of this, the present invention provides a kind of mobile edge calculations unloading sides of the multiserver based on Lyapunov optimization
Method and device, to solve or at least partly to solve method in the prior art limited there are computing resource and task is caused to be prolonged
The technical issues of late and energy consumption is high.
First aspect present invention provides the mobile edge calculations discharging method of multiserver based on Lyapunov optimization, packet
It includes:
Step S1: task buffer queue is constructed for the task of the generation of each mobile device, and is needed for each mobile device
It unloads but not processed task creation virtual task buffer queue；
Step S2: power consumption of terminal model, Bandwidth Model and QoE model are established；
Step S3: in each preset time piece, task buffer queuing message, the virtual cache queue letter of mobile device are collected
The bandwidth situation of breath and server；
Step S4: according to the bandwidth feelings of the task buffer queuing message of collection, virtual cache queuing message and server
Condition, the drift of building Liapunov plus penalty；
Step S5: based on Liapunov drift plus penalty, solving the calculating unloading decision in current time piece,
Wherein, unloading decision is calculated to characterize the task amount and be unloaded to side that the calculating task of mobile terminal generation is performed locally
The task amount that edge server executes.
In one embodiment, after step s 5, the method also includes:
Step S6: the Liapunov drift of building plus penalty are split and is solved, for different calculating
It unloads decision and carries out resource allocation.
In one embodiment, step S1 is specifically included:
Step S1.1: each mobile device is the one task buffer queue of application build being each currently running, and task is slow
Deposit queue Qi(t) iterative formula is formula (1):
Qi(t+1)=max { Qi(t)-DΣ, i(t),0}+Ai(t) (1)
Wherein, DΣ,i(t) indicate that i-th of mobile device locally executes and be unloaded to Edge Server in t-th of timeslice
On the sum of task amount, Qi(t) it indicates in t-th of timeslice, the local task queue length of i-th of mobile device, Ai(t) table
Show the task amount generated in t-th of timeslice, i-th of mobile device；Qi(t+1) it indicates in the t+1 timeslice, i-th of shifting
The local task queue length of dynamic equipment；
Step S1.2: virtual task buffer queue H is created for each mobile devicei(t), it is stored in the virtual queue
It is unloaded also not processed task, iterative formula is formula (2):
Hi(t+1)=max { Hi(t)-Ds,i(t),0}+Dr,i (2)
Wherein, Ds,i(t) it indicates in t-th of timeslice, the task amount for belonging to i-th of mobile device executed by server,
Dr,i(t) expression i-th of mobile device in t-th of timeslice is unloaded to the task amount of Edge Server, Hi(t) it indicates in t
A timeslice, the virtual queue length of i-th of mobile device, Hi(t+1) it indicates in the t+1 timeslice, i-th of mobile device
Virtual queue length.
In one embodiment, in step S2,
Power consumption of terminal model includes the CPU power consumption locally executed and signal transmitting power consumption, wherein the local of mobile device
Cpu frequency depending on the task amount locally executed, specific formula is as follows:
fl,i(t)=LiDl,i(t)τ-1 (3)
Wherein, fl,i(t) cpu frequency in t-th of timeslice, i-th of mobile device, D are indicatedl,i(t) it indicates at t-th
The task amount that i-th of mobile device of timeslice locally executes, LiIndicate the cpu cycle that the every processing 1bit required by task of CPU is wanted；It moves
Dynamic equipment locally executes power consumption pl,i(t) directly proportional to the frequency of CPU, formula is as follows:
Wherein, kmod,iIndicate effective switching capacity of the CPU of i-th of mobile device；
The radio transmitted power of mobile device is pr,i(t) formula is as follows:
Dr,i(t)=∑j∈Sri,j(t)τ (6)
Wherein, ri,j(t) message transmission rate between mobile device i to server j, B are indicatedjIndicate server to every
The amount of bandwidth of a mobile device distribution, Γi,j(t) indicate that channel fading power between mobile device i and server j, τ indicate
The interval of timeslice, in t-th of timeslice, xi,j(t)=1 indicate that i-th of mobile device unloads task to j-th of server, it is no
It then indicates not unload；
Channel is divided into different subchannels by frequency division multiple access technology by Bandwidth Model；
QoE model, including task delay and mobile device power consumption, according to preset rules to task delay and mobile device function
Consumption carries out the QoE evaluation criterion that linear weighted function constitutes user, wherein the task decay part of QoECalculation formula
Are as follows:
Wherein, α is weighting parameters, and α ∈ [0,1],
The power consumption sections of QoECalculation formula are as follows:
Synthesis obtains the calculation formula of QoE are as follows:
Wherein, β is weighting parameters, and β ∈ [0,1].
In one embodiment, step S4 is specifically included:
Step S4.1: building Lyapunov function, specific building formula are as follows:
Wherein, Θ (t) was indicated within t-th of period, the state of all task buffer queues, Qi(t) when indicating t-th
Between in section i-th of mobile device local task buffer queue length, Hi(t) i-th of mobile device in t-th of period is indicated
Virtual task length of buffer queue on the server；
Step S4.2: building Lyapunov drift function, specific formula is as follows:
Wherein,Expectation is asked in expression；
Step S4.3: building Lyapunov drift plus penalty:
Wherein, V indicates that penalty factor, ξ (t) indicate that the QoE of user is spent, which has a upper limit, is limited to thereon:
Wherein, C is constant.
In one embodiment, step S6 is specifically included:
For the task amount locally executed, mobile device is obtained after Lyapunov drift plus penalty are unfolded and are split
Local cpu frequency regulation scheme, specific formulation indicate are as follows:
It finds out formula (14) acquirement minimum value Shi Zewei and locally executes cpu frequency, locally execute cpu frequency expression are as follows:
For being unloaded to the task amount of Edge Server execution, Lyapunov drift plus penalty are unfolded and are split
Afterwards, it obtains and calculates unloading and mobile device transimission power decision scheme, formulation indicates are as follows:
minP(t),x(t)∑i∈U[-(Qi(t)-Hi(t)+V·αβ)Dr,i(t)+V·(1-β)·pr,i(t)] (16)。
In one embodiment, it obtains and calculates unloading and mobile device transimission power decision scheme, specifically include:
Formula (16) is split as the wireless signal transmission power of mobile device and calculates unloading decision, wherein mobile device
Wireless signal transmission power, optimization after fractionation, which formulates, to be indicated are as follows:
The wireless transmitted power for being when formula (17) obtain minimum value, can formulate expression are as follows:
The optimization formula for calculating unloading decision may be expressed as:
Wherein, for each mobile device, work as Qi(t)-Hi(t) when+V α β≤0, i-th of mobile device unloading one is indicated
On partial task to Edge Server.
In one embodiment, the task amount executed for being unloaded to Edge Server, the method also includes:
Server computational resource allocation is carried out, optimization formula may be expressed as:
minD(t)∑i∈U[-[V·(1-α)β+Hi(t)]·Ds,i(t)] (20)。
Based on same inventive concept, it is mobile that second aspect of the present invention provides the multiserver based on Lyapunov optimization
Edge calculations discharge mechanism, comprising:
Queue constructs module, and the task for the generation for each mobile device constructs task buffer queue, and is each
Mobile device needs to unload but not processed task creation virtual task buffer queue；
Model building module, for establishing power consumption of terminal model, Bandwidth Model and QoE model；
Information collection module, in each preset time piece, collecting the task buffer queuing message, virtual of mobile device
The bandwidth situation of buffer queue information and server；
Function constructs module, for the task buffer queuing message, virtual cache queuing message and service according to collection
The bandwidth situation of device, the drift of building Liapunov plus penalty；
Decision-making module is unloaded, for solving the meter in current time piece based on Liapunov drift plus penalty
Calculate unloading decision, wherein calculate the task amount that unloading decision is performed locally to characterize the calculating task of mobile terminal generation
And it is unloaded to the task amount of Edge Server execution.
Based on same inventive concept, third aspect present invention provides a kind of computer equipment, including memory, processing
On a memory and the computer program that can run on a processor, when processor execution described program, is realized for device and storage
Method as described in relation to the first aspect.
Said one or multiple technical solutions in the embodiment of the present application at least have following one or more technology effects
Fruit:
In method provided by the invention, task buffer queue is constructed for the task of the generation of each mobile device first, and
Unloading is needed for each mobile device but not processed task creation virtual task buffer queue, and establish power consumption of terminal mould
Type, Bandwidth Model and QoE model；Then in each preset time piece, the task buffer queuing message, virtual of mobile device is collected
The bandwidth situation of buffer queue information and server；Task buffer queuing message, virtual cache queue then according to collection
The bandwidth situation of information and server, the drift of building Liapunov plus penalty；Added again based on Liapunov drift
Penalty solves the calculating unloading decision in current time piece, calculates unloading based on multiserver MEC to realize, can
To effectively reduce the deadline of task, and reduce energy consumption of mobile equipment.
In compared with the existing technology by traditional mobile cloud computing mode for, proposed by the present invention is that one kind is based on
The multiserver MEC of Lyapunov optimization calculates discharging method, and the scope of application is the mobile edge meter of multiserver multi-user
Environment is calculated, by collecting the task buffer queuing message of mobile device, the task buffer queuing message and clothes of server
The bandwidth situation of business device, and by the model of the information being collected into and building, calculating task unloading is carried out based on Lyapunov optimization
Decision, obtain the calculating unloading decision in current time piece, wherein the task amount that is performed locally of calculating task and unload
It is downloaded to the task amount of Edge Server execution.
Further, using Liapunov optimisation technique is based on, resource is carried out for the calculating unloading decision solved
Distribution, to be optimized to resource.
Further, by being further reduced task delay to the adjusting to computing resource and the distribution of radio resource
With terminal device energy consumption, to promote user experience quality.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is the present invention
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.
Fig. 1 is the stream of the mobile edge calculations discharging method of multiserver based on Lyapunov optimization in the embodiment of the present invention
Cheng Tu；
Fig. 2 is the task queue model schematic constructed in the embodiment of the present invention；
Fig. 3 is the knot of the mobile edge calculations discharge mechanism of multiserver based on Lyapunov optimization in the embodiment of the present invention
Structure block diagram；
Fig. 4 is the mobile edge calculations unloading of multiserver in specific example a kind of in the embodiment of the present invention and resource allocation
Frame diagram；
Fig. 5 is the structure chart of computer equipment in the embodiment of the present invention.
Specific embodiment
The technical issues of leading to task delay it is an object of the invention to limited for existing computing resource and energy consumption is high,
A kind of mobile edge calculations discharging method of the multiserver based on Lyapunov optimization provided, the scope of application are service more
The mobile edge calculations environment of device multi-user.By collecting the task buffer queuing message of mobile device, being offloaded to server
The bandwidth situation of virtual task buffer queue information and server, using the information being collected into as the input of decision submodule.
Decision submodule uses being directed to for mobile edge calculations environment based on the Lyapunov multi-user's multiserver collaboration optimized to prolong
The calculating unloading of tolerance application late and resource allocation joint decision algorithm handle data, obtain decision scheme.
Further, resource allocation is carried out according to decision-making technique, especially by to computing resource and radio resource
The adjusting of distribution is further reduced task delay and terminal device energy consumption, to promote user experience quality.
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
Embodiment one
A kind of mobile edge calculations discharging method of the multiserver based on Lyapunov optimization is present embodiments provided, please be join
See Fig. 1, this method comprises:
Step S1 is first carried out: constructing task buffer queue for the task of the generation of each mobile device, and is each shifting
Dynamic equipment needs to unload but not processed task creation virtual task buffer queue.
Specifically, for the application of delay-tolerant, generating for task will not be performed immediately, but be first stored in slow
It deposits in queue.The needing to be unloaded to server execution similarly, for mobile device in generating for task of the task, can construct one
Virtual task buffer queue.
Present inventor is by largely studying and practicing discovery: mobile edge calculations (Mobile Edge Compute
It MEC) is that the IT new example of service and cloud computing function is provided in Radio Access Network, in future, mobile edge calculations service
Device will be distributed over by super-intensive small-sized honeycomb base station.In this environment, the calculating task of mobile subscriber can be unloaded to side
On edge server, to reduce its service delay and power consumption.Mobile subscriber can be benefited from this mode.For example, multiple base stations
The mobile subscriber of covering can select from multiple mobile edge calculations servers, according to radio transmission conditions and available
Edge calculations resource unloads its calculating task.When there is great amount of terminals to submit unloading request to Edge Server, Edge Server
Unloading decision can be made for request and distributes reasonable resource, but since Edge Server is deployed in the network edge of mobile terminal
Edge, calculates and the resources such as storage are limited, can not always be immediately request distribution resource, while the unloading strategy meeting between user
It influences each other.Therefore, unload problem in the resource-constrained calculating of Edge Server, how to reduce application deadline and
The energy consumption for reducing terminal is urgently to be resolved.Based under this background, the invention proposes a kind of based on Lyapunov optimization
Multiserver moves edge calculations discharging method.
In one embodiment, step S1 is specifically included:
Step S1.1: each mobile device is the one task buffer queue of application build being each currently running, and task is slow
Deposit queue Qi(t) iterative formula is formula (1):
Qi(t+1)=max { Qi(t)-DΣ,i(t),0}+Ai(t) (1)
Wherein, DΣ,i(t) indicate that i-th of mobile device locally executes and be unloaded to Edge Server in t-th of timeslice
On the sum of task amount, Qi(t) it indicates in t-th of timeslice, the local task queue length of i-th of mobile device, Ai(t) table
Show the task amount generated in t-th of timeslice, i-th of mobile device；Qi(t+1) it indicates in the t+1 timeslice, i-th of shifting
The local task queue length of dynamic equipment；
Step S1.2: virtual task buffer queue H is created for each mobile devicei(t), it is stored in the virtual queue
It is unloaded also not processed task, iterative formula is formula (2):
Hi(t+1)=max { Hi(t)-Ds,i(t),0}+Dr,i (2)
Wherein, Ds,i(t) it indicates in t-th of timeslice, the task amount for belonging to i-th of mobile device executed by server,
Dr,i(t) expression i-th of mobile device in t-th of timeslice is unloaded to the task amount of Edge Server, Hi(t) it indicates in t
A timeslice, the virtual queue length of i-th of mobile device, Hi(t+1) it indicates in the t+1 timeslice, i-th of mobile device
Virtual queue length.
2 specifically are referred to, for the task queue schematic diagram constructed in the embodiment of the present invention.
Then it executes step S2: establishing power consumption of terminal model, Bandwidth Model and QoE model.
Specifically, the model of building is for subsequent calculating unloading and resource allocation.
In one embodiment, in step S2,
Power consumption of terminal model includes the CPU power consumption locally executed and signal transmitting power consumption, wherein the local of mobile device
Cpu frequency depending on the task amount locally executed, specific formula is as follows:
fl,i(t)=LiDl,i(t)τ-1 (3)
Specifically, the present invention uses dynamic voltage scaling technology and dynamic voltage regulation technology to the local of mobile device
Cpu frequency and radio transmission power are adjusted, to reduce terminal energy consumption.The local cpu frequency of mobile device is according to this
Depending on the task amount that ground executes.Wherein, fl,i(t) cpu frequency in t-th of timeslice, i-th of mobile device, D are indicatedl,i(t)
Indicate the task amount locally executed in t-th of timeslice, i-th of mobile device, LiIndicate that the every processing 1bit required by task of CPU is wanted
Cpu cycle；Mobile device locally executes power consumption
pl,i(t) directly proportional to the frequency of CPU, formula is as follows:
Wherein, kmod,iIndicate effective switching capacity of the CPU of i-th of mobile device；
The radio transmitted power of mobile device is pr,i(t) formula is as follows:
Dr,i(t)=∑j∈Sri,j(t)τ (6)
Wherein, ri,j(t) message transmission rate between mobile device i to server j, B are indicatedjIndicate server to every
The amount of bandwidth of a mobile device distribution, Γi,j(t) indicate that channel fading power between mobile device i and server j, τ indicate
The interval of timeslice, in t-th of timeslice, xi,j(t)=1 indicate that i-th of mobile device unloads task to j-th of server, it is no
It then indicates not unload；
Channel is divided into different subchannels by frequency division multiple access technology by Bandwidth Model；
Specifically, the wireless channel of frequency division multiple access can be directed in present embodiment, by frequency division multiple access technology by channel
It is divided into multiple subchannels.Each mobile device monopolizes a sub-channels when carrying out calculating unloading.
QoE model, including task delay and mobile device power consumption, according to preset rules to task delay and mobile device function
Consumption carries out the QoE evaluation criterion that linear weighted function constitutes user, wherein the task decay part of QoECalculation formula
Are as follows:
Wherein, α is weighting parameters, and α ∈ [0,1],
The power consumption sections of QoECalculation formula are as follows:
Synthesis obtains the calculation formula of QoE are as follows:
Wherein, β is weighting parameters, and β ∈ [0,1].
Specifically, for different tasks, the standard that QoE stresses is different, can be joined by adjusting the weighting of QoE
It counts to adjust the ratio between task delay and mobile device power consumption.Task delay can be divided into local task delay and server again
Task delay, the present invention postpones local task and server task delay has carried out linear weighted function, can be by adjusting weighting ginseng
Number changes the weight of local task delay and server task delay, meets the needs of different task.According to Little's Law,
In the system of one stable non-preemptive, the residence time of the average latency of task and task in the queue is directly proportional,
Therefore available task queue length indicates the delay of task.
Next it executes step S3: in each preset time piece, collecting the task buffer queuing message, virtual of mobile device
The bandwidth situation of buffer queue information and server.
Specifically, can be responsible for collecting the task of all mobile devices in each timeslice by information collection module
Task queuing message on queuing message, quality of wireless channel information and each server.All information being collected into will conduct
The input of decision submodule is used for subsequent calculating unloading scheme.Wherein, quality of wireless channel information refers to wireless channel decline function
Rate, i.e. fl,i(t), it is used when calculating optimal transmission power.
Step S4 is executed again: according to the task buffer queuing message of collection, virtual cache queuing message and server
Bandwidth situation, the drift of building Liapunov plus penalty.
Specifically, Liapunov stability (Lyapunov stability or Liapunov stability) can be with
For describing the stability of a dynamical system.If track of any primary condition of this system near equilibrium state can maintain
Near equilibrium state, then being properly termed as in place's Lyapunov stability.
In one embodiment, step S4 is specifically included:
Step S4.1: building Lyapunov function, specific building formula are as follows:
Wherein, Θ (t) was indicated within t-th of period, the state of all task buffer queues, Qi(t) when indicating t-th
Between in section i-th of mobile device local task buffer queue length, Hi(t) i-th of mobile device in t-th of period is indicated
Virtual task length of buffer queue on the server；
Step S4.2: building Lyapunov drift function, specific formula is as follows:
Wherein,Expectation is asked in expression；
Step S4.3: building Lyapunov drift plus penalty:
Wherein, V indicates that penalty factor, ξ (t) indicate that the QoE of user is spent, which has a upper limit, is limited to thereon:
Wherein, C is constant.
Step S5 is executed again: being unloaded based on Liapunov drift plus penalty, the calculating solved in current time piece
Carry decision, wherein calculate unloading decision to characterize mobile terminal generation the task amount that is performed locally of calculating task and
It is unloaded to the task amount of Edge Server execution.
In one embodiment, after step s 5, the method also includes:
Step S6: the Liapunov drift of building plus penalty are split and is solved, for different calculating
It unloads decision and carries out resource allocation.
Specifically, it after obtaining current unloading decision by abovementioned steps, can further be floated by Liapunov
It moves and adds penalty, carry out resource allocation.
Specifically, resource allocation includes the distribution of computational resource allocation and radio channel resource, wherein computing resource
Distribution can be divided into two parts: the 1) distribution of local computing resource；2) distribution of server computing resource.Local computing resource point
Determine that each mobile device is the local task queue of oneself distributes how many computing resource with responsible.Server computational resource allocation
It is responsible for determining each server to the data between each virtual task queue assignment how many computing resource and different server
Transmission.Radio channel resource distribution includes two parts: 1) mobile device transmission power adjustment；2) server bandwidth is distributed.It moves
Dynamic equipment transmission power adjustment is responsible for the determination of the signal transmission power of each mobile device.Server bandwidth, which is distributed, is responsible for decision
How many bandwidth resources distributed to mobile device for each server.
In one embodiment, step S6 is specifically included:
For the task amount locally executed, mobile device is obtained after Lyapunov drift plus penalty are unfolded and are split
Local cpu frequency regulation scheme, specific formulation indicate are as follows:
It finds out formula (14) acquirement minimum value Shi Zewei and locally executes cpu frequency, locally execute cpu frequency expression are as follows:
For being unloaded to the task amount of Edge Server execution, Lyapunov drift plus penalty are unfolded and are split
Afterwards, it obtains and calculates unloading and mobile device transimission power decision scheme, formulation indicates are as follows:
minp(t),x(t)∑i∈U[-(Qi(t)-Hi(t)+V·αβ)Dr,i(t)+V·(1-β)·pr,i(t)] (16)。
Specifically, for the task amount locally executed, main includes the distribution of local computing resource, i.e. mobile device sheet
Ground cpu frequency regulation scheme.It then include the distribution of server computing resource for being unloaded to the task amount of Edge Server execution
And the distribution of radio channel resource.
In one embodiment, it obtains and calculates unloading and mobile device transimission power decision scheme, specifically include:
Formula (16) is split as the wireless signal transmission power of mobile device and calculates unloading decision, wherein mobile device
Wireless signal transmission power, optimization after fractionation, which formulates, to be indicated are as follows:
The wireless transmitted power for being when formula (17) obtain minimum value, can formulate expression are as follows:
The optimization formula for calculating unloading decision may be expressed as:
Wherein, for each mobile device, work as Qi(t)-Hi(t) when+V α β≤0, i-th of mobile device unloading one is indicated
On partial task to Edge Server.
Specifically, each mobile device can traverse its Servers-all that can connect to, and be unloaded to which edge of table
Edge server is more excellent, then offloads tasks on these Edge Servers.
In one embodiment, the task amount executed for being unloaded to Edge Server, the method also includes:
Server computational resource allocation is carried out, optimization formula may be expressed as:
minD(t)∑i∈U[-[V·(1-α)β+Hi(t)]·Ds,i(t)] (20)。
It specifically, can be according to V (1- α) β+Hi(t) size is from big to small ranked up mobile device, preferentially
Discharge in front mobile device task distribute computing resource.The scheme of its resource allocation are as follows: if server resource is insufficient, own
Computing resource be all assigned to current mobile device, if server resource is sufficient, what all current mobile devices may be connected to
Server distributes computing resource to current mobile device according to the ratio of remaining computing resource.It, can basis after selected server
The task buffer queue length of server end determines that how many task of server-side is performed.It can lead between adjacent Edge Server
It crosses data exchange and carries out load balancing.
It should be noted that formula involved in the present invention and character meaning are as shown in table 1.
Table 1
Based on the same inventive concept, present invention also provides in embodiment one it is a kind of based on Lyapunov optimization more clothes
The corresponding device of the mobile edge calculations discharging method of business device, detailed in Example two.
Embodiment two
The mobile edge calculations discharge mechanism of multiserver based on Lyapunov optimization is present embodiments provided, figure is referred to
3, which includes:
The mobile edge calculations discharge mechanism of multiserver based on Lyapunov optimization characterized by comprising
Queue constructs module 201, and the task for the generation for each mobile device constructs task buffer queue, and is every
A mobile device needs to unload but not processed task creation virtual task buffer queue；
Model building module 202, for establishing power consumption of terminal model, Bandwidth Model and QoE model；
Information collection module 203, in each preset time piece, collect mobile device task buffer queuing message,
The bandwidth situation of virtual cache queuing message and server；
Function constructs module 204, for the task buffer queuing message, virtual cache queuing message and clothes according to collection
The bandwidth situation of business device, the drift of building Liapunov plus penalty；
Decision-making module 205 is unloaded, for solving in current time piece based on Liapunov drift plus penalty
Calculate unloading decision, wherein calculate the task that unloading decision is performed locally to characterize the calculating task of mobile terminal generation
Measure and be unloaded to the task amount of Edge Server execution.
In one embodiment, described device further includes resource distribution module, for solving in current time piece
Calculating unloading decision after:
The Liapunov drift of building plus penalty are split and solved, unloads decision for different calculating
Carry out resource allocation.
In one embodiment, queue building module 201 is specifically used for executing following step:
Step S1.1: each mobile device is the one task buffer queue of application build being each currently running, and task is slow
Deposit queue Qi(t) iterative formula is formula (1):
Qi(t+1)=max { Qi(t)-DΣ,i(t),0}+Ai(t) (1)
Wherein, DΣ,i(t) indicate that i-th of mobile device locally executes and be unloaded to Edge Server in t-th of timeslice
On the sum of task amount, Qi(t) it indicates in t-th of timeslice, the local task queue length of i-th of mobile device, Ai(t) table
Show the task amount generated in t-th of timeslice, i-th of mobile device；Qi(t+1) it indicates in the t+1 timeslice, i-th of shifting
The local task queue length of dynamic equipment；
Step S1.2: virtual task buffer queue H is created for each mobile devicei(t), it is stored in the virtual queue
It is unloaded also not processed task, iterative formula is formula (2):
Hi(t+1)=max { Hi(t)-DS, i(t), 0 }+DR, i (2)
Wherein, DS, i(t) it indicates in t-th of timeslice, the task amount for belonging to i-th of mobile device executed by server,
DR, i(t) expression i-th of mobile device in t-th of timeslice is unloaded to the task amount of Edge Server, Hi(t) it indicates in t
A timeslice, the virtual queue length of i-th of mobile device, Hi(t+1) it indicates in the t+1 timeslice, i-th of mobile device
Virtual queue length.
In one embodiment, in model building module 202,
Power consumption of terminal model includes the CPU power consumption locally executed and signal transmitting power consumption, wherein the local of mobile device
Cpu frequency depending on the task amount locally executed, specific formula is as follows:
fl,i(t)=LiDl,i(t)τ-1 (3)
Wherein, fl,i(t) cpu frequency in t-th of timeslice, i-th of mobile device, D are indicatedl,i(t) it indicates at t-th
The task amount that i-th of mobile device of timeslice locally executes, LiIndicate the cpu cycle that the every processing 1bit required by task of CPU is wanted；It moves
Dynamic equipment locally executes power consumption pl,i(t) directly proportional to the frequency of CPU, formula is as follows:
Wherein, kMod, iIndicate effective switching capacity of the CPU of i-th of mobile device；
The radio transmitted power of mobile device is pR, i(t) formula is as follows:
DR, i(t)=∑j∈SrI, j(t)τ (6)
Wherein, ri,j(t) message transmission rate between mobile device i to server j, B are indicatedjIndicate server to every
The amount of bandwidth of a mobile device distribution, Γi,j(t) indicate that channel fading power between mobile device i and server j, τ indicate
The interval of timeslice, in t-th of timeslice, xi,j(t)=1 indicate that i-th of mobile device unloads task to j-th of server, it is no
It then indicates not unload；
Channel is divided into different subchannels by frequency division multiple access technology by Bandwidth Model；
QoE model, including task delay and mobile device power consumption, according to preset rules to task delay and mobile device function
Consumption carries out the QoE evaluation criterion that linear weighted function constitutes user, wherein the task decay part of QoECalculation formula
Are as follows:
Wherein, α is weighting parameters, and α ∈ [0,1],
The power consumption sections of QoECalculation formula are as follows:
Synthesis obtains the calculation formula of QoE are as follows:
Wherein, β is weighting parameters, and β ∈ [0,1].
In one embodiment, function building module 204 is specifically used for executing following step:
Step S4.1: building Lyapunov function, specific building formula are as follows:
Wherein, Θ (t) was indicated within t-th of period, the state of all task buffer queues, Qi(t) when indicating t-th
Between in section i-th of mobile device local task buffer queue length, Hi(t) i-th of mobile device in t-th of period is indicated
Virtual task length of buffer queue on the server；
Step S4.2: building Lyapunov drift function, specific formula is as follows:
Wherein,Expectation is asked in expression；
Step S4.3: building Lyapunov drift plus penalty:
Wherein, V indicates that penalty factor, ξ (t) indicate that the QoE of user is spent, which has a upper limit, is limited to thereon:
Wherein, C is constant.
In one embodiment, resource distribution module is specifically used for:
For the task amount locally executed, mobile device is obtained after Lyapunov drift plus penalty are unfolded and are split
Local cpu frequency regulation scheme, specific formulation indicate are as follows:
It finds out formula (14) acquirement minimum value Shi Zewei and locally executes cpu frequency, locally execute cpu frequency expression are as follows:
For being unloaded to the task amount of Edge Server execution, Lyapunov drift plus penalty are unfolded and are split
Afterwards, it obtains and calculates unloading and mobile device transimission power decision scheme, formulation indicates are as follows:
minP (t), x (t)∑i∈U[-(Qi(t)-Hi(t)+V·αβ)DR, i(t)+V·(1-β)·pr,i(t)] (16)。
In one embodiment, resource distribution module is also used to:
Formula (16) is split as the wireless signal transmission power of mobile device and calculates unloading decision, wherein mobile device
Wireless signal transmission power, optimization after fractionation, which formulates, to be indicated are as follows:
The wireless transmitted power for being when formula (17) obtain minimum value, can formulate expression are as follows:
The optimization formula for calculating unloading decision may be expressed as:
Wherein, for each mobile device, work as Qi(t)-Hi(t) when+V α β≤0, i-th of mobile device unloading one is indicated
On partial task to Edge Server.
In one embodiment, the task amount executed for being unloaded to Edge Server, resource distribution module are also used to:
Server computational resource allocation is carried out, optimization formula may be expressed as:
minD(t)∑i∈U[-[V·(1-α)β+Hi(t)]·Ds,i(t)] (20)。
In order to illustrate more clearly of the specific embodiment of device in the present invention, give in detail below by a model example
It is thin to introduce, referring specifically to Fig. 4.
On the whole, the technical solution that the present invention mainly takes are as follows: provide a kind of more clothes based on Lyapunov optimization
The mobile edge calculations of device of being engaged in unload and resource allocation device, which, which specifically includes, calculates Unload module and resource distribution module.
In Fig. 4, calculating Unload module includes information collection module, decision-making module.Information collection module collection is responsible for every
Task queue information, quality of wireless channel information and each server for being responsible for collecting all mobile devices in a timeslice are taken up an official post
Business queuing message.All information being collected into will finally obtain as the input of decision-making module and calculate unloading scheme.
Resource distribution module is responsible for the distribution of computing resource and radio channel resource.The distribution of computing resource can be divided into two
Part: the 1) distribution of local computing resource；2) distribution of server computing resource.Local computing resource allocation is responsible for determining each
Mobile device is the local task queue of oneself distributes how many computing resource.Server computational resource allocation is responsible for determining each clothes
Device be engaged in the data transmission between each virtual task queue assignment how many computing resource and different server.Wireless channel
Resource allocation includes two parts: 1) mobile device transmission power adjustment；2) server bandwidth is distributed.Mobile device transimission power
Adjust the determination for being responsible for the signal transmission power of each mobile device.Server bandwidth distribution is responsible for determining each server to shifting
Dynamic equipment distributes how many bandwidth resources.
Multiserver MEC provided by the invention based on Lyapunov optimization calculates unloading and resource allocation algorithm, fits
It is the mobile edge calculations environment of multiserver multi-user with range.The collecting mobile device by information collection module of the task is slow
Queuing message, the task buffer queuing message of server and the bandwidth situation of server are deposited, using the information being collected into as determining
The input of plan module, decision-making module use the mobile edge calculations ring based on the Lyapunov multi-user's multiserver collaboration optimized
The calculating unloading for delay-tolerant application in border handles data, obtains decision scheme, then give decision scheme to money
Source distribution module is executed.By to the adjusting to computing resource and the distribution of radio resource, come reduce task delay and
Terminal device energy consumption, to promote user experience quality.
By the device that the embodiment of the present invention two is introduced, to implement to optimize in the embodiment of the present invention one based on Lyapunov
The mobile edge calculations discharging method of multiserver used by device, so the side introduced based on the embodiment of the present invention one
Method, the affiliated personnel in this field can understand specific structure and the deformation of the device, so details are not described herein.All present invention are real
It applies device used by the method for example one and belongs to the range of the invention to be protected.
Embodiment three
Based on the same inventive concept, present invention also provides a kind of computer equipment, Fig. 5 is referred to, including storage 401,
On a memory and the computer program 403 that can run on a processor, processor 402 executes above-mentioned for processor 402 and storage
The method in embodiment one is realized when program.
Since the computer equipment that the embodiment of the present invention three is introduced is to implement to be based in the embodiment of the present invention one
Computer equipment used by the mobile edge calculations discharging method of the multiserver of Lyapunov optimization, so it is real based on the present invention
The method that example one is introduced is applied, the affiliated personnel in this field can understand specific structure and the deformation of the computer equipment, so
This is repeated no more.Computer equipment used by method belongs to the model of the invention to be protected in all embodiment of the present invention one
It encloses.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the present invention
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the present invention, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
Although preferred embodiments of the present invention have been described, it is created once a person skilled in the art knows basic
Property concept, then additional changes and modifications can be made to these embodiments.So it includes excellent that the following claims are intended to be interpreted as
It selects embodiment and falls into all change and modification of the scope of the invention.
Obviously, those skilled in the art can carry out various modification and variations without departing from this hair to the embodiment of the present invention
The spirit and scope of bright embodiment.In this way, if these modifications and variations of the embodiment of the present invention belong to the claims in the present invention
And its within the scope of equivalent technologies, then the present invention is also intended to include these modifications and variations.
Claims (10)
1. the mobile edge calculations discharging method of multiserver based on Lyapunov optimization characterized by comprising
Step S1: task buffer queue is constructed for the task of the generation of each mobile device, and is unloaded for each mobile device needs
Load but not processed task creation virtual task buffer queue；
Step S2: power consumption of terminal model, Bandwidth Model and QoE model are established；
Step S3: in each preset time piece, collect task buffer queuing message, the virtual cache queuing message of mobile device with
And the bandwidth situation of server；
Step S4: according to the bandwidth situation of the task buffer queuing message of collection, virtual cache queuing message and server, structure
Build Liapunov drift plus penalty；
Step S5: based on Liapunov drift plus penalty, solving the calculating unloading decision in current time piece,
In, unloading decision is calculated to characterize the task amount and be unloaded to edge that the calculating task of mobile terminal generation is performed locally
The task amount that server executes.
2. the method as described in claim 1, which is characterized in that after step s 5, the method also includes:
Step S6: the Liapunov drift of building plus penalty are split and is solved, is unloaded for different calculating
Decision carries out resource allocation.
3. the method as described in claim 1, which is characterized in that step S1 is specifically included:
Step S1.1: each mobile device is the one task buffer queue of application build being each currently running, task buffer team
Arrange Qi(t) iterative formula is formula (1):
Qi(t+1)=max { Qi(t)-DΣ,i(t),0}+Ai(t) (1)
Wherein, DΣ,i(t) indicate that i-th of mobile device is locally executed and be unloaded on Edge Server in t-th of timeslice
The sum of task amount, Qi(t) it indicates in t-th of timeslice, the local task queue length of i-th of mobile device, Ai(t) it indicates
T-th of timeslice, the task amount that i-th of mobile device generates；Qi(t+1) indicate that i-th of movement is set in the t+1 timeslice
Standby local task queue length；
Step S1.2: virtual task buffer queue H is created for each mobile devicei(t), be stored in the virtual queue by
Also not processed task is unloaded, iterative formula is formula (2):
Hi(t+1)=max { Hi(t)-DS, i(t), 0 }+DR, i (2)
Wherein, DS, i(t) it indicates in t-th of timeslice, the task amount for belonging to i-th of mobile device executed by server, Dr,i
(t) expression i-th of mobile device in t-th of timeslice is unloaded to the task amount of Edge Server, Hi(t) it indicates at t-th
Timeslice, the virtual queue length of i-th of mobile device, Hi(t+1) it indicates in the t+1 timeslice, i-th mobile device
Virtual queue length.
4. the method as described in claim 1, which is characterized in that in step S2,
Power consumption of terminal model includes the CPU power consumption locally executed and signal transmitting power consumption, wherein the local cpu frequency of mobile device
Rate depending on the task amount locally executed, specific formula is as follows:
fl,i(t)=LiDl,i(t)τ-1 (3)
Wherein, fL, i(t) cpu frequency in t-th of timeslice, i-th of mobile device, D are indicatedL, i(t) it indicates t-th of time
The task amount that i-th of mobile device of piece locally executes, LiIndicate the cpu cycle that the every processing 1bit required by task of CPU is wanted；Movement is set
Standby locally executes power consumption pL, i(t) directly proportional to the frequency of CPU, formula is as follows:
Wherein, kmod,iIndicate effective switching capacity of the CPU of i-th of mobile device；
The radio transmitted power of mobile device is pr,i(t) formula is as follows:
Dr,i(t)=∑j∈Sri,j(t)τ (6)
Wherein, ri,j(t) message transmission rate between mobile device i to server j, B are indicatedjIndicate server to each movement
The amount of bandwidth of equipment distribution, Γi,j(t) indicate that channel fading power between mobile device i and server j, τ indicate timeslice
Interval, in t-th of timeslice, xi,j(t)=1 it indicates that i-th of mobile device unloads task to j-th of server, otherwise indicates
It does not unload；
Channel is divided into different subchannels by frequency division multiple access technology by Bandwidth Model；
QoE model, including task delay and mobile device power consumption, according to preset rules to task postpone and mobile device power consumption into
The QoE evaluation criterion of row linear weighted function composition user, wherein the task decay part of QoECalculation formula are as follows:
Wherein, α is weighting parameters, and α ∈ [0,1],
The power consumption sections of QoECalculation formula are as follows:
Synthesis obtains the calculation formula of QoE are as follows:
Wherein, β is weighting parameters, and β ∈ [0,1].
5. the method as described in claim 1, which is characterized in that step S4 is specifically included:
Step S4.1: building Lyapunov function, specific building formula are as follows:
Wherein, Θ (t) was indicated within t-th of period, the state of all task buffer queues, Qi(t) t-th of period is indicated
The task buffer queue length of interior i-th of mobile device local, Hi(t) indicate that i-th of mobile device is taking in t-th of period
The virtual task length of buffer queue being engaged on device；
Step S4.2: building Lyapunov drift function, specific formula is as follows:
Wherein,Expectation is asked in expression；
Step S4.3: building Lyapunov drift plus penalty:
Wherein, V indicates that penalty factor, ξ (t) indicate that the QoE of user is spent, which has a upper limit, is limited to thereon:
Wherein, C is constant.
6. method according to claim 2, which is characterized in that step S6 is specifically included:
For the task amount locally executed, mobile device local is obtained after Lyapunov drift plus penalty are unfolded and are split
Cpu frequency regulation scheme, specific formulation indicate are as follows:
It finds out formula (14) acquirement minimum value Shi Zewei and locally executes cpu frequency, locally execute cpu frequency expression are as follows:
For being unloaded to the task amount of Edge Server execution, after Lyapunov drift plus penalty are unfolded and are split, obtain
Unloading and mobile device transimission power decision scheme must be calculated, formulation indicates are as follows:
minP (t), x (t)∑i∈U[-(Qi(t)-Hi(t)+V·αβ)Dr,i(t)+V·(1-β)·pR, i(t)] (16)。
7. method as claimed in claim 6, which is characterized in that obtain and calculate unloading and mobile device transimission power decision-making party
Case specifically includes:
Formula (16) is split as the wireless signal transmission power of mobile device and calculates unloading decision, wherein the nothing of mobile device
Line signal transmission power, the optimization after fractionation are formulated and are indicated are as follows:
The wireless transmitted power for being when formula (17) obtain minimum value, can formulate expression are as follows:
The optimization formula for calculating unloading decision may be expressed as:
Wherein, for each mobile device, work as Qi(t)-Hi(t) when+V α β≤0, i-th of mobile device unloading a part is indicated
In task to Edge Server.
8. the method for claim 7, which is characterized in that described for being unloaded to the task amount of Edge Server execution
Method further include:
Server computational resource allocation is carried out, optimization formula may be expressed as:
minD(t)∑i∈U[-[V·(1-α)β+Hi(t)]·Ds,i(t)] (20)。
9. the mobile edge calculations discharge mechanism of multiserver based on Lyapunov optimization characterized by comprising
Queue constructs module, and the task for the generation for each mobile device constructs task buffer queue, and is each movement
Equipment needs to unload but not processed task creation virtual task buffer queue；
Model building module, for establishing power consumption of terminal model, Bandwidth Model and QoE model；
Information collection module, for collecting task buffer queuing message, the virtual cache of mobile device in each preset time piece
The bandwidth situation of queuing message and server；
Function constructs module, for according to the task buffer queuing message of collection, virtual cache queuing message and server
Bandwidth situation, the drift of building Liapunov plus penalty；
Decision-making module is unloaded, for unloading based on Liapunov drift plus penalty, the calculating solved in current time piece
Carry decision, wherein calculate unloading decision to characterize mobile terminal generation the task amount that is performed locally of calculating task and
It is unloaded to the task amount of Edge Server execution.
10. a kind of computer equipment including memory, processor and stores the meter that can be run on a memory and on a processor
Calculation machine program, which is characterized in that realized when the processor executes described program as any one of claims 1 to 8 right is wanted
Seek the method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910004742.4A CN109857546B (en) | 2019-01-03 | 2019-01-03 | Multi-server mobile edge computing unloading method and device based on Lyapunov optimization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910004742.4A CN109857546B (en) | 2019-01-03 | 2019-01-03 | Multi-server mobile edge computing unloading method and device based on Lyapunov optimization |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109857546A true CN109857546A (en) | 2019-06-07 |
| CN109857546B CN109857546B (en) | 2021-02-02 |
Family
ID=66893760
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910004742.4A Active CN109857546B (en) | 2019-01-03 | 2019-01-03 | Multi-server mobile edge computing unloading method and device based on Lyapunov optimization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109857546B (en) |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110287034A (en) * | 2019-07-04 | 2019-09-27 | 重庆大学 | The dynamic task allocation method of energy-delay balance in a kind of chargeable mobile edge calculations |
| CN110290011A (en) * | 2019-07-03 | 2019-09-27 | 中山大学 | Dynamic Service laying method based on Lyapunov control optimization in edge calculations |
| CN110351760A (en) * | 2019-07-19 | 2019-10-18 | 重庆邮电大学 | A kind of mobile edge calculations system dynamic task unloading and resource allocation methods |
| CN110662221A (en) * | 2019-09-10 | 2020-01-07 | 杭州电子科技大学 | Resource allocation method for security and performance perception of enterprise multimedia in MEC |
| CN110688221A (en) * | 2019-09-09 | 2020-01-14 | 北京邮电大学 | Dynamic task scheduling method for augmented reality application in edge computing |
| CN111132235A (en) * | 2019-12-27 | 2020-05-08 | 东北大学秦皇岛分校 | Mobile offload migration algorithm based on improved HRRN algorithm and multi-attribute decision |
| CN111124639A (en) * | 2019-12-11 | 2020-05-08 | 安徽大学 | Operation method and system of edge computing system and electronic equipment |
| CN111258677A (en) * | 2020-01-16 | 2020-06-09 | 重庆邮电大学 | Task unloading method for heterogeneous network edge computing |
| CN111511028A (en) * | 2020-04-13 | 2020-08-07 | 北京邮电大学 | Multi-user resource allocation method, device, system and storage medium |
| CN111757354A (en) * | 2020-06-15 | 2020-10-09 | 武汉理工大学 | Multi-user slicing resource allocation method based on competitive game |
| CN112600921A (en) * | 2020-12-15 | 2021-04-02 | 重庆邮电大学 | Heterogeneous mobile edge network-oriented dynamic task unloading method |
| CN112601197A (en) * | 2020-12-18 | 2021-04-02 | 重庆邮电大学 | Resource optimization method in train-connected network based on non-orthogonal multiple access |
| CN112612610A (en) * | 2020-12-18 | 2021-04-06 | 广州竞远安全技术股份有限公司 | SOC service quality guarantee system and method based on Actor-Critic deep reinforcement learning |
| WO2021103991A1 (en) * | 2019-11-25 | 2021-06-03 | 华为技术有限公司 | Resource allocation method and communication device |
| CN113010317A (en) * | 2021-03-30 | 2021-06-22 | 华南理工大学 | Method, device, computer equipment and medium for joint service deployment and task unloading |
| CN113064665A (en) * | 2021-03-18 | 2021-07-02 | 四川大学 | Multi-server computing unloading method based on Lyapunov optimization |
| CN113114733A (en) * | 2021-03-24 | 2021-07-13 | 重庆邮电大学 | Distributed task unloading and computing resource management method based on energy collection |
| CN113326130A (en) * | 2021-05-31 | 2021-08-31 | 北京邮电大学 | Resource allocation method and device |
| CN113342514A (en) * | 2021-05-11 | 2021-09-03 | 武汉理工大学 | Edge calculation model based on near-earth orbit and service placement method thereof |
| CN113377447A (en) * | 2021-05-28 | 2021-09-10 | 四川大学 | Multi-user computing unloading method based on Lyapunov optimization |
| CN113950081A (en) * | 2021-10-08 | 2022-01-18 | 东北大学 | Dynamic service migration and request routing method facing microservice in multi-unit mobile edge computing |
| CN114051266A (en) * | 2021-11-08 | 2022-02-15 | 首都师范大学 | Wireless body area network task unloading method based on mobile cloud-edge computing |
| CN114302233A (en) * | 2021-12-10 | 2022-04-08 | 网络通信与安全紫金山实验室 | Video compression and network service quality joint decision method and device |
| CN114691362A (en) * | 2022-03-22 | 2022-07-01 | 重庆邮电大学 | Edge calculation method for compromising time delay and energy consumption |
| CN117545017A (en) * | 2024-01-09 | 2024-02-09 | 大连海事大学 | Online computing and unloading method for wireless energy supply mobile edge network |
| CN114691362B (en) * | 2022-03-22 | 2024-04-30 | 重庆邮电大学 | Edge computing method for time delay and energy consumption compromise |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050240745A1 (en) * | 2003-12-18 | 2005-10-27 | Sundar Iyer | High speed memory control and I/O processor system |
| US20170164237A1 (en) * | 2015-12-03 | 2017-06-08 | The Trustees Of The Stevens Institute Of Technology | System Apparatus And Methods For Cognitive Cloud Offloading In A Multi-Rat Enabled Wireless Device |
| CN107682443A (en) * | 2017-10-19 | 2018-02-09 | 北京工业大学 | Joint considers the efficient discharging method of the mobile edge calculations system-computed task of delay and energy expenditure |
| CN107708152A (en) * | 2017-11-28 | 2018-02-16 | 重庆邮电大学 | The task discharging method of isomery cellular network |
| CN108809695A (en) * | 2018-04-28 | 2018-11-13 | 国网浙江省电力有限公司电力科学研究院 | A kind of distribution uplink unloading strategy towards mobile edge calculations |
-
2019
- 2019-01-03 CN CN201910004742.4A patent/CN109857546B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050240745A1 (en) * | 2003-12-18 | 2005-10-27 | Sundar Iyer | High speed memory control and I/O processor system |
| US20170164237A1 (en) * | 2015-12-03 | 2017-06-08 | The Trustees Of The Stevens Institute Of Technology | System Apparatus And Methods For Cognitive Cloud Offloading In A Multi-Rat Enabled Wireless Device |
| CN107682443A (en) * | 2017-10-19 | 2018-02-09 | 北京工业大学 | Joint considers the efficient discharging method of the mobile edge calculations system-computed task of delay and energy expenditure |
| CN107708152A (en) * | 2017-11-28 | 2018-02-16 | 重庆邮电大学 | The task discharging method of isomery cellular network |
| CN108809695A (en) * | 2018-04-28 | 2018-11-13 | 国网浙江省电力有限公司电力科学研究院 | A kind of distribution uplink unloading strategy towards mobile edge calculations |
Non-Patent Citations (2)
| Title |
|---|
| YUYI MAO ET AL: "Stochastic Joint Radio and Computational Resource Management for Multi-User Mobile-Edge Computing Systems", 《IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS》 * |
| ZILING WEI ET.AL: "Optimal Offloading in Fog Computing System With Non-Orthogonal Multiple Access", 《IEEE ACCESS》 * |
Cited By (43)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110290011A (en) * | 2019-07-03 | 2019-09-27 | 中山大学 | Dynamic Service laying method based on Lyapunov control optimization in edge calculations |
| CN110287034A (en) * | 2019-07-04 | 2019-09-27 | 重庆大学 | The dynamic task allocation method of energy-delay balance in a kind of chargeable mobile edge calculations |
| CN110351760A (en) * | 2019-07-19 | 2019-10-18 | 重庆邮电大学 | A kind of mobile edge calculations system dynamic task unloading and resource allocation methods |
| CN110351760B (en) * | 2019-07-19 | 2022-06-03 | 重庆邮电大学 | Dynamic task unloading and resource allocation method for mobile edge computing system |
| CN110688221A (en) * | 2019-09-09 | 2020-01-14 | 北京邮电大学 | Dynamic task scheduling method for augmented reality application in edge computing |
| CN110688221B (en) * | 2019-09-09 | 2022-01-21 | 北京邮电大学 | Dynamic task scheduling method for augmented reality application in edge computing |
| CN110662221A (en) * | 2019-09-10 | 2020-01-07 | 杭州电子科技大学 | Resource allocation method for security and performance perception of enterprise multimedia in MEC |
| WO2021103991A1 (en) * | 2019-11-25 | 2021-06-03 | 华为技术有限公司 | Resource allocation method and communication device |
| CN111124639A (en) * | 2019-12-11 | 2020-05-08 | 安徽大学 | Operation method and system of edge computing system and electronic equipment |
| CN111124639B (en) * | 2019-12-11 | 2023-05-23 | 安徽大学 | Operation method and system of edge computing system and electronic equipment |
| CN111132235A (en) * | 2019-12-27 | 2020-05-08 | 东北大学秦皇岛分校 | Mobile offload migration algorithm based on improved HRRN algorithm and multi-attribute decision |
| CN111132235B (en) * | 2019-12-27 | 2023-04-07 | 东北大学秦皇岛分校 | Mobile offload migration algorithm based on improved HRRN algorithm and multi-attribute decision |
| 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 |
| CN111258677A (en) * | 2020-01-16 | 2020-06-09 | 重庆邮电大学 | Task unloading method for heterogeneous network edge computing |
| CN111511028A (en) * | 2020-04-13 | 2020-08-07 | 北京邮电大学 | Multi-user resource allocation method, device, system and storage medium |
| US11716748B2 (en) | 2020-06-15 | 2023-08-01 | Wuhan University Of Technology | Multi-user slice resource allocation method based on competitive game |
| CN111757354A (en) * | 2020-06-15 | 2020-10-09 | 武汉理工大学 | Multi-user slicing resource allocation method based on competitive game |
| CN112600921A (en) * | 2020-12-15 | 2021-04-02 | 重庆邮电大学 | Heterogeneous mobile edge network-oriented dynamic task unloading method |
| CN112600921B (en) * | 2020-12-15 | 2022-05-10 | 重庆邮电大学 | Heterogeneous mobile edge network-oriented dynamic task unloading method |
| CN112612610A (en) * | 2020-12-18 | 2021-04-06 | 广州竞远安全技术股份有限公司 | SOC service quality guarantee system and method based on Actor-Critic deep reinforcement learning |
| CN112601197A (en) * | 2020-12-18 | 2021-04-02 | 重庆邮电大学 | Resource optimization method in train-connected network based on non-orthogonal multiple access |
| CN112612610B (en) * | 2020-12-18 | 2021-08-03 | 广州竞远安全技术股份有限公司 | SOC service quality guarantee system and method based on Actor-Critic deep reinforcement learning |
| CN112601197B (en) * | 2020-12-18 | 2022-04-05 | 重庆邮电大学 | Resource optimization method in train-connected network based on non-orthogonal multiple access |
| CN113064665A (en) * | 2021-03-18 | 2021-07-02 | 四川大学 | Multi-server computing unloading method based on Lyapunov optimization |
| CN113114733A (en) * | 2021-03-24 | 2021-07-13 | 重庆邮电大学 | Distributed task unloading and computing resource management method based on energy collection |
| CN113010317A (en) * | 2021-03-30 | 2021-06-22 | 华南理工大学 | Method, device, computer equipment and medium for joint service deployment and task unloading |
| CN113010317B (en) * | 2021-03-30 | 2023-08-22 | 华南理工大学 | Combined service deployment and task offloading method and device, computer equipment and medium |
| CN113342514B (en) * | 2021-05-11 | 2023-11-07 | 武汉理工大学 | Edge calculation model based on near-earth orbit and service placement method thereof |
| CN113342514A (en) * | 2021-05-11 | 2021-09-03 | 武汉理工大学 | Edge calculation model based on near-earth orbit and service placement method thereof |
| CN113377447A (en) * | 2021-05-28 | 2021-09-10 | 四川大学 | Multi-user computing unloading method based on Lyapunov optimization |
| CN113377447B (en) * | 2021-05-28 | 2023-03-21 | 四川大学 | Multi-user computing unloading method based on Lyapunov optimization |
| CN113326130A (en) * | 2021-05-31 | 2021-08-31 | 北京邮电大学 | Resource allocation method and device |
| CN113326130B (en) * | 2021-05-31 | 2022-07-12 | 北京邮电大学 | Resource allocation method and device |
| CN113950081B (en) * | 2021-10-08 | 2024-03-22 | 东北大学 | Dynamic service migration and request routing method for micro service |
| CN113950081A (en) * | 2021-10-08 | 2022-01-18 | 东北大学 | Dynamic service migration and request routing method facing microservice in multi-unit mobile edge computing |
| CN114051266A (en) * | 2021-11-08 | 2022-02-15 | 首都师范大学 | Wireless body area network task unloading method based on mobile cloud-edge computing |
| 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 |
| CN114302233A (en) * | 2021-12-10 | 2022-04-08 | 网络通信与安全紫金山实验室 | Video compression and network service quality joint decision method and device |
| CN114302233B (en) * | 2021-12-10 | 2023-10-27 | 网络通信与安全紫金山实验室 | Video compression and network service quality joint decision method and device |
| CN114691362A (en) * | 2022-03-22 | 2022-07-01 | 重庆邮电大学 | Edge calculation method for compromising time delay and energy consumption |
| CN114691362B (en) * | 2022-03-22 | 2024-04-30 | 重庆邮电大学 | Edge computing method for time delay and energy consumption compromise |
| CN117545017A (en) * | 2024-01-09 | 2024-02-09 | 大连海事大学 | Online computing and unloading method for wireless energy supply mobile edge network |
| CN117545017B (en) * | 2024-01-09 | 2024-03-19 | 大连海事大学 | Online computing and unloading method for wireless energy supply mobile edge network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109857546B (en) | 2021-02-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109857546A (en) | The mobile edge calculations discharging method of multiserver and device based on Lyapunov optimization | |
| CN109829332B (en) | Joint calculation unloading method and device based on energy collection technology | |
| CN109788069B (en) | Computing unloading method based on mobile edge computing in Internet of things | |
| Zhang et al. | A hierarchical game framework for resource management in fog computing | |
| Nguyen et al. | Price-based resource allocation for edge computing: A market equilibrium approach | |
| Xu et al. | Game theory for distributed IoV task offloading with fuzzy neural network in edge computing | |
| Dai et al. | A probabilistic approach for cooperative computation offloading in MEC-assisted vehicular networks | |
| CN104850450B (en) | A kind of load-balancing method and system towards mixed cloud application | |
| Lin et al. | Stochastic digital-twin service demand with edge response: An incentive-based congestion control approach | |
| Xiao et al. | System delay optimization for mobile edge computing | |
| CN102711129B (en) | The determination method and device of net planning parameter | |
| Xie et al. | Distributed multi-dimensional pricing for efficient application offloading in mobile cloud computing | |
| Chamola et al. | Latency aware mobile task assignment and load balancing for edge cloudlets | |
| Li et al. | Optimal pricing and service selection in the mobile cloud architectures | |
| CN110149657A (en) | A kind of method and apparatus of determining QoS description information | |
| CN110662238A (en) | Reinforced learning scheduling method and device for burst request under edge network | |
| CN108900628A (en) | Thin cloud computational resource allocation method in edge calculations environment based on pricing mechanism | |
| Jia et al. | Delay-sensitive multiplayer augmented reality game planning in mobile edge computing | |
| Shu et al. | An edge computing offloading mechanism for mobile peer sensing and network load weak balancing in 5G network | |
| CN110069342A (en) | Net cast channel dispositions method is merged under a kind of mobile cloud computing environment | |
| Alencar et al. | Dynamic microservice allocation for virtual reality distribution with qoe support | |
| Tian et al. | User preference-based hierarchical offloading for collaborative cloud-edge computing | |
| Wang et al. | Wireless network aware cloud scheduler for scalable cloud mobile gaming | |
| Zhang et al. | Theoretical analysis on edge computation offloading policies for IoT devices | |
| Ma et al. | Dynamic task scheduling in cloud-assisted mobile edge computing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |