CN110351145A - A kind of radio network functions method of combination of the virtualization based on economic benefit - Google Patents

A kind of radio network functions method of combination of the virtualization based on economic benefit Download PDF

Info

Publication number
CN110351145A
CN110351145A CN201910648218.0A CN201910648218A CN110351145A CN 110351145 A CN110351145 A CN 110351145A CN 201910648218 A CN201910648218 A CN 201910648218A CN 110351145 A CN110351145 A CN 110351145A
Authority
CN
China
Prior art keywords
particle
resource
virtual functions
cost
network
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201910648218.0A
Other languages
Chinese (zh)
Other versions
CN110351145B (en
Inventor
邹赛
许磊
田淋风
肖蕾
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing College of Electronic Engineering
Original Assignee
Chongqing College of Electronic Engineering
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chongqing College of Electronic Engineering filed Critical Chongqing College of Electronic Engineering
Priority to CN201910648218.0A priority Critical patent/CN110351145B/en
Publication of CN110351145A publication Critical patent/CN110351145A/en
Application granted granted Critical
Publication of CN110351145B publication Critical patent/CN110351145B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/142—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
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/06—Testing, supervising or monitoring using simulated traffic
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/02—Protocols based on web technology, e.g. hypertext transfer protocol [HTTP]

Abstract

The invention discloses the virtualization radio network functions based on economic benefit to compile method.This method establishes virtualization radio network functions layout mathematic optimal model, proposes a kind of virtualization radio network functions layout process (VNFPSO) based on economic benefit.The characteristics of occurring for the opening, the discreteness of network function, network load exponentially for virtualizing Radio Access Network framework, to classical particle colony optimization algorithm inertia weight, particle variations, in terms of corrected, accelerate the solving speed of global approximate optimal solution.The present invention helps to reduce the service denial rate of virtualization Radio Access Network, improves the utilization rate of network system resources.

Description

A kind of radio network functions method of combination of the virtualization based on economic benefit
Technical field
The present invention relates to the communications fields, and in particular to a kind of radio network functions layout of the virtualization based on economic benefit Method.
Background technique
As network function virtualizes (NFV) and software defined network (SDN) process, network function layout is virtualized (NFVO) be considered as future network soul, will become operator's flexible management network, maximize play new technology advantage Key.NFVO helps to standardize the function of virtual network, to improve the interoperability of software-defined network element.Its energy Enough execute resource layout, network service orchestration and other function layout.It can set independently of any specific Virtual base Manager is applied to coordinate, authorize, issue and use resource, the management of the VNF example of shared NFV infrastructure resources is also provided. European Telecommunications Standards Institute (ETSI) specially develops NFVO composer.Therefore, to the radio network functions layout side of virtualization Method research is of great significance and practical value.
Existing NFVO is broadly divided into two types.(1) carry out layout to the function of node (Middlebox): it is one Special network hardware equipment is substituted in platform server with software.Md Faizul Bari and Shihabur Rahman Chowdhury is in 2016 in " IEEE Transactions on Network and Service Management " periodical On publish thesis " Orchestrating Virtualized Network Functions ", propose it is a kind of utilize integer line Property planing method carry out the virtualization network function of layout core network, reduce network operation cost and improve resource utilization.Meanwhile VNFO problem is solved using the heuritic approach of nested bin packing.(2) to the link of virtualization (Service Function Chaining) carry out layout: it, which refers to, is dynamically integrated into one or more service functions in service function link, such as content Distribution network.Tarik Taleb and Adlen Ksentini are in 2016 in " IEEE Transactions On Wireless Communications " publish thesis on periodical " Coping With Emerging Mobile Social Media Applications Through Dynamic Service Function Chaining ", proposes a kind of service function chain Layout process.It identifies mobile weblication and the user positioned at same neighborhood first, then defines to related data flow The correlated process of application, and in mobile domains handle multicast procedures foundation.
Existing NFVO process is mainly specifically sometime specifically basic for specific a certain application layout Facility, thus the maximization of optimized integration facility overall value, but existing NFVO process can not achieve the warp of infrastructure Ji maximizing the benefits, Provider Equipment purchase cost and O&M expenditure are very high.
Summary of the invention
It is an object of the invention to overcome the deficiencies of the prior art and provide one kind can reduce virtualization Radio Access Network Service denial rate, improve the virtualization radio network functions layout side based on economic benefit of the utilization rate of network system resources Method.
The object of the present invention is achieved like this:
S1: identification Web service selects the component of virtualization network function according to the total resources of resource pool, respectively to Web The corresponding resource of required function, the function of physical network, the function of physical network are applied using required function, Web Corresponding resource, virtualization network function layout process founding mathematical models；
S2: the mathematical model that S1 step is established is converted into particle swarm algorithm model；
S3: inertia weight parameter, particle variations parameter, the Studying factors ginseng of the particle swarm algorithm model in optimization S2 step Number；
S4: wireless access network is carried out using the model parameter after optimizing in S3 step and virtualizes network function layout.
Further, the step S1 includes the following steps,
S101: carrying out mathematical modeling using required function to Web includes,
Definition is R using the virtual functions request function of ii=(Fi,QoSi),
Wherein, FiIt is the virtual functions set of an application i request,
Fi={ f1,fj,fn}
FiMathematical expression form be: where fj={ id, name, description, note }
Wherein, j indicates FiJth item function, n indicate FiN-th function, i.e., the number of function in functional resources pond, j, N is natural number, j < n,
Id indicates virtual functions identification number,
Name indicates virtual functions title,
Description indicates the explanation of virtual functions,
Note indicates annotation information；
QoSiIt is the virtual functions f of the request of an application ijCorresponding attribute, QoSiMathematical expression form be:
Wherein, fjIndicate a specific virtual functions, akIndicate a virtual functions fjThe attribute being had, k indicate empty Quasi- function fjKth item attribute, m indicate virtual functions fjM attributes, it indicate information resources, b indicate bandwidth resources, p Power resource in radio frequency；
S102: carrying out mathematical modeling to physical network resource, including,
The cost for defining unit frequency spectrum resource is cs, using i virtual functions fjAttribute akThe number of required bandwidth resources ForS indicates frequency spectrum, and B is the size of per bit rate,
Define wireless access links ηsThe cost of unit frequency spectrum benefit are as follows:
Wherein, γiIt is signal-to-noise ratio, the cost for defining the power resource in unit radio frequency is cp, wireless access links ηpUnit The cost of power resource benefit in radio frequency is ηp=pk×cp,
The cost for defining unit information resource is cit, service ηitThe cost of consumption information resource is ηp=itk×cit,
The deployment benefit η of VNFdCost be ηd=σb×ηs+σp×ηp+σit×ηit,
Wherein, σb、σp、σitIt is weight coefficient, 0≤σb,σp,σit≤ 1, and σb+σp+σit=1；
S103: carrying out mathematical modeling to virtualization network function layout process, including,
Building virtualization network function layout process mathematical model:
Wherein μi,jIt is cost function, indicates that there is attribute akVirtual functions module fjThe required cost paid,
xj',k'Indicate the functional module of selection, Ri,j,k→it≤N×xj',k'→ it, Ri,j,k→p≤N×xj',k'→ p, Ri,j,k→b≤N×xj',k'→ b, which indicates that selection is N number of, has attribute akVirtual functions module fjIT resource value, frequency spectrum resource, Transmitting-receiving power resource is greater than or waits the resource value requested using i, and N is natural number,Indicate virtual functions module fi With virtual functions module fi+yThere are dependence, fi≠fi+yIndicate virtual functions module fiWith virtual functions module fi+yThere are rows Reprimand relationship, costj,kRefer to multiple with attribute ajVirtual functions module fiThe cost paid when using side by side, Mcostj,kRefer to It is multiple that there is attribute ajVirtual functions module fiIt is used in conjunction with the cost paid when a resource, δs, δp, δitFor combination At functional module xj',k'Coefficient；
S104: optimizing virtualization network function layout process mathematical model, include the following steps,
S1041: addition constraint condition: Wherein s', p', it' indicate that the related resource used, all indicate all money Source；
S1042: virtualization network function layout process mathematical model is optimized for for the first time
Virtual functions module f is increased using following formula in optimization processiCost cost, μj,k'=μj,k+μj+y,k, y For temporary variable；
S1043: it is by quasi-ization network function layout process mathematical model optimizing
Further, the step S2 includes the following steps,
S201: in total dimension D=m*n of Virtual Service Arrangement solution space, appoint and L particle is taken to form a group Body, i are less than L, wherein i-th of particle kth for when described using two indices, using the value of virtual functions attribute as PSO Position, be expressed asD dimensional vector, it is fast as the flight of PSO that virtual functions attribute value changes speed Degree, is expressed asD dimensional vector, wherein t indicate particle t time fly；
S202: i-th particle search to kth for when individual history optimal location be Search for kth for when the history optimal location g of entire population beAt+1 generation of kth, The iteration more new formula of jth the dimension speed and position of i-th of particle are as follows:Its In, ω is inertia weight, the influence mobile to next time of the speed for measuring previous moment, c1And c2For Studying factors, r1And r2 For the random number in [0,1].
Further, the step S3 includes the following steps,
S301: the inertia weight ω of dynamic regulation traditional PS O algorithm, when particle position close to individual history optimal location or It improves convergent resolution ratio when the entire population history optimal location of person and reduces the flying speed of particle, when particle position is separate Accelerate convergence rate when individual history optimal location or the history optimal location of entire population and improves the flight speed of particle Degree, inertia weight ω are given by:
Wherein, tv1And tv2It is threshold value, c1iAnd c2iFor the Studying factors of i-th, r1iAnd r2iFor the random number of i-th；
S302: the design for the particle that makes a variation, comprising the following steps:
S3021: ifThen determine psiFor the particle that makes a variation, wherein tp is threshold values,
S3022: set variation the smooth moving distance of particle as
Wherein Bd (k) is the smooth moving distance of variation particle in kth generation, and L is the interval number between two particles；
S3023: set the variation equation of particle as
The beneficial effects of the present invention are:
1. the present invention can reduce Provider Equipment purchase cost and O&M branch under the conditions of guaranteeing flexibility and opening Out, the present invention does not consider the network function sequencing of each virtualization from the angle of economic benefit, studies wireless dummy Change Internet resources layout.In the practical O&M of operator, the variation of network and the adjustment demand of resource are often derived from service layer, The control and analysis of business simultaneously also be unable to do without the understanding to resource level, and the two has complicated connection, present invention virtualization The function that application may be implemented in network function layout realizes the maximization of economic benefit of " customized " and infrastructure.
2. the present invention helps to reduce the service denial rate of virtualization Radio Access Network, the benefit of network system resources is improved With rate.
Detailed description of the invention
Fig. 1 is one embodiment of the invention flow chart.
Fig. 2 is one embodiment of the invention system architecture schematic diagram.
Specific embodiment
The present invention solves invention thinking of problems in background technique, and the present invention is by establishing virtualization wireless network Function layout mathematic optimal model proposes a kind of virtualization radio network functions layout process based on economic benefit (VNFPSO).Go out for the opening, the discreteness of network function, network load exponentially of virtualization Radio Access Network framework Existing feature, to classical particle colony optimization algorithm inertia weight, particle variations, in terms of corrected, add The fast solving speed of global approximate optimal solution, facilitates the service denial rate for reducing virtualization Radio Access Network, improves net The utilization rate of network system resource.
As shown in Figure 1, the present invention provides a kind of radio network functions method of combination packet of virtualization based on economic benefit Include following steps,
S1: identification Web service selects the group of virtualization network function according to the total resources of the characteristic of business and resource pool Part, respectively to Web application, physical network resource and virtualization network function layout process founding mathematical models；
S2: the mathematical model that S1 step is established is converted into particle swarm algorithm model；
S3: inertia weight parameter, particle variations parameter, the Studying factors ginseng of the particle swarm algorithm model in optimization S2 step Number；
S4: wireless access network is carried out using the model parameter after optimizing in S3 step and virtualizes network function layout.
Step S1 is specifically described below:
Step S1 includes needing first to identify Web service for virtualization network function layout, further according to the resource of resource pool The component and link of the required virtualization network function of total amount selection carry out Arrangement, provide respectively to Web application, physical network Source and virtualization network function layout process founding mathematical models.
Step S1 specifically includes following steps,
The modeling of S101:Web applied mathematics
The virtual functions request function of i is applied in definition are as follows: Ri=(Fi,QoSi)
Wherein, FiIt is the virtual functions set of an application i request.FiMathematical expression form be: Fi={ f1,fj,fn}
where fj={ id, name, description, note }.
QoSiIt is the virtual functions f of the request of an application ijCorresponding attribute.QoSiMathematical expression form be:
Wherein, fjIndicate a specific virtual functions, akIndicate a virtual functions fjThe attribute being had, k indicate empty Quasi- function fjKth item attribute, m indicate virtual functions fjM attributes, it indicate information resources, b indicate bandwidth resources, p Power resource in radio frequency, n are the numbers of function in resource pool；
S102: physical network resource mathematical modeling
The cost for defining unit frequency spectrum resource is cs.Using i virtual functions fjAttribute akThe number of required bandwidth resources
B is the size of per bit rate.Define wireless access links ηsThe cost of unit frequency spectrum benefit is:
γiIt is signal-to-noise ratio.The cost for defining the power resource in unit radio frequency is cp.Wireless access links ηpUnit radio frequency In the cost of power resource benefit be: ηp=pk×cp。
The cost for defining unit information resource is cit.Service ηitThe cost of consumption information resource is ηp=itk×cit。
The deployment benefit of VNF is that cost is used in combination in frequency spectrum resource, power resource, information resources.The deployment benefit η of VNFd Cost be ηd=σb×ηs+σp×ηp+σit×ηit。
Wherein, σb、σp, be weight coefficient, 0≤σb,σp,σit≤ 1, and σb+σp+σit=1.
S103: virtualization network function layout process mathematical modeling
Virtualizing network function layout process mathematical model is
Wherein μi,jIt is cost function, it indicates there is attribute akVirtual functions module fjThe required cost paid. xj',k'Indicate the functional module of selection.Ri,j,k→it≤N×xj',k'→ it, Ri,j,k→p≤N×xj',k'→ p, Ri,j,k→b≤N ×xj',k'→ b, which indicates that selection is N number of, has attribute akVirtual functions module fjIT resource value, frequency spectrum resource, transmitting-receiving power money Source is greater than or waits the resource value requested using i.Indicate virtual functions module fiWith virtual functions module fi+yIt deposits In dependence, if fiIn the presence of then fi+yIt there will necessarily be.fi≠fi+yIndicate virtual functions module fiWith virtual functions module fi+y There are exclusion relations, if fiIn the presence of then fi+yIt must be not present.costj,kRefer to multiple with attribute ajVirtual functions module fi The cost paid when using side by side.Mcostj,kRefer to multiple with attribute ajVirtual functions module fiIt is used in conjunction with a money The cost paid when source.δs, δp, δitDenotes is combined into functional module xj',k'Coefficient
S104: virtualization network function layout process mathematical model optimizing
S1041: Virtual Service layout substantially concentrates from virtual functions and selects sub- virtual functions.When construction cost is equal When, specific selection scheme has diversity.In order to reduce the difficulty of solution, while resource anxiety Cheng Du is embodied, added as follows Constraint condition: Wherein s', p', it' indicate the related resource used.All indicates all resources.
S1042: quasi-ization network function layout process mathematical model is optimized for for the first time
Virtual functions module fiWith virtual functions module fi+yThere are exclusion relations, can embody in service request.Cause This, in solution procedure,if fi≠fi+yConstraint condition can delete.Virtual functions module simultaneously fiWith virtual functions module fi+yThere are dependences, only increase virtual functions module f in solution procedureiCost cost, Increment is shown below: μj,k'=μj,k+μj+y,k。
S1043: quasi-ization network function layout process mathematical model final optimization pass is
Step S2 is illustrated below:
Step S2 includes that the S1 mathematical model established is converted into classical particle swarm algorithm.
Step S2 specifically includes following steps,
S201: in total dimension D=mn of Virtual Service Arrangement solution space, appointing and L particle taken to form a group, Wherein i-th (i < L) a particle is described in kth for Shi Keyong two indices: the value of virtual functions attribute can regard the position of PSO as It sets, is expressed asD dimensional vector；Virtual functions attribute value variation speed can regard the flying speed of PSO as, It is expressed asD dimensional vector.
S202: i-th particle search to kth for when individual history optimal location be Search for kth for when the history optimal location of entire population beThen at+1 generation of kth, i-th The jth dimension speed of particle and the iteration more new formula of position are original: Wherein, ω is inertia weight, it measures influence of the speed of previous moment to movement next time.c1And c2For Studying factors, r1With r2For the random number in [0,1].
Step S3 is illustrated below:
Step S3 includes inertia weight, particle variations, the Studying factors of the particle swarm algorithm in modification S2 step.
Step S3 specifically includes following steps,
S301: according to searching process, the inertia weight ω of dynamic regulation traditional PS O algorithm.When the close individual of particle position It when history optimal location or the history optimal location of entire population, needs to improve convergent resolution ratio, reduces flying for particle Scanning frequency degree.When history optimal location of the particle position far from individual history optimal location or entire population, need to accelerate Convergence rate improves the flying speed of particle.To guarantee the preferable local search ability of algorithm and convergence rate, inertia weight ω by It is given below:
Wherein, tv1And tv2It is threshold value, c1iAnd c2iFor the Studying factors of i-th, r1iAnd r2iFor the random number of i-th.
S302: the design for the particle that makes a variation is as follows:
S3021: ifThen determine psiFor the particle that makes a variation, wherein tp is threshold values.
S3022: variation the smooth moving distance of particle be
Wherein Bd (k) is the smooth moving distance of variation particle in kth generation.
S3023: the variation equation of particle is
The invention has the advantages that
1. the present invention can reduce Provider Equipment purchase cost and O&M branch under the conditions of guaranteeing flexibility and opening Out, the present invention does not consider the network function sequencing of each virtualization from the angle of economic benefit, studies wireless dummy Change Internet resources layout.In the practical O&M of operator, the variation of network and the adjustment demand of resource are often derived from service layer, The control and analysis of business simultaneously also be unable to do without the understanding to resource level, and the two has complicated connection, present invention virtualization The function that application may be implemented in network function layout realizes the maximization of economic benefit of " customized " and infrastructure.
2. the present invention helps to reduce the service denial rate of virtualization Radio Access Network, the benefit of network system resources is improved With rate.
Finally, it is stated that preferred embodiment above is only used to illustrate the technical scheme of the present invention and not to limit it, although logical It crosses above preferred embodiment the present invention is described in detail, however, those skilled in the art should understand that, can be Various changes are made to it in form and in details, without departing from claims of the present invention limited range.

Claims (4)

1. a kind of radio network functions method of combination of the virtualization based on economic benefit, which is characterized in that include the following steps,
S1: identification Web service selects the component of virtualization network function according to the total resources of resource pool, applies respectively to Web Required function, Web are corresponding using the corresponding resource of required function, the function of physical network, the function of physical network Resource, virtualization network function layout process founding mathematical models；
S2: the mathematical model that S1 step is established is converted into particle swarm algorithm model；
S3: inertia weight parameter, particle variations parameter, the Studying factors parameter of the particle swarm algorithm model in optimization S2 step；
S4: wireless access network is carried out using the model parameter after optimizing in S3 step and virtualizes network function layout.
2. a kind of radio network functions method of combination of the virtualization based on economic benefit as described in claim 1, feature It is, the step S1 includes the following steps,
S101: carrying out mathematical modeling using required function to Web includes,
Definition is R using the virtual functions request function of ii=(Fi,QoSi),
Wherein, FiIt is the virtual functions set of an application i request,
FiMathematical expression form be:
Wherein, j indicates FiJth item function, n indicate FiN-th function, j, n are natural number, j < n；
QoSiIt is the virtual functions f of the request of an application ijCorresponding attribute, QoSiMathematical expression form be:
Wherein, fjIndicate a specific virtual functions, akIndicate a virtual functions fjThe attribute being had, k indicate virtual function It can fjKth item attribute, m indicate virtual functions fjM attributes, it indicate information resources, b indicate bandwidth resources, p be penetrate Power resource in frequency；
S102: carrying out mathematical modeling to physical network resource, including,
The cost for defining unit frequency spectrum resource is cs, using i virtual functions fjAttribute akThe number of required bandwidth resources isS indicates frequency spectrum, and B is the size of per bit rate,
Define wireless access links ηsThe cost of unit frequency spectrum benefit are as follows:Wherein, γi It is signal-to-noise ratio, the cost for defining the power resource in unit radio frequency is cp, wireless access links ηpPower resource in unit radio frequency The cost of benefit is ηp=pk×cp,
The cost for defining unit information resource is cit, service ηitThe cost of consumption information resource is ηp=itk×cit, the portion of VNF Affix one's name to benefit ηdCost be ηd=σb×ηs+σp×ηp+σit×ηit,
Wherein, σb、σp、σitIt is weight coefficient, 0≤σb,σp,σit≤ 1, and σb+σp+σit=1；
S103: carrying out mathematical modeling to virtualization network function layout process, including,
Building virtualization network function layout process mathematical model:
s.t.Ri,j,k→it≤N×xj',k'→it
Ri,j,k→p≤N×xj',k'→p
Ri,j,k→b≤N×xj',k'→b
Wherein μi,jIt is cost function, indicates that there is attribute akVirtual functions module fjThe required cost paid, xj',k'It indicates The functional module of selection, Ri,j,k→it≤N×xj',k'→ it, Ri,j,k→p≤N×xj',k'→ p, Ri,j,k→b≤N×xj',k'→ B, which indicates that selection is N number of, has attribute akVirtual functions module fjIT resource value, frequency spectrum resource, transmitting-receiving power resource be greater than or The resource value of the application i request such as person, N is natural number,Indicate virtual functions module fiWith virtual functions module fi+yIt deposits In dependence, fi≠fi+yIndicate virtual functions module fiWith virtual functions module fi+yThere are exclusion relations, costj,kRefer to multiple With attribute ajVirtual functions module fiThe cost paid when using side by side, Mcostj,kRefer to multiple with attribute ajIt is virtual Functional module fiIt is used in conjunction with the cost paid when a resource, δs, δp, δitTo be combined into functional module xj',k'Be Number；
S104: optimizing virtualization network function layout process mathematical model, include the following steps,
S1041: addition constraint condition: Wherein s', p', it' indicate that the related resource used, all indicate all money Source；
S1042: virtualization network function layout process mathematical model is optimized for for the first time
s.t.Ri,j,k→it≤N×xj',k'→it
Ri,j,k→p≤N×xj',k'→p
Ri,j,k→b≤N×xj',k'→b
Virtual functions module f is increased using following formula in optimization processiCost cost, μj,k'=μj,k+μj+y,k, y is to face Variations per hour；
S1043: it is by quasi-ization network function layout process mathematical model optimizing
s.t.Ri,j,k→it≤N×xj',k'→it
Ri,j,k→p≤N×xj',k'→p
Ri,j,k→b≤N×xj',k'→b。
3. a kind of radio network functions method of combination of the virtualization based on economic benefit as described in claim 1, feature It is, the step S2 includes the following steps,
S201: in total dimension D=m*n of Virtual Service Arrangement solution space, appoint and L particle is taken to form a group, i is small In L, wherein i-th of particle kth for when described using two indices, using the value of virtual functions attribute as the position of PSO, It is expressed asD dimensional vector, virtual functions attribute value changes flying speed of the speed as PSO, expression ForD dimensional vector, wherein t indicate particle t time fly；
S202: i-th particle search to kth for when individual history optimal location beSearch To kth for when the history optimal location g of entire population beAt+1 generation of kth, i-th The iteration more new formula of jth the dimension speed and position of a particle are as follows:
Wherein, ω is inertia weight, previous for measuring The influence mobile to next time of the speed at moment, c1And c2For Studying factors, r1And r2For the random number in [0,1].
4. a kind of radio network functions method of combination of the virtualization based on economic benefit as described in claim 1, feature It is, the step S3 includes the following steps,
S301: the inertia weight ω of dynamic regulation traditional PS O algorithm, when particle position is close to individual history optimal location or whole Convergent resolution ratio is improved when a population history optimal location and reduces the flying speed of particle, when particle position is far from individual Accelerate convergence rate when history optimal location or the history optimal location of entire population and improve the flying speed of particle, is used to Property weights omega is given by:
Wherein, tv1And tv2It is threshold value, c1iAnd c2iFor the Studying factors of i-th, r1iAnd r2iFor the random number of i-th；
S302: the design for the particle that makes a variation, comprising the following steps:
S3021: ifThen determine psiFor the particle that makes a variation, wherein tp is threshold values,
S3022: set variation the smooth moving distance of particle as
Wherein Bd (k) is the smooth moving distance of variation particle in kth generation, and L is the interval number between two particles；
S3023: set the variation equation of particle as
CN201910648218.0A 2019-07-18 2019-07-18 Virtualized wireless network function arrangement method based on economic benefits Active CN110351145B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910648218.0A CN110351145B (en) 2019-07-18 2019-07-18 Virtualized wireless network function arrangement method based on economic benefits

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910648218.0A CN110351145B (en) 2019-07-18 2019-07-18 Virtualized wireless network function arrangement method based on economic benefits

Publications (2)

Publication Number Publication Date
CN110351145A true CN110351145A (en) 2019-10-18
CN110351145B CN110351145B (en) 2020-03-20

Family

ID=68175699

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910648218.0A Active CN110351145B (en) 2019-07-18 2019-07-18 Virtualized wireless network function arrangement method based on economic benefits

Country Status (1)

Country Link
CN (1) CN110351145B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110996334A (en) * 2019-12-09 2020-04-10 衡阳师范学院 Virtualized wireless network function arrangement strategy
CN111082997A (en) * 2019-12-30 2020-04-28 西安电子科技大学 Network function arrangement method based on service identification in mobile edge computing platform
CN111510334A (en) * 2020-04-21 2020-08-07 中国电子科技集团公司第五十四研究所 Particle swarm algorithm-based VNF online scheduling method
CN113904923A (en) * 2021-09-27 2022-01-07 重庆电子工程职业学院 Service function chain joint optimization method based on software defined network

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1474356A (en) * 2003-06-19 2004-02-11 山东华泽精密模塑有限公司 Use of arithmetic model in integrated digitized digitized manufacture technology for enterprise
US20140277599A1 (en) * 2013-03-13 2014-09-18 Oracle International Corporation Innovative Approach to Distributed Energy Resource Scheduling
CN105430706A (en) * 2015-11-03 2016-03-23 国网江西省电力科学研究院 WSN (Wireless Sensor Networks) routing optimization method based on improved PSO (particle swarm optimization)
CN107104826A (en) * 2017-04-05 2017-08-29 电子科技大学 A kind of network control coding method based on SDN and NFV
CN108174394A (en) * 2018-01-12 2018-06-15 西安邮电大学 A kind of Arrangement algorithm of 5G networks slice
US20190089780A1 (en) * 2017-09-15 2019-03-21 Nec Europe Ltd. Application function management using nfv mano system framework
CN109818865A (en) * 2019-03-11 2019-05-28 江苏君英天达人工智能研究院有限公司 A kind of SDN enhancing path boxing apparatus and method
US20190196852A1 (en) * 2015-08-25 2019-06-27 Futurewei Technologies, Inc. System and Method for Network Function Virtualization Resource Management

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1474356A (en) * 2003-06-19 2004-02-11 山东华泽精密模塑有限公司 Use of arithmetic model in integrated digitized digitized manufacture technology for enterprise
US20140277599A1 (en) * 2013-03-13 2014-09-18 Oracle International Corporation Innovative Approach to Distributed Energy Resource Scheduling
US20190196852A1 (en) * 2015-08-25 2019-06-27 Futurewei Technologies, Inc. System and Method for Network Function Virtualization Resource Management
CN105430706A (en) * 2015-11-03 2016-03-23 国网江西省电力科学研究院 WSN (Wireless Sensor Networks) routing optimization method based on improved PSO (particle swarm optimization)
CN107104826A (en) * 2017-04-05 2017-08-29 电子科技大学 A kind of network control coding method based on SDN and NFV
US20190089780A1 (en) * 2017-09-15 2019-03-21 Nec Europe Ltd. Application function management using nfv mano system framework
CN108174394A (en) * 2018-01-12 2018-06-15 西安邮电大学 A kind of Arrangement algorithm of 5G networks slice
CN109818865A (en) * 2019-03-11 2019-05-28 江苏君英天达人工智能研究院有限公司 A kind of SDN enhancing path boxing apparatus and method

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
SAI ZOU等: ""Resource allocation mechanism based on two-step mapping for saving energy in wireless network virtualization"", 《2015 IEEE 9TH INTERNATIONAL CONFERENCE ON ANTI-COUNTERFEITING, SECURITY, AND IDENTIFICATION (ASID)》 *
VAHID ABEDIFAR等: ""An optimized virtual network mapping using PSO in cloud computing"", 《2013 21ST IRANIAN CONFERENCE ON ELECTRICAL ENGINEERING (ICEE) ON IEEE》 *
XUEZHI LEI: ""Optimization of Intelligent Neural Network Prediction Based on Particle Swarm"", 《2018 INTERNATIONAL CONFERENCE ON VIRTUAL REALITY AND INTELLIGENT SYSTEMS (ICVRIS) ON IEEE》 *
王改堂等: ""基于自适应变异的动态粒子群优化算法"", 《科技通报》 *
马丁等: ""基于离散粒子群优化的多目标服务路径构建算法"", 《通信学报》 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110996334A (en) * 2019-12-09 2020-04-10 衡阳师范学院 Virtualized wireless network function arrangement strategy
CN110996334B (en) * 2019-12-09 2022-10-11 衡阳师范学院 Virtualized wireless network function arrangement strategy
CN111082997A (en) * 2019-12-30 2020-04-28 西安电子科技大学 Network function arrangement method based on service identification in mobile edge computing platform
CN111082997B (en) * 2019-12-30 2021-05-14 西安电子科技大学 Network function arrangement method based on service identification in mobile edge computing platform
CN111510334A (en) * 2020-04-21 2020-08-07 中国电子科技集团公司第五十四研究所 Particle swarm algorithm-based VNF online scheduling method
CN113904923A (en) * 2021-09-27 2022-01-07 重庆电子工程职业学院 Service function chain joint optimization method based on software defined network
CN113904923B (en) * 2021-09-27 2023-08-22 重庆电子工程职业学院 Service function chain joint optimization method based on software defined network

Also Published As

Publication number Publication date
CN110351145B (en) 2020-03-20

Similar Documents

Publication Publication Date Title
Chen et al. DNNOff: offloading DNN-based intelligent IoT applications in mobile edge computing
CN110351145A (en) A kind of radio network functions method of combination of the virtualization based on economic benefit
Lin et al. A time-driven data placement strategy for a scientific workflow combining edge computing and cloud computing
He et al. Blockchain-based edge computing resource allocation in IoT: A deep reinforcement learning approach
Scutari et al. Parallel and distributed methods for constrained nonconvex optimization—Part I: Theory
Wen et al. Federated dropout—a simple approach for enabling federated learning on resource constrained devices
CN113435472A (en) Vehicle-mounted computing power network user demand prediction method, system, device and medium
Hao et al. Deep reinforcement learning for edge service placement in softwarized industrial cyber-physical system
Xiao et al. System delay optimization for mobile edge computing
CN107579518A (en) Power system environment economic load dispatching method and apparatus based on MHBA
Liu et al. Communication-efficient asynchronous federated learning in resource-constrained edge computing
Li et al. Joint optimization of bandwidth for provider and delay for user in software defined data centers
Zhou et al. Resource allocation of federated learning for the metaverse with mobile augmented reality
CN114172820A (en) Cross-domain SFC dynamic deployment method, device, computer equipment and storage medium
Li et al. A self-learning bee colony and genetic algorithm hybrid for cloud manufacturing services
CN110351144A (en) A kind of Internet of Things access of virtual resource method of combination based on CRO correction
Yang et al. Federated learning with nesterov accelerated gradient
Mollahasani et al. Energy-aware dynamic DU selection and NF relocation in O-RAN using actor–critic learning
Deng et al. Tailorfl: Dual-personalized federated learning under system and data heterogeneity
Song et al. Adaptive and collaborative edge inference in task stream with latency constraint
Huang et al. Cost efficient offloading strategy for DNN-based applications in edge-cloud environment
Gogulan et al. An multiple pheromone algorithm for cloud scheduling with various QOS requirements
Yi et al. DMADRL: A distributed multi-agent deep reinforcement learning algorithm for cognitive offloading in dynamic MEC networks
CN115660895A (en) Water resource management method for improving deep learning based on width learning
Baccour et al. RL-PDNN: Reinforcement learning for privacy-aware distributed neural networks in IoT systems

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