CN111405587B - Virtual wireless network resource allocation method based on channel pricing - Google Patents

Virtual wireless network resource allocation method based on channel pricing Download PDF

Info

Publication number
CN111405587B
CN111405587B CN202010205407.3A CN202010205407A CN111405587B CN 111405587 B CN111405587 B CN 111405587B CN 202010205407 A CN202010205407 A CN 202010205407A CN 111405587 B CN111405587 B CN 111405587B
Authority
CN
China
Prior art keywords
user
game
power
virtual network
matching
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202010205407.3A
Other languages
Chinese (zh)
Other versions
CN111405587A (en
Inventor
曾菊玲
张春雷
蒋砺思
夏凌
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.)
China Three Gorges University CTGU
Original Assignee
China Three Gorges University CTGU
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 China Three Gorges University CTGU filed Critical China Three Gorges University CTGU
Priority to CN202010205407.3A priority Critical patent/CN111405587B/en
Publication of CN111405587A publication Critical patent/CN111405587A/en
Application granted granted Critical
Publication of CN111405587B publication Critical patent/CN111405587B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/22—Traffic simulation tools or models
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/50—Allocation or scheduling criteria for wireless resources
    • H04W72/53—Allocation or scheduling criteria for wireless resources based on regulatory allocation policies
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/50—Allocation or scheduling criteria for wireless resources
    • H04W72/54—Allocation or scheduling criteria for wireless resources based on quality criteria
    • H04W72/542—Allocation or scheduling criteria for wireless resources based on quality criteria using measured or perceived quality
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Quality & Reliability (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention discloses a wireless virtual network resource allocation strategy, which comprises the following steps: channel pricing based matching/Stackelberg layered gaming. And respectively taking the user satisfaction based on the stream bandwidth, the system bandwidth and the slicing power as reward functions, establishing a joint optimization model for the three-layer architecture virtual network, and solving by adopting a matching/Stackelberg layered game. Game low level, defining MVNOm-InPn pairsm n And one-to-one matching game between the game and UEs, pairm n Defining a slicing power price based on channel average information, coupling upper-layer game information, reducing iteration times, and providing a rejection-reception algorithm with low calculation complexity, so that a user can select the optimal slicing power under the condition of equal slicing powerm n . The game high layer is InPs andm n and in the inter-Stackelberg game, a power pricing and allocation strategy based on channel local information is given, so that the system utility and the resource utilization rate are optimal. Double-layer cycling and stabilization conditions are given.

Description

Virtual wireless network resource allocation method based on channel pricing
Technical Field
The invention belongs to the field of wireless communication, and particularly relates to a virtual wireless network resource allocation method based on channel pricing.
Background
Because the wireless network virtualization has the characteristics of good flexibility, programmability, easy customization, isolation and the like, the wireless network virtualization is proposed as a main networking mode for realizing multiple services and large connection targets in the future 5G. In the wireless network virtualization technology, a proper network architecture and an effective resource allocation strategy are key technologies for improving resource utilization rate and user satisfaction. In the secondary architecture model in which the basic resource providing layers inp directly provide resources to the user layer UEs, the computational complexity is high and the on-chip instantiation cannot be realized because the virtual network operation layer MVNOs is not included. In the virtual network with the three-level architecture, the InPs are abstracted into slices to provide resources for the MVNOs in a service form, and the MVNOs of the middle virtual network operation layer provides services for the user layer UEs which generate virtual resource requests, so that the InPs of the basic resource abstraction layer and the user UEs are decoupled and shared, and the resource utilization rate, operation, control experience and customization are greatly improved, thereby being widely adopted. A plurality of layered joint optimization resource distribution models based on a three-layer structure are provided, and the models are solved by a layered game. In an optimization model, especially at a user level, it is currently common to use bandwidth gains as a utility function, replace user requirements with a network performance function, which does not truly reflect user experience, and also does not relate user experience to the supply of physical resources, so that channels cannot be tracked. By adopting the user satisfaction based on the stream bandwidth as the utility function, the problems can be solved, and the method can adapt to the future 5G stream service mode. In the layered game solving of the model, a mechanism for realizing virtualization through opportunity spectrum sharing is commonly used, but the practicability is reduced due to incomplete isolation, and the MVNOs are based on a resource allocation algorithm of a bankruptcy game, but the optimization of the whole virtual network is difficult to realize only considering a low-level mechanism, and the most of the layered auction mechanisms with the combination of UEs and MVNOs and the secondary auctions of MVNOs and InPs can ensure the maximum overall utility and the suboptimum resource allocation, but a centralized control mode of a third party is required, so that the system is complex and the maximum individual utility cannot be ensured. The layered matching game can overcome the defects of a layered auction mechanism, the layered matching game mechanism between UEs and MVNOs and between MVNOs and InPs is adopted, the problem of two-stage joint optimization of service selection and resource purchase under a three-stage architecture is solved, a decentralized realization method enables a system to be relatively simple, but the utility function and the price in an optimization model of the system ignore wireless channel factors and cannot adapt to a wireless environment, particularly a high-dynamic wireless environment with fast channel change, and meanwhile, the problems of large-computation-amount cyclic iteration, incapability of tracking a channel state, low resource utilization rate and the like caused by the lack of coupling information between the upper layer and the lower layer of the layered game, the lack of a random pricing mechanism of each layer and an on-chip resource optimization distribution mechanism are solved.
The method has the advantages that the wireless channel state can be self-adapted in the wireless virtual network with the three-level architecture, and the upper-layer game and the lower-layer game have certain coupling information, low cycle times, high benefit and high efficiency.
Disclosure of Invention
The invention discloses a wireless virtual network resource allocation strategy, which comprises the following steps: channel pricing based matching/Stackelberg layered gaming. And respectively taking the user satisfaction based on the stream bandwidth, the system bandwidth and the slicing power as reward functions, establishing a joint optimization model for the three-layer architecture virtual network, and solving by adopting a matching/Stackelberg layered game. Gaming lower layer, defining MVNOm-InPn to m n And one-to-one matching game between the game and UEs, pair m n Defining a slicing power price based on channel average information, coupling upper-layer game information, reducing iteration times, and providing a rejection-reception algorithm with low calculation complexity, so that a user can select the optimal m under the condition of equal slicing power n . The game is based on the user and m n InPs and m in connection relation n And in the inter-Stackelberg game, a power pricing and allocation strategy based on channel local information is given, so that the utility and resource utilization rate of the system are optimal. The upper and lower double-layer circulation and stable conditions are given.
The technical scheme of the invention is a virtual wireless network resource allocation method based on channel pricing, a wireless virtual network adopts a three-level architecture, wherein a basic resource providing layer InPs is abstracted into slices to provide resources to a virtual network operation layer MVNOs in a service form, a middle layer virtual network operation layer MVNOs purchases bandwidth from the basic resource providing layer InPs and provides service to user layer UEs generating virtual resource requests, and the virtual wireless network resource allocation method comprises the following steps:
1) Establishing a three-level joint optimization model based on a three-level architecture: and establishing an optimization model with maximum utility for the user layer UEs by taking the user satisfaction based on the stream bandwidth as a reward function and buying the bandwidth to the middle layer MVNOs as a cost respectively, establishing an optimization model with maximum utility for the middle layer MVNOs by taking the benefit of providing the bandwidth to the user layer UEs as the reward function and buying the power to the InPs as a cost, and establishing an optimization model with maximum utility for the InPs by taking the benefit of providing the power to the middle layer MVNOs as a reward to form a three-level joint optimization model.
2) Solving the optimization model by adopting a matching/Stackelberg hierarchical game: defining m in low-level game formed by user layer UEs and virtual network operation layer MVNOs n Represents MVNO m —InP n To, wherein MVNO m For the mth virtual network operator, inP n For the nth infrastructure provider, converting a many-to-one matching game of UEs (user Equipment) and MVNOs (virtual network operation layer) of a user layer into users and m n One-to-one matching game between m n Mutually independent optimized slice power prices based on channel average information are defined, upper layer game information is coupled, the number of times of upper and lower layer loop iteration is reduced, and users can adaptively select optimal m related to channels under the condition that slice power is equal n The isolation among slices is kept, and a distributed rejection-reception algorithm with low calculation complexity is provided; at the base resource providing layer InPs and m n The upper layer game is formed based on the users and m n Forming base resource providing layers InPs and m in connection relation n In the Stackelberg game, an optimized power pricing and allocation strategy based on local information of the channel is given, so that the utility and the resource utilization rate of the system are optimal and the channel is self-adapted.
3) And giving an upper and lower double-layer game circulation flow diagram and a stable condition thereof.
Preferably, the three-level joint optimization model establishes a three-level joint optimization model for a wireless virtual network with a three-level architecture of user layer UEs, virtual network operation layer MVNOs and basic resource providing layer inp by taking user satisfaction based on stream bandwidth, system bandwidth and slice power as reward functions; solving the optimization model by adopting a matching/Stackelberg hierarchical game;
the three-level joint optimization model is used for any user UE k K belongs to K, K is a user set, and the optimization goal is as follows:
Figure GDA0003761286190000031
Figure GDA0003761286190000032
the formula (1) indicates that the user layer optimization target is maximum in utility, the utility of the user is the user satisfaction minus the cost of buying bandwidth, and the constraint of the formula (2) indicates that each user UE can only be served by one virtual network operator MVNO; wherein x is k,m =1 denotes user UE k Selecting a virtual network operator MVNO m Service if user UE k Not selecting virtual network operator MVNO m Then x is k,m ＝0；
Figure GDA00037612861900000312
Satisfaction obtained by purchasing bandwidth for the user; d k Is a user UE k The number of required slices; lambda k,m For virtual network operator MVNO m To the user UE k A price to provide bandwidth;
Figure GDA0003761286190000033
representing the bandwidth purchased by the user;
Figure GDA0003761286190000034
is as follows
Figure GDA0003761286190000035
Function U data For a stream bandwidth based user satisfaction function
Figure GDA0003761286190000036
Wherein B is max Bandwidth to maximum effect, s 1 Is a normal number, B is the bandwidth provided to the user；
Figure GDA0003761286190000037
Is as follows
Figure GDA0003761286190000038
Wherein L is the frame length and,
Figure GDA0003761286190000039
buying the nth slice of the base resource provider InP for the virtual network operator MVNOm provides the data rate of the service to user k,
Figure GDA00037612861900000310
wherein
Figure GDA00037612861900000311
As the signal-to-noise ratio
Figure GDA0003761286190000041
Wherein h is k,n ,σ 2 Are each m n Channel gain to user k, noise power spectral density;
for any base resource provider InP, the optimization goals are as follows:
Figure GDA0003761286190000042
Figure GDA0003761286190000043
Figure GDA0003761286190000044
Figure GDA0003761286190000045
wherein y is m,n =1 base resource provider InP agreed virtual network operator MVNO m Purchasing a slice n; d m Represents MVNO m Total number of admitted user UE demands, mu m,n For virtual network operator MVNO m Price for purchasing slice n, P m,n Is m n The power of (c); p is the total power of the InP of a single base resource provider; n is a radical of s To represent the total number of slices a single base resource provider InP contains; wherein
Figure GDA0003761286190000046
Equation (5) indicates that the base resource provider InP allocated power gets the most profitable, equation (6) indicates that the total number of all virtual network operator application slices admitted by each base resource provider InP is limited, and equation (7) indicates that the virtual network operator MVNO m The number of slices requested is greater than the user requirement to guarantee service, and equation (8) indicates that the total power is limited;
for virtual network operator MVNOs, the optimization objectives are as follows:
Figure GDA0003761286190000047
Figure GDA0003761286190000048
Figure GDA0003761286190000051
x k,m =1 denotes user UE k Virtual network operator MVNO m Receiving; y is m,n =1 virtual network operationCommercial MVNO m Section n, mu of InP to be purchased as a base resource provider m,n To purchase power p m,n The price of (c); l. the k,n Implementation d representing virtual network operator computation k Number of slices required, λ m,k For virtual network operator MVNO m Price when selling bandwidth to user k;
equation (10) ensures that user k is served by at most one virtual network operator, and equation (11) ensures that the resources allocated on the slice are smaller than those provided to the virtual network operator MVNO m The volume of the slice.
Preferably, the user and m n One-to-one matching game for user UE k A preference function of
Figure GDA0003761286190000052
Wherein λ is k,m Representing the price, lambda, of the bandwidth purchased by the user k,m Is a uniformly distributed random number, beta k,m Purchasing a virtual network operator MVNO for user k m Is required to be satisfied by the service of (c),
Figure GDA0003761286190000053
to the virtual network operator MVNO for user k m The bandwidth purchased;
for m n A preference function of
Figure GDA0003761286190000054
Wherein mu m,n Is m n Price of power purchased, mu m,n Is calculated as follows
Figure GDA0003761286190000055
σ 2 Representing the noise power, P represents the total power of the InP of a single underlying resource provider, N s Representing a single underlying resourceTotal number of slices contained in provider InP, λ denotes m n The price of bandwidth to the user is offered and h represents the channel gain between the slice and the user.
Preferably, the user is associated with m n One-to-one matching game, reject-receive algorithm:
the one-to-one matching rejection/reception algorithm includes: assuming equal power for each slice in the infrastructure provider InP, the prices for the user UE to purchase bandwidth from the virtual network operator MVNO and the prices for the virtual network operator MVNO to provide bandwidth to the user UE are subject to uniform distribution,
(1) Each user UE towards the best preference m acceptable n Submitting a lease application; each m n Selecting the user UE with the best preference among all the received applications k Rejecting all other applications; assume each m n Only one user can be selected;
(2) If rejection occurs, each user who has not yet matched rejects its most preferred m n Submitting a lease application; each m n The most preferred application is left in all the received applications, and all other applications are rejected;
and when all the user UEs finish matching, the algorithm is ended.
Preferably, the base resource providing layers inp and m n In the Stackelberg game, the revenue function of the InP sales power of the base resource provider is:
Figure GDA0003761286190000061
Figure GDA0003761286190000062
in the formula of m,n For virtual network operators MVNO m Price for purchasing slice n, P m.n Is m n P represents the total power of the single base resource provider InP; n is a radical of s Representing the total number of slices a single underlying resource provider InP contains
For the virtual network operator MVNO, the revenue obtained from purchasing power is:
Figure GDA0003761286190000063
wherein λ k,m The price at which the bandwidth is purchased for the user,
Figure GDA0003761286190000064
to a virtual network operator MVNO for a user k m The bandwidth purchased;
for MVNO m —InP n To m n Let us order
Figure GDA0003761286190000065
m n The strategy for buying or distributing the power of the InP of the basic resource provider is
Figure GDA0003761286190000066
m n The power price strategy for buying or selling InP of the basic resource provider is
Figure GDA0003761286190000071
If and only if
Figure GDA0003761286190000072
Order to
Figure GDA0003761286190000073
Figure GDA0003761286190000074
Preferably, the optimization model is solved by adopting a layered game, and the cyclic process and the stable conditions of the layered game are as follows:
stability definition: let Ω (k, m) n ) Is UEs and m n The existing matching pair set formed by one-to-one matching, if the new matching pair set psi (k, m) n )＝Ω(k,m n ) Then the match is blocked and stable is achieved.
An upper and a lower double-layer circulation process:
(1) First, assume that the power of each slice in InP is equal for the infrastructure providers, UE and m n The game is matched one by one, wherein the prices of the user UE for buying the bandwidth to the virtual network operator MVNO and the virtual network operator MVNO for providing the bandwidth to the user UE are subjected to uniform distribution, and the prices of the virtual network operator MVNO for buying the slicing power are based on average channel information;
(2) The connection relation is sent to the upper layer.
(3) Infrastructure provider InP and multiple m n Performing a Stackelberg game to obtain a power price and a power strategy based on local channel information, and performing power distribution; m after power distribution n And performing one-to-one matching with the user UE, and performing loop iteration until the matching results of 2 times are completely the same.
The beneficial effects of the invention are:
(1) The method comprises the steps of respectively taking user satisfaction, system bandwidth and slice power based on stream bandwidth as reward functions, establishing a three-level joint optimization model for a wireless virtual network with a three-layer structure of a User (UEs), a virtual network operation layer (MVNOs) and a basic resource providing layer (InPs), and taking slice physical resources as arguments in the reward functions of the user satisfaction, the system bandwidth and the slice power, so that the maximum utility and the resource utilization rate of adaptive channel characteristics can be better guaranteed compared with the traditional matching layered game.
(2) Solving the optimization model by adopting a matching/Stackelberg hierarchical game, wherein the lower layer is a one-to-one matching game, the upper layer is a Stackelberg game and the result passes through m n And coupling of power pricing thereof into up and down gamingCompared with the traditional 2-layer many-to-one hierarchical matching game, the method is easier to form accurate matching and reduce iteration times, in the upper and lower two-layer games, the preference function and the power pricing both adopt the channel characteristics as the characteristics, and compared with the hierarchical matching game, when the channel changes, the method is superior in both channel tracking capability and resource utilization rate.
(3) At the lower layer of game formed by UEs and MVNOs, by defining MVNOm-InPn to m n Converting the many-to-one matching game of UEs and MVNOs into UEs and m n One-to-one matching game between games, UEs and m n The accurate matching of the channel characteristics enables the channel characteristics to be directly reflected in a user layer utility function, and the user utility can quickly track the channel change.
(4) In the bottom layer game, for m n Defining a mutually independent optimized slice power price based on channel average information to ensure that a user selects the optimal m related to a channel under the condition of equal slice power n The price and m n The purchase power prices (in the average channel information) in the upper layer game have similarity, so that the lower layer game and the upper layer game can easily achieve consistent convergence, the cycle iteration times are reduced, and meanwhile, the channel characteristics are reflected in the matching price and preference functions, so that the utility of the user layer and the MVNOs can track the channel change.
(5) The distributed rejection-reception algorithm with low computational complexity is provided, and the computational complexity is low.
(6) At the upper level of the game, based on the users and m n The connection relation is formed between InPs and m n The Stackelberg game among the Stackelberg games can adapt to the channel change.
(7) Compared with a Stackelberg game which adopts a Stackelberg game and a power distribution strategy but does not adopt a mechanism of optimizing the power pricing, namely the Stackelberg game of power random pricing, the optimized power pricing and the optimized power distribution strategy not only ensure that the system utility and the resource utilization rate are optimal and self-adaptive to the channel to realize the on-chip optimization, but also are easy to match m n The power price based on average channel information adopted in the low-level game forms consistent convergence, and the number of cycle iteration is reduced.
(8) And finally, defining the stability of the matching/Stackelberg hierarchical game, giving a process diagram of equal power matching, stackelberg power distribution and optimized power matching, and ensuring that the whole resource distribution process is executed.
Simulation results show that when channels change, the matching/Stackelberg layered game strategy based on channel pricing is superior to a power random pricing mechanism in a matching/Stackelberg layered game and a traditional layered matching game in the aspects of tracking channels, resource utilization rate and effectiveness.
Drawings
The invention is further illustrated by the following examples in conjunction with the drawings.
Fig. 1 is a schematic diagram of an embodiment of a matching/Stackelberg hierarchical game based on channel pricing.
Fig. 2 is a schematic diagram of the structure and lower and upper layers of a wireless virtual network according to an embodiment.
Fig. 3 is a schematic diagram illustrating a comparison of tracking channel capabilities of a matching/Stackelberg hierarchical game based on channel pricing and a matching/Stackelberg hierarchical game based on random pricing.
Fig. 4 is a schematic diagram of comparison of summation rates of matching/stagelberg hierarchical gaming based on channel pricing and different resource allocation methods.
Fig. 5 is a schematic diagram showing the utility comparison of the matching/Stackelberg hierarchical game based on channel pricing and the matching/Stackelberg hierarchical game based on non-channel pricing.
Detailed Description
The downlink of a wireless virtual network implemented in a three-tier architecture according to the present invention is shown in fig. 2. The highest level of the model is infrastructure providers InP, which can provide N, assuming without loss of generality that each infrastructure provider InP contains a base station, each channel of which is abstracted into one slice S Each slice has the same bandwidth and adjustable power, and supposing that one slice can only serve one virtual network operator at the same time, different rewards are charged for different virtual network operators. The middle layer of the model is mobile virtual network operator MVNOs or mobile virtual network operator MVNOs, and the virtual network operator MVNOs pass through an independent contract according toNetwork conditions and resource prices, and the like, dynamically purchasing wireless resources from infrastructure providers InPs, and one virtual network operator may purchase multiple slices, the virtual network operator providing services to end users at different prices, the mth virtual network operator MVNO m M is equal to M, and is sent to a group of users K m The provision of the service(s) is made, K is the total number of users, K = & m K m The radix of the set K is represented by the symbol | K |. The lowest layer of the model is end Users (UEs) which can make bandwidth requests to virtual network operators (MVNOs) in a service flow mode, and one user can make service applications to a plurality of virtual network operators. The system work flow is as follows: the end user UEs apply for the bandwidth to the virtual network operator MVNOs, and then the virtual network operator MVNOs applies for the network slice to the infrastructure provider InPs.
As shown in fig. 1, the method for allocating virtual wireless network resources based on channel pricing comprises the following steps:
s1: establishing a three-level joint optimization model; establishing a three-level joint optimization model for a wireless virtual network of a three-level architecture of user layer UEs, virtual network operation layer MVNOs and basic resource providing layer InPs by taking user satisfaction based on stream bandwidth, system bandwidth and slice power as reward functions; and solving the optimization model by adopting a matching/Stackelberg hierarchical game. Giving out a three-level combined optimization model under double-layer circulation and stable conditions;
for any user K belonging to K, wherein K is a user set, the optimization target is as follows:
Figure GDA0003761286190000101
Figure GDA0003761286190000102
the formula (1) represents that the user layer optimization target is maximum in utility, the utility of the user is the user satisfaction minus the cost of purchasing bandwidth, and the constraint of the formula (2) represents that each user UE can only be served by one virtual network operator MVNO;
wherein x k,m =1 denotes user UE k Selecting a virtual network operator MVNO m Service if user UE k Not selecting virtual network operator MVNO m Then x is k,m ＝0；
Figure GDA0003761286190000108
Satisfaction obtained by purchasing bandwidth for the user; d k Is a user UE k The number of required slices; lambda k,m For virtual network operator MVNO m To the user UE k A price to provide bandwidth;
Figure GDA0003761286190000103
representing the bandwidth purchased by the user;
Figure GDA0003761286190000104
is as follows
Figure GDA0003761286190000105
Function U data For a stream bandwidth based user satisfaction function
Figure GDA0003761286190000106
Wherein B is max Bandwidth to maximum effect, s 1 Is a normal number, B is the bandwidth provided to the user;
Figure GDA0003761286190000107
is expressed as follows
Figure GDA0003761286190000111
Wherein L is the frame length and,
Figure GDA0003761286190000112
for virtual network operators MVNO m Purchasing an nth slice of the base resource provider InP provides the data rate of the service to user k,
Figure GDA0003761286190000113
wherein
Figure GDA0003761286190000114
As the signal-to-noise ratio
Figure GDA0003761286190000115
Wherein h is k,n ,σ 2 Are each m n Channel gain to user k, noise power spectral density;
for any base resource provider InP, the optimization goals are as follows:
Figure GDA0003761286190000116
Figure GDA0003761286190000117
Figure GDA0003761286190000118
Figure GDA0003761286190000119
wherein y is m,n =1 represents base resource provider InP agreeing virtual network operator MVNO m Purchasing a slice n; d m Represents MVNO m Total number of admitted user UE demands, mu m,n For virtual network operators MVNO m Purchasing slicesPrice of n, P m,n Is m n The power of (c); p is the total power of the InP of a single base resource provider; n is a radical of hydrogen s To represent the total number of slices a single base resource provider InP contains; wherein
Figure GDA00037612861900001110
Equation (5) indicates that the base resource provider InP allocated power gets the most profitable, equation (6) indicates that the total number of all virtual network operator application slices admitted by each base resource provider InP is limited, and equation (7) indicates that the virtual network operator MVNO m The number of slices applied is greater than the user's demand to guarantee service, and equation (8) represents that the total power is limited;
for virtual network operator MVNOs, the optimization goals are as follows:
Figure GDA0003761286190000121
Figure GDA0003761286190000122
Figure GDA0003761286190000123
x k,m =1 denotes user UE k Virtual network operator MVNO m Receiving; y is m,n =1 for virtual network operator MVNO m Section n, mu of InP to be purchased as a base resource provider m,n To purchase power p m,n The price of (c); l k,n Implementation d representing virtual network operator computation k Number of slices required, λ m,k For virtual network operators MVNO m Price when bandwidth is sold to user k;
equation (10) ensures that user k is served by at most one virtual network operator, and equation (11) ensures that the resources allocated on the slice are smaller than those provided to the virtual network operatorMVNO m The volume of the slice.
S2: solving the optimization model by adopting a matching/Stackelberg hierarchical game;
s201: slice equal power condition, user and m n Match the game one by one;
as shown in FIG. 2, assuming that the virtual network operators MVNOs can lease all slices, slices of infrastructure provider InPs can accommodate any virtual network operator, combine the virtual network operator and slices into a unit pair, define m n Represents MVNO m —InP n To, wherein MVNO m For the mth virtual network operator, inP n For the nth infrastructure provider, all m that may occur n Form a set M n And converting a many-to-one matching game between the UEs of the user layer and the MVNOs of the virtual network operator into the UEs and the M of the user layer n One-to-one matching games.
For user UE k A preference function of
Figure GDA0003761286190000124
Without loss of generality, let λ m,k Are uniformly distributed random numbers.
For m n Let its preference function be
Figure GDA0003761286190000131
Wherein mu m,n Is calculated as follows
Figure GDA0003761286190000132
Figure GDA0003761286190000133
σ 2 Representing noise power, P representing unitTotal power of InP of individual underlying resource providers, N s Denotes the total number of slices that a single underlying resource provider InP contains, and λ denotes m n The price of bandwidth to the user is offered and h represents the channel gain between the slice and the user.
Figure GDA0003761286190000134
Respectively, the price of providing bandwidth to users by virtual network operators and the channel gain are averaged in the whole network range, and the price lambda k,n And channel h k,n When the statistical distribution of (a) is known,
Figure GDA0003761286190000135
can solve and is applicable to all users and m n Are all the same definite quantity, therefore, mu m,n Only with matching both users UE k And m n And a price lambda k,n Channel h, channel h k,n Correlation, each pair of UE-m n Are independent of each other, user UE k And m n M is not subject to other users n The impact, and therefore, with the above power pricing, slice isolation is guaranteed. Meanwhile, equation (12) is an average optimal pricing function, and the user UE determined according to the pricing function k And m n The matching can realize the user UE under the condition of equal power k Virtual network operator MVNO m And optimal selection between slices n.
User UE and m n The game is matched one by one, and a one-to-one matching rejection/reception algorithm is adopted:
per user UE k To the best preference of acceptable m n Submitting a lease application; each m n Selecting the user UE with the best preference among all the received applications k Rejecting all other applications; suppose each m n Only one user can be selected;
if rejection occurs, each user who has not yet matched rejects its most preferred m n Submitting a lease application; each m n The most preferred application is left in all the received applications, and the most preferred application is rejectedThere are other applications. If m is n Original existing selection user UE k But the matching utility brought by the new applicant is larger, the latter is selected;
and finishing the algorithm when all the user UE finishes matching.
S202: user UEs and m n After the connection relation is determined, the InP of the infrastructure provider and a plurality of m n Performing a Stackelberg game;
as shown in FIG. 1, when users UEs and m n After establishing a stable matching relation under the condition of equal power, assuming that each slice bandwidth is constant and the same, a virtual network operator MVNOs buys power to infrastructure provider InPs and then provides services to user UEs, each infrastructure provider can accept a plurality of virtual network operators, because the InP power of the infrastructure provider is limited, the accepted MVNOs of the plurality of virtual network operators share the power, and the buying process is modeled by taking the InP of the infrastructure provider as a leader and m n For the follower master-slave game, because the infrastructure providers are independent of each other, only any infrastructure provider and the m accepted by the infrastructure provider need to be considered n Game of formation, research infrastructure provider and m n Power and price policies in between. Without loss of generality, consider m n To contain N s Individual slice infrastructure provider purchase power, mu m,n Is m n The price for purchasing power, the revenue function for the infrastructure provider to sell power is:
Figure GDA0003761286190000141
Figure GDA0003761286190000142
for the virtual network operator MVNO, the revenue obtained from purchasing power is:
Figure GDA0003761286190000143
for MVNO m —InP n To m is aligned with n Let us order
Figure GDA0003761286190000144
The Stackelberg game forms a non-cooperative game on a virtual network operator, and the balance point can be obtained through individual optimization to define a Nash balance point:
Figure GDA0003761286190000145
formula (17) relates to P m,n The strict concave function of (2), from the KKT condition, can yield P m,n Is equalized to
Figure GDA0003761286190000146
Mu can be obtained by transformation m,n Is equalized to
Figure GDA0003761286190000151
If and only if
Figure GDA0003761286190000152
Order to
Figure GDA0003761286190000153
Figure GDA0003761286190000154
S3: after power reallocation, users UEs and m n A one-to-one matching is performed again,and circularly iterating until the matching result is not changed any more.
Through the process, the wireless virtual network resource allocation based on channel pricing is completed, and the method greatly improves the tracking of channels, the system, the speed, the utility and the like due to the full utilization of channel information, and is beneficial to the specific implementation in the environment of rapid change of the channels.
The beneficial effects of the invention are proved by the following comparison of channel tracking capability and system resource utilization rate between the matching/Stackelberg hierarchical game based on channel pricing and the matching/Stackelberg hierarchical game based on power random pricing, and the comparison of effectiveness of three parts of a system between the matching/Stackelberg hierarchical game and the matching/Stackelberg hierarchical game based on power random pricing.
Considering an area with uniformly distributed users, 3 virtual network operators MVNOs lease resources to 1 infrastructure provider InP to provide services to 8 users UEs, the channel of the infrastructure provider InP is abstracted into 8 slices, the channel from slice to user contains large-scale fading and small-scale fading, and the large-scale fading is represented by d -2 Determining that d is the distance between users and slices, the loss of 3 users is 1, the loss of 3 users is 0.1, the loss of 3 users is 0.05, the small-scale fading of all channels is subject to Rayleigh fading with equal mean value according to the distance between the users and the base station, and supposing that any virtual network operator can rent any slice, but each slice can only accept one virtual network operator, so 24 MVNOs can be formed m —InP n To m n The user UEs purchases bandwidth at a random price to the virtual network operator MVNO, but once determined, does not change throughout the game. In the embodiment, a hierarchical matching/Stackelberg game based on channel pricing is compared, the hierarchical matching/Stackelberg game based on the Stackelberg power strategy and random power price is adopted, a resource allocation method based on the matching strategy adopted by the upper layer and the lower layer and random slicing power and price is adopted by the upper layer and the lower layer, the channel tracking capability of the 3 methods is shown in figure 3, and when the channel changes, the hierarchical matching/Stackelberg game based on the channel pricing circulates the upper layer and the lower layerThe game can be stabilized after 2-3 iterations, the upper and lower layer game loop iterations are 3-6 times for stabilizing by adopting the Stackelberg power strategy and the hierarchical matching/Stackelberg game with random power price, the upper and lower layer matching games with random power price and random power price are adopted, the proportion of the hundred times of iteration to achieve the stabilization is about 2/10 in the simulation, and the stability can be achieved by adopting more cycles, so the graph 3 is not shown. It can be seen that the hierarchical matching/Stackelberg game based on channel pricing proposed by the present invention has the best ability to track channels.
Fig. 4 shows that, in the hierarchical matching/palletizerg game based on channel pricing, the hierarchical matching/palletizerg game based on the palletizerg power policy with random power price, the resource allocation method in which the upper and lower layers both adopt the matching policy and the slicing power and price are random under the average power condition, and the unmatched worst channel utilization method under the average power condition, and the systems and rates of the 4 methods are analyzed, it can be found that, compared with the unmatched worst channel utilization, the systems and rates of the 3 methods adopting the matching method are greatly improved, and among the 3 methods adopting the matching method, the system and rate of the hierarchical matching/palletizerg game based on channel pricing proposed by the present invention are optimal and are about 15% higher than that of the randomly priced hierarchical matching/palletizerg game, and the system and rate of the resource allocation method in which the upper and lower layers both adopt the matching policy and the slicing power and price are random under the average power condition are the lowest.
Fig. 5 compares the utility of the channel pricing-based hierarchical matching/stacking game and the randomly priced hierarchical matching/stacking game system, and it can be found that the utility of the former game is significantly higher than that of the latter game.
In summary, the virtual wireless network resource allocation method based on the hierarchical matching/Stackelberg game of the channel pricing provided by the invention is superior to other methods in channel tracking, system, speed and effectiveness, the hierarchical matching/Stackelberg game based on the random price is inferior, the resource allocation method that the upper layer and the lower layer both adopt the matching strategy and the slice power and the price are random is worst under the condition of average power, but the virtual wireless network resource allocation method provided by the invention requires a virtual network operator MVNOs to know the average channel information, so the complexity is highest.
The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and substitutions can be made without departing from the technical principle of the present invention, and these modifications and substitutions should also be regarded as the protection scope of the present invention.

Claims (10)

1. The virtual wireless network resource allocation method based on channel pricing is characterized by comprising the following steps:
the wireless virtual network adopts a three-level architecture, wherein a basic resource providing layer InPs is abstracted into slices to provide resources for a virtual network operation layer MVNOs in a service form, and a middle layer virtual network operation layer MVNOs purchases bandwidth for the basic resource providing layer InPs and provides service for user layer UEs generating virtual resource requests;
establishing a three-level joint optimization model based on a three-level architecture: establishing an optimization model with maximum utility for user layer UEs respectively by taking user satisfaction based on stream bandwidth as a reward function and purchasing bandwidth from middle layer MVNOs as cost, establishing an optimization model with maximum utility for middle layer MVNOs by taking benefit of providing bandwidth for user layer UEs as a reward function and purchasing power from basic resource providing layer InPs as cost, and establishing an optimization model with maximum utility for basic resource providing layer InPs by taking benefit of providing power to middle layer MVNOs as reward, so as to form a three-level joint optimization model;
solving the three-level joint optimization model by adopting a matching/Stackelberg hierarchical game: defining a virtual network operator-an infrastructure provider to form a slicing pair, wherein a user layer UEs and the slicing pair form a low-layer one-to-one matching game, the slicing pair is defined with mutually independent optimized slicing power prices based on channel average information, and a distributed rejection-reception algorithm with low computational complexity is adopted; an upper-layer Stackelberg game is formed by the basic resource providing layer InPs and the slice pairs, and the power pricing and allocation strategy based on the channel local information is adopted to solve based on the existing connection relation between the user and the slice pairs; and the lower layer slicing pairs are equal in matching initial power, receive power re-matching distributed by the upper layer Stackelberg game strategy, finish upper and lower double-layer circulation when matching stable conditions are met, and completely solve the combined optimization model.
2. The method of claim 1, wherein the channel pricing-based virtual wireless network resource allocation is performed by a user,
the solution of the three-level joint optimization model by adopting the matching/Stackelberg hierarchical game is as follows: defining m in a low-layer game formed by UEs (user equipments) and MVNOs (virtual network operation layers) n Represents MVNO m —InP n To, wherein MVNO m For the mth virtual network operator, inP n For the nth infrastructure provider, converting the many-to-one matching game of the UEs and the MVNOs of the virtual network operation layer into the users and the m n One-to-one matching game between the two, by pair m n Mutually independent optimized slice power price based on channel average information is defined, and upper-layer game information is coupled, so that users can adaptively select optimal m related to channels under the condition that slice power is equal n Keeping the isolation among slices and providing a distributed rejection-reception algorithm with lower computation complexity; at the base resource providing layer InPs and m n The upper layer game is formed based on the users and m n Forming base resource providing layers InPs and m in connection relation n In the inter-Stackelberg game, a power pricing and allocation strategy based on channel local information is given, so that the utility and resource utilization rate of the system are optimal, and the lower layer m is n The initial matching power is equal, the power distributed by the upper layer Stackelberg game is received for re-matching, the upper and lower double-layer circulation is finished when the matching stability condition is met, and the combined optimization model is completely solved.
3. The method for allocating virtual wireless network resources based on channel pricing according to claim 1, wherein:
the three-level joint optimization model takes the user satisfaction degree, the system bandwidth and the slice power based on the stream bandwidth as a reward function, and carries out optimization on UEs (user equipments), MVNOs (virtual network operation layers) and basic resourcesEstablishing a three-level joint optimization model by using a wireless virtual network with a three-level architecture of source providing layer InPs; solving the optimization model by adopting a matching/Stackelberg hierarchical game; giving double-layer circulation and stable conditions, and the three-level joint optimization model is used for any user UE k K belongs to K, K is a user set, and the optimization goal is as follows:
Figure FDA0003761286180000021
Figure FDA0003761286180000022
the formula (1) represents that the user layer optimization target is maximum in utility, the utility of the user is the user satisfaction minus the cost of purchasing bandwidth, and the constraint of the formula (2) represents that each user UE can only be served by one virtual network operator MVNO;
wherein x k,m =1 denotes user UE k Selecting a virtual network operator MVNO m Service if user UE k Non-selection of virtual network operator MVNO m Then x k,m ＝0；β k,m Satisfaction obtained by purchasing bandwidth for the user; d k Is a user UE k The number of required slices; lambda [ alpha ] k,m For virtual network operators MVNO m To user UE k A price to provide bandwidth;
Figure FDA0003761286180000023
representing the bandwidth purchased by the user;
Figure FDA0003761286180000024
is expressed as follows
Figure FDA0003761286180000025
Function U data (B) As a function of user satisfaction based on stream bandwidth
Figure FDA0003761286180000026
Wherein B is max Bandwidth to maximum effect, s 1 Is a normal number, and B is the bandwidth provided by the MVNO to the user;
Figure FDA0003761286180000027
is as follows
Figure FDA0003761286180000028
Wherein L is the frame length and,
Figure FDA0003761286180000029
for virtual network operators MVNO m Buying an nth slice of the underlying resource provider InP to a user UE k The data rate of the service is provided,
Figure FDA00037612861800000210
wherein
Figure FDA0003761286180000031
As the signal-to-noise ratio
Figure FDA0003761286180000032
Wherein h is k,n ,σ 2 Are each m n Channel gain to user k, noise power spectral density;
for any base resource provider InP, the optimization objectives are as follows:
Figure FDA0003761286180000033
Figure FDA0003761286180000034
Figure FDA0003761286180000035
Figure FDA0003761286180000036
in the formula y m,n =1 represents base resource provider InP agreeing virtual network operator MVNO m Purchasing a slice n; d is a radical of m Represents MVNO m Total number of admitted user UE demands, mu m,n For virtual network operators MVNO m Price for purchasing slice n, P m,n Is m n The power of (d); p is the total power of the InP of a single base resource provider; n is a radical of s To represent the total number of slices a single base resource provider InP contains; wherein
Figure FDA0003761286180000037
Equation (5) indicates that the base resource provider InP allocated power gets the most profitable, equation (6) indicates that the total number of all virtual network operator application slices admitted by each base resource provider InP is limited, and equation (7) indicates that the virtual network operator MVNO m The number of slices requested is greater than the user requirement to guarantee service, and equation (8) indicates that the total power is limited;
for virtual network operator MVNOs, the optimization goals are as follows:
Figure FDA0003761286180000038
Figure FDA0003761286180000041
Figure FDA0003761286180000042
x k,m =1 denotes user UE k Virtual network operator MVNO m Receiving; y is m,n =1 for virtual network operator MVNO m Section n, mu of InP to be purchased as a base resource provider m,n To purchase power p m,n The price of (c); l. the k,n Implementation d representing virtual network operator computation k Number of slices required, λ k,m For virtual network operator MVNO m To user UE k The price at which the bandwidth is sold;
equation (10) ensures user UE k At most, one virtual network operator provides services, equation (11) ensures that the resources allocated on the slice are smaller than those provided to the virtual network operator MVNO m The volume of the slice.
4. The method for allocating virtual wireless network resources based on channel pricing according to claim 1, wherein: adopting a matching/Stackelberg layered game to solve, wherein the lower layer of the game is UEs and m n The game is matched one by one, and the upper layer of the game is a Stackelberg game between InPs and MVNOs.
5. The method for allocating virtual wireless network resources based on channel pricing according to claim 4, wherein: the user and m n One-to-one matching game between users, for UE k A preference function of
Figure FDA0003761286180000043
Wherein λ k,m Representing the price, lambda, of the bandwidth purchased by the user k,m Being uniformly distributed random numbers, beta k,m For a user UE k Purchasing virtual network operator MVNO m Is required to be satisfied by the service of (c),
Figure FDA0003761286180000044
for a user UE k To virtual network operators MVNOs m The bandwidth purchased;
for m n A preference function of
Figure FDA0003761286180000045
6. The method of claim 5, wherein the method comprises: the user and m n In the one-to-one matching game, mu m,n Is m n Price of power purchased, mu m,n Is calculated as follows
Figure FDA0003761286180000051
σ 2 Representing the noise power, P the total power of the InP of the single underlying resource provider, N s Denotes the total number of slices that a single underlying resource provider InP contains, and λ denotes m n The price of bandwidth is offered to the user and h represents the channel gain between the slice and the user.
7. The method for allocating virtual wireless network resources based on channel pricing according to claim 4, wherein: the user UE and m n The game is matched one by one, and a one-to-one matching rejection/reception algorithm is adopted;
the one-to-one matching rejection/reception algorithm includes: assuming equal power for each slice in the infrastructure provider InP, the prices for the user UE to purchase bandwidth from the virtual network operator MVNO and the prices for the virtual network operator MVNO to provide bandwidth to the user UE are subject to uniform distribution,
(1) Each user UE towards the best preference m acceptable n Submitting a lease application; each m n Selecting the user UE with the best preference among all the received applications k Rejecting all other applications; suppose each m n Only one user can be selected;
(2) If rejection occurs, each user who has not yet matched rejects its most preferred m n Submitting a lease application; each m n The most preferred application is left in all the received applications, and all other applications are rejected;
and when all the user UEs finish matching, the algorithm is ended.
8. The method of claim 3, wherein the method comprises:
lower level of the matching/Stackelberg hierarchical game: the base resource providing layers InPs and m n In the Stackelberg game, the revenue function of the InP sale power of the base resource provider is as follows:
Figure FDA0003761286180000052
Figure FDA0003761286180000053
in the formula of m,n For virtual network operators MVNO m Price for purchasing slice n, P m.n Is m n P represents the total power of the single base resource provider InP; n is a radical of hydrogen s Representing the total number of slices a single base resource provider InP contains,
for the virtual network operator MVNO, the revenue obtained from purchasing power is:
Figure FDA0003761286180000061
wherein λ k,m The price at which the bandwidth is purchased for the user,
Figure FDA0003761286180000062
for a user UE k To virtual network operators MVNOs m The bandwidth purchased;
for MVNO m —InP n To m n Let us order
Figure FDA0003761286180000063
9. The method for allocating virtual wireless network resources based on channel pricing according to claim 5, wherein: the base resource providing layers InPs and m n The Stackelberg games between, the power price and the allocation policy are:
m n buying or base resource provider InP to distribute power when P m,n Satisfies the following formula
Figure FDA0003761286180000064
μ m,n Satisfies the following formula
Figure FDA0003761286180000065
The utility and the resource utilization rate are optimal.
10. The method for allocating virtual wireless network resources based on channel pricing according to claim 1 or 2, characterized in that: the matching/Stackelberg layered game double-layer circulation and stable condition based on channel pricing is as follows:
defining: matching blocking setsΨ(k,m n )：
Let Ω (k, m) n ) Is UEs and m n The existing matching pair set formed by one-to-one matching, if the new matching pair set psi (k, m) n )＝Ω(k,m n ) If so, the matching is blocked to achieve stability;
the process of matching/Stackelberg layered game includes:
(1) First, the user UE and the equal power m n A one-to-one matching game between them, wherein the price of the virtual network operator MVNO to buy the sliced power is based on the average channel information;
sending the connection relation to an upper layer;
(2) Infrastructure provider InP and multiple m n Performing a Stackelberg game to obtain a power price and a power strategy based on local channel information, and performing power distribution;
(3) M after power allocation n And performing one-to-one matching with the user UE, and performing loop iteration until the matching results of 2 times are completely the same.
CN202010205407.3A 2020-03-20 2020-03-20 Virtual wireless network resource allocation method based on channel pricing Active CN111405587B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010205407.3A CN111405587B (en) 2020-03-20 2020-03-20 Virtual wireless network resource allocation method based on channel pricing

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010205407.3A CN111405587B (en) 2020-03-20 2020-03-20 Virtual wireless network resource allocation method based on channel pricing

Publications (2)

Publication Number Publication Date
CN111405587A CN111405587A (en) 2020-07-10
CN111405587B true CN111405587B (en) 2022-10-21

Family

ID=71436463

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010205407.3A Active CN111405587B (en) 2020-03-20 2020-03-20 Virtual wireless network resource allocation method based on channel pricing

Country Status (1)

Country Link
CN (1) CN111405587B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112601257B (en) * 2020-12-02 2022-04-08 北京邮电大学 Radio access network RAN slice resource allocation method and related equipment
CN113543228A (en) * 2021-06-10 2021-10-22 中国电子科技集团公司电子科学研究院 Satellite-ground converged network slice resource scheduling method, system and storage medium
CN114339892B (en) * 2022-03-17 2022-05-24 山东科技大学 DQN and joint bidding based two-layer slice resource allocation method
CN115209555B (en) * 2022-06-10 2023-04-07 广西壮族自治区通信产业服务有限公司技术服务分公司 Game theory-based network slicing and computing resource allocation method and system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7787494B1 (en) * 2004-12-07 2010-08-31 Nortel Networks Limited Method and apparatus for assigning and allocating network resources to packet-based virtual private networks
CN103781081A (en) * 2014-02-28 2014-05-07 清华大学 Opportunity sharing-based wireless virtual resource allocation method
CN106846031A (en) * 2016-11-18 2017-06-13 大连理工大学 Credible P 2 P Streaming Media bandwidth pricing method based on credit system and Stackelberg games
CN108990067A (en) * 2018-07-09 2018-12-11 常熟理工学院 A kind of energy efficiency controlling method applied to super-intensive heterogeneous network

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7787494B1 (en) * 2004-12-07 2010-08-31 Nortel Networks Limited Method and apparatus for assigning and allocating network resources to packet-based virtual private networks
CN103781081A (en) * 2014-02-28 2014-05-07 清华大学 Opportunity sharing-based wireless virtual resource allocation method
CN106846031A (en) * 2016-11-18 2017-06-13 大连理工大学 Credible P 2 P Streaming Media bandwidth pricing method based on credit system and Stackelberg games
CN108990067A (en) * 2018-07-09 2018-12-11 常熟理工学院 A kind of energy efficiency controlling method applied to super-intensive heterogeneous network

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
基于双层博弈的虚拟网络带宽分配模型;贾晓光等;《小型微型计算机系统》;20160315(第03期);全文 *
基于非合作博弈论的多小区OFDMA系统动态资源分配算法研究;仲崇显等;《电子与信息学报》;20090815(第08期);全文 *
虚拟环境中基于Stackelberg博弈的资源分配;杨宇等;《华中科技大学学报(自然科学版)》;20121215;全文 *

Also Published As

Publication number Publication date
CN111405587A (en) 2020-07-10

Similar Documents

Publication Publication Date Title
CN111405587B (en) Virtual wireless network resource allocation method based on channel pricing
US11716748B2 (en) Multi-user slice resource allocation method based on competitive game
Du et al. Contract design for traffic offloading and resource allocation in heterogeneous ultra-dense networks
Zhu et al. Virtualization of 5G cellular networks as a hierarchical combinatorial auction
CN112860350A (en) Task cache-based computation unloading method in edge computation
Bae et al. Sequential bandwidth and power auctions for distributed spectrum sharing
CN111866601B (en) Cooperative game-based video code rate decision method in mobile marginal scene
Liwang et al. Let’s trade in the future! A futures-enabled fast resource trading mechanism in edge computing-assisted UAV networks
CN102781072A (en) Network selecting method based on multi-attribute decision-making and group decision-making
Ni et al. Spectrum Allocation Based on Game Theory in Cognitive Radio Networks.
CN108990067B (en) Energy efficiency control method applied to ultra-dense heterogeneous network
CN110234155A (en) A kind of super-intensive network insertion selection method based on improved TOPSIS
Li et al. Computation offloading and service allocation in mobile edge computing
Kazmi et al. Distributed radio slice allocation in wireless network virtualization: Matching theory meets auctions
Fan et al. A collaborative task offloading scheme in D2D-assisted fog computing networks
Le et al. Joint cache allocation with incentive and user association in cloud radio access networks using hierarchical game
Gomez et al. Market-driven stochastic resource allocation framework for wireless network virtualization
Xu et al. Resource allocation based on double auction for cloud computing system
Ou et al. Two-Tier Resource Allocation for Multi-Tenant Network Slicing: A Federated Deep Reinforcement Learning Approach
Liu et al. Scalable traffic management for mobile cloud services in 5G networks
Zhu et al. Game-theory based buffer-space and transmission-rate allocations for optimal energy-efficiency over wireless virtual networks
Sterz et al. Multi-stakeholder service placement via iterative bargaining with incomplete information
CN109327494A (en) A kind of service quality adaptive excitation method towards multitask collaboration application
CN107027124B (en) Queuing theory-based cognitive radio network channel pricing and service selection joint optimization method
Zhang et al. Constrained VCG auction with multi-level channel valuations for spatial spectrum reuse in non-symmetric networks

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