CN105657750A - Network dynamic resource calculating method and device - Google Patents

Abstract

The embodiment of the invention discloses a network dynamic resource calculating method and device. According to the scheme provided by the embodiment of the invention, firstly, a cloud service provider and a client with a resource supply and demand relationship are converted into a multi-master multi-slave two-stage Stackelberg game model; according to the multi-master multi-slave two-stage Stackelberg game model, a cloud utility function and a client utility function are built; according to the multi-master multi-slave two-stage Stackelberg game model, an optimum cloud bandwidth price strategy and an optimum client bandwidth allocation strategy are computed for allocating dynamic resources of a network to the client. In application of the method provided by the embodiment of the invention, the dynamic network resource demands of the client are computed; the network performance is maximized; the dynamic network resource demands of the client are satisfied; and the user experience rank is promoted.

Description

The method of calculation of the dynamic resource of a kind of network and device

Technical field

The present invention relates to wireless communication technology field, in particular to method of calculation and the device of the dynamic resource of a kind of network.

Background technology

Along with the development of mobile Internet and intelligent terminal technology, the application program of mobile intelligent terminal also presents explosive growth. The computing power that present stage moves service end still cannot meet application program and optimal user experience requirements in mobile intelligent terminal. Mobile cloud computing (MCC, MobileCloudComputing), namely task intensive to calculated amount and energy consumption is unloaded to cloud computing center by mobile Internet by mobile intelligent terminal, by cloud computing center load capacity calculation function, to meet the calculation services demand of mobile intelligent terminal. Mobile cloud computing is that the concept based on cloud computing puts forward. Mobile cloud computing is mobile subscriber/terminal by mobile Internet as required, easily to obtain a kind of IT resource of required Infrastructure, platform, software or application etc. or the payment of Information services and using forestland in the way of expansion, and it is the new technique becoming and integrating mobile computing, mobile network and cloud computing.

Between application and network, communication quality usually includes time delay, shake, the bandwidth of network, misses the indexs such as code, and these indexs are also the important indicators weighing communication quality over-all properties. Traditional QoS of customer (QoS) lays particular emphasis on the measurement to communication quality and statistics, objectively react the performance of equipment, its starting point is from network perspective to estimate the impression of user, not being that the subjective demand according to user carrys out allocation of network resources, therefore QoS is difficult to embody the satisfactory degree of user. And user experience quality (QoE, QualityofExperience) is felt as value core with user, directly carry out configuration network resource by the subjective demand of user feedback, relevant to user's behavior. QoE can be understood as Consumer's Experience or user awareness intuitively, namely goes out the quality sent and weigh communication quality from the angle of user. Based on this, user experience quality (QoE) becomes the focus of industry competition and development gradually, and QoE becomes a kind of important indicator evaluating radio communication overall system performance, and becomes one of leading force of radio resource management techniques innovation gradually.

In mobile cloud computing, the current research of the QoE based on MCC mainly lays particular emphasis on how to set up QoE model, thus the resource management based on QoE is also a matter of opening, secondly, due to the characteristic of user's end and wireless network, such as the defect of user's end, the low bandwidth of network and unstable etc., make resource management problem more complicated, need more fine granularity in conjunction with the resource in high in the clouds and user's end, and, in actual applications, the dynamic resource of mobile system for cloud computing often multiple cloud service provider to the relation of multiple user's end, there is no an effectively method in the prior art to the dynamic resource requirement that calculates the network of user's end, the method of calculation needing a better dynamic resource of network badly carry out maximization network performance, meet the dynamic resource requirement of network of user's end, strengthen the experience value of user.

Summary of the invention

The embodiment of the invention discloses the method for calculation of the dynamic resource of a kind of network and device, it is applied to mobile cloud computing environment, with maximization network performance, strengthen the experience value of user.

In order to achieve the above object, embodiments provide the method for calculation of the dynamic resource of a kind of network, comprise step:

According to the resource supplydemand relationship between cloud service provider and user's end, set up how main many from the Staenberg betting model of two benches;

According to how main many from the Staenberg betting model of two benches, set up user's end utility function and high in the clouds utility function;

According to described much much more main from Staenberg betting model calculate high in the clouds optimum bandwidth price strategy and user's end optimal bandwidth allocation strategy, for being user's dynamic resource of end distribution network.

Preferably, described many masters are many from the Staenberg betting model of two benches is:

G=(p, b, U_T(p,b),U_C(p, b));

Wherein, U_T(p, b) be much much more main from the Staenberg betting model of two benches user's end utility function, U_C(p, b) be much much more main from the Staenberg betting model of two benches high in the clouds utility function, p={p₁,p₂,...,p_mIt it is high in the clouds price strategy; B={b₁,b₂,...,b_IIt it is user's end band-width tactics.

Preferably, described user's end utility function is:

$U_{Ti}(p,b_i)=R_i\left(\underset{j=1}{\overset{m}{\Σ}}{b}_{ij}\right)-\underset{j=1}{\overset{m}{\Σ}}{C}_j^B(b_{ij},p_j)-\underset{j=1}{\overset{m}{\Σ}}{C}_j^D\left(b_{ij}\right);$

Wherein, m is the total number of cloud service provider, b_ijIt is the bandwidth that i-th user's end obtains from jth cloud service provider,It is the total bankwidth that i-th user's end obtains from cloud service provider,It is i-th user's end distributionThe income that bandwidth obtains,For the expense paid to jth cloud service provider,For the time delay expense that cloud service provider j gives i-th user's end band next, wherein, $R_i\left(\underset{j=1}{\overset{m}{\Σ}}{b}_{ij}\right)=B_i{\mathrm{\η}}_i{r}_i{M}_i,$ In formula $B_i=\underset{j=1}{\overset{m}{\Σ}}{b}_{ij}{B}_{tot}/\underset{j=1}{\overset{m}{\Σ}}\underset{i=1}{\overset{I}{\Σ}}{b}_{ij},$ I is the total number of user's end, B_totFor network total bankwidth, ��_iIt is the spectrum effectiveness of i-th user's end, r_iIt is i-th user's end unit transfer rate income, M_iIt is the satisfaction of users of i-th user's termination by MCC service;In formula, p_jFor the unit bandwidth price that jth cloud service provider is formulated,In formula, c_iIt is the unit time delay overhead value of i-th user's end, d_iIt is that i-th user's end obtains total time lag when MCC serves;

Described high in the clouds utility function is:

U_Cj=Q_j*p_j;

Wherein, Q_jFor the total bankwidth to all user's ends required by cloud service provider j, p_jFor the unit bandwidth price that jth cloud service provider is formulated;

Described user's end utility function at least comprises the income based on bandwidth and cost; Described cost comprises the expense to cloud service provider payment and time delay expense.

Preferably, described according to described much much more main from Staenberg betting model calculate high in the clouds optimum bandwidth price strategy and user's end optimal bandwidth allocation strategy, comprising:

The cloud service provider in high in the clouds, based on the state of loading of current network, calculates high in the clouds price strategy p={p₁,p₂,...,p_m, and it is broadcast to all user's ends that high in the clouds covers;

User's termination receives the high in the clouds price strategy p={p of different cloud service provider broadcast₁,p₂,...,p_mAfter, calculate user end optimum bandwidth demand strategy b according to the price strategy of current network^*(p), and by described user end optimum bandwidth demand strategy b^*P () feeds back to all cloud service provider, wherein, and described user end optimum bandwidth demand strategy b^*P () represents:

$b^{*}\left(p\right)=\frac{r_i{\mathrm{\η}}_i{B}_{tot}{M}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b}_{nj}}{{\left({\Σ}_{i=1}^I{\Σ}_{j=1}^m{b}_{ij}\right)}^2}-p+\frac{c_i{T}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b}_{nj}}{{\mathrm{\η}}_i{\left({\Σ}_{j=1}^m{b}_{ij}\right)}^2};$

In formulaThe all user's terminations represented except i-th user's end receive the total bankwidth of all cloud service provider,Representing that all user's terminations receive the total bankwidth of all cloud service provider, p is the price strategy of the corresponding cloud service provider that i-th user's end receives;

High in the clouds cloud service provider is according to the user end optimum bandwidth demand strategy b of all client feeds back^*P () calculates the high in the clouds optimum bandwidth price strategy p meeting user's end band width demand^*, and it is broadcast to all user's ends, and wherein, described high in the clouds optimum bandwidth price strategy p^*Expression is:

$p^{*}=\frac{r_i{\mathrm{\η}}_i{B}_{tot}{M}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b^{*}}_{nj}}{{\left({\Σ}_{i=1}^I{\Σ}_{j=1}^m{b^{*}}_{ij}\right)}^2}+\frac{c_i{T}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b^{*}}_{nj}}{{\mathrm{\η}}_i{\left({\Σ}_{j=1}^m{b^{*}}_{ij}\right)}^2};$

In formulaRepresent, the optimum bandwidth demand summation that all user's ends met except i-th user's end receive from all cloud service provider,For the optimum bandwidth demand summation that all user's ends receive from all cloud service provider,It it is the optimum bandwidth demand summation that i-th user's end receives from all cloud service provider;

User's end is according to the high in the clouds optimum bandwidth price strategy p of all cloud service provider broadcast received^*Recalculate user end optimal bandwidth allocation strategy b^*(p^*), wherein, described user end optimal bandwidth allocation strategy b^*(p^*) represent be:

$b^{*}\left(p^{*}\right)=\frac{r_i{\mathrm{\η}}_i{B}_{tot}{M}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b}_{nj}}{{\left({\Σ}_{i=1}^I{\Σ}_{j=1}^m{b}_{ij}\right)}^2}-p^{*}+\frac{c_i{T}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b}_{nj}}{{\mathrm{\η}}_i{\left({\Σ}_{j=1}^m{b}_{ij}\right)}^2};$

In formula, p^*It it is the high in the clouds best price strategy of the corresponding cloud service provider that i-th user's end receives.

In order to achieve the above object, embodiments provide the calculating device of the dynamic resource of a kind of network, comprising:

Many masters are many sets up module from Staenberg betting model, for according to the resource supplydemand relationship between cloud service provider and user's end, setting up how main how from the Staenberg betting model of two benches;

Utility function sets up module, for according to how main many from the Staenberg betting model of two benches, setting up user's end utility function and high in the clouds utility function;

Optimal strategy calculates module, for according to described much much more main from Staenberg betting model calculate high in the clouds optimum bandwidth price strategy and user's end optimal bandwidth allocation strategy, for being user's dynamic resource of end distribution network.

Preferably, setting up module from Staenberg betting model more than described many masters is:

G=(p, b, U_T(p,b),U_C(p, b));

Wherein, U_T(p, b) be much much more main from the Staenberg betting model of two benches user's end utility function, U_C(p, b) be much much more main from the Staenberg betting model of two benches high in the clouds utility function, p={p₁,p₂,...,p_mIt it is high in the clouds price strategy; B={b₁,b₂,...,b_IIt it is user's end band-width tactics.

Preferably, described user's end utility function is:

$U_{Ti}(p,b_i)=R_i\left(\underset{j=1}{\overset{m}{\Σ}}{b}_{ij}\right)-\underset{j=1}{\overset{m}{\Σ}}{C}_j^B(b_{ij},p_j)-\underset{j=1}{\overset{m}{\Σ}}{C}_j^D\left(b_{ij}\right);$

Wherein, m is the total number of cloud service provider, b_ijIt is the bandwidth that i-th user's end obtains from jth cloud service provider,It is the total bankwidth that i-th user's end obtains from cloud service provider,It is i-th user's end distributionThe income that bandwidth obtains,For the expense paid to jth cloud service provider,For the time delay expense that cloud service provider j gives i-th user's end band next, wherein, $R_i\left(\underset{j=1}{\overset{m}{\Σ}}{b}_{ij}\right)=B_i{\mathrm{\η}}_i{r}_i{M}_i,$ In formula $B_i=\underset{j=1}{\overset{m}{\Σ}}{b}_{ij}{B}_{tot}/\underset{j=1}{\overset{m}{\Σ}}\underset{i=1}{\overset{I}{\Σ}}{b}_{ij},$ I is the total number of user's end, B_totFor network total bankwidth, ��_iIt is the spectrum effectiveness of i-th user's end, r_iIt is i-th user's end unit transfer rate income, M_iIt is the satisfaction of users of i-th user's termination by MCC service;In formula, p_jFor the unit bandwidth price that jth cloud service provider is formulated,In formula, c_iIt is the unit time delay overhead value of i-th user's end, d_iIt is that i-th user's end obtains total time lag when MCC serves;

Described high in the clouds utility function is:

U_Cj=Q_j*p_j;

Wherein, Q_jFor the total bankwidth to all user's ends required by cloud service provider j, p_jFor the unit bandwidth price that jth cloud service provider is formulated;

Described user's end utility function at least comprises the income based on bandwidth and cost; Described cost comprises the expense to cloud service provider payment and time delay expense.

Preferably, described optimal strategy calculates module, comprising:

High in the clouds price strategy calculates and broadcast submodule block, for the state of loading according to current network, calculates high in the clouds price strategy p={p₁,p₂,...,p_m, and it is broadcast to all user's ends that high in the clouds covers;

User's end optimum bandwidth demand strategy calculates and feedback submodule block, for receiving the high in the clouds price strategy p={p of different cloud service provider broadcast in user's termination₁,p₂,...,p_mAfter, calculate user end optimum bandwidth demand strategy b according to the price strategy of current network^*(p), and by described user end optimum bandwidth demand strategy b^*P () feeds back to all cloud service provider, wherein, and described user end optimum bandwidth demand strategy b^*P () represents:

$b^{*}\left(p\right)=\frac{r_i{\mathrm{\η}}_i{B}_{tot}{M}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b}_{nj}}{{\left({\Σ}_{i=1}^I{\Σ}_{j=1}^m{b}_{ij}\right)}^2}-p+\frac{c_i{T}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b}_{nj}}{{\mathrm{\η}}_i{\left({\Σ}_{j=1}^m{b}_{ij}\right)}^2};$

In formulaThe all user's terminations represented except i-th user's end receive the total bankwidth of all cloud service provider,Representing that all user's terminations receive the total bankwidth of all cloud service provider, p is the price strategy of the corresponding cloud service provider that i-th user's end receives;

High in the clouds optimum bandwidth price strategy calculates and broadcast submodule block, for the user end optimum bandwidth demand strategy b according to all client feeds back^*P (), cloud service provider formulates the high in the clouds optimum bandwidth price strategy p meeting user's end band width demand^*, and it is broadcast to all user's ends, and wherein, described high in the clouds optimum bandwidth price strategy p^*Expression is:

$p^{*}=\frac{r_i{\mathrm{\η}}_i{B}_{tot}{M}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b^{*}}_{nj}}{{\left({\Σ}_{i=1}^I{\Σ}_{j=1}^m{b^{*}}_{ij}\right)}^2}+\frac{c_i{T}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b^{*}}_{nj}}{{\mathrm{\η}}_i{\left({\Σ}_{j=1}^m{b^{*}}_{ij}\right)}^2};$

In formulaRepresent, the optimum bandwidth demand summation that all user's ends met except i-th user's end receive from all cloud service provider,For the optimum bandwidth demand summation that all user's ends receive from all cloud service provider,It it is the optimum bandwidth demand summation that i-th user's end receives from all cloud service provider;

User's end optimal bandwidth allocation policy calculation submodule block, for the high in the clouds optimum bandwidth price strategy p according to all cloud service provider broadcast received^*After, user's end recalculates user end optimal bandwidth allocation strategy b^*(p^*), wherein, described user end optimal bandwidth allocation strategy b^*(p^*) represent be:

$b^{*}\left(p^{*}\right)=\frac{r_i{\mathrm{\η}}_i{B}_{tot}{M}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b}_{nj}}{{\left({\Σ}_{i=1}^I{\Σ}_{j=1}^m{b}_{ij}\right)}^2}-p^{*}+\frac{c_i{T}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b}_{nj}}{{\mathrm{\η}}_i{\left({\Σ}_{j=1}^m{b}_{ij}\right)}^2};$

In formula, p^*It it is the high in the clouds best price strategy of the corresponding cloud service provider that i-th user's end receives. In the scheme that the embodiment of the present invention provides, first the cloud service provider and user's end with resource supplydemand relationship are converted into how main many from two benches Staenberg betting model; According to how main many from two benches Staenberg betting model, set up high in the clouds utility function and user's end utility function; According to described much much more main from Staenberg betting model calculate high in the clouds optimum bandwidth price strategy and user's end optimal bandwidth allocation strategy, for the dynamic resource for user's end distribution network. The application method that provides of the embodiment of the present invention, to the dynamic resource requirement that calculates the network of user's end, maximises network performance, meets the dynamic resource requirement of network of user's end, strengthen the experience value of user.

Accompanying drawing explanation

In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, it is briefly described to the accompanying drawing used required in embodiment or description of the prior art below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, it is also possible to obtain other accompanying drawing according to these accompanying drawings.

The schematic flow sheet of the method for calculation of the dynamic resource of a kind of network that Fig. 1 provides for the embodiment of the present invention;

The schematic flow sheet of the method for calculation of the dynamic resource of another kind of network that Fig. 2 provides for the embodiment of the present invention;

The structural representation of the calculating device of the dynamic resource of a kind of network that Fig. 3 provides for the embodiment of the present invention;

The structural representation of the calculating device of the dynamic resource of another kind of network that Fig. 4 provides for the embodiment of the present invention;

Fig. 5 is the mobile cloud network system model schematic of the embodiment of the present invention;

The Staenberg game that Fig. 6 is the embodiment of the present invention solves process schematic diagram;

Fig. 7 be the embodiment of the present invention under different bandwidth prices, the utility value variation diagram of cloud service provider;

Fig. 8 is the embodiment of the present invention under being worth situation at different MOS, and the result that user's end adopts the method for calculation of two kinds of dynamic resources of network to calculate carries out the user's end utility value variation diagram distributed.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is only the present invention's part embodiment, instead of whole embodiments. Based on the embodiment in the present invention, those of ordinary skill in the art, not making other embodiments all obtained under creative work prerequisite, belong to the scope of protection of the invention.

Embodiments provide method of calculation and the device of the dynamic resource of a kind of network, first the cloud service provider and user's end with resource supplydemand relationship are converted into how main many from two benches Staenberg betting model; According to how main many from two benches Staenberg betting model, set up high in the clouds utility function and user's end utility function; According to described much much more main from Staenberg betting model calculate high in the clouds optimum bandwidth price strategy and user's end optimal bandwidth allocation strategy, for the dynamic resource for user's end distribution network. The application method that provides of the embodiment of the present invention, to the dynamic resource requirement that calculates the network of user's end, maximises network performance, meets the dynamic resource requirement of network of user's end, strengthen the experience value of user.

Below by specific embodiment, the present invention is described in detail.

The schematic flow sheet of the method for calculation of the dynamic resource of a kind of network that Fig. 1 provides for the embodiment of the present invention, comprises the steps:

S101: according to the resource supplydemand relationship between cloud service provider and user's end, sets up how main many from the Staenberg betting model of two benches;

In mobile system for cloud computing, dynamically the supply and demand of resource relate to cloud service provider and user's end usually, and multiple user's end competition bandwidth resources obtain maximum bandwidth resources and maximumization self benefits; Multiple cloud service provider is by constantly regulating own bandwidth price simultaneously, the bandwidth demand affecting mobile subscriber is in the hope of maximumization self benefits, by the resource supplydemand relationship between this kind of cloud service provider and user's end, it is abstracted into how main many from the Staenberg betting model of two benches;

It will be understood by those skilled in the art that, when setting up Staenberg betting model, usually, the game player taken action at first is called leader by us, and claim to follow the action of leader and take the player of countermeasure to be follower, and such a game being made up of player leader and follower is claimed to be Staenberg game. The prediction of follower's countermeasure can be formulated action strategy according to own situation and oneself by leader in gambling process in advance. After the action observing leader, follower can formulate countermeasure according to own situation and observation result thereof and respond.

In the embodiment of the present invention, leader is high in the clouds, and follower is user's end, more than main more than can setting up from the Staenberg betting model of two benches is:

G=(p, b, U_T(p,b),U_C(p, b));

Wherein, U_T(p, b) be much much more main from the Staenberg betting model of two benches user's end utility function, U_C(p, b) be much much more main from the Staenberg betting model of two benches high in the clouds utility function, p={p₁,p₂,...,p_mIt it is high in the clouds price strategy; B={b₁,b₂,...,b_IIt it is user's end band-width tactics.

First high in the clouds cloud service provider formulates high in the clouds price strategy p={p according to networking load₁,p₂,...,p_m, and by this high in the clouds price strategy p={p₁,p₂,...,p_mIt is broadcast to all user's ends that high in the clouds covers, when user's termination receives the high in the clouds price strategy p={p of high in the clouds cloud service provider broadcast₁,p₂,...,p_mAfter, user's end by according to do receive strategy formulation to should strategy user end band-width tactics b={b₁,b₂,...,b_I}��

S102: according to how main many from the Staenberg betting model of two benches, set up user's end utility function and high in the clouds utility function.

It will be understood by those skilled in the art that, need when carrying out Staenberg betting model first all to need first to declare in gambling process who be leader who be follower, from S101, defining in the present invention network resource is by leader, i.e. high in the clouds, definition follower is user's end, and the user's end utility function as follower user's end is:

$U_{Ti}(p,b_i)=R_i\left(\underset{j=1}{\overset{m}{\Σ}}{b}_{ij}\right)-\underset{j=1}{\overset{m}{\Σ}}{C}_j^B(b_{ij},p_j)-\underset{j=1}{\overset{m}{\Σ}}{C}_j^D\left(b_{ij}\right);$

Wherein, m is the total number of cloud service provider; b_ijIt is the bandwidth that i-th user's end obtains from jth cloud service provider,It it is the total bankwidth that i-th user's end obtains from cloud service provider;It is i-th user's end distributionThe income that bandwidth obtains,For the expense paid to jth cloud service provider,For the time delay expense that cloud service provider j gives i-th user's end band next.

User's end utility function can be understood as the gross earnings of user's end, and this user's end utility function comprises the income based on bandwidth and cost; Described cost comprises the expense to cloud service provider payment and time delay expense, and the selection cloud service provider of each user's independence formulates user's end band width demand strategy, and the bandwidth demand of user's end is dynamic change value.

Wherein, the revenue function of user is as follows:

$R_i\left(\underset{j=1}{\overset{m}{\Σ}}{b}_{ij}\right)=B_i{\mathrm{\η}}_i{r}_i{M}_i;$

In formula,I is the total number of user's end, B_totFor network total bankwidth; ��_iIt it is the spectrum effectiveness of i-th user's end; r_iIt is i-th user's end unit transfer rate income; M_iBeing the satisfaction of users of i-th user's termination by MCC service, this satisfaction of users MOS (MeanOpinionStore) value is weighed under normal circumstances.

��_iIt is the spectrum effectiveness of i-th user's end:

${\mathrm{\η}}_i=\frac{R_c}{W}=\frac{{\mathrm{Wlog}}_2(1+\frac{P}{N_{0}W})}{W}={\log }_2(1+\mathrm{\γ}),$ Wherein, P, W and N₀It is respectively bandwidth, through-put power and noise power spectral density;For the maximum rate of the channel under Gaussian white noise channel; �� is the signal to noise ratio of user's end;

M_iBeing the satisfaction of users accepting MCC service of i-th user's end, this satisfaction of users MOS value is weighed. Base station provides three kinds of business to user's end, is respectively audio frequency stream, data stream and video flowing;

Audio gauge stream MOS formula is:

$M_i=0.5*{\log }_2\left(\frac{1}{1+60*PEP}\right)+4.3;$ Wherein PEP is packet error probability;

The MOS formula weighing data stream is:

Wherein S is file size, T_cFor the throughput capacity of user's end;

The MOS formula weighing video flowing is:

Wherein k is about video shifter factor, and �� is sdi video character factor, R_aFor the transfer rate of user's end, u is code rate.

The allocated bandwidth spending function of user's end is:

$C_j^B(b_{ij},p_j)=p_j{b}_{ij};$

The time delay expense function of user's end is:

$C_j^D\left(h_{ij}\right)=c_i{d}_i;$

In upper formula, c_iIt is the unit time delay overhead value of i-th user's end, d_iIt is that i-th user's end obtains total time lag when MCC serves,Wherein T_iIt it is the transport delay of i-th user's end.

Then, user's end utility function is:

$U_{Ti}(p,b_i)=\frac{B_{tot}{\mathrm{\η}}_i{r}_i{M}_i{\Σ}_{j=1}^m{b}_{ij}}{{\Σ}_{i=1}^I{\Σ}_{j=1}^m{b}_{ij}}-\underset{j=1}{\overset{m}{\Σ}}{p}_j{b}_{ij}-c_i\frac{B_{tot}{T}_i{\Σ}_{i=1}^I{\Σ}_{j=1}^m{b}_{ij}}{\underset{j=1}{\overset{m}{\Σ}}{\mathrm{\η}}_i{b}_{ij}{B}_{tot}}.$

High in the clouds utility function as leader high in the clouds is:

U_Cj=Q_j*p_j;

Wherein, Q_jFor the total bankwidth to all user's ends required by cloud service provider j, p_jFor the unit bandwidth price that jth cloud service provider is formulated.

In the process of the Dynamic Resource Allocation for Multimedia of mobile system for cloud computing, it is exactly the demand to the dynamic resource of system for cloud computing concerning user's end, it is possible to be following formula by the dynamic resource requirement question variation of user's end:

$\underset{b_i^{*}}{max}{U}_{Ti}(p,b_i),$

$s.t.M_i\≥M_{{\min }^i},\∀i\∈(1,I),$

$\underset{i=1}{\overset{I}{\Σ}}{b}_{ij}\≤Q_j,\∀j\∈(1,m),$

$b_{ij}\≥0,\∀i\∈(1,I),\∀j\∈(1,m),$

Above-mentioned formula shows, takes the maximum value of family end utility function, and the s.tMi of i-th user's end is greater than the minimum flow rate equaling i-th user's end, the total bankwidth that i-th user's end obtains from cloud service providerIt is less than or equal to the total bankwidth Q of all user's ends that cloud service provider provides_j, the bandwidth b that i-th user's end obtains from jth cloud service provider_ijIt is more than or equal to 0.

Can understand, be exactly resource provision for leader high in the clouds, it is possible to be following formula by the resource provision question variation in high in the clouds:

$\underset{p_j^{*}}{max}{U}_{c_j}$

Above-mentioned formula shows, gets the maximum value of high in the clouds utility function, the s.t.p of jth cloud service provider_jIt is greater than and equals 0.

By to the conversion in high in the clouds and the resource problem of user's end, user's end utility function and high in the clouds utility function being substituted into how main many from Staenberg betting model.

S103: according to described much much more main from Staenberg betting model calculate high in the clouds optimum bandwidth price strategy and user's end optimal bandwidth allocation strategy, for the dynamic resource for user's end distribution network.

Being understood that, in Staenberg betting model, be owner and the decision maker of network resource as the high in the clouds of leader, its strategy will affect the income of follower user's end. In whole competition gambling process in the embodiment of the present invention, the high in the clouds price strategy of self is formulated in high in the clouds first, and announce to user's end, calculate through competition game repeatedly, final acquisition high in the clouds optimum bandwidth price strategy and user's end optimal bandwidth allocation strategy, for the dynamic resource for user's end distribution network.

The application embodiment of the present invention, is first converted into how main many from two benches Staenberg betting model by the cloud service provider and user's end with resource supplydemand relationship; According to how main many from two benches Staenberg betting model, set up high in the clouds utility function and user's end utility function; According to described much much more main from Staenberg betting model calculate high in the clouds optimum bandwidth price strategy and user's end optimal bandwidth allocation strategy, for the dynamic resource for user's end distribution network. The application method that provides of the embodiment of the present invention, to the dynamic resource requirement that calculates the network of user's end, maximises network performance, meets the dynamic resource requirement of network of user's end, strengthen the experience value of user.

The schematic flow sheet of the method for calculation of the dynamic resource of another kind of network that Fig. 2 provides for the embodiment of the present invention, compared with embodiment illustrated in fig. 1, step S103 embodiment illustrated in fig. 1 is realized by the S1031 to S1034 in the present embodiment, specifically comprises:

S1031: the cloud service provider in high in the clouds, based on the state of loading of current network, calculates high in the clouds and goes out high in the clouds price strategy p={p₁,p₂,...,p_m, and it is broadcast to all user's ends that high in the clouds covers.

S1032: user's termination receives the high in the clouds price strategy p={p of different cloud service provider broadcast₁,p₂,...,p_mAfter, calculate user end optimum bandwidth demand strategy b according to the price strategy of current network^*(p), and by described user end optimum bandwidth demand strategy b^*P () feeds back to all cloud service provider, wherein, and described user end optimum bandwidth demand strategy b^*P () represents:

$b^{*}\left(p\right)=\frac{r_i{\mathrm{\η}}_i{B}_{tot}{M}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b}_{nj}}{{\left({\Σ}_{i=1}^I{\Σ}_{j=1}^m{b}_{ij}\right)}^2}-p+\frac{c_i{T}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b}_{nj}}{{\mathrm{\η}}_i{\left({\Σ}_{j=1}^m{b}_{ij}\right)}^2};$

In formulaThe all user's terminations represented except i-th user's end receive the total bankwidth of all cloud service provider,Representing that all user's terminations receive the total bankwidth of all cloud service provider, p is the price strategy of the corresponding cloud service provider that i-th user's end receives.

In actual gambling process, when user's termination receives the price strategy p={p of different cloud service provider issue₁,p₂,...,p_mAfter, calculate user end optimum bandwidth demand strategy b according to the price strategy of current network^*(p), wherein,Concrete, it is assumed that when moment t, cloud service provider broadcasts its price strategy p (t), at this moment user's end according to the bandwidth demand of self dynamic adjustments band-width tactics so that obtain optimal strategy, it is assumed that the time variable of user's end is ��, thenFrom time instant �� to time instant ��+1,v_iIt it is the bandwidth iteration step length of i-th user's end. User's end passes through iterative formulaSolve, calculate user end optimum bandwidth demand strategy b^*(p), and fed back to high in the clouds.

S1033: high in the clouds cloud service provider is according to the user end optimum bandwidth demand strategy b of all client feeds back^*P () calculates the high in the clouds optimum bandwidth price strategy p meeting user's end band width demand^*, and it is broadcast to all user's ends, and wherein, described high in the clouds optimum bandwidth price strategy p^*Expression is:

$p^{*}=\frac{r_i{\mathrm{\η}}_i{B}_{tot}{M}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b^{*}}_{nj}}{{\left({\Σ}_{i=1}^I{\Σ}_{j=1}^m{b^{*}}_{ij}\right)}^2}+\frac{c_i{T}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b^{*}}_{nj}}{{\mathrm{\η}}_i{\left({\Σ}_{j=1}^m{b^{*}}_{ij}\right)}^2};$

In formulaRepresent, the optimum bandwidth demand summation that all user's ends met except i-th user's end receive from all cloud service provider,For the optimum bandwidth demand summation that all user's ends receive from all cloud service provider,It it is the optimum bandwidth demand summation that i-th user's end receives from all cloud service provider.

In the gambling process of reality, high in the clouds cloud service provider is according to the user end optimum bandwidth demand strategy b receiving client feeds back^*P, after (), cloud service provider is by iterative formulaSolve, calculate high in the clouds optimum bandwidth price strategy p^*, the w in formula_jFor the price iteration step length of jth cloud service provider.

S1034: user's end is according to the high in the clouds optimum bandwidth price strategy p of all cloud service provider broadcast received^*Recalculate user end optimal bandwidth allocation strategy b^*(p^*), wherein, described user end optimal bandwidth allocation strategy b^*(p^*) represent be:

$b^{*}\left(p^{*}\right)=\frac{r_i{\mathrm{\η}}_i{B}_{tot}{M}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b}_{nj}}{{\left({\Σ}_{i=1}^I{\Σ}_{j=1}^m{b}_{ij}\right)}^2}-p^{*}+\frac{c_i{T}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b}_{nj}}{{\mathrm{\η}}_i{\left({\Σ}_{j=1}^m{b}_{ij}\right)}^2};$

In formula, p^*It it is the high in the clouds best price strategy of the corresponding cloud service provider that i-th user's end receives.

It is not difficult to obtain the maximum utility function value of user's end to represent by above-mentioned each formula be:

$U_{Ti}(p^{*},{b^{*}}_i)=\frac{B_{tot}{\mathrm{\η}}_i{r}_i{M}_i{\Σ}_{j=1}^m{b^{*}}_{ij}}{{\Σ}_{i=1}^I{\Σ}_{j=1}^m{b^{*}}_{ij}}-\underset{j=1}{\overset{m}{\Σ}}{p^{*}}_j{b^{*}}_{ij}-c_i\frac{B_{tot}{T}_i{\Σ}_{i=1}^I{\Σ}_{j=1}^m{b^{*}}_{ij}}{\underset{j=1}{\overset{m}{\Σ}}{\mathrm{\η}}_i{b^{*}}_{ij}{B}_{tot}};$

The process that solves of its dynamic game can intuitively with reference to the schematic diagram in figure 6.

The application embodiment of the present invention, by to described much much more main from Staenberg betting model competition gambling process repeatedly and iterative computation repeatedly, finally calculate high in the clouds optimum bandwidth price strategy and user's end optimum bandwidth strategy of the dynamic resource requirement of the network meeting user's end, to maximise network performance.

The structural representation of the calculating device of the dynamic resource of a kind of network that Fig. 3 provides for the embodiment of the present invention, corresponding with Fig. 1 method flow diagram, described device comprises:

Many masters are many sets up module 101 from Staenberg betting model, for according to the resource supplydemand relationship between cloud service provider and user's end, setting up how main how from the Staenberg betting model of two benches.

Utility function sets up module 102, for according to how main many from the Staenberg betting model of two benches, setting up user's end utility function and high in the clouds utility function.

Optimal strategy obtains module 103, for according to described much much more main from Staenberg betting model calculate high in the clouds optimum bandwidth price strategy and user's end optimal bandwidth allocation strategy, for the dynamic resource for user's end distribution network.

The application embodiment of the present invention, is first converted into how main many from two benches Staenberg betting model by the cloud service provider and user's end with resource supplydemand relationship; According to how main many from two benches Staenberg betting model, set up high in the clouds utility function and user's end utility function; According to described much much more main from Staenberg betting model calculate high in the clouds optimum bandwidth price strategy and user's end optimal bandwidth allocation strategy, for the dynamic resource for user's end distribution network. The application method that provides of the embodiment of the present invention, to the dynamic resource requirement that calculates the network of user's end, maximises network performance, meets the dynamic resource requirement of network of user's end, strengthen the experience value of user.

The structural representation of the calculating device of the dynamic resource of another kind of network that Fig. 4 provides for the embodiment of the present invention, compared with embodiment illustrated in fig. 3, structure 103 embodiment illustrated in fig. 3 is realized by the structure 1031 to 1034 in the present embodiment, specifically comprises:

High in the clouds price strategy calculates and broadcast submodule block 1031, for the state of loading according to current network, calculates high in the clouds price strategy p={p₁,p₂,...,p_m, and it is broadcast to all user's ends that high in the clouds covers.

User's end optimum bandwidth demand strategy calculates and feedback submodule block 1032, for receiving the high in the clouds price strategy p={p of different cloud service provider broadcast in user's termination₁,p₂,...,p_mAfter, calculate user end optimum bandwidth demand strategy b according to the price strategy of current network^*(p), and by described user end optimum bandwidth demand strategy b^*P () feeds back to all cloud service provider, wherein, and described user end optimum bandwidth demand strategy b^*P () represents:

$b^{*}\left(p\right)=\frac{r_i{\mathrm{\η}}_i{B}_{tot}{M}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b}_{nj}}{{\left({\Σ}_{i=1}^I{\Σ}_{j=1}^m{b}_{ij}\right)}^2}-p+\frac{c_i{T}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b}_{nj}}{{\mathrm{\η}}_i{\left({\Σ}_{j=1}^m{b}_{ij}\right)}^2};$

In formulaThe all user's terminations represented except i-th user's end receive the total bankwidth of all cloud service provider,Representing that all user's terminations receive the total bankwidth of all cloud service provider, p is the price strategy of the corresponding cloud service provider that i-th user's end receives.

The optimum wide lattice policy calculation of valence band in high in the clouds and broadcast submodule block 1033, for the user end optimum bandwidth demand strategy b according to all client feeds back^*P (), cloud service provider formulates the high in the clouds optimum bandwidth price strategy p meeting user's end band width demand^*, and it is broadcast to all user's ends, and wherein, described high in the clouds optimum bandwidth price strategy p^*Expression is:

$p^{*}=\frac{r_i{\mathrm{\η}}_i{B}_{tot}{M}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b^{*}}_{nj}}{{\left({\Σ}_{i=1}^I{\Σ}_{j=1}^m{b^{*}}_{ij}\right)}^2}+\frac{c_i{T}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b^{*}}_{nj}}{{\mathrm{\η}}_i{\left({\Σ}_{j=1}^m{b^{*}}_{ij}\right)}^2};$

In formulaRepresent, the optimum bandwidth demand summation that all user's ends met except i-th user's end receive from all cloud service provider,For the optimum bandwidth demand summation that all user's ends receive from all cloud service provider,It it is the optimum bandwidth demand summation that i-th user's end receives from all cloud service provider.

User's end optimal bandwidth allocation policy calculation submodule block 1034, for the high in the clouds optimum bandwidth price strategy p according to all cloud service provider broadcast received^*After, user's end recalculates user end optimal bandwidth allocation strategy b^*(p^*), wherein, described user end optimal bandwidth allocation strategy b^*(p^*) represent be:

$b^{*}\left(p^{*}\right)=\frac{r_i{\mathrm{\η}}_i{B}_{tot}{M}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b}_{nj}}{{\left({\Σ}_{i=1}^I{\Σ}_{j=1}^m{b}_{ij}\right)}^2}-p^{*}+\frac{c_i{T}_i{\Σ}_{\∀n\≠i}{\Σ}_{j=1}^m{b}_{nj}}{{\mathrm{\η}}_i{\left({\Σ}_{j=1}^m{b}_{ij}\right)}^2};$

In formula, p^*It it is the high in the clouds best price strategy of the corresponding cloud service provider that i-th user's end receives.

It will be understood by those skilled in the art that, needing when carrying out bandwidth resource allocation to know that who is the leader arranging whole network resource based on mobile cloud network, who is the person of sharing of network resource, based on this, embodiments provide a kind of mobile cloud network system model, see Fig. 7, this model comprises cloud layer, i.e. high in the clouds, network layer and equipment layer, wherein high in the clouds is exactly the domination person of the dynamic resource of network, and network layer is articulamentum, and equipment layer is also exactly user's end layer.

The application embodiment of the present invention, by multipair many dynamic Resource Calculations of network, maximising network performance, meet the dynamic resource requirement of network of user's end, strengthens the experience value of user.

Fig. 7 be the embodiment of the present invention under different bandwidth prices, the utility value variation diagram of cloud service provider;

In figure, X-axis represents the bandwidth price of cloud service provider 1, Y-axis represents the bandwidth price of cloud service provider 2, Z axle represents high in the clouds utility value, figure can find intuitively, when the price of cloud service provider 1 is 5, when the price of cloud service provider 2 is 10, the utility value of whole cloud service provider reaches maximum.

Fig. 8 is the embodiment of the present invention under being worth situation at different MOS, and the result that user's end adopts the method for calculation of two kinds of dynamic resources of network to calculate carries out the user's end utility value variation diagram distributed.

In whole computation process, it is worth increasing along with MOS, the utility value of user's end also increases, two straight lines in figure represent two kinds of different Resource Allocation Formulas, wherein, the result that the numerical procedure of the dynamic resource of network that a straight line above provides for adopting the embodiment of the present invention calculates carries out resources allocation, the utility value change curve of user's end; The result that a straight line below is the average computation computation schemes of the dynamic resource of network adopting prior art carries out resources allocation, the utility value change curve of user, from then on schemes obviously to see that dynamic resource is distributed the result being obviously better than adopting average computation to calculate and dynamic resource distributed by the result adopting out the present embodiment to calculate.

It should be noted that, herein, the such as relational terms of first and second grades and so on is only used for separating an entity or operation with another entity or operational zone, and not necessarily requires or imply to there is any this kind of actual relation or sequentially between these entities or operation. And, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, so that comprise the process of a series of key element, method, article or equipment not only comprise those key elements, but also comprise other key elements clearly do not listed, or also comprise the key element intrinsic for this kind of process, method, article or equipment. When not more restrictions, the key element limited by statement " comprising ... ", and be not precluded within process, method, article or the equipment comprising described key element and also there is other identical element.

Each embodiment in this specification sheets all adopts relevant mode to describe, and what between each embodiment, identical similar part illustrated see, each embodiment emphasis mutually is the difference with other embodiments. Especially, for device embodiment, owing to it is substantially similar to embodiment of the method, so what describe is fairly simple, relevant part illustrates see the part of embodiment of the method.

One of ordinary skill in the art will appreciate that all or part of step realized in aforesaid method enforcement mode can be completed by the hardware that program carrys out instruction relevant, described program can be stored in computer read/write memory medium, here the alleged storage media obtained, as: ROM/RAM, magnetic disc, CD etc.

The foregoing is only the better embodiment of the present invention, it is not intended to limit protection scope of the present invention. All do within the spirit and principles in the present invention any amendment, equivalent replacement, improvement etc., be all included in protection scope of the present invention.

Claims (8)

1. method of calculation for the dynamic resource of network, are applied to mobile cloud computing environment, it is characterised in that, described method comprises:

According to the resource supplydemand relationship between cloud service provider and user's end, set up how main many from the Staenberg betting model of two benches;

According to how main many from the Staenberg betting model of two benches, set up user's end utility function and high in the clouds utility function;

According to described much much more main from Staenberg betting model calculate high in the clouds optimum bandwidth price strategy and user's end optimal bandwidth allocation strategy, for being user's dynamic resource of end distribution network.

2. method according to claim 1, it is characterised in that, described many main many from the Staenberg betting model of two benches is:

G=(p, b, U_T(p,b),U_C(p, b));

Wherein, U_T(p, b) be much much more main from the Staenberg betting model of two benches user's end utility function, U_C(p, b) be much much more main from the Staenberg betting model of two benches high in the clouds utility function, p={p₁,p₂,...,p_mIt it is high in the clouds price strategy; B={b₁,b₂,...,b_IIt it is user's end band-width tactics.

3. method according to claim 1 or 2, it is characterised in that,

Described user's end utility function is:

$U_{Ti}(p,b_i)=R_i\left(\underset{j=1}{\overset{m}{\Σ}}{b}_{ij}\right)-\underset{j=1}{\overset{m}{\Σ}}{C}_j^B(b_{ij},p_j)-\underset{j=1}{\overset{m}{\Σ}}{C}_j^D\left(b_{ij}\right);$

Wherein, m is the total number of cloud service provider, b_ijIt is the bandwidth that i-th user's end obtains from jth cloud service provider,It is the total bankwidth that i-th user's end obtains from cloud service provider,It is i-th user's end distributionThe income that bandwidth obtains,For the expense paid to jth cloud service provider,For the time delay expense that cloud service provider j gives i-th user's end band next, wherein,In formulaI is the total number of user's end, B_totFor network total bankwidth, ��_iIt is the spectrum effectiveness of i-th user's end, r_iIt is i-th user's end unit transfer rate income, M_iIt is the satisfaction of users of i-th user's termination by MCC service;In formula, p_jFor the unit bandwidth price that jth cloud service provider is formulated,In formula, c_iIt is the unit time delay overhead value of i-th user's end, d_iIt is that i-th user's end obtains total time lag when MCC serves;

Described high in the clouds utility function is:

U_Cj=Q_j*p_j;

Wherein, Q_jFor the total bankwidth to all user's ends required by cloud service provider j, p_jFor the unit bandwidth price that jth cloud service provider is formulated;

Described user's end utility function at least comprises the income based on bandwidth and cost; Described cost comprises the expense to cloud service provider payment and time delay expense.

4. method according to claim 1, it is characterised in that, described according to described many main many from Staenberg betting model calculate high in the clouds optimum bandwidth price strategy and user's end optimal bandwidth allocation strategy, comprising:

The cloud service provider in high in the clouds, based on the state of loading of current network, calculates high in the clouds price strategy p={p₁,p₂,...,p_m, and it is broadcast to all user's ends that high in the clouds covers;

User's termination receives the high in the clouds price strategy p={p of different cloud service provider broadcast₁,p₂,...,p_mAfter, calculate user end optimum bandwidth demand strategy b according to the price strategy of current network^*(p), and by described user end optimum bandwidth demand strategy b^*P () feeds back to all cloud service provider, wherein, and described user end optimum bandwidth demand strategy b^*P () represents:

$b^{*}\left(p\right)=\frac{r_i{\mathrm{\η}}_i{B}_{tot}{M}_i{\mathrm{\Σ}}_{\∀n\≠i}{\mathrm{\Σ}}_{j=1}^m{b}_{nj}}{{\left({\mathrm{\Σ}}_{i=1}^I{\mathrm{\Σ}}_{j=1}^m{b}_{ij}\right)}^2}-p+\frac{c_i{T}_i{\mathrm{\Σ}}_{\∀n\≠i}{\mathrm{\Σ}}_{j=1}^m{b}_{nj}}{{\mathrm{\η}}_i{\left({\mathrm{\Σ}}_{j=1}^m{b}_{ij}\right)}^2};$

In formulaThe all user's terminations represented except i-th user's end receive the total bankwidth of all cloud service provider,Representing that all user's terminations receive the total bankwidth of all cloud service provider, p is the price strategy of the corresponding cloud service provider that i-th user's end receives;

High in the clouds cloud service provider is according to the user end optimum bandwidth demand strategy b of all client feeds back^*P () calculates the high in the clouds optimum bandwidth price strategy p meeting user's end band width demand^*, and it is broadcast to all user's ends, and wherein, described high in the clouds optimum bandwidth price strategy p^*Expression is:

$p^{*}=\frac{r_i{\mathrm{\η}}_i{B}_{tot}{M}_i{\mathrm{\Σ}}_{\∀n\≠i}{\mathrm{\Σ}}_{j=1}^m{b^{*}}_{nj}}{{\left({\mathrm{\Σ}}_{i=1}^I{\mathrm{\Σ}}_{j=1}^m{b^{*}}_{ij}\right)}^2}+\frac{c_i{T}_i{\mathrm{\Σ}}_{\∀n\≠i}{\mathrm{\Σ}}_{j=1}^m{b^{*}}_{nj}}{{\mathrm{\η}}_i{\left({\mathrm{\Σ}}_{j=1}^m{b^{*}}_{ij}\right)}^2};$

In formulaRepresent, the optimum bandwidth demand summation that all user's ends met except i-th user's end receive from all cloud service provider,For the optimum bandwidth demand summation that all user's ends receive from all cloud service provider,It it is the optimum bandwidth demand summation that i-th user's end receives from all cloud service provider;

User's end is according to the high in the clouds optimum bandwidth price strategy p of all cloud service provider broadcast received^*Recalculate user end optimal bandwidth allocation strategy b^*(p^*), wherein, described user end optimal bandwidth allocation strategy b^*(p^*) represent be:

$b^{*}\left(p^{*}\right)=\frac{r_i{\mathrm{\η}}_i{B}_{tot}{M}_i{\mathrm{\Σ}}_{\∀n\≠i}{\mathrm{\Σ}}_{j=1}^m{b}_{nj}}{{\left({\mathrm{\Σ}}_{i=1}^I{\mathrm{\Σ}}_{j=1}^m{b}_{ij}\right)}^2}-p^{*}+\frac{c_i{T}_i{\mathrm{\Σ}}_{\∀n\≠i}{\mathrm{\Σ}}_{j=1}^m{b}_{nj}}{{\mathrm{\η}}_i{\left({\mathrm{\Σ}}_{j=1}^m{b}_{ij}\right)}^2};$

In formula, p^*It it is the high in the clouds best price strategy of the corresponding cloud service provider that i-th user's end receives.

5. a calculating device for the dynamic resource of network, is applied to mobile cloud computing environment, it is characterised in that, described device comprises:

Many masters are many sets up module from Staenberg betting model, for according to the resource supplydemand relationship between cloud service provider and user's end, setting up how main how from the Staenberg betting model of two benches;

Utility function sets up module, for according to how main many from the Staenberg betting model of two benches, setting up user's end utility function and high in the clouds utility function;

Optimal strategy calculates module, for according to described much much more main from Staenberg betting model calculate high in the clouds optimum bandwidth price strategy and user's end optimal bandwidth allocation strategy, for being user's dynamic resource of end distribution network.

6. device according to claim 5, it is characterised in that, described many main many set up module from Staenberg betting model and is:

G=(p, b, U_T(p,b),U_C(p, b));

Wherein, U_T(p, b) be much much more main from the Staenberg betting model of two benches user's end utility function, U_C(p, b) be much much more main from the Staenberg betting model of two benches high in the clouds utility function, p={p₁,p₂,...,p_mIt it is high in the clouds price strategy; B={b₁,b₂,...,b_IIt it is user's end band-width tactics.

7. device according to claim 5 or 6, it is characterised in that,

Described user's end utility function is:

$U_{Ti}(p,b_i)=R_i\left(\underset{j=1}{\overset{m}{\Σ}}{b}_{ij}\right)-\underset{j=1}{\overset{m}{\Σ}}{C}_j^B(b_{ij},p_j)-\underset{j=1}{\overset{m}{\Σ}}{C}_j^D\left(b_{ij}\right);$

Wherein, m is the total number of cloud service provider, b_ijIt is the bandwidth that i-th user's end obtains from jth cloud service provider,It is the total bankwidth that i-th user's end obtains from cloud service provider,It is i-th user's end distributionThe income that bandwidth obtains,For the expense paid to jth cloud service provider,For the time delay expense that cloud service provider j gives i-th user's end band next, wherein,In formulaI is the total number of user's end, B_totFor network total bankwidth, ��_iIt is the spectrum effectiveness of i-th user's end, r_iIt is i-th user's end unit transfer rate income, M_iIt is the satisfaction of users of i-th user's termination by MCC service;In formula, p_jFor the unit bandwidth price that jth cloud service provider is formulated,In formula, c_iIt is the unit time delay overhead value of i-th user's end, d_iIt is that i-th user's end obtains total time lag when MCC serves;

Described high in the clouds utility function is:

U_Cj=Q_j*p_j;

Wherein, Q_jFor the total bankwidth to all user's ends required by cloud service provider j, p_jFor the unit bandwidth price that jth cloud service provider is formulated;

Described user's end utility function at least comprises the income based on bandwidth and cost; Described cost comprises the expense to cloud service provider payment and time delay expense.

8. device according to claim 5, it is characterised in that, described optimal strategy calculates module, comprising:

High in the clouds price strategy calculates and broadcast submodule block, for the state of loading according to current network, calculates high in the clouds price strategy p={p₁,p₂,...,p_m, and it is broadcast to all user's ends that high in the clouds covers;

User's end optimum bandwidth demand strategy calculates and feedback submodule block, for receiving the high in the clouds price strategy p={p of different cloud service provider broadcast in user's termination₁,p₂,...,p_mAfter, calculate user end optimum bandwidth demand strategy b according to the price strategy of current network^*(p), and by described user end optimum bandwidth demand strategy b^*P () feeds back to all cloud service provider, wherein, and described user end optimum bandwidth demand strategy b^*P () represents:

$b^{*}\left(p\right)=\frac{r_i{\mathrm{\η}}_i{B}_{tot}{M}_i{\mathrm{\Σ}}_{\∀n\≠i}{\mathrm{\Σ}}_{j=1}^m{b}_{nj}}{{\left({\mathrm{\Σ}}_{i=1}^I{\mathrm{\Σ}}_{j=1}^m{b}_{ij}\right)}^2}-p+\frac{c_i{T}_i{\mathrm{\Σ}}_{\∀n\≠i}{\mathrm{\Σ}}_{j=1}^m{b}_{nj}}{{\mathrm{\η}}_i{\left({\mathrm{\Σ}}_{j=1}^m{b}_{ij}\right)}^2};$

In formulaThe all user's terminations represented except i-th user's end receive the total bankwidth of all cloud service provider,Representing that all user's terminations receive the total bankwidth of all cloud service provider, p is the price strategy of the corresponding cloud service provider that i-th user's end receives;

High in the clouds optimum bandwidth price strategy calculates and broadcast submodule block, for the user end optimum bandwidth demand strategy b according to all client feeds back^*P (), cloud service provider formulates the high in the clouds optimum bandwidth price strategy p meeting user's end band width demand^*, and it is broadcast to all user's ends, and wherein, described high in the clouds optimum bandwidth price strategy p^*Expression is:

$p^{*}=\frac{r_i{\mathrm{\η}}_i{B}_{tot}{M}_i{\mathrm{\Σ}}_{\∀n\≠i}{\mathrm{\Σ}}_{j=1}^m{b^{*}}_{nj}}{{\left({\mathrm{\Σ}}_{i=1}^I{\mathrm{\Σ}}_{j=1}^m{b^{*}}_{ij}\right)}^2}+\frac{c_i{T}_i{\mathrm{\Σ}}_{\∀n\≠i}{\mathrm{\Σ}}_{j=1}^m{b^{*}}_{nj}}{{\mathrm{\η}}_i{\left({\mathrm{\Σ}}_{j=1}^m{b^{*}}_{ij}\right)}^2};$

In formulaRepresent, the optimum bandwidth demand summation that all user's ends met except i-th user's end receive from all cloud service provider,For the optimum bandwidth demand summation that all user's ends receive from all cloud service provider,It it is the optimum bandwidth demand summation that i-th user's end receives from all cloud service provider;

User's end optimal bandwidth allocation policy calculation submodule block, for the high in the clouds optimum bandwidth price strategy p according to all cloud service provider broadcast received^*After, user's end recalculates user end optimal bandwidth allocation strategy b^*(p^*), wherein, described user end optimal bandwidth allocation strategy b^*(p^*) represent be:

$b^{*}\left(p^{*}\right)=\frac{r_i{\mathrm{\η}}_i{B}_{tot}{M}_i{\mathrm{\Σ}}_{\∀n\≠i}{\mathrm{\Σ}}_{j=1}^m{b}_{nj}}{{\left({\mathrm{\Σ}}_{i=1}^I{\mathrm{\Σ}}_{j=1}^m{b}_{ij}\right)}^2}-p^{*}+\frac{c_i{T}_i{\mathrm{\Σ}}_{\∀n\≠i}{\mathrm{\Σ}}_{j=1}^m{b}_{nj}}{{\mathrm{\η}}_i{\left({\mathrm{\Σ}}_{j=1}^m{b}_{ij}\right)}^2};$

In formula, p^*It it is the high in the clouds best price strategy of the corresponding cloud service provider that i-th user's end receives.