CN101534508B - Dynamic resource scheduling method with introduced heterogeneous customer service executing coefficient - Google Patents

Abstract

A heterogeneous customer service executing coefficient introduced dynamic resource scheduling method relates to a method and strategy for dynamically scheduling time and frequency spectrum resources (wireless channel) for a plurality of customers with different service requirements in a heterogeneous wireless network environment containing cognitive radio technical equipment, in particular to a method and strategy for dynamically scheduling and allocating opportunistic frequency spectrum on time and frequency band and making the customers with the different service requirements in the heterogeneous network fairly and effectively use finite wireless network resources. The dynamic resource scheduling method comprises the following steps: considering customer service characteristics into network resources for the first time, including dynamic allocation of the time and frequency spectrum resources; defining a customer service executed priority coefficient by various parameters deciding customer service types, and sorting heterogeneous customer services according to the priority coefficient so as to obtain basis for effective allocating finite available channel resources; on the basis, allocating available channels according to the principle of maximizing the frequency spectrum utilization rate; and obtaining service quality guarantee for the customers with the different service requirements.

Description

A kind of dynamic resource scheduling method of introducing heterogeneous customer service executing coefficient

Technical field

The present invention relates generally to a kind of in comprising the heterogeneous wireless network environment of cognitive radio technology equipment, the user of a plurality of different business demands is carried out the method and the strategy of the dynamic dispatching of time and frequency spectrum resource (wireless channel), more particularly, relate to a kind of dynamic dispatching and distribute opportunistic spectrum on time and frequency band, and make fair between the user of different business demand in the heterogeneous network, effectively use the method and the strategy of limited wireless Internet resources.

Background technology

At present radio spectrum resources utilize situation extremely uneven: some authorize frequency range degree of taking very high, and some authorizes frequency range then often idle.One of effective way that solves this phenomenon is the opportunistic spectrum access technology.The opportunistic spectrum access technology allows in the network with cognitive radio (CR) system, exploitation of chance ground and the idle mandate frequency spectrum resource of use.

Realize that the key technology that opportunistic spectrum inserts comprises frequency spectrum detection and analytical technology, opportunistic spectrum management and dispatching technique.Opportunistic spectrum management and dispatching technique relate to the problem that the frequency range (channel) of the authorized user how will not be used is fairly and reasonably distributed between the unauthorized user of different business demand, promptly

Which available channel does 1. how authoring system determine to distribute give the user of which kind of business demand when?

2. the how fairness and the reasonability of assurance time and channel allocation between the user of different business demand of authoring system?

As the solution of these problems, wireless network resource dynamic dispatching algorithm and scheme have been produced.The scheme that the present invention studied just is being based on above background, overcomes the above problems.The heterogeneous network environment of being studied as shown in Figure 1.Comprise authoring system base station (BS), access node (AP) and consumer premise equipment (CPE).

Fig. 1 shows in same authorisation network coverage, the unauthorized user that has a plurality of not same areas, unauthorized user is according to the current available channel conditions request of authorized user base station access service, and the case of the service business type isomery of the interior unauthorized user of each subdomain.

With reference to Fig. 1, the zone in the authoring system base station range is divided into plurality of sub-regions, and each subregion has several access nodes, comprises N the isomery user that the different business demand is arranged.Number of available channels is K in each zone, and K channel position and bandwidth on frequency spectrum becomes when being, and along with the difference of subregion difference.Number of users, number of available channels, channel position, change when channel width all is change with the region in each subregion.The problem to be solved in the present invention is, in each time slot under this environment, make BS distribute rationally and effectively existing Internet resources according to situations such as number of users, customer service type, number of available channels, channel position, channel width in each sub regions, communicating by letter between the satisfied and different user of assurance customer service demand do not clash.

Summary of the invention

Technical problem: the purpose of this invention is to provide a kind of dynamic resource scheduling method of introducing heterogeneous customer service executing coefficient, address the above problem at least basically and/or shortcoming, and following advantage is provided at least.Therefore, the invention provides a kind of heterogeneous network system based on cognitive techniques, obtain the priority of resource according to the different definition user of customer service demand, dynamic dispatching and distribute opportunistic spectrum on time and frequency band, and make fair between the user of different business demand in the heterogeneous network, effectively use the method and the strategy of limited wireless Internet resources.

Technical scheme: the invention provides a kind of dynamic dispatching and the method and the strategy that distribute opportunistic spectrum on time and frequency band.

At first, here Ding Yi term definition is as follows:

γ: channel signal noise ratio CSNR (channel-signal-to-noise ratio)

Λ: customer service executing coefficient

λ: the arrival rate of customer service

μ: the transmission rate of customer service

P: customer service is the probability of real-time

τ: the duration of customer service

B: system's available channel bandwidth

K: the current number of available channels of system

N: the number of users of different business demand in current certain sub regions of system

R: the channel capacity on system's available channel

X: channel allocation variable

T: base station assigns is given the length of user's time slot

α: time coefficient

The dynamic resource scheduling scheme is as follows:

This scheme provides a kind of can distribute foundation according to the resource dynamic of heterogeneous network user's different business requirement definition.This method takes into account Internet resources with service characteristics first, comprises in the dynamic assignment of time and frequency spectrum resource.According to the type of customer service the enforceability of customer service is carried out priority and divide, thereby be effective distribution acquisition foundation of limited available channel resources.On this basis, according to availability of frequency spectrum maximization principle, carry out available channel and distribute.Have characteristics such as Resource Allocation in Networks fairness, validity, self adaptation, obtain different business demand quality of services for users simultaneously and guarantee.

At first, at the isomery characteristic of customer service, definition user's business is carried out coefficient

For the user in the heterogeneous network, the parameter that characterizes its service feature comprises professional arrival pattern, service rate (message transmission rate), professional duration, professional real-time requirement, under the environment of limited, a plurality of user's competition of frequency spectrum resource, user's service feature parameter has not only determined the priority that customer service self is performed, and also influences and has determined the user to obtain the priority of Internet resources.For this reason, the different parameters that influences user performance characteristic is integrated consideration, it is as follows that definition user's business is carried out coefficient:

Λ(t)＝λ(t)μ(t)P(t)

Wherein, parameter lambda is represented the arrival rate of customer service, and parameter μ represents the transmission rate of customer service, and parameter P represents that customer service is the probability of real-time, has expressed professional requirement to real-time.Parameter lambda, μ and P are time dependent.

Secondly, because at a time, number of available channels K is less than the number of users N that business demand is arranged, for guaranteeing that all users use the fair and reasonable property of Internet resources, guarantee that different business demand user can both in time obtain required Internet resources, user's business is carried out coefficient and need be performed situation according to the variation of time, customer service and business and upgrade, and more is newly defined as

Λ(t)＝Λ(t-1)D(t-1)+λ(t)μ(t)P(t)

Wherein,

$D\left(t\right)=\left\{\begin{array}{cc}\mathrm{\α}=\frac{T-\mathrm{\τ}}{T},& \mathrm{\τ}>T\\ 0,& \mathrm{\τ}\≤T\end{array}\right.$

Be initialized as Λ (0)=λ (0) μ (0) P (0).That is, the business of user's current time is carried out the state that coefficient not only depends on user's current time business, depends on that also the business of previous moment is carried out coefficient, and closely related with the timeliness requirement of customer service.If slot length is T, the professional duration τ of user's previous moment, when τ＞T, the business of this user's current time carry out coefficient be the professional product of carrying out coefficient and time coefficient α of previous moment again with current business execution coefficient sum; When τ≤T, the business of this user's current time is carried out coefficient only by the service condition decision of user's current time, and is irrelevant with previous moment.

At last, by ordering to N customer service executing coefficient, the user of K the coefficient correspondence that the value of obtaining is bigger participates in the distribution of current time authoring system to K available channel, the availability of frequency spectrum with maximization current time whole system is the optimal solution that target is tried to achieve time slot and channel allocation parameter, and specific algorithm is:

$\max \underset{n\∈V}{\mathrm{\Σ}}\underset{k\∈V}{\mathrm{\Σ}}{C}_{\mathrm{nk}}{X}_{\mathrm{nk}}$

Wherein, C _Nk=B _kLog ₂(1+ γ _Nk), V={1,2 ..., K} is the channel capacity on the channel k, γ _NkBe the channel signal noise ratio of user n on channel k, B _kBandwidth for channel k.

K channel allocation principle is:

1. same channel difference is distributed to same user simultaneously;

2.K each user distributes a channel at least among the individual user.

For this reason, the channel defined variable X that distributes for each user _Nk, if X _Nk=1 means that channel k distributes to user n, otherwise X _Nk=0.

In view of the above, more than ask the constraints of optimal solution to be:

$\underset{n\∈V}{\mathrm{\Σ}}\underset{k\∈V}{\mathrm{\Σ}}{X}_{\mathrm{nk}}\≤K$

$\underset{k\∈V}{\mathrm{\Σ}}{X}_{\mathrm{nk}}=1,$ $\∀n\∈V$

Thus, obtain the current time base station to the optimal allocation of K available channel between K unauthorized user.

Beneficial effect: the invention provides a kind of dynamic wireless network resource scheduling scheme based on the customer service state.

The invention provides a kind ofly under isomerous environment, at first the coefficient that is performed according to different demand users' service condition definition user corresponding service determines the user to obtain the priority of wireless network resource thus.

The invention provides a kind of under heterogeneous network environment, according to the time variation of customer service demand, algorithm and scheme that the time variation of network available resources dynamically updates and dispatches the heterogeneous wireless network resource.

According to a further aspect in the invention, the dynamic resource scheduling scheme has been guaranteed the most effective utilization of wireless network resource on time and frequency spectrum resource, and takes into account the fairness between different business demand user, comprehensive and customer service service quality assurance.

Description of drawings

By below in conjunction with the accompanying drawing and the detailed description of carrying out, it is clearer that above and other objects of the present invention, characteristics and advantage will become, wherein;

Fig. 1 shows the heterogeneous network environment case of multi-user wireless network scheduling of resource.

Fig. 2 shows the operational flowchart that wireless network resource is dispatched in heterogeneous network environment on time and frequency spectrum.

Embodiment

Preferential embodiment of the present invention is here described below with reference to accompanying drawings.Because known function and structure make the present invention fuzzy in unnecessary details, therefore they are described no longer in the following description.

The invention provides make the heterogeneous network Radio Resource in time with frequency spectrum on the method for dynamic assignment and management, characterization parameter according to the different business demand, the decision different user obtains the priority of wireless network resource, isomerism, time variation, the isomerism of customer service, time variation with the network available resources serve as according to wireless network resource dynamically being adjusted and distributed, accomplishing fairness and validity that user resources are shared.Specific implementation method is as follows

A. customer service executing coefficient

With reference to Fig. 1, in zone 1, there be N unauthorized user request base station (BS) to distribute time slot and channel resource with different business demand and characteristic, at current time, BS number of available channels in zone 1 is K.With reference to Fig. 2, at first according to the different service types of N the unauthorized user that exists in the zone 1, the parameter of type of service will be characterized, be customer service arrival rate λ (t), delivery of user traffic speed μ (t) and customer service real-time probability P (t), the priority factor that integrating representation is performed for this business, i.e. customer service executing coefficient Λ (t).The arrival of this coefficient explanation customer service more early, delivery of user traffic speed is high more, and the real-time requirement is high more can preferentially to obtain required Radio Resource more.Afterwards, current all users' business is carried out coefficient sort, preceding K the user that coefficient value is bigger is chosen as the distribution that participates in K current available authorization channel, and all the other N-K user waits for and participates in next channel allocation constantly.

B. channel optimization distributes

With reference to Fig. 2, be target function with the availability of frequency spectrum of current time global system, by right

$\max \underset{n\∈V}{\mathrm{\Σ}}\underset{k\∈V}{\mathrm{\Σ}}{C}_{\mathrm{nk}}{X}_{\mathrm{nk}}$

Find the solution, obtain current time BS to the time slot of K user in N unauthorized user in the zone 1 and the optimal solution of channel allocation, promptly to concrete n and k assignment.

C. customer service executing coefficient upgrades

With reference to Fig. 2, share fairness for adapted to isomerous network time variation, available channel resources time variation, customer service time variation, user distribution time variation and user resources, need upgrade in time to customer service executing coefficient.It is low that principle is that the customer service executing coefficient that has been assigned to the user of wireless network resource is carried out coefficient than the user's who is not assigned to wireless network resource business; The customer service executing coefficient that has new business to arrive is lower than the customer service executing coefficient that does not have new business to arrive; And customer service executing coefficient is a cumulant, comprises the historical data information of customer service implementation status.Under above principle, the update algorithm that the business of the N of a current time t unauthorized user is carried out coefficient is as follows.On the one hand, in previous moment, promptly t-1 participate in K channel allocation constantly K user's the renewal of professional coefficient according to formula:

Λ(t)＝Λ(t-1)D(t-1)+λ(t)μ(t)P(t)

And have

$D\left(t\right)=\left\{\begin{array}{cc}\mathrm{\α}=\frac{T-\mathrm{\τ}}{T},& \mathrm{\τ}>T\\ 0,& \mathrm{\τ}\≤T\end{array}\right.$

On the other hand, t-1 have neither part nor lot in K channel allocation constantly N-K user's the renewal of professional coefficient according to formula:

Λ(t)＝λ(t)μ(t)P(t)

Carrying out the final stage of coefficient update, N user's upgrading business execution coefficient is being reentered into the formation of execution coefficient update, the line ordering of going forward side by side.

D. dynamic resource scheduling circulation

With reference to Fig. 2, customer service executing coefficient upgrade and rearrangement after, select again in the distribution of K channel that the customer service of K coefficient maximum wherein participates in a new moment, from then on system enters the next cyclic process of scheduling of resource and reorganization.

Claims (1)

1. dynamic resource scheduling method of introducing heterogeneous customer service executing coefficient is characterized in that this dynamic resource scheduling method is specific as follows:

First user performance characteristic is taken into account Internet resources, comprise in the dynamic assignment of time and frequency spectrum resource; Various parameters with the decision user performance characteristic define the preferred number that customer service is performed, and with this heterogeneous customer service are sorted, thereby be effective distribution acquisition foundation of limited available channel resources; On this basis, according to availability of frequency spectrum maximization principle, carry out available channel and distribute; Obtaining different business demand quality of services for users guarantees;

A. at the isomery characteristic of customer service, definition user's business is carried out coefficient

For the user in the heterogeneous network, the parameter that characterizes its service feature comprises that professional arrival pattern, service rate are message transmission rate, professional duration, professional real-time requirement, under the environment of limited, a plurality of user's competition of frequency spectrum resource, the different parameters that influences user performance characteristic is integrated consideration, and it is as follows that definition user's business is carried out coefficient:

Λ(t)＝λ(t)μ(t)P(t)

Wherein, parameter lambda is represented the arrival rate of customer service, and parameter μ represents the transmission rate of customer service, and parameter P represents that customer service is the probability of real-time, has expressed professional requirement to real-time, and parameter lambda, μ and P are time dependent;

B. because at a time, number of available channels K is less than the number of users N that business demand is arranged, for guaranteeing that all users use the fair and reasonable property of Internet resources, guarantee that different business demand user can both in time obtain required Internet resources, user's business is carried out coefficient and need be performed situation according to the variation of time, customer service and business and upgrade, and more is newly defined as

Λ(t)＝Λ(t-1)D(t-1)+λ(t)μ(t)P(T)

Wherein,

$D\left(t\right)=\left\{\begin{array}{c}\mathrm{\α}=\frac{T-\mathrm{\τ}}{T},\mathrm{\τ}>T\\ 0,\mathrm{\τ}\≤T\end{array}\right.$

Be initialized as Λ (0)=λ (0) μ (0) P (0), promptly, the business of user's current time is carried out the state that coefficient not only depends on user's current time business, depends on that also the business of previous moment is carried out coefficient, and closely related with the timeliness requirement of customer service; If slot length is T, the professional duration τ of user's previous moment, when τ＞T, the business of this user's current time carry out coefficient be the professional product of carrying out coefficient and time coefficient α of previous moment again with current business execution coefficient sum; When τ≤T, the business of this user's current time is carried out coefficient only by the service condition decision of user's current time, and is irrelevant with previous moment;

C. by ordering to N customer service executing coefficient, preceding K the user that coefficient value is bigger participates in the distribution of current time authoring system to K available channel, the availability of frequency spectrum with maximization current time whole system is the optimal solution that target is tried to achieve time slot and channel allocation parameter, and concrete grammar is:

$\max \underset{n\∈V}{\mathrm{\Σ}}\underset{k\∈V}{\mathrm{\Σ}}{C}_{\mathrm{nk}}{X}_{\mathrm{nk}}$

Wherein, C _Nk=B _kLog ₂(1+ γ _Nk), V={1,2 ..., K} is the channel capacity on the channel k, γ _NkBe the channel signal noise ratio of user n on channel k, B _kBandwidth for channel k; X _NkChannel defined variable for each user's distribution;

K channel allocation principle is:

Distribute to same user when same channel is different;

Each user distributes a channel at least among K the user;

D. the channel defined variable X that distributes for each user _Nk, if X _Nk=1 means that channel k distributes to user n, otherwise X _Nk=0,

In view of the above, more than ask the constraints of optimal solution to be:

$\underset{n\∈V}{\mathrm{\Σ}}\underset{k\∈V}{\mathrm{\Σ}}{X}_{\mathrm{nk}}\≤K$

$\underset{k\∈V}{\mathrm{\Σ}}{X}_{\mathrm{nk}}=1,\∀n\∈V$

Thus, obtain the current time base station to the optimal allocation of K available channel between K unauthorized user.

Images