CN110418356A - A kind of calculating task discharging method, device and computer readable storage medium - Google Patents
A kind of calculating task discharging method, device and computer readable storage medium Download PDFInfo
- Publication number
- CN110418356A CN110418356A CN201910527427.XA CN201910527427A CN110418356A CN 110418356 A CN110418356 A CN 110418356A CN 201910527427 A CN201910527427 A CN 201910527427A CN 110418356 A CN110418356 A CN 110418356A
- Authority
- CN
- China
- Prior art keywords
- individual
- task
- user terminal
- base station
- micro
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/142—Network analysis or design using statistical or mathematical methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- 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
- H04W16/22—Traffic simulation tools or models
Abstract
A kind of disclosed calculating task discharging method, device and computer readable storage medium according to embodiments of the present invention calculate separately the call duration time of user terminal unloading calculating task under mobile edge calculations environmentWith unloading energy consumptionUser terminal executes the task execution time of local calculating taskWith locally execute energy consumptionMobile edge calculations MEC server executes the task execution time of the calculating task of unloadingIt is based onWithThe average energy consumption of user is calculated, and is based onWithThe average response of calculating task postpones;Model for Multi-Objective Optimization is established in the average response delay of average energy consumption and task for user;Model for Multi-Objective Optimization is solved, the calculating unloading decision scheme of multiple satisfactions is obtained.The present invention solves after converting multi-objective optimization question for calculating task unloading problem, obtain the unloading decision scheme of multiple satisfactions, under fast-changing network environment, policymaker can be made to possess more selections, and more conducively adapt to the demand changed at any time.
Description
Technical field
The present invention relates to field of communication technology more particularly to a kind of calculating task discharging methods, device and computer-readable
Storage medium.
Background technique
With the continuous expansion of Internet of Things scale, the quantity and data volume rapid growth of smart machine, but hardware technology
It muchly stagnates, also therefore results in the bad difficult situation of most Internet of Things application Quality of experience.Mobile edge
The appearance of (Mobile Edge Computing, the MEC) technology of calculating provides completely new solution to the alleviation of this difficult situation
Scheme.Mobile edge calculations can provide more strong big resource by calculating Unloading Technology for equipment, prolong meeting application
The purpose for reducing energy consumption and improving user experience is also achieved while Shi Yaoqiu.
In general, the unloading decision index system for calculating Unloading Technology includes reducing delay, reducing energy consumption and delay and energy
Three kinds of balance between consumption.For the task of delay-sensitive, it is natural that preferential selection, which reduces delay,;It is close for calculating
Collection type task, in order to extend the life cycle of equipment, reducing energy consumption is better choice;For the task of some complexity, such as people
Face identification, car networking etc., delay and energy consumption all directly affect the Quality of experience of user, it is therefore desirable to consider simultaneously.Further, since
The unstable factors such as the variation of variation and user demand of channel quality, the priority for energy consumption and delay are different.
Currently, being considered in the calculating discharging method of delay and energy consumption at the same time, being usually all will using the method for weighting
Consumption and time delay are unified into single-objective problem to be solved, however are delayed in such way with the weight of energy consumption with very big
Subjectivity, and unique solution can only be found out, to cause optimum results excessively single, it can not consider global optimization, be unable to satisfy
The demand that user changes at any time.
Summary of the invention
The main purpose of the embodiment of the present invention is to provide a kind of calculating task discharging method, device and computer-readable deposits
Storage media is at least able to solve and energy consumption and time delay is unified into single-objective problem to carry out using the method for weighting in the related technology
It solves, caused calculating unloading decision is more single, can not consider global unloading decision and be unable to satisfy user with time-varying
The problem of demand of change.
To achieve the above object, first aspect of the embodiment of the present invention provides a kind of calculating task discharging method, is applied to
Multiuser mobile communication network includes multiple communication junior units and leads to the multiple in the multiuser mobile communication network
The macro base station that is connected by cable network of letter junior unit, the communication junior unit include multiple micro-base stations and with each micro- base
The user terminal stood through wireless network connection is provided with mobile edge calculations MEC server, this method packet on the macro base station
It includes:
Calculate separately the call duration time of the user terminal unloading calculating taskWith unloading energy consumptionThe user
Terminal executes the task execution time of local calculating taskWith locally execute energy consumptionThe MEC server executes
The task execution time of the calculating task to come from user terminal unloading
Based on describedWithThe average energy consumption of user is calculated, and based on describedWithIt calculates
The average response of task postpones;
Model for Multi-Objective Optimization is established in the average response delay of average energy consumption and the task for the user;It is described
Model for Multi-Objective Optimization indicates are as follows:
Wherein, the P1 is optimization aim, and described C1, C2 and C3 are constraint condition, the Om,j,kIt is described micro- for m-th
The unloading decision of k-th of calculating task of j-th of user terminal of base station connection, the E are the flat of the user
Equal energy consumption, the average response that the T is the task postpone, the fm,jFor described in j-th of m-th of micro-base station connection
The task processing speed of user terminal, the PRm,jDescribed in j-th of user terminal for m-th of micro-base station connection
Calculating task is performed locally shared percentage, the λm,jJ-th of user for m-th of micro-base station connection is whole
The average arrival rate of the calculating task at end, the F are the task processing speed of the MEC server, and the NS is micro- base
The quantity stood, the NmFor the user terminal total quantity of m-th of micro-base station connection;
The Model for Multi-Objective Optimization is solved, the unloading decision scheme of multiple satisfactions is obtained.
To achieve the above object, second aspect of the embodiment of the present invention provides a kind of calculating task discharge mechanism, is applied to
Multiuser mobile communication network includes multiple communication junior units and leads to the multiple in the multiuser mobile communication network
The macro base station that is connected by cable network of letter junior unit, the communication junior unit include multiple micro-base stations and with each micro- base
The user terminal stood through wireless network connection is provided with MEC server on the macro base station, which includes:
First computing module, for calculating separately the call duration time of the user terminal unloading calculating taskAnd unloading
Energy consumptionThe user terminal executes the task execution time of local calculating taskWith locally execute energy consumption
The MEC server executes the task execution time of the calculating task to come from user terminal unloading
Second computing module, for based on describedWithThe average energy consumption of user is calculated, and based on describedWithThe average response of calculating task postpones;
Module is established, multiple target is established in the average response delay for the average energy consumption and the task for the user
Optimized model;The Model for Multi-Objective Optimization indicates are as follows:
Wherein, the P1 is optimization aim, and described C1, C2 and C3 are constraint condition, the Om,j,kIt is described micro- for m-th
The unloading decision of k-th of calculating task of j-th of user terminal of base station connection, the E are the flat of the user
Equal energy consumption, the average response that the T is the task postpone, the fm,jFor described in j-th of m-th of micro-base station connection
The task processing speed of user terminal, the PRm,jDescribed in j-th of user terminal for m-th of micro-base station connection
Calculating task is performed locally shared percentage, the λm,jJ-th of user for m-th of micro-base station connection is whole
The average arrival rate of the calculating task at end, the F are the task processing speed of the MEC server, and the NS is micro- base
The quantity stood, the NmFor the user terminal total quantity of m-th of micro-base station connection;
It solves module and obtains the unloading decision scheme of multiple satisfactions for solving to the Model for Multi-Objective Optimization.
To achieve the above object, the third aspect of the embodiment of the present invention provides a kind of electronic device, which includes:
Processor, memory and communication bus;
The communication bus is for realizing the connection communication between the processor and memory;
The processor is above-mentioned any one to realize for executing one or more program stored in the memory
The step of kind calculating task discharging method.
To achieve the above object, fourth aspect of the embodiment of the present invention provides a kind of computer readable storage medium, the meter
Calculation machine readable storage medium storing program for executing is stored with one or more program, and one or more of programs can be by one or more
It manages device to execute, the step of to realize any one of the above calculating task discharging method.
Calculating task discharging method, device and the computer readable storage medium provided according to embodiments of the present invention, respectively
Calculate the call duration time of user terminal unloading calculating task under mobile edge calculations environmentWith unloading energy consumptionUser is whole
End executes the task execution time of local calculating taskWith locally execute energy consumptionMEC server executes the meter of unloading
The task execution time of calculation taskIt is based onWithThe average energy consumption of user is calculated, and is based on
WithThe average response of calculating task postpones;Multiple target is established in the average response delay of average energy consumption and task for user
Optimized model;Model for Multi-Objective Optimization is solved, the unloading decision scheme of multiple satisfactions is obtained.Reality through the invention
It applies, is solved after converting multi-objective optimization question for calculating task unloading problem, and obtain the unloading decision of multiple satisfactions
Scheme can make policymaker possess more selections under fast-changing network environment, and be more conducive to adapt to change at any time
Demand.
Other features of the invention and corresponding effect are described in the aft section of specification, and should be appreciated that
At least partly effect is apparent from from the record in description of the invention.
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 only this
Some embodiments of invention for those skilled in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.
Fig. 1 is the system architecture schematic diagram for the multiuser mobile communication network that first embodiment of the invention provides;
Fig. 2 is the basic procedure schematic diagram for the calculating task discharging method that first embodiment of the invention provides;
Fig. 3 is the basic procedure schematic diagram for the method for solving based on NSGA-II algorithm that first embodiment of the invention provides;
Fig. 4 is the composition schematic diagram for the initial population that first embodiment of the invention provides;
Fig. 5 is that the crowding distance that first embodiment of the invention provides calculates schematic diagram;
Fig. 6 is the mutation operation schematic diagram that first embodiment of the invention provides;
Fig. 7 is the crossover operation schematic diagram that first embodiment of the invention provides;
Fig. 8 is a kind of structural schematic diagram for calculating task discharge mechanism that second embodiment of the invention provides;
Fig. 9 is the structural schematic diagram for the electronic device that third embodiment of the invention provides.
Specific embodiment
In order to make the invention's purpose, features and advantages of the invention more obvious and easy to understand, below in conjunction with the present invention
Attached drawing in embodiment, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described reality
Applying example is only a part of the embodiment of the present invention, and not all embodiments.Based on the embodiments of the present invention, those skilled in the art
Member's every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
First embodiment:
In order to solve energy consumption and delay are unified into single-objective problem using the method for weighting in the related technology to ask
Solution, caused calculating unloading decision is more single, can not consider that the overall situation unloads decision and is unable to satisfy user changes at any time
Demand the technical issues of, the present embodiment proposes a kind of calculating task discharging method, be applied to include multiple communication junior units
And the multiuser mobile communication networks of macro base station connect with multiple communication junior units by cable network, communication junior unit packet
Multiple micro-base stations and the user terminal with each micro-base station by wireless network connection are included, is provided with mobile edge meter on macro base station
Calculate MEC server.
It is as shown in Figure 1 the system architecture schematic diagram of multiuser mobile communication network provided in this embodiment, the present embodiment
Suitable for the mobile edge calculations system under multi-user scene.Wherein, have in the communication junior unit in network (Small cell)
Many micro-base stations (Small-cell eNodeB, SeNB), each micro-base station is connected with several Intelligent mobile equipments (Smart
Mobile Devices, SMD), i.e., the user terminal of the present embodiment, each user terminal only connect a micro-base station, and user is whole
It holds and passes through wireless network connection between corresponding micro-base station, and all micro-base stations are all connected to macro base station in wired form
(Macro eNodeB, MeNB) deploys mobile edge calculations (MEC, Mobile Edge Computing) clothes on the macro base station
Business device, the computing capability of MEC server are F.If NS is the quantity of micro-base station, m-th of micro-base station is known as Sm(m=1,
2,...,NS}).If the user terminal sum of m-th of micro-base station connection is Nm, remember that j-th of user of m-th of micro-base station connection is whole
End is Um,j.Each user terminal is referred to as 3 tuple Um,j=(fm,j,pm,j,em,j), wherein fm,jIt is Um,jComputing capability, here
Computing capability be alternatively referred to as task processing speed, pm,jIt is Um,jTransimission power, em,jIt is that subscriber terminal equipment executes calculating times
Coefficient of energy dissipation when business.
In addition, in this example, it is assumed that Um,jThe calculating for needing to handle several computation-intensives and delay-sensitive is appointed
Business, it is λ that the generation of these tasks, which follows average arrival rate,m,jPoisson process, and between these tasks independently of one another, cannot
It is divided again.Assuming that Um,jThe task amount of generation is Km,j, each task modeling is 2 tuple τ by the present embodimentm,j,k={ dm,j,k,
cm,j,k, wherein dm,j,kIt is task τm,j,kSize of data, cm,j,kIt is to execute task τm,j,kRequired calculated performance.It is each to appoint
Business can be performed locally or be unloaded to the execution of MEC server, and unloading decision when unloading task is denoted as Om,j,kIf, wherein
Om,j,k=1, which will be discharged into MEC server, whereas if Om,j,k=0, then task will be in the local of user terminal
It executes.
It is illustrated in figure 2 the basic procedure schematic diagram of calculating task discharging method provided in this embodiment, the present embodiment mentions
Calculating task discharging method out includes the following steps:
Step 201, the call duration time for calculating separately user terminal unloading calculating taskWith unloading energy consumptionUser
Terminal executes the task execution time of local calculating taskWith locally execute energy consumptionMEC server execute from
The task execution time for the calculating task that terminal unloading in family comes
Specifically, in the present embodiment, in entire calculating task uninstall process, the energy consumption of user terminal is divided into two portions
Point: locally execute the calculating energy consumption of task and the communication energy consumption of unloading required by task;And the operating lag of task is by three parts group
At: it locally executes delay, transmission delay and MEC server and executes delay.Based on this analysis, the present embodiment was respectively to communicating
Journey, local computing process and MEC server calculating process are modeled.
It should be noted that in this example, it is assumed that channel total bandwidth is B, and have N number of channel, wherein each letter
The bandwidth in road is equal, and the wireless communication in each communication junior unit between user terminal and micro-base station follows Lognormal shadowing and declines
Stamping die type.The distribution for communicating channel in junior unit is completely random, and each user terminal only distributes a channel, communicates small list
Communication Jamming is not present in user terminal in member, and Communication Jamming is only possible in other junior units using same channel
User terminal.It is to be further understood that the present embodiment only considers the process from user terminal unloading task to micro-base station, because micro-
By wired connection, transmission time almost be can be ignored for base station and macro base station, the task action result returned from micro-base station
Data volume it is usually very small, therefore, the time of transmission is far smaller than the time of upload procedure, and the time returned to result can be with
Ignored.
In the present embodiment, for user Um,jFor, it, can be according to Shannon- if there is the task that needs unload
The message transmission rate R of Hartley theorem calculating user terminalm,j, calculation formula expression are as follows:
Wherein, ω is that subband is wide, pm,jFor the transimission power of j-th of user terminal of m-th of micro-base station connection, Gm,jIt is
Channel gain between j-th of the user terminal and m-th of micro-base station of m micro-base station connection, σ2For ambient noise, Im,jFor m
The channel disturbance that j-th of user terminal of a micro-base station connection is subject to.
It should be noted that channel disturbance Im,jIt is obtained by following calculation formula:
Wherein, al,i,m,jWhether ∈ { 0,1 } is a binary variable, represent with the presence of interference.
Thus, it is possible to calculate the call duration time of unloading taskAre as follows:
Also, the energy consumption that the task of unloading is lostAre as follows:
In this example, it is assumed that the implementation procedure of task follows M/M/1 queuing model.Assuming that locally being held in user terminal
The percentage of capable task is PRm,j, then the mean residence time of these tasks is locally executed(including queueing delay and hold
Row delay) it is defined as follows:
And locally execute energy consumed by calculatingAre as follows:
Moreover, it is assumed that the task execution process of MEC server is also in compliance with M/M/1 queuing model.For MEC server, appoint
The arrival rate of business can calculate as follows:
To execute the required by task time in MECAre as follows:
Step 202 is based onWithThe average energy consumption of user is calculated, and is based onWithIt calculates
The average response of task postpones.
In the present embodiment, the total energy consumption of user terminal is the summation for locally executing energy consumption and transmission power consumption,
Then the average energy consumption of user can calculate as follows:
Wherein, em,jCoefficient of energy dissipation when task, K are executed for j-th of user terminal that m-th of micro-base station connectsm,jFor m
The task quantity that j-th of user terminal of a micro-base station connection generates, dm,j,kJ-th of user for m-th of micro-base station connection is whole
The data volume of k-th of calculating task at end, ω are the occupied channel width of j-th of user terminal of m-th of micro-base station connection,
pm,jFor the data transmission utilization measure of j-th of user terminal of m-th of micro-base station connection, Gm,jIt is j-th of the connection of m-th of micro-base station
Channel gain between user terminal and m-th of micro-base station, σ2For ambient noise, Im,jIt is j-th of the connection of m-th of micro-base station
The channel disturbance that user terminal is subject to.
In addition, in the present embodiment, the total delay time of task is transmission delay and MEC executes the sum of delay, or local
Execute delay.Then the average response delay of task can calculate as follows:
Step 203, for user average energy consumption and task average response delay establish Model for Multi-Objective Optimization.
Specifically, in the present embodiment, unloading decision process is modeled as the multiple-objection optimization comprising two objective functions
Problem, it is therefore intended that find the balance between response delay and energy consumption.Wherein, Model for Multi-Objective Optimization is indicated in the form of following:
Wherein, P1 is optimization aim, and C1, C2 and C3 are constraint condition, Om,j,kIt is j-th of the connection of m-th of micro-base station
The unloading decision of k-th of calculating task of user terminal, the average response delay that E is the average energy consumption of user, T is task, fm,j
For the task processing speed of j-th of user terminal of m-th of micro-base station connection, PRm,jFor percentage shared by local calculating task
Than λm,jFor the average arrival rate of the calculating task of j-th of user terminal of m-th of micro-base station connection, F is appointing for MEC server
Business processing speed, NS are the quantity of micro-base station, NmFor the user terminal total quantity of m-th of micro-base station connection.
Additionally, it should be noted that " s.t. " expression is confined to, is constrained in constraint condition C1 guarantee locally executes
Task arrival rate be less than user task processing speed;Constraint condition C2 ensures the task arrival rate of MEC server lower than MEC
Task processing speed;It is binary that constraint condition C3, which ensures to unload decision,.
Step 204 solves Model for Multi-Objective Optimization, obtains the unloading decision scheme of multiple satisfactions.
Optionally, in order to reach optimization aim P1, the present embodiment can use NSGA-II algorithm and solve to it.It is first
First, by the decision variable O in P1m,j,kThe individual being encoded into population.Assuming that the total number of users in network is
User terminal Um,jThere is Km,jA task needs to carry out decision, then the dimension of each individual is in populationPer one-dimensional
The unloading decision of all tasks of the corresponding user terminal of the value of degree.Due to unloading decision variable Om,j,kFor binary variable,
The value of every dimension corresponds to K simultaneously againm,jThe unloading decision of a task.Therefore, the present embodiment is by this Km,jThe group of a unloading decision
Conjunction is converted into decimal representation, as corresponds to the value of dimension.Example: user terminal Um,jThere is Km,j=3 tasks need to unload, corresponding
Unloading decision be { 1,0,1 }, then the user terminal correspond to the value of dimension in individual as 101 (binary system)=5 (decimal system).
Obviously, user terminal Um,jThe value range that the value of dimension is corresponded in individual isCorrespondingly,
After Evolution of Population is completed to obtain Pareto disaggregation, we need to be only decoded individual, i.e., convert two for decimal number
System indicates that the unloading decision of all tasks can be obtained.After decision variable is encoded into individual, it need to only utilize NSGA-II's
Algorithm mechanism finds out Pareto disaggregation, and multiple satisfactory solutions of problem P1 can be obtained.Specific NSGA-II algorithm operating mechanism is retouched
It states as follows:
Such as the flow diagram that Fig. 3 is the method for solving provided in this embodiment based on NSGA-II algorithm, the present embodiment exists
When being solved to Model for Multi-Objective Optimization, specifically includes the following steps:
Each unloading decision is encoded into the individual in population, and generates initial population by step 301.
It wherein, include NP individual in population, the dimension of each individual is the use in multiuser mobile communication network in population
The total quantity of family terminal, the unloading decision of all calculating tasks of the corresponding user terminal of the value of every dimension, m-th of micro- base
Stand j-th of the user terminal U connectedm,jThe value range that the value of dimension is corresponded in individual is
Additionally, it should be understood that the major parameter of NSGA-II is first initialized in the present embodiment before initialization of population, including
Population Size NP, maximum number of iterations G, crossover probability CR and mutation probability MP.
It is illustrated in figure 4 a kind of composition schematic diagram of initial population provided in this embodiment.Optionally, initial population is generated
It include: to generate two first kind individuals, the value of all dimensions of one of first kind individual is maximized Gx, another
The value of all dimensions of a kind of individual is minimized 0;Wherein,A is multi-user's shifting
The total quantity of user terminal in dynamic communication network,The value for generating A one of dimensions is 0, codimension
The value of degreeFor the second class individual;NP-A-2 third class individual is generated, all dimensions of all individuals in third class individual
The value of degree generates at random all in accordance with respective value range;It is made of just first kind individual, the second class individual and third class individual
Beginning population.
Step 302 calculates separately individual domination level corresponding to each individual.
Optionally, calculating separately individual domination level corresponding to each individual can be accomplished by the following way: respectively
The target function value of each individual is calculated, target function value includes the average response delay of the average energy consumption and task of user;It will
Meet the individual composition feasible zone of constraint condition, and the individual for being unsatisfactory for constraint condition is formed into non-feasible zone;Based on feasible zone
The functional value of each of interior individual each of determines in feasible zone that individual domination corresponding to individual is horizontal;It calculates in non-feasible zone
Each of individual constraint violation value, and based on each of in non-feasible zone individual constraint violation value and functional value, determine non-
Individual corresponding to individual dominates horizontal each of in feasible zone;Constraint violation value is used to characterize the individual in non-feasible zone and can
The distance in row domain.
Further, the constraint violation value for each of calculating in non-feasible zone individual includes:
The constraint violation value CV of individual each of is calculated in non-feasible zone using preset constraint violation value calculation formula, about
Beam violation value calculation formula indicates are as follows:
Wherein,
It should also be noted that the superiority and inferiority of each individual is assessed in the present embodiment by objective function, namely by aforementioned
E and T calculation formula come calculate unloading decision corresponding to energy consumption and delay size.It should be noted that may have
Individual is unsatisfactory for constraint condition, we claim these individuals to be infeasible solution, all individuals for being unsatisfactory for constraint condition
Constitute non-feasible zone.The value of CV can reflect them at a distance from feasible zone.In feasible zone, functional value is smaller, individual branch
With horizontal higher;Individual in feasible zone always dominates than the individual in non-feasible zone horizontal high;In non-feasible zone, the value of CV
It is smaller, dominate horizontal higher, when identical CV, functional value is smaller, and individual domination level is higher.
Step 303 dominates level based on all individuals calculated, carries out layer sorting to all individuals in population.
Specifically, the present embodiment is layered population, all individuals will all be distributed in different levels.The foundation of layering
It is the domination level of individual.Firstly, finding all individuals not dominated by any individual, in the entire population to form first
Layer.Then, we find the individual not dominated by any individual in remaining individual, to form the second layer.Repeat this always
Process, until level is all assigned in all individuals.
Step 304, in each level after layer sorting individual carry out layer internal sort, and using the population after sequence as
Parent population.
Optionally, the mode for carrying out layer internal sort to the individual of same level in the present embodiment can be real in the following manner
It is existing:
The crowding distance of the individual in each level after being calculated layer sorting using preset crowding distance calculation formula, is gathered around
Squeezing distance calculation formula indicates are as follows:
Wherein, CDiFor individual xiCrowding distance, l be in same level individual quantity, M be objective function number
Amount,WithThe maximum value and minimum value of respectively k-th target function value, target function value include the average energy of user
The average response of consumption and task postpones;
Carry out layer internal sort based on the crowding distance of the individual in each level calculated, to the individual in each level.
Specifically, the present embodiment passes through crowding distance (Crowding Distance, CD) after being layered to population
To assess the diversity of individual.It is illustrated in figure 5 crowding distance provided in this embodiment and calculates schematic diagram, wherein abscissa f1
With ordinate f2Energy consumption and delay are respectively indicated, circle represents the individual for being in same level, xiFor target individual, xi+1And
xi-1For two individuals nearest with target individual.For target individual xi, consider and it be in two of same level nearest
The rectangle that body is constituted, when constructing the rectangle, the length of rectangle and wide parallel with two reference axis respectively, and it is nearest with target individual
Two same layer individuals be located at rectangle it is diagonal on, thus in target individual is enclosed in by the rectangle, and individual xi
CD be the rectangle average side length.
Additionally, it should be noted that for any two individual x1And x2, x1Compare x2Good (x1Dominate x2) and if only if satisfaction
One of two conditions:
1)x1Level ratio x2It is higher;
2)x1And x2In same layer but
The individual of same level can be ranked up according to this principle, then all individuals will all obtain final row
Sequence.Level serial number is smaller, and the level is not higher, and it is individual better to represent, and the probability for being chosen as next-generation individual is higher.
Step 305 carries out intersection or mutation operation to the individual in parent population, obtains offspring flocks.
Specifically, the present embodiment randomly chooses two individuals in population and generates random in (0, a 1) range
Number.If this value is greater than crossover probability CR, mutation operation will be carried out to the two individuals.Otherwise, crossover operation will be executed.
It is illustrated in figure 6 mutation operation schematic diagram provided in this embodiment, the present embodiment is according to following formula to individual
Per one-dimensional carry out mutation operation:
Wherein, x 'iIt is xiOffspring, j ∈ [1, A] is individual xiDimension, j=jrand ensures at least one dimension meeting
It changes.
It is illustrated in figure 7 crossover operation schematic diagram provided in this embodiment, the present embodiment is according to following formula to individual
Per one-dimensional carry out mutation operation:
Wherein x '1With x '2It is all offspring, and θ ∈ (0,1), for controlling degree (Fig. 7 of entire algorithm exploration and development
Middle θ=0.2)
Step 306 calculates separately individual domination level corresponding to each individual in offspring flocks, and based on including parent
The individual of all individuals dominates horizontal in population and the whole population of offspring flocks, carries out to all individuals in whole population
Layer sorting, and layer internal sort is carried out to the individual in each level after the layering of whole population, obtain the sequence knot of whole population
Fruit.
Step 307, according to the ranking results of whole population, select NP optimal individual to form next-generation parent population,
The number of iterations adds one.Step 305 is gone to, step 305 is repeated to step 307 to next-generation parent population,.
Step 308, when reaching preset the number of iterations, using the first layer individual of finally obtained population as disaggregation
Output, and each individual in output result is decoded respectively, obtain the unloading decision scheme of multiple satisfactions.
Specifically, after generation for population after, the present embodiment continues to estimate that the individual domination of offspring flocks is horizontal, then,
Layer sorting and level internal sort are carried out to the entire population for including parent population and offspring flocks, according to ranking results, selection
Best preceding NP individual continuously forms next-generation population and continues algorithm iteration, when algorithm follows always badly until reaching maximum
The number of iterations then regard the individual output of first layer as Pareto disaggregation, and these solutions is converted to binary form as most
Good unloading decision scheme, it is best to unload decision scheme, that is, multiple satisfactions unloading decision scheme.
The calculating task discharging method provided according to embodiments of the present invention calculates separately user terminal unloading calculating task
Call duration timeWith unloading energy consumptionUser terminal executes the task execution time of local calculating taskThe local and
Execute energy consumptionMEC server executes the task execution time of the calculating task of unloadingIt is based onWithIt calculates
The average energy consumption of user, and be based onWithThe average response of calculating task postpones;For being averaged for user
Model for Multi-Objective Optimization is established in the average response delay of energy consumption and task;Model for Multi-Objective Optimization is solved, is obtained multiple
Satisfied unloading decision scheme.Implementation through the invention, after converting multi-objective optimization question for calculating task unloading problem
It is solved, and obtains the unloading decision scheme of multiple satisfactions, under fast-changing network environment, can policymaker be gathered around
There are more selections, and more conducively adapts to the demand changed at random.
Second embodiment:
In order to solve energy consumption and time delay are unified into single-objective problem using the method for weighting in the related technology to ask
Solution, caused calculating unloading decision is more single, can not consider that the overall situation unloads decision and is unable to satisfy user changes at any time
Demand the technical issues of, present embodiment illustrates a kind of calculating task discharge mechanism, be applied to include multiple communication junior units
And the multiuser mobile communication networks of macro base station connect with multiple communication junior units by cable network, communication junior unit packet
Multiple micro-base stations and the user terminal with each micro-base station by wireless network connection are included, is provided with mobile edge meter on macro base station
MEC server is calculated, specifically refers to Fig. 8, the calculating task discharge mechanism of the present embodiment includes:
First computing module 801, for calculating separately the call duration time of user terminal unloading calculating taskWith unloading energy
ConsumptionUser terminal executes the task execution time of local calculating taskWith locally execute energy consumptionMEC clothes
Business device executes the task execution time of the calculating task to come from user terminal unloading
Second computing module 802, for being based onWithThe average energy consumption of user is calculated, and is based on
WithThe average response of calculating task postpones;
Module 803 is established, multiple-objection optimization is established in the average response delay for the average energy consumption and task for user
Model;Model for Multi-Objective Optimization indicates are as follows:
Wherein, P1 is optimization aim, and C1, C2 and C3 are constraint condition, Om,j,kIt is j-th of the connection of m-th of micro-base station
The unloading decision of k-th of calculating task of user terminal, the average response delay that E is the average energy consumption of user, T is task, fm,j
For the task processing speed of j-th of user terminal of m-th of micro-base station connection, PRm,jFor percentage shared by local calculating task
Than λm,jFor the average arrival rate of the calculating task of j-th of user terminal of m-th of micro-base station connection, F is appointing for MEC server
Business processing speed, NS are the quantity of micro-base station, NmFor the user terminal total quantity of m-th of micro-base station connection;
Module 804 is solved, for solving to Model for Multi-Objective Optimization, obtains multiple satisfied unloading decisions.
In some embodiments of the present embodiment, the second computing module 802 is specifically used for based on following calculation formula
Calculate the average energy consumption of user:
And postponed based on the average response of following calculation formula calculating task:
In some embodiments of the present embodiment, solves module 804 and specifically include: generating submodule, for that will unload
Decision is encoded into the individual in population, and generates initial population, wherein and it include NP individual in population, each individual in population
Dimension is the total quantity of the user terminal in multiuser mobile communication network, the institute of the corresponding user terminal of the value of every dimension
There are the unloading decision of calculating task, j-th of user terminal U of m-th of micro-base station connectionm,jThe value of dimension is corresponded in individual
Value range isComputational submodule, for calculating separately individual branch corresponding to each individual
With level;First sorting sub-module, it is horizontal for being dominated based on all individuals calculated, all individuals in population are carried out
Layer sorting;Second sorting sub-module, for carrying out layer internal sort to the individual in each level after layer sorting, and will sequence
Population afterwards is as parent population;Submodule is operated, for carrying out intersection or mutation operation to the individual in parent population, is obtained
Offspring flocks;Third sorting sub-module, it is horizontal for calculating separately individual domination corresponding to each individual in offspring flocks, and
Individual based on individuals all in the whole population including parent population and offspring flocks dominates level, in whole population
All individuals carry out layer sorting, and carry out layer internal sort to the individual in each level after the layering of whole population, obtain entirety
The ranking results of population;Submodule is selected, for the ranking results according to whole population, is selected under NP optimal individual composition
Generation parent population, the number of iterations add one;Iteration submodule is repeated for next-generation parent population to be input to operation submodule
Corresponding step is executed, until preset the number of iterations is reached, the first layer individual of finally obtained population is defeated as disaggregation
Out, and to each individual in output result it is decoded respectively, obtains the unloading decision scheme of multiple satisfactions.
Further, in some embodiments of the present embodiment, submodule is generated when generating initial population, it is specific to use
In generating two first kind individuals, the value of all dimensions of one of first kind individual is maximized Gx, another first
The value of all dimensions of class individual is minimized 0;Wherein,A is mobile for multi-user
The total quantity of user terminal in communication network,The value for generating A one of dimensions is 0, remaining dimension
ValueFor the second class individual;NP-A-2 third class individual is generated, all dimensions of all individuals in third class individual
The value of degree generates at random all in accordance with respective value range;It is made of just first kind individual, the second class individual and third class individual
Beginning population.
Further, in some embodiments of the present embodiment, computational submodule is specifically used for calculating separately per each and every one
The target function value of body, target function value include E and T;The individual composition feasible zone of constraint condition will be met, and will be unsatisfactory for about
The individual of beam condition forms non-feasible zone;Each individual in feasible zone is determined based on the functional value of individual each of in feasible zone
Corresponding individual dominates horizontal;The constraint violation value of individual each of is calculated in non-feasible zone, and based in non-feasible zone
The constraint violation value and functional value of each individual each of determine in non-feasible zone that individual domination corresponding to individual is horizontal;About
Beam violation value is used to characterize the individual in non-feasible zone at a distance from feasible zone.
Further, in some embodiments of the present embodiment, computational submodule is every in the non-feasible zone of calculating
When the constraint violation value of individual, specifically for each of being calculated in non-feasible zone using preset constraint violation value calculation formula
The constraint violation value CV of individual, constraint violation value calculation formula indicate are as follows:
Wherein,
Further, the second sorting sub-module is specifically used for calculating the layering using preset crowding distance calculation formula
The crowding distance of individual in each level after sequence;The crowding distance calculation formula indicates are as follows:
Wherein, the CDiFor the crowding distance of individual i, the l is the quantity of the individual in same level, and the M is
The quantity of target function value,WithThe maximum value and minimum value of respectively k-th objective function, the target function value
Including the E and the T;Based on the crowding distance of the individual in each level calculated, layer is carried out to the individual in each level
Internal sort.
It should be noted that the calculating task discharging method in previous embodiment can be based on calculating provided in this embodiment
Task discharge mechanism realizes that those of ordinary skill in the art can be clearly understood that, for convenience and simplicity of description, this
The specific work process of calculating task discharge mechanism described in embodiment, can be with reference to the correspondence in preceding method embodiment
Process, details are not described herein.
Using calculating task discharge mechanism provided in this embodiment, the communication of user terminal unloading calculating task is calculated separately
TimeWith unloading energy consumptionUser terminal executes the task execution time of local calculating taskWith locally execute
Energy consumptionMEC server executes the task execution time of the calculating task of unloadingIt is based onWithCalculate user
Average energy consumption, and be based onWithThe average response of calculating task postpones;For the average energy consumption of user
Model for Multi-Objective Optimization is established in average response delay with task;Model for Multi-Objective Optimization is solved, multiple satisfactions are obtained
Unloading decision scheme.Implementation through the invention carries out after converting multi-objective optimization question for calculating task unloading problem
It solves, and obtains the unloading decision scheme of multiple satisfactions, under fast-changing network environment, can policymaker be possessed more
More selections, and more conducively adapt to the demand changed at any time.
3rd embodiment:
A kind of electronic device is present embodiments provided, it is shown in Figure 9 comprising processor 901, memory 902 and logical
Believe bus 903, in which: communication bus 903 is for realizing the connection communication between processor 901 and memory 902;Processor
901 for executing one or more computer program stored in memory 902, to realize the calculating in above-described embodiment one
At least one step in task discharging method.
The present embodiment additionally provides a kind of computer readable storage medium, which, which is included in, is used for
Store any method or skill of information (such as computer readable instructions, data structure, computer program module or other data)
The volatibility implemented in art or non-volatile, removable or non-removable medium.Computer readable storage medium includes but not
It is limited to RAM (Random Access Memory, random access memory), ROM (Read-Only Memory, read-only storage
Device), EEPROM (Electrically Erasable Programmable read only memory, band electric erazable programmable
Read-only memory), flash memory or other memory technologies, (Compact Disc Read-Only Memory, CD is only by CD-ROM
Read memory), digital versatile disc (DVD) or other optical disc storages, magnetic holder, tape, disk storage or other magnetic memory apparatus,
Or any other medium that can be used for storing desired information and can be accessed by a computer.
Computer readable storage medium in the present embodiment can be used for storing one or more computer program, storage
One or more computer program can be executed by processor, with realize the method in above-described embodiment one at least one step
Suddenly.
The present embodiment additionally provides a kind of computer program, which can be distributed in computer-readable medium
On, by can computing device execute, to realize at least one step of the method in above-described embodiment one;And in certain situations
Under, at least one shown or described step can be executed using the described sequence of above-described embodiment is different from.
The present embodiment additionally provides a kind of computer program product, including computer readable device, the computer-readable dress
It sets and is stored with computer program as shown above.The computer readable device may include calculating as shown above in the present embodiment
Machine readable storage medium storing program for executing.
As it can be seen that those skilled in the art should be understood that whole or certain steps in method disclosed hereinabove, be
Functional module/unit in system, device may be implemented as the software (computer program code that can be can be performed with computing device
To realize), firmware, hardware and its combination appropriate.In hardware embodiment, the functional module that refers in the above description/
Division between unit not necessarily corresponds to the division of physical assemblies;For example, a physical assemblies can have multiple functions, or
One function of person or step can be executed by several physical assemblies cooperations.Certain physical assemblies or all physical assemblies can be by realities
It applies as by processor, such as the software that central processing unit, digital signal processor or microprocessor execute, or is implemented as hard
Part, or it is implemented as integrated circuit, such as specific integrated circuit.
In addition, known to a person of ordinary skill in the art be, communication media generally comprises computer-readable instruction, data knot
Other data in the modulated data signal of structure, computer program module or such as carrier wave or other transmission mechanisms etc, and
It and may include any information delivery media.So the present invention is not limited to any specific hardware and softwares to combine.
The above content is combining specific embodiment to be further described to made by the embodiment of the present invention, cannot recognize
Fixed specific implementation of the invention is only limited to these instructions.For those of ordinary skill in the art to which the present invention belongs,
Without departing from the inventive concept of the premise, a number of simple deductions or replacements can also be made, all shall be regarded as belonging to the present invention
Protection scope.
Claims (10)
1. a kind of calculating task discharging method, it is applied to multiuser mobile communication network, in the multiuser mobile communication network
It is small including multiple communication junior units and the macro base station being connect with the multiple communication junior unit by cable network, the communication
Unit includes multiple micro-base stations and the user terminal with each micro-base station by wireless network connection, is set on the macro base station
It is equipped with mobile edge calculations MEC server characterized by comprising
Calculate separately the call duration time of the user terminal unloading calculating taskWith unloading energy consumptionThe user terminal
Execute the task execution time of local calculating taskWith locally execute energy consumptionThe MEC server is executed from institute
State the task execution time for the calculating task that user terminal unloading comes
Based on describedWithThe average energy consumption of user is calculated, and based on describedWithCalculating task
Average response delay;
Model for Multi-Objective Optimization, more mesh are established in the average response delay of average energy consumption and the task for the user
Marking Optimized model indicates are as follows:
Wherein, the P1 is optimization aim, and described C1, C2 and C3 are constraint condition, the Om,j,kFor m-th of micro-base station
The unloading decision of k-th of calculating task of j-th of user terminal of connection, the E are the average energy of the user
Consumption, the average response that the T is the task postpone, the fm,jFor j-th of user of m-th of micro-base station connection
The task processing speed of terminal, the PRm,jFor the calculating of j-th of user terminal of m-th of micro-base station connection
Task is performed locally shared percentage, the λm,jFor j-th of user terminal of m-th micro-base station connection
The average arrival rate of calculating task, the F are the task processing speed of the MEC server, and the NS is the micro-base station
Quantity, the NmFor the user terminal total quantity of m-th of micro-base station connection;
The Model for Multi-Objective Optimization is solved, the unloading decision scheme of multiple satisfactions is obtained.
2. calculating task discharging method as described in claim 1, which is characterized in that described based on describedWithIt calculates
The average energy consumption of user, and based on describedWithThe average response of calculating task postpones
Based on describedWithThe average energy consumption of user is calculated, calculation formula indicates are as follows:
Wherein, the em,jWhen executing local calculating task for j-th of user terminal of m-th of micro-base station connection
Coefficient of energy dissipation, the Km,jIt is described for the task quantity that j-th of user terminal of m-th of micro-base station connection generates
dm,j,kIt is described for the data volume of k-th of calculating task of j-th of user terminal of m-th of micro-base station connection
ω is j-th of the occupied channel width of user terminal of m-th of micro-base station connection, pm,jFor m-th of micro- base
Stand connection j-th of user terminal data transmission utilization measure, the Gm,jFor j-th of institute of m-th of micro-base station connection
State the channel gain between user terminal and m-th of micro-base station, the σ2For ambient noise, the Im,jDescribed in m-th
The channel disturbance that j-th of user terminal of micro-base station connection is subject to;
Based on describedWithThe average response of calculating task postpones, and calculation formula indicates are as follows:
3. calculating task discharging method as described in claim 1, which is characterized in that it is described to the Model for Multi-Objective Optimization into
Row solves, and the unloading decision scheme for obtaining multiple satisfactions includes:
Step A, the unloading decision is encoded into the individual in population, and generates initial population;It wherein, include NP in the population
Individual, the dimension of each individual is the total quantity of the user terminal in the multiuser mobile communication network in the population,
The unloading decision of all calculating tasks of the corresponding user terminal of the value of every dimension, m-th of micro-base station connection
J-th of user terminal Um,jThe value range of value for corresponding to dimension in the individual is
Step B, it is horizontal to calculate separately individual domination corresponding to each individual in population;
Step C, level is dominated based on all individuals calculated, layering row is carried out to all individuals in the population
Sequence;
Step D, layer internal sort is carried out to the individual in each level after the layer sorting, and the population after sequence is made
For parent population;
Step E, intersection or mutation operation are carried out to the individual in the parent population, obtains offspring flocks;
Step F, it is horizontal to calculate separately individual domination corresponding to each individual in the offspring flocks, and based on including the father
Level is dominated for the individual of individuals all in population and the whole population of the offspring flocks, to the institute in the whole population
There is individual to carry out layer sorting, and layer internal sort is carried out to the individual in each level after the whole population layering, obtains institute
State the ranking results of whole population;
Step G, according to the ranking results of the whole population, NP wherein optimal individual is selected to form next-generation parent kind
Group, the number of iterations add one;
Step H, the step E to the step G is repeated to the next-generation parent population, until reaching preset iteration
Number, using the first layer individual of finally obtained population as disaggregation export, and to output result in it is each individual respectively into
Row decoding, obtains the unloading decision scheme of multiple satisfactions.
4. calculating task discharging method as claimed in claim 3, which is characterized in that the generation initial population includes:
Two first kind individuals are generated, the value of all dimensions of one of first kind individual is maximized Gx, another first
The value of all dimensions of class individual is minimized 0;Wherein,The A is described more
The total quantity of the user terminal in user's mobile communications network,
A the second class individuals are generated, the value of a dimension of all individuals is 0 in the second class individual, the value of remaining dimension
For
NP-A-2 third class individual is generated, the value of all dimensions of all individuals is all in accordance with respective in the third class individual
Value range generates at random;
Initial population is formed by the first kind individual, the second class individual and third class individual.
5. calculating task discharging method as claimed in claim 3, which is characterized in that described to calculate separately corresponding to each individual
Individual domination level include:
Calculate separately the target function value of each individual;The target function value include the user average energy consumption and described
The average response of business postpones;
The individual composition feasible zone of the constraint condition will be met, and the individual composition for being unsatisfactory for the constraint condition is non-feasible
Domain;
Individual corresponding to individual each of is determined in the feasible zone based on the functional value of individual each of in the feasible zone
It dominates horizontal;
The constraint violation value of individual each of is calculated in the non-feasible zone, and based on individual each of in the non-feasible zone
Constraint violation value and functional value each of determine in the non-feasible zone that individual domination corresponding to individual is horizontal;The constraint
Violation value is used to characterize the individual in the non-feasible zone at a distance from feasible zone.
6. calculating task discharging method as claimed in claim 5, which is characterized in that every in the calculating non-feasible zone
Individual constraint violation value include:
The constraint violation value CV of individual, institute each of are calculated in the non-feasible zone using preset constraint violation value calculation formula
State the expression of constraint violation value calculation formula are as follows:
Wherein,
7. calculating task discharging method as claimed in claim 3, which is characterized in that each layer to after the layer sorting
Individual in grade carries out layer internal sort
The crowding distance of individual in each level after calculating the layer sorting using preset crowding distance calculation formula;Institute
State the expression of crowding distance calculation formula are as follows:
Wherein, the CDiFor individual xiCrowding distance, the l be in same level individual quantity, the M be target
The quantity of functional value,WithThe maximum value and minimum value of respectively k-th objective function, the target function value include
The average response of the average energy consumption of the user and the task postpones;
Based on the crowding distance of the individual in each level calculated, the individual in each level is carried out in layer
Sequence.
8. a kind of calculating task discharge mechanism, it is applied to multiuser mobile communication network, in the multiuser mobile communication network
It is small including multiple communication junior units and the macro base station being connect with the multiple communication junior unit by cable network, the communication
Unit includes multiple micro-base stations and the user terminal with each micro-base station by wireless network connection, is set on the macro base station
It is equipped with mobile edge calculations MEC server characterized by comprising
First computing module, for calculating the call duration time of the user terminal unloading calculating taskWith unloading energy consumption
The user terminal executes the task execution time of local calculating taskWith locally execute energy consumptionThe MEC clothes
Business device executes the task execution time of the calculating task to come from user terminal unloading
Second computing module, for based on describedWithThe average energy consumption of user is calculated, and based on described
WithThe average response of calculating task postpones;
Module is established, multiple-objection optimization is established in the average response delay for the average energy consumption and the task for the user
Model, the Model for Multi-Objective Optimization indicate are as follows:
Wherein, the P1 is optimization aim, and described C1, C2 and C3 are constraint condition, the Om,j,kFor m-th of micro-base station
The unloading decision of k-th of calculating task of j-th of user terminal of connection, the E are the average energy of the user
Consumption, the average response that the T is the task postpone, the fm,jFor j-th of user of m-th of micro-base station connection
The task processing speed of terminal, the PRm,jFor the calculating of j-th of user terminal of m-th of micro-base station connection
Task is performed locally shared percentage, the λm,jFor j-th of user terminal of m-th micro-base station connection
The average arrival rate of calculating task, the F are the task processing speed of the MEC server, and the NS is the micro-base station
Quantity, the NmFor the user terminal total quantity of m-th of micro-base station connection;
It solves module and obtains the unloading decision scheme of multiple satisfactions for solving to the Model for Multi-Objective Optimization.
9. a kind of electronic device characterized by comprising processor, memory and communication bus;
The communication bus is for realizing the connection communication between the processor and memory;
The processor is for executing one or more program stored in the memory, to realize such as claim 1 to 7
Any one of described in calculating task discharging method the step of.
10. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage have one or
Multiple programs, one or more of programs can be executed by one or more processor, to realize such as claim 1 to 7
Any one of described in calculating task discharging method the step of.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910527427.XA CN110418356A (en) | 2019-06-18 | 2019-06-18 | A kind of calculating task discharging method, device and computer readable storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910527427.XA CN110418356A (en) | 2019-06-18 | 2019-06-18 | A kind of calculating task discharging method, device and computer readable storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110418356A true CN110418356A (en) | 2019-11-05 |
Family
ID=68359238
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910527427.XA Pending CN110418356A (en) | 2019-06-18 | 2019-06-18 | A kind of calculating task discharging method, device and computer readable storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110418356A (en) |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111405569A (en) * | 2020-03-19 | 2020-07-10 | 三峡大学 | Calculation unloading and resource allocation method and device based on deep reinforcement learning |
CN111694664A (en) * | 2020-06-04 | 2020-09-22 | 海南大学 | Calculation unloading distribution method of edge server |
CN111930435A (en) * | 2020-07-13 | 2020-11-13 | 兰州理工大学 | Task unloading decision method based on PD-BPSO technology |
CN112083967A (en) * | 2020-08-18 | 2020-12-15 | 深圳供电局有限公司 | Unloading method of cloud edge computing task, computer equipment and storage medium |
CN112105062A (en) * | 2020-09-16 | 2020-12-18 | 云南大学 | Mobile edge computing network energy consumption minimization strategy method under time-sensitive condition |
CN112272102A (en) * | 2020-09-11 | 2021-01-26 | 北京工业大学 | Method and device for unloading and scheduling edge network service |
CN112380008A (en) * | 2020-11-12 | 2021-02-19 | 天津理工大学 | Multi-user fine-grained task unloading scheduling method for mobile edge computing application |
CN112732359A (en) * | 2021-01-14 | 2021-04-30 | 广东技术师范大学 | Multi-user hybrid computing unloading method and device, electronic equipment and storage medium |
CN112969163A (en) * | 2021-02-22 | 2021-06-15 | 安徽大学 | Cellular network computing resource allocation method based on self-adaptive task unloading |
WO2021139273A1 (en) * | 2020-07-28 | 2021-07-15 | 平安科技(深圳)有限公司 | Cdn scheduling method and apparatus based on pareto algorithm, and computer device and storage medium |
CN113407251A (en) * | 2021-05-26 | 2021-09-17 | 北京工业大学 | Method and device for unloading mobile edge network service |
CN113515324A (en) * | 2021-07-16 | 2021-10-19 | 广东工业大学 | Collaborative edge calculation method and device for unloading decision based on directed acyclic graph, electronic equipment and storage medium |
CN113590339A (en) * | 2021-08-16 | 2021-11-02 | 重庆大学 | Internet of things terminal cooperation method for batch computing tasks |
CN113727391A (en) * | 2021-09-07 | 2021-11-30 | 深圳大学 | Method, device and equipment for controlling system operation and storage medium |
CN114398106A (en) * | 2022-01-21 | 2022-04-26 | 中国联合网络通信集团有限公司 | Unloading strategy determination method and device, electronic equipment and storage medium |
CN115866626A (en) * | 2023-02-23 | 2023-03-28 | 安徽思高智能科技有限公司 | NSGA-II and simulated annealing based service deployment method under edge environment |
CN116582873A (en) * | 2023-07-13 | 2023-08-11 | 湖南省通信建设有限公司 | System for optimizing offloading tasks through 5G network algorithm to reduce delay and energy consumption |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180183855A1 (en) * | 2016-12-28 | 2018-06-28 | Intel Corporation | Application computation offloading for mobile edge computing |
CN108920279A (en) * | 2018-07-13 | 2018-11-30 | 哈尔滨工业大学 | A kind of mobile edge calculations task discharging method under multi-user scene |
CN109814951A (en) * | 2019-01-22 | 2019-05-28 | 南京邮电大学 | The combined optimization method of task unloading and resource allocation in mobile edge calculations network |
-
2019
- 2019-06-18 CN CN201910527427.XA patent/CN110418356A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180183855A1 (en) * | 2016-12-28 | 2018-06-28 | Intel Corporation | Application computation offloading for mobile edge computing |
CN108920279A (en) * | 2018-07-13 | 2018-11-30 | 哈尔滨工业大学 | A kind of mobile edge calculations task discharging method under multi-user scene |
CN109814951A (en) * | 2019-01-22 | 2019-05-28 | 南京邮电大学 | The combined optimization method of task unloading and resource allocation in mobile edge calculations network |
Non-Patent Citations (1)
Title |
---|
LAIZHONG CUI,ET.AL: "《Joint Optimization of Energy Consumption and Latency in Mobile Edge Computing for Internet of Things》", 《IEEE INTERNET OF THINGS JOURNAL》 * |
Cited By (25)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111405569A (en) * | 2020-03-19 | 2020-07-10 | 三峡大学 | Calculation unloading and resource allocation method and device based on deep reinforcement learning |
CN111694664A (en) * | 2020-06-04 | 2020-09-22 | 海南大学 | Calculation unloading distribution method of edge server |
CN111694664B (en) * | 2020-06-04 | 2021-12-21 | 海南大学 | Calculation unloading distribution method of edge server |
CN111930435A (en) * | 2020-07-13 | 2020-11-13 | 兰州理工大学 | Task unloading decision method based on PD-BPSO technology |
WO2021139273A1 (en) * | 2020-07-28 | 2021-07-15 | 平安科技(深圳)有限公司 | Cdn scheduling method and apparatus based on pareto algorithm, and computer device and storage medium |
CN112083967A (en) * | 2020-08-18 | 2020-12-15 | 深圳供电局有限公司 | Unloading method of cloud edge computing task, computer equipment and storage medium |
CN112083967B (en) * | 2020-08-18 | 2023-10-20 | 深圳供电局有限公司 | Cloud edge computing task unloading method, computer equipment and storage medium |
CN112272102A (en) * | 2020-09-11 | 2021-01-26 | 北京工业大学 | Method and device for unloading and scheduling edge network service |
CN112105062A (en) * | 2020-09-16 | 2020-12-18 | 云南大学 | Mobile edge computing network energy consumption minimization strategy method under time-sensitive condition |
CN112105062B (en) * | 2020-09-16 | 2022-05-31 | 云南大学 | Mobile edge computing network energy consumption minimization strategy method under time-sensitive condition |
CN112380008A (en) * | 2020-11-12 | 2021-02-19 | 天津理工大学 | Multi-user fine-grained task unloading scheduling method for mobile edge computing application |
CN112380008B (en) * | 2020-11-12 | 2022-04-15 | 天津理工大学 | Multi-user fine-grained task unloading scheduling method for mobile edge computing application |
CN112732359A (en) * | 2021-01-14 | 2021-04-30 | 广东技术师范大学 | Multi-user hybrid computing unloading method and device, electronic equipment and storage medium |
CN112969163A (en) * | 2021-02-22 | 2021-06-15 | 安徽大学 | Cellular network computing resource allocation method based on self-adaptive task unloading |
CN113407251A (en) * | 2021-05-26 | 2021-09-17 | 北京工业大学 | Method and device for unloading mobile edge network service |
CN113515324A (en) * | 2021-07-16 | 2021-10-19 | 广东工业大学 | Collaborative edge calculation method and device for unloading decision based on directed acyclic graph, electronic equipment and storage medium |
CN113590339A (en) * | 2021-08-16 | 2021-11-02 | 重庆大学 | Internet of things terminal cooperation method for batch computing tasks |
CN113590339B (en) * | 2021-08-16 | 2023-06-23 | 重庆大学 | Internet of things terminal collaboration method for batch computing tasks |
CN113727391A (en) * | 2021-09-07 | 2021-11-30 | 深圳大学 | Method, device and equipment for controlling system operation and storage medium |
CN114398106A (en) * | 2022-01-21 | 2022-04-26 | 中国联合网络通信集团有限公司 | Unloading strategy determination method and device, electronic equipment and storage medium |
CN114398106B (en) * | 2022-01-21 | 2023-11-21 | 中国联合网络通信集团有限公司 | Unloading strategy determining method and device, electronic equipment and storage medium |
CN115866626A (en) * | 2023-02-23 | 2023-03-28 | 安徽思高智能科技有限公司 | NSGA-II and simulated annealing based service deployment method under edge environment |
CN115866626B (en) * | 2023-02-23 | 2023-05-12 | 安徽思高智能科技有限公司 | Service deployment method based on NSGA-II and simulated annealing in edge environment |
CN116582873A (en) * | 2023-07-13 | 2023-08-11 | 湖南省通信建设有限公司 | System for optimizing offloading tasks through 5G network algorithm to reduce delay and energy consumption |
CN116582873B (en) * | 2023-07-13 | 2023-09-08 | 湖南省通信建设有限公司 | System for optimizing offloading tasks through 5G network algorithm to reduce delay and energy consumption |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110418356A (en) | A kind of calculating task discharging method, device and computer readable storage medium | |
CN109684075B (en) | Method for unloading computing tasks based on edge computing and cloud computing cooperation | |
Li et al. | Joint optimization strategy of computation offloading and resource allocation in multi-access edge computing environment | |
CN111445111B (en) | Electric power Internet of things task allocation method based on edge cooperation | |
CN110798849A (en) | Computing resource allocation and task unloading method for ultra-dense network edge computing | |
CN112286677B (en) | Resource-constrained edge cloud-oriented Internet of things application optimization deployment method | |
CN111953758B (en) | Edge network computing unloading and task migration method and device | |
CN111800828B (en) | Mobile edge computing resource allocation method for ultra-dense network | |
CN111246586B (en) | Method and system for distributing smart grid resources based on genetic algorithm | |
CN109710374A (en) | The VM migration strategy of task unloading expense is minimized under mobile edge calculations environment | |
Hassan et al. | A novel energy model for renewable energy-enabled cellular networks providing ancillary services to the smart grid | |
CN111866601B (en) | Cooperative game-based video code rate decision method in mobile marginal scene | |
Guo et al. | Energy efficient computation offloading for multi-access MEC enabled small cell networks | |
CN106095529B (en) | A kind of carrier wave emigration method under C-RAN framework | |
CN109548044B (en) | DDPG (distributed data group pg) -based bit rate optimization method for energy-collectable communication | |
CN113573363B (en) | MEC calculation unloading and resource allocation method based on deep reinforcement learning | |
CN113590279A (en) | Task scheduling and resource allocation method for multi-core edge computing server | |
Cheng et al. | Energy-efficient resource allocation for UAV-empowered mobile edge computing system | |
Huang et al. | A DRL-based automated algorithm selection framework for cross-layer QoS-aware scheduling and antenna allocation in massive MIMO systems | |
CN114615705B (en) | Single-user resource allocation strategy method based on 5G network | |
CN113747450A (en) | Service deployment method and device in mobile network and electronic equipment | |
Sen | Intelligent Admission and Placement of O-RAN Slices Using Deep Reinforcement Learning | |
Bendaoud et al. | Network selection using game theory | |
CN110392377A (en) | A kind of 5G super-intensive networking resources distribution method and device | |
CN102845028A (en) | Method and device for resource allocation |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191105 |
|
RJ01 | Rejection of invention patent application after publication |