CN103781101B - The artificial frequency spectrum cavity-pocket acquisition methods and device driven in a kind of cellular network based on user behavior - Google Patents

The artificial frequency spectrum cavity-pocket acquisition methods and device driven in a kind of cellular network based on user behavior Download PDF

Info

Publication number
CN103781101B
CN103781101B CN201410004408.6A CN201410004408A CN103781101B CN 103781101 B CN103781101 B CN 103781101B CN 201410004408 A CN201410004408 A CN 201410004408A CN 103781101 B CN103781101 B CN 103781101B
Authority
CN
China
Prior art keywords
application
user
subcarrier
terminal system
background application
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
CN201410004408.6A
Other languages
Chinese (zh)
Other versions
CN103781101A (en
Inventor
单杭冠
倪志风
黄爱苹
王玮
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.)
Zhejiang University ZJU
Original Assignee
Zhejiang University ZJU
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 Zhejiang University ZJU filed Critical Zhejiang University ZJU
Priority to CN201410004408.6A priority Critical patent/CN103781101B/en
Publication of CN103781101A publication Critical patent/CN103781101A/en
Application granted granted Critical
Publication of CN103781101B publication Critical patent/CN103781101B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

The invention discloses the artificial frequency spectrum cavity-pocket acquisition methods driven in a kind of cellular network based on user behavior, it is by considering the behavior of people in the use to frequency spectrum resource and the optimization to communication system, frequency spectrum quantity shared by the corresponding connection of application/business of the reduction in the non-interesting phase, artificially vacates some frequency spectrums in other words;Include the network resource manager of user profile acquisition module, user profile reporting module and base station the invention also discloses the device for realizing the above method.Behavioural information of the invention by obtaining user, and take running time-frequency resource to change the application that user is not concerned with using the information, reduction is to its resource provisioning, artificially vacate some frequency spectrums, and devise the Access Control Policy matched with the strategy, cell allow while meet the traffic demand of more users, or realizes in network congestion the load moved and mitigate network of frequency spectrum in time.

Description

A kind of artificial frequency spectrum cavity-pocket acquisition methods driven in cellular network based on user behavior And device
Technical field
The invention belongs to mobile communication technology field, and in particular to the people driven in a kind of cellular network based on user behavior Work frequency spectrum cavity-pocket acquisition methods and device.
Background technology
Mobile broadband technology(Such as WiFi, LTE)High speed development bring great convenience to our daily life and Comfortably.Increase with people to various demands for services, is limited to limited Radio Resource, and the load of mobile network just becomes to get over Come heavier, therefore Consumer's Experience can not also be effectively guaranteed.A main cause for causing this situation is exactly people To premium terminals(Such as smart mobile phone, net book, tablet device)Use rapid growth.Report display, 92% in 2012 Global mobile phone flow occupied by smart mobile phone.In addition, it has been observed that being not only created more using the user of premium terminals High average load, also than consuming more times on their device using the user of plain old telephone.Therefore, it is following to become Gesture is exactly, with increasing for the number of users using premium terminals, and the consumption of Radio Resource becomes big, causes Radio Resource shortage very Probability of happening to network congestion is big also with change.Therefore, design one kind can successfully manage this premium terminals quantity surge feelings The radio resource management policy of shape turns into the task of top priority.
In order to resolve RRM problem in the network of congestion, " Differentiated Services " strategy is widely adopted. When the service of network load burdensome and high priority needs more Radio Resources, in order to improve the frequency spectrum of whole network Efficiency, service or user's reduction service quality of certain particular traffic type that this strategy would generally be to low priority (Quality of Service,QoS)It is required that.For example, in order to accommodate more connection requests in an overload system, towards The Access Control scheme of Differentiated Services, such as document " Handoff prioritization and decision schemes in wireless cellular networks:A survey”(Author A.Sgora, is published in IEEE Communications Surveys & Tutorials,vol.11,no.4,pp.57-77,2009)In mention classification grading strategy, document " An efficient admission control scheme for adaptive multimedia services in IEEE802.16e networks”(Author Y.Ge, is published in Proc.IEEE VTC-2006Fall, 2006)In the service mentioned Degrade strategy and document " Radio resource management games in wireless networks:An approach to bandwidth allocation and admission control for polling service in IEEE802.16”(Author D.Niyato, is published in IEEE Wireless Commun., vol.14, no.1, pp.27-35, Feb.2007)In the bandwidth mentioned borrow strategy etc., can forcibly make some low priorities in being currently running be connected to low clothes Business requires to run under pattern.It is similar, when network because when a part possesses the user of big flow and occurs congestion, towards differentiation The congestion control of service can punish them by imposing high cost to this certain customers, so as to alleviate congestion problems and possible Higher income can be also brought to Virtual network operator.
But, directly use Differentiated Services strategy actually effect to be not so good in the case of premium terminals quantity is increased sharply 's.The network that premium terminals quantity is increased sharply has very big difference with traditional speech network.In former network, equipment is usual Multiple tasks are supported to work simultaneously(Such as IOS and Android), and the equipment in latter network is then without this function.Use The user of premium terminals can produce multiple requests simultaneously or have multiple application receiving services simultaneously.If it is not known that user institute Some specific application of concern, and the supply of Radio Resource is simply reduced according to service type, then it can produce extra And unnecessary user experience quality(Quality of Experience,QoE)Loss.For example, a user is with net book Baidu video websites are accessed, he can open multiple video clippings interested simultaneously.A video being most interested in except him It is outer that he will give priority to for editing, and other editings all first suspend, and allow them first to buffer a period of time.When network congestion, Using the tactful QoS that will force all web applications in reduction system is distinguished, user has to by severe Quality of experience.So And, if network can perceive the behavior of user, the resource shared by withdrawing application that certain amount user is not concerned with, distribution The sufficient Internet resources application of interest to user, user QoE can be obtained by lifting.Therefore, for premium terminals user Usage behavior design a kind of effective resource distribution mode for solve current network increasingly congestion the problem of it is most important.
The content of the invention
For the above-mentioned technical problem present in prior art, the invention provides user's row is based in a kind of cellular network For the artificial frequency spectrum cavity-pocket acquisition methods and device of driving, user is not concerned with to apply institute by taking running time-frequency resource conversion to reduce The resource provisioning of correspondence connection, artificially vacates the traffic demand that some frequency spectrums meet more users.
The behavior of people is considered in the use to frequency spectrum resource and the optimization to communication system that reduction is in by the present invention Frequency spectrum quantity shared by the corresponding connection of application/business of non-interesting phase, artificially vacates some frequency spectrums in other words.We This operation driven by user behavior is referred to as " running time-frequency resource conversion ", and the vacant frequency spectrum obtained by it is referred to as " artificial frequency spectrum sky Hole ".It is noted that when the application/business changed by running time-frequency resource regains the concern of user, it would be desirable to recovery pair in time Its resource provisioning.
A kind of artificial frequency spectrum cavity-pocket acquisition methods driven in cellular network based on user behavior, including:
The periodically business information in collection base station institute coverage cell in each client terminal system, and then by these business Information is reported to the network resource manager of base station;Network resource manager is according to described business information to each user terminal system Background application in system carries out running time-frequency resource conversion, and to obtain artificial frequency spectrum cavity-pocket, detailed process is as follows:
(1)For any user, by each background application in historical statistics analysis acquisition present terminal system next Cycle still needs to the probability of happening of buffered data;
(2)According to the described probability of happening, determine that each background application should be cut down in present terminal system by below equation Subcarrier supply number;So as to next cycle to background application carry out subcarrier recovery and to being asked with resource recovery before Platform application provides full subcarrier supply;
φ(ri)=H(ri)-ω·J(ri)
s.t. 0≤ri≤Ri
Wherein:riNumber, R are supplied for the subcarrier that i-th of background application in present terminal system should be cut downiFor current end The full subcarrier supply number (it is depending on application type) of i-th of background application in end system, H (ri) it is on riNormalizing Change frequency spectrum cavity-pocket function, J (ri) it is on riNormalization QoE risk functions, ω is weight factor;I be natural number and 1≤i≤ N, n are the number of background application in present terminal system;
(3)According to step(1)With(2), travel through all users in base station institute coverage cell.
Described business information is included on the foreground application run in active user's terminal system and background application Information, the amount of buffered data of each background application and the foreground application asked with resource recovery.Described has resource extensive The foreground application asked again is as running background in the previous cycle and switches to the application of front stage operation in current period.
Described normalized spatial spectrum cavity function H (ri) tried to achieve according to relationship below:
H(ri)=ri/Ri
Described normalization QoE risk function J (ri) tried to achieve according to relationship below:
Wherein:α, β and γ are default QoE measurement indexs coefficient (nonnegative constant), PiFor in present terminal system I-th of background application still needs to the probability of happening of buffered data, Z (r in next cyclei) be present terminal system in i-th after Platform applies the buffered data total amount when collection is reported next time.
Described buffered data total amount Z (ri) tried to achieve according to relationship below:
Z(ri)=Si+(Ri-ri)vτ
Wherein:τ is collection period, and v is the mean data rate of subcarrier, SiAfter i-th in present terminal system The amount of buffered data of platform application.
Described network resource manager is according to application corresponding connection distribution subcarrier of the following priority for each user: The corresponding connection > of application asked with resource recovery switches into the connection > intra-cell users of this cell from neighbor cell The corresponding connection of application newly opened;
The described application that there is resource recovery to ask is corresponding to be connected as having in the last week in base station institute coverage cell Phase is running background and switches to the corresponding connection of the application of front stage operation in current period;
The described corresponding connection of application for switching into the connection of this cell from neighbor cell to have been opened by user and The user is moved into base station institute coverage cell from neighbor cell;
The corresponding user being connected as in base station institute coverage cell of application that described intra-cell users are newly opened newly opens The corresponding connection of application.
Described network resource manager is according to application corresponding connection distribution subcarrier of the following strategy for each user:
If the corresponding sub-carrier number taken that connects of all applications reaches R-R in current base station institute coverage cella, then network The corresponding connection distribution subcarrier of application that explorer is no longer newly opened for intra-cell users;
If the corresponding sub-carrier number taken that connects of all applications reaches R-R in current base station institute coverage cellb, then network Explorer is no longer that the connection for switching into this cell from neighbor cell distributes subcarrier;
For the corresponding connection of application asked with resource recovery, as long as network resource manager also has residue unallocated Subcarrier, just for such using it is corresponding connection distribution subcarrier;
If user has multiple while the application connection opened, the subcarrier shared by the corresponding connection of application of FEFO The corresponding connection of application that can keep for this user that there is resource recovery to ask;
Wherein:R is the specified supply number of subcarrier of base station institute coverage cell, RaPlease with resource recovery to be kept in R The corresponding connection of application asked and the subcarrier supply number for the connection that this cell is switched into from neighbor cell, RbFor list in R The subcarrier for the corresponding connection of application asked with resource recovery is solely kept for supply number.
The artificial frequency spectrum cavity-pocket acquisition device driven in a kind of cellular network based on user behavior, including user profile collection The network resource manager of module, user profile reporting module and base station;
Described user profile acquisition module and the terminal device of user are connected, and it is used to periodically gather user terminal Business information in system;Described business information includes the foreground application run in active user's terminal system and backstage should The foreground application asked with, the amount of buffered data of each background application and with resource recovery;
Described user profile reporting module is used for the business information of each client terminal system in base station institute coverage cell It is reported to the network resource manager of base station;
Described network resource manager includes:
Message processing module, for obtaining each background application in current each client terminal system by historical statistics analysis The probability of happening of buffered data is still needed in next cycle;And then calculate current each by below equation according to the described probability of happening The subcarrier supply number that each background application should be cut down in client terminal system;
φ(ri)=H(ri)-ω·J(ri)
s.t. 0≤ri≤Ri
H(ri)=ri/Ri Z(ri)=Si+(Ri-ri)vτ
Wherein:riNumber, R are supplied for the subcarrier that i-th of background application in active user's terminal system should be cut downiTo work as The full subcarrier supply number of i-th of background application in preceding client terminal system, ω is weight factor;I is natural number and 1≤i ≤ n, n are the number of background application in active user's terminal system;α, β and γ are default QoE measurement indexs coefficient, PiFor I-th of background application in active user's terminal system still needs to the probability of happening of buffered data in next cycle;τ is collection week Phase, v is the mean data rate of subcarrier, SiFor the buffered data of i-th of background application in active user's terminal system Amount;
Resource distribution module, the subcarrier for that should be cut down according to background application supplies number, by dispatching in next cycle Subcarrier recovery is carried out to the background application in each client terminal system and the foreground application asked with resource recovery is provided Full subcarrier supply.
The present invention takes time-frequency to provide the application that user is not concerned with by obtaining the behavioural information of user using the information Source is changed, and is reduced to its resource provisioning, is artificially vacateed some frequency spectrums, and devise the Access Control matched with the strategy Strategy so that cell can meet the traffic demand of more users simultaneously, or realize frequency spectrum in time in network congestion The load moved and mitigate network.
Brief description of the drawings
Fig. 1 is the step flow chart of the inventive method.
Fig. 2 is the schematic flow sheet of running time-frequency resource conversion in the inventive method.
Fig. 3 is the structural representation of apparatus of the present invention.
Fig. 4 is the emulation schematic diagram on obstruction/call drop probability of three kinds of connection requests using the inventive method.
Fig. 5 is not use running time-frequency resource switch technology to be used with that can be supported using maximum after running time-frequency resource switch technology of the present invention The emulation comparison schematic diagram of amount.
Embodiment
In order to more specifically describe the present invention, below in conjunction with the accompanying drawings and embodiment is to technical scheme It is described in detail.
Embodiment one
Fig. 1 is a kind of " artificial frequency spectrum for supporting to drive using user behavior in multiple services LTE type cells wireless network Cavity " acquisition methods flow chart.It gives " artificial frequency spectrum cavity-pocket " acquisition methods that the present invention is driven using user behavior Implement process.This method includes:First, user terminal operations system is periodically adopted to user behavior information Collection;Then, the user behavior information collected is reported to network resource management by user terminal operations system by feedback mechanism Device;Finally, network resource manager is according to the user behavior information of collection, the corresponding connection of application to each user's non-interesting Running time-frequency resource conversion is carried out, so that " artificial frequency spectrum cavity-pocket " is obtained, meanwhile, to each user's foreground application with recovery request Corresponding connection, recovers to supply its full subcarrier.
Fig. 2 wraps to calculate the flow chart of sub-carrier number reduction amount when each connection is changed by running time-frequency resource in present embodiment Include step:
(1)Calculate user's focus switching probability Prf
Keep record part E1、E2、E3And E4Respectively in hypomere period tau:User opens new opplication and opens new connection, preceding Platform terminates using corresponding connection, and user have selected an application, the background application pair that user chooses from background application set The connection answered also needs to transmit data.Then have:
(1)
When user j new be connected to up to speed is obeyed is λuPoisson process when, then have:
Obeying average when the service duration that user j foreground application correspondence is connected isExponential distribution when, then have:
When user j chooses an application in equiprobable mode from background application set is switched to foreground, then have:
WhereinFor the set of user j all background applications,Represent the number of all background applications of the user.
Non- running time-frequency resource conversion is connected to when the background application that user j chooses is corresponding(Expire subcarrier supply)When clothes Business duration obeys averageExponential distribution when, then have:
Wherein rpShared sub-carrier number before being connected to this running time-frequency resource conversion for this.
Finally, formula is utilized(1)Calculate Prf
(2)Calculate normalization QoE risks;
Assuming that the son that the corresponding connection i of the background application changed by running time-frequency resource is reduced altogether in next section of period tau Carrier wave supply number is ri, the data volume that the connection has currently been buffered is Si, then it is total to the data that it can be buffered when reporting next time Measure and be
Z(ri)=Si+(Ri-ri)vτ (6)
Formula(6)Middle RiNumber is supplied for the full subcarrier of the connection, v is the mean data rate of subcarrier.
Then calculating the corresponding potential QoE influence functions of the connection is:
Formula(7)Middle α, β and γ are the nonnegative constants depending on QoE measurement indexs.
Federated user focus switching probability PrfWith potential QoE influence functions Q (ri), calculate this running time-frequency resource conversion pair The QoE risk functions that the connection is produced:
G(ri)=Prf·Q(ri) (8)
And the maximum QoE risks that the connection can suffer from are:
Then corresponding normalization QoE risk functions:
By taking video flowing as an example, QoE risk functions can be represented with the form of video breaks probability, such as:
Wherein αTAnd βTIt is document " Impact of network dynamics on users ' video quality: Analytical framework and QoS provision”(Author T.H.Luan etc., is published in IEEE Trans.Multimedia,vol.12,no.1,pp.64-78,Jan.2010)In draw sign network dynamic characteristic(With Wrap arrival time relevant with the variance of service time)With the net rate of rise wrapped in buffering area(Downloading rate subtracts playback rate) Desired constant.
(3)Calculate the normalized spatial spectrum cavity function H (r that this running time-frequency resource conversion is produced by the connectioni);
H(ri)=ri/Ri(12)
(4)Calculate optimal sub-carrier number reduction amount;
According to normalized spatial spectrum cavity function H (ri) and normalization QoE risk function J (ri), find one and cause both property The r of optimum balance can be reachedi, it is that this running time-frequency resource conversion supplies number to the subcarrier that the connection is reduced.
Such as both weighting function φ (r of definitioni)=H(ri)-w·J(ri), wherein ω>0 is a weight factor.It is logical φ (r can be caused by crossing solving-optimizing problem and finding onei) the maximum r of valuei, it is that this running time-frequency resource conversion is dropped to the connection Low subcarrier supply number.
By taking video flowing as an example, solve:
(13)
s.t. 0≤ri≤Ri,ri∈Ζ
Wherein Z represents nonnegative integer domain, can obtain:
WhereinRepresent in no 0≤r of constraintsi≤RiAnd ri During ∈ Ζ, make function phi (ri) obtain maximum when riValue.
Embodiment two
The technical scheme that the present invention is provided can also be realized using the device shown in Fig. 3.Device includes:
User information collection module:The terminal device of user is connected, the behavioural information of user is periodically gathered(Including with Application number that family is opened, after each is temporarily not concerned with using corresponding connection type, user foreground application of interest, user Platform application, the data volume that the corresponding connection of each background application has been buffered and the foreground application with recovery request);
User profile reporting module:Connect user information collection module, the user that user information collection module collection is arrived Behavioural information is reported to network resource manager according to feedback mechanism;
Data processing module:Network resource manager is connected, the user behavior being collected into according to network resource manager is believed Cease, the corresponding connection of the application being not concerned with to user calculates its subcarrier changed in next period tau planted agent by running time-frequency resource Number;
Resource distribution module:Connect data processing module, the connection to that can be changed by running time-frequency resource, according to data processing The result of calculation of module, distributes it corresponding subcarrier;To there is the connection of recovery request, recover to supply its full subcarrier Should;User is being paid close attention to and any operation is not taken in the corresponding connection of application in full subcarrier supply status then.
Data processing module further comprises:
User's focus switching probability calculating sub module:Network resource manager is connected, calculating is used in next section of period tau The focus at family from foreground application be switched to background application set in any application corresponding to probability Prf
User's QoE Risk Calculation submodules:Network resource manager is connected, calculates in next section of period tau that user can be with The normalization QoE risks born by the connection that running time-frequency resource is changed;
User's frequency spectrum cavity-pocket manufacture efficiency calculating sub module:Network resource manager is connected, is calculated in next section of period tau Normalized spatial spectrum cavity produced by the connection that user can be changed by running time-frequency resource;
Sub-carrier number reduction amount calculating sub module:Connect user's focus switching probability calculating sub module, user's QoE risks Calculating sub module and user's frequency spectrum cavity-pocket manufacture efficiency calculating sub module, the connection that calculating can be changed by running time-frequency resource are dropped Low sub-carrier number optimum value.
The beneficial effect of present embodiment technical scheme can be verified by emulating.
What Fig. 4 was represented is to work as average user number in total number of sub-carriers R=78, cell to beKeep for money The connection of source recovery request and switched into from neighbor cell the cell connection sub-carrier number(Subchannel)Number is Ra=16, Individually keep for the sub-carrier number for the connection asked with resource recovery(Subchannel)Number is RbWhen=8, network load can to three kinds Talk about the QoS performances of rank(New connection request blocking probability PNB, switching connection request call drop probability PHD, recover connection request call drop Probability PRD)Influence.It can be seen that when network load is from gently to when changing again(Each user's i.e. shown in figure Newly it is connected up to speed λuWhen changing from small to large), three kinds of probability can all increase, but recover connection request cutting off rate and protect all the time Hold the degree of about two orders of magnitude lower than other two kinds of probability.
The contrast for the maximum number of user that network can be accommodated before and after the use running time-frequency resource switch technology that Fig. 5 is represented, here In order to increase the probability for alloing network to accommodate more users number, the total number of sub-carriers of cell is increased to R=320 by us.Full Sufficient PHD≤ 0.02 and PRDIn the case of≤0.001, to minimize PNBR is set using the method for traversal search for targetaAnd Rb。 From the figure, it can be seen that when network load gradually becomes weight, in spite of running time-frequency resource switch technology is used, cell can be supported Maximum number of user all taper into, this can consume more frequency spectrum resources mainly due to each user.But, when employing After running time-frequency resource switch technology, especially when the load of network tends to be heavy, cell can always be accommodated more than original Number of users.
Obviously, those skilled in the art can carry out various changes and the essence deformed without departing from the present invention to the present invention God and scope.So, if these modification and variation of the present invention belong to the scope of the claims in the present invention and its equivalent technologies Within, then the present invention is also intended to comprising including these changes and deformation.The present invention is by frequency with being based on exemplified by time resource conversion The artificial frequency spectrum cavity-pocket acquisition methods of user behavior driving;In the case where not changing the essence of the present invention, it can be obtained using same method Yard resource is able to change to obtain the artificial code resource acquiring method driven based on user behavior with time resource.

Claims (5)

1. the artificial frequency spectrum cavity-pocket acquisition methods driven in a kind of cellular network based on user behavior, including:
The periodically business information in collection base station institute coverage cell in each client terminal system, and then by these business information It is reported to the network resource manager of base station;Network resource manager is according to described business information in each client terminal system Background application carry out running time-frequency resource conversion, to obtain artificial frequency spectrum cavity-pocket, detailed process is as follows:
(1) for any user, by each background application in historical statistics analysis acquisition present terminal system in next cycle The probability of happening of buffered data is still needed to, specific algorithm is as follows:
Keep record part E1、E2、E3And E4Respectively in hypomere period tau, event E1Represent that user opens new opplication and opens new company Connect, event E2Represent that the corresponding connection of foreground application terminates, event E3Represent that user have selected one from background application set and answer With event E4Represent that the corresponding connection of background application that user chooses also needs to transmit data;The described probability of happeningWherein:
When new be connected to up to speed is obeyed of user is λuPoisson process when, then have
It is 1/ μ to obey average when the service duration that the foreground application correspondence of user is connectedj2Exponential distribution when, then have
When user chooses an application in equiprobable mode from background application set is switched to foreground, then have Represent the number of all background applications of the user;
The service duration being connected to when the background application that user chooses is corresponding when non-running time-frequency resource is changed obeys average into 1/ μj1 Exponential distribution when, then haverpShared son before being connected to this running time-frequency resource conversion for this Carrier number, rj1Sub-carrier number shared when non-running time-frequency resource is changed is connected to for this;
The non-running time-frequency resource conversion i.e. correspondence represents full subcarrier supply;
(2) according to the described probability of happening, the son that each background application should be cut down in present terminal system is determined by below equation Carrier wave supplies number, so as to carry out subcarrier recovery to background application in next cycle and the foreground asked with resource recovery is answered Supplied with full subcarrier is provided, the described foreground application that there is resource recovery to ask be in the previous cycle running background and The application of front stage operation is switched in current period;
m a x r i φ ( r i )
s.t. 0≤ri≤Ri φ(ri)=H (ri)-ω·J(ri)
Wherein:riNumber, R are supplied for the subcarrier that i-th of background application in present terminal system should be cut downiFor present terminal system The full subcarrier supply number of i-th of background application in system, H (ri) it is on riThe empty function of normalized spatial spectrum, J (ri) be On riNormalization QoE risk functions, ω is weight factor;I be natural number and 1≤i≤n, n be in present terminal system after The number of platform application;
Described normalized spatial spectrum cavity function H (ri) tried to achieve according to relationship below:
H(ri)=ri/Ri
Described normalization QoE risk function J (ri) tried to achieve according to relationship below:
J ( r i ) = P i αe - β Z ( r i ) + γ α + γ
Wherein:α, β and γ are default QoE measurement indexs coefficient, PiExist for i-th of background application in present terminal system Next cycle still needs to the probability of happening of buffered data, Z (ri) adopted for i-th of background application in present terminal system next time Buffered data total amount when collection is reported;
Described buffered data total amount Z (ri) tried to achieve according to relationship below:
Z(ri)=Si+(Ri-ri)vτ
Wherein:τ is collection period, and v is the mean data rate of subcarrier, SiFor i-th of background application in present terminal system Amount of buffered data;
(3) according to step (1) and (2), all users in traversal base station institute coverage cell.
2. artificial frequency spectrum cavity-pocket acquisition methods according to claim 1, it is characterised in that:Described business information includes closing Foreground application and the information of background application, the amount of buffered data of each background application run in active user's terminal system And the foreground application asked with resource recovery.
3. artificial frequency spectrum cavity-pocket acquisition methods according to claim 1, it is characterised in that:Described network resource manager According to application corresponding connection distribution subcarrier of the following priority for each user:The application asked with resource recovery is corresponding The corresponding connection of application that the connection > intra-cell users that connection > switches into this cell from neighbor cell are newly opened;
The described application that there is resource recovery to ask is corresponding to be connected as with being in the previous cycle base station institute the coverage cell in Running background and switch to the corresponding connection of the application of front stage operation in current period;
The described application for switching into the connection of this cell from neighbor cell to have been opened by user is corresponding to be connected and the use Family is moved into base station institute coverage cell from neighbor cell;
What the corresponding user being connected as in base station institute coverage cell of application that described intra-cell users are newly opened newly opened should With corresponding connection.
4. artificial frequency spectrum cavity-pocket acquisition methods according to claim 3, it is characterised in that:Described network resource manager According to application corresponding connection distribution subcarrier of the following strategy for each user:
If the corresponding sub-carrier number taken that connects of all applications reaches R-R in current base station institute coverage cella, then Internet resources The corresponding connection distribution subcarrier of application that manager is no longer newly opened for intra-cell users;
If the corresponding sub-carrier number taken that connects of all applications reaches R-R in current base station institute coverage cellb, then Internet resources Manager is no longer that the connection for switching into this cell from neighbor cell distributes subcarrier;
For the corresponding connection of application asked with resource recovery, as long as network resource manager also has remaining unappropriated son Carrier wave is just such using corresponding connection distribution subcarrier;
If user has multiple while the application opened, the subcarrier shared by the corresponding connection of application of FEFO can be kept for This user has the corresponding connection of application that resource recovery is asked;
Wherein:R is the specified supply number of subcarrier of base station institute coverage cell, RaTo be kept in R with answering that resource recovery is asked The subcarrier that the connection of this cell is switched into corresponding connection and from neighbor cell supplies number, RbIndividually to be reserved in R Subcarrier to the corresponding connection of application asked with resource recovery supplies number.
5. the artificial frequency spectrum cavity-pocket acquisition device driven in a kind of cellular network based on user behavior, it is characterised in that:Including with The network resource manager of family information acquisition module, user profile reporting module and base station;
Described user profile acquisition module and the terminal device of user are connected, and it is used to periodically gather client terminal system In business information;
Described user profile reporting module is used to report the business information of each client terminal system in base station institute coverage cell Network resource manager to base station;
Described network resource manager includes:
Message processing module, for obtaining in current each client terminal system each background application under by historical statistics analysis One cycle still needed to the probability of happening of buffered data;And then current each user is calculated by below equation according to the described probability of happening The subcarrier supply number that each background application should be cut down in terminal system;
m a x r i φ ( r i )
s.t. 0≤ri≤Ri φ(ri)=H (ri)-ω·J(ri)
H(ri)=ri/Ri Z(ri)=Si+(Ri-ri)vτ
Wherein:riNumber, R are supplied for the subcarrier that i-th of background application in active user's terminal system should be cut downiFor current use The full subcarrier supply number of i-th of background application in the terminal system of family, ω is weight factor;I is natural number and 1≤i≤n, n For the number of background application in active user's terminal system;α, β and γ are default QoE measurement indexs coefficient, PiTo be current I-th of background application in client terminal system still needs to the probability of happening of buffered data in next cycle;τ is collection period, and v is The mean data rate of subcarrier, SiFor the amount of buffered data of i-th of background application in active user's terminal system;
Resource distribution module, the subcarrier for that should be cut down according to background application supplies number, by dispatching in next cycle to each Background application in client terminal system carries out subcarrier recovery and provides full son to the foreground application asked with resource recovery Carrier wave is supplied;
Described business information include on run in active user's terminal system foreground application and the information of background application, The amount of buffered data of each background application and the foreground application asked with resource recovery;Described has resource recovery request Foreground application be in the previous cycle be running background and to switch to the application of front stage operation in current period;
Described message processing module obtains each background application in current each client terminal system by historical statistics analysis and existed Next cycle still needs to the probability of happening of buffered data, and specific algorithm is as follows:
For any user, account part E1、E2、E3And E4Respectively in hypomere period tau, event E1Represent that user opens new opplication Open new connection, event E2Represent that the corresponding connection of foreground application terminates, event E3Represent user from background application set It has selected an application, event E4Represent that the corresponding connection of background application that user chooses also needs to transmit data;Described event ProbabilityWherein:
When new be connected to up to speed is obeyed of user is λuPoisson process when, then have
It is 1/ μ to obey average when the service duration that the foreground application correspondence of user is connectedj2Exponential distribution when, then have
When user chooses an application in equiprobable mode from background application set is switched to foreground, then have Represent the number of all background applications of the user;
The service duration being connected to when the background application that user chooses is corresponding when non-running time-frequency resource is changed obeys average into 1/ μj1 Exponential distribution when, then haverpShared son before being connected to this running time-frequency resource conversion for this Carrier number, rj1Sub-carrier number shared when non-running time-frequency resource is changed is connected to for this;
The non-running time-frequency resource conversion i.e. correspondence represents full subcarrier supply.
CN201410004408.6A 2014-01-02 2014-01-02 The artificial frequency spectrum cavity-pocket acquisition methods and device driven in a kind of cellular network based on user behavior Active CN103781101B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410004408.6A CN103781101B (en) 2014-01-02 2014-01-02 The artificial frequency spectrum cavity-pocket acquisition methods and device driven in a kind of cellular network based on user behavior

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410004408.6A CN103781101B (en) 2014-01-02 2014-01-02 The artificial frequency spectrum cavity-pocket acquisition methods and device driven in a kind of cellular network based on user behavior

Publications (2)

Publication Number Publication Date
CN103781101A CN103781101A (en) 2014-05-07
CN103781101B true CN103781101B (en) 2017-07-21

Family

ID=50572812

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410004408.6A Active CN103781101B (en) 2014-01-02 2014-01-02 The artificial frequency spectrum cavity-pocket acquisition methods and device driven in a kind of cellular network based on user behavior

Country Status (1)

Country Link
CN (1) CN103781101B (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102487324A (en) * 2010-12-02 2012-06-06 中国移动通信集团吉林有限公司 Customer information sending method and device and system thereof
CN102880676A (en) * 2012-09-10 2013-01-16 新浪网技术(中国)有限公司 Statistical method for user behavior data and user behavior data statistical system
CN103024825A (en) * 2011-09-26 2013-04-03 中国移动通信集团公司 Method and device of distributing network source among multiple applications of terminal

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102487324A (en) * 2010-12-02 2012-06-06 中国移动通信集团吉林有限公司 Customer information sending method and device and system thereof
CN103024825A (en) * 2011-09-26 2013-04-03 中国移动通信集团公司 Method and device of distributing network source among multiple applications of terminal
CN102880676A (en) * 2012-09-10 2013-01-16 新浪网技术(中国)有限公司 Statistical method for user behavior data and user behavior data statistical system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
"Context-Aware Resource Allocation to Improve the Quality of Service of Heterogeneous Traffic";Magnus Proebster, Matthias Kaschub,Stefan Valentin;《IEEE》;20110609;全文 *

Also Published As

Publication number Publication date
CN103781101A (en) 2014-05-07

Similar Documents

Publication Publication Date Title
US8130655B2 (en) Systems and methods for network congestion management using radio access network congestion indicators
CN105407486B (en) A kind of method and device of the network capacity extension
CN103119899B (en) Reduce the method and apparatus of data transfer overhead
US20180098332A1 (en) Resource allocation method, apparatus, and system, and base station
CN112868265B (en) Network resource management method, management device, electronic equipment and storage medium
CN106095529B (en) A kind of carrier wave emigration method under C-RAN framework
Zhou et al. Learning from peers: Deep transfer reinforcement learning for joint radio and cache resource allocation in 5G RAN slicing
CN109996248A (en) Electronic equipment and method and computer readable storage medium for wireless communication
CN104980934A (en) Method and device for scheduling multi-service resources
CN106775990A (en) Request scheduling method and device
CN107027134A (en) A kind of user-defined radio communication network side method and system
CN105227396B (en) A kind of inferior commending contents dissemination system and its method towards mobile communications network
CN108076531A (en) A kind of wireless network section resource dynamic distributing method of multi-service oriented business
CN114007225A (en) BWP allocation method, apparatus, electronic device and computer readable storage medium
Alotaibi et al. Linking handover delay to load balancing in SDN-based heterogeneous networks
CN105744592A (en) Service Access Method, Device and System Based on Heterogeneous Wireless Network
CN102752757A (en) Method for optimizing frequency spectrum allocation according to minimal waste criterion in frequency spectrum aggregation process
CN106211290A (en) The power-economizing method in Base-Band Processing pond under a kind of C RAN framework
CN101541086B (en) Method and device for distributing carrier number
CN103781101B (en) The artificial frequency spectrum cavity-pocket acquisition methods and device driven in a kind of cellular network based on user behavior
CN107005849A (en) A kind of cell power is shared and method of adjustment and base station
Chang et al. Resource allocation for wireless virtualized hetnet with caching and hybrid energy supply
CN107846702A (en) Strengthen the resource allocation method of Physical Downlink Control Channel
Liang et al. Video rate adaptation and traffic engineering in mobile edge computing and caching-enabled wireless networks
CN101945448A (en) Service quality control method and equipment

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant