CN102413570B - Base station in a kind of wireless communication system and resource regulating method - Google Patents

Abstract

The invention provides base station and resource regulating method in a kind of wireless communication system. This base station comprises: logic link control module, for upgrading the residue token quantity of the each user i administering this base station, offers media Access Control module by the residue token quantity of described user i; Described media Access Control module, determines scheduling priority factor for the residue token quantity according to described user i, and according to one or more scheduling priority factor of determining, described not scheduling resource block k is distributed to corresponding user. Adopt base station provided by the invention and method, resource is distributed in to a certain extent and more optimizes.

Description

Base station in a kind of wireless communication system and resource regulating method

Technical field

The present invention relates to wireless communication technology, base station and resource regulating method in espespecially a kind of wireless communication system.

Background technology

In wireless communication system, the service quality (QualityofService, QoS) of broad sense refer to network to carryOne for the grade of service (serviceclass) is promised to undertake. The definition that it may be noted that service quality can be a relative concept(for example definition based on priority) can be also an absolute probability (the specific performance parameter of for example service Network Based).

In the present invention, service quality can refer to that in radio communication, each user (UE) wants the one of data throughoutAsk. Particularly, this requirement can be described as: under the prerequisite requiring at all users' of As soon as possible Promising Policy data throughout, makeThe average throughput of system maximizes, and its mathematical expression is:

1, object function: $max\underset{n=1}{\overset{N}{\Σ}}\underset{k=1}{\overset{K}{\Σ}}\underset{i=1}{\overset{T_w}{\Σ}}{r}_{k,n}\[i\]\·a_{k,n}\[i\]$

2, constraints:

(1) $\underset{n=1}{\overset{N}{\Σ}}\underset{i=1}{\overset{T_w}{\Σ}}{r}_{k,n}\[i\]\·a_{k,n}\[i\]\≥R_{QoS}\[k\],\∀k\∈K$

(2) $\underset{k=1}{\overset{K}{\Σ}}{a}_{k,n}\[i\]=1,\∀n,is.t.1\≤n\≤N,1\≤i\≤T_w$

(3)

Wherein, r_k,n[i]＝r′_k,n[i]·F_signal·F_payload·(1-BLER)。

In above-mentioned mathematical expression, N is sets of sub-channels, and K is user's set, T_wFor scheduling window size; r_k,n[i] isHigh level (such as Layer3) the equivalent data handling capacity of user k on the subchannel n of i frame (or time slot); a_k,n[i] is booleanVariable, characterizes subchannel n and whether distributes to user k at i frame; R_QoS[k] is the qos requirement of the average throughput of user k;r′_k,n[i] is the physical layer actual data throughput of user k on the subchannel n of i frame; F_signalFor the packet in systemScale parameter, is generally 70%; F_payloadFor the proportionality coefficient of True Data load in high level data bag, be generally 96%;BLER is user's average Block Error Rate.

Requirement described above is a problem for service quality to be solved in wireless communication system. Normal conditionsUnder, this problem solves by the optimization of dispatching method, and token bucket algorithm (TokenBucketAlgorithm, TBA) is exactlyA kind of traditional dispatching method.

Particularly, token bucket algorithm is the application to token concept, in 3GPP standardization, have corresponding specification (such as3GPPTS23.107) it is defined. The realization flow of TBA comprises: token arrives user according to given pace, and user willOriginal token quantity adds that newly-increased token number measures current token quantity (or claiming residue token quantity). When user needs transmissionWhen packet, token quantity required data to be transmitted bag and user's current token quantity is compared, by comparative resultWhether determination data bag transmits. Particularly, TBA has two kinds of implementations:

1, maximum rate token (MaximumbitrateToken). In which, what token characterized is that user canThe maximum amount of data (or being called quota) of transmission, therefore the required token quantity of data to be transmitted bag can not be greater than user's residueToken quantity. If be greater than, the transmission request of packet is rejected; If be less than or equal to, packet is transmitted, correspondingThe token quantity that transmit packet uses need to be deducted in ground from user's token bucket.

2, guaranteed rate (Guaranteedbitrate, GBR) token. In which, token characterize be user extremelyFew data volume that needs transmission, i.e. user's guaranteed rate requirement, and residue token quantity in this user's token bucket is mainFor the current performance to user's guaranteed rate of characterization system, system should consume all users' guarantee speed as far as possibleRate token. Therefore,, even if user's residue token quantity is 0 to be even less than 0, also there is the possibility of transmission packet.

Fig. 1 is a rough schematic of TBA realization flow. Wherein, L1, L2, L3 be respectively user time point t1,T2, the required token quantity of t3 data to be transmitted bag; B is token bucket size, stores the upper limit of token quantity; V is token speedRate, common token rate has " maximum rate ", " guaranteed rate " etc. It may be noted that token is to continue to arrive, packetIt is uncertain arriving.

Can find out, at time point t1, due to L1 < TBC, user this data packet transmission require can be met,Through above-mentioned transmission, TBC reduces to b-L1. At time point t2, TBC increases to b-L1+v* Δ T, and now user occurs that again packet passesDefeated requirement, due to L2<TBC, this data packet transmission requires still can be met. But, at time point t3, due to L3>TBC, for maximum rate token mode, this data packet transmission requires cannot meet.

In addition, can also adopt other traditional dispatching methods, such as max carrier to interference (MaxCI) algorithm, wheel are named scheduling(Round-Robin, RR) algorithm, equitable proportion (ProportionalFairness, PF) algorithm etc.

In max carrier to interference algorithm, system by have max carrier to interference resource block assignments give corresponding user. The partyMethod can maximize the handling capacity of system, but there is no the consideration of fairness aspect completely.

Cry in dispatching algorithm at wheel, system, according to certain permanent order, is given each user by resource block assignments, no successivelyThe resource quantity that same user obtains equates. The method from distribute resource quantity aspect ensure absolute fairness (andThe fairness of non-handling capacity), but owing to not considering the information of any channel quality, therefore the handling capacity of system is conventionally lower.

In equitable proportion algorithm, system withMode calculate preferential at each Resource Block of each userLevel, then gives corresponding user by resource block assignments the highest priority. Wherein, P_i,jBe j Transmission Time IntervalEquitable proportion (PF) factor of user i in (TransmissionTimeInterval, TTI); r_i,jBe user in j TTIThe estimation handling capacity (or data estimator transmission quantity) of i; R_i,jBe that (or average reportedly for the average throughput of user i in j TTIThroughput rate). The effect of the method is to ensure under the prerequisite of handling capacity fairness, by the resource block assignments with high throughput toCorresponding user. This is a kind of method between MaxCI and RR, due to simple and practical and usually adopt at engineering field.

In practical application, also several traditional dispatching methods can be combined with.

Such as, in scheduling of resource, equitable proportion algorithm is combined with token restriction, max carrier to interference algorithm, this scheduling pointIt is two stages.

First,, in whole user-Resource Block PF factor matrix, get rid of token quantity and be 0 or be less than 0 and buffer memory(buffer) user of countless certificates in, and find PF factor the greater, user-Resource Block unit of answering according to this PF factor pair entersRow distributes, by resource block assignments corresponding to this element to user corresponding to this element. After every sub-distribution, by each user i'sAverage throughput R_i,jUpgrade, then recalculate this user-Resource Block PF factor matrix. It may be noted that PF in the method because ofSon is traditional PF factor, that is: $P_{i,j}=\frac{r_{i,j}}{R_{i,j}}.$

Secondly,, after having all users' the token of data all to consume in buffer, in all buffer, there are dataUser all may be excluded outside PF scheduling time, proceed to the MaxCI stage, namely distribute according to the principle of max carrier to interferenceRemaining Resource Block. Certainly, distribution now also can be considered the remaining data amount of buffer in the lump.

The problem of the method is, the user that token has just reached guaranteed rate for getting rid of some, i.e. only impactParticipate in user's set of PF scheduling, and can not be embodied in the priority calculating of resource distribution. Therefore, the method is divided resourceJoin and can not more optimize, thereby occur the waste of resource.

In addition, also can adopt equitable proportion algorithm+token restriction+max carrier to interference algorithm of considering buffer to provideSource scheduling. The method is similar with " equitable proportion algorithm+token restriction+max carrier to interference algorithm ", and difference is that it is at PFWhen the factor is calculated, can consider the remaining data amount of buffer, resource be distributed more excellent. , the PF factor in the method isWherein Buffer_i,jIt is the buffer memory remaining data amount of user i in j TTI.

But, because the PF factor adopting in the method is not considered the service condition of token therefore can cause resource yetDistribution can not more be optimized, and can waste equally resource.

Summary of the invention

The invention provides base station and resource regulating method in a kind of wireless communication system, so as to resource distribute intoRow is optimized.

Particularly, the invention provides the base station in a kind of wireless communication system, comprising:

Logic link control module, for upgrading the residue token quantity of the each user i administering this base station, described in inciting somebody to actionThe residue token quantity of user i offers media Access Control module;

Described media Access Control module, for not dispatching each according to j the each user i of Transmission Time IntervalThe residue token quantity of the estimation handling capacity of Resource Block k and described user i, and the buffer memory remaining data amount of each user i, reallyDetermine scheduling priority factor, and according to one or more scheduling priority factor of determining, described not scheduling resource block k is dividedThe user that dispensing is corresponding, wherein, in described scheduling priority factor and j Transmission Time Interval, each user i is eachThe buffer memory remaining data amount of the estimation handling capacity of scheduling resource block k and/or each user i is directly proportional.

Further, described media Access Control module is used for:

According to each user i in j Transmission Time Interval the estimation handling capacity of each not scheduling resource block k and described inThe residue token quantity of user i, and average throughput and the each user i of each user i in this j Transmission Time IntervalBuffer memory remaining data amount, determine one or more scheduling priority factor.

Further, described media Access Control module is used for: according to formula $P_{i,j,k}=\frac{min({\mathrm{Token}}_{i,j},{\mathrm{Buffer}}_{i,j})\&CenterDot;r_{i,j,k}}{B_{Bucket\_size}\&CenterDot;R_{i,j}},k\&Element;U$ Determine equitable proportion factor P_i,j,kAs described dispatching priority because ofSon;

Wherein, Token_i,jFor the residue token quantity of each user i; Buffer_i,jFor the buffer memory remainder of each user iAccording to amount; B_{Bucket_size}For token bucket size; r_i,j,kBe in j Transmission Time Interval each user i at each not scheduling resourceThe estimation handling capacity of piece k; R_i,jIt is the average throughput of each user i in j Transmission Time Interval; U is scheduling resource block notSet.

Described media Access Control module is further used for: determine described scheduling priority factor according to token bucket size.

Described media Access Control module is further used for:

From one or more scheduling priority factor of determining, find maximum scheduling priority factor, judge this maximumWhether scheduling priority factor is greater than 0, if be greater than 0, by first resource piece k ' corresponding this maximum scheduling priority factor pointThe first user i ' that this maximum scheduling priority factor of dispensing is corresponding, carries out more the residue token quantity of this first user i 'Newly, and scheduling resource set of blocks is not upgraded, described first resource piece k ' is never deleted in scheduling resource set of blocks.

Described media Access Control module is further used for:

Described in judging, whether scheduling resource set of blocks is not empty, if be not empty, according to the residue token quantity after upgradingRedefine scheduling priority factor, and according to the one or more scheduling priority factor that redefine out by upgrade after notThe not scheduling resource block k comprising in scheduling resource set of blocks distributes to corresponding user.

Described media Access Control module is further used for:

First resource piece k ' corresponding maximum scheduling priority factor is distributed to this maximum scheduling priority factor correspondenceFirst user i ' time, the buffer memory remaining data amount of this first user i ' is upgraded, and according to the buffer memory residue after upgradingData volume redefines scheduling priority factor.

Described media Access Control module is further used for:

Be not more than at 0 o'clock in described maximum scheduling priority factor, be dispatching algorithm according to max carrier to interference algorithm or wheelDescribed first resource piece k ' is distributed to corresponding user; Or, according to formulaDetermine the second equitable proportion factorP′_i,j, and according to the second equitable proportion factor P ' determining_i,jDescribed first resource piece k ' is distributed to corresponding user;

Wherein, described r_i,jBe in j Transmission Time Interval each user i in the estimation handling capacity of first resource piece k ';R_i,jIt is the average throughput of each user i in j Transmission Time Interval.

Described media Access Control module is further used for:

In the time that described not scheduling resource set of blocks is sky, determine that in j Transmission Time Interval, the actual of each user i gulps downThe amount of telling, provides it to described logic link control module, and determines each user i's in j+1 Transmission Time IntervalAverage throughput.

Described logic link control module is further used for: in the time that j Transmission Time Interval starts, upgrade this base station instituteThe residue token quantity of each user i of administration;

After the goodput that receives each user i and before this j Transmission Time Interval finish, according to described everyThe goodput of individual user i upgrades the residue token quantity of described user i.

Particularly, the invention provides the resource regulating method in a kind of radio communication, comprising:

The residue token quantity of the each user i administering is determined in base station;

This base station according to each user i in j Transmission Time Interval in the estimation handling capacity of each not scheduling resource block kWith the residue token quantity of described user i, and the buffer memory remaining data amount of each user i, determine scheduling priority factor, andAccording to one or more scheduling priority factor of determining, described not scheduling resource block k is distributed to corresponding user, wherein,In described scheduling priority factor and j Transmission Time Interval, each user i handles up in the estimation of each not scheduling resource block kThe buffer memory remaining data amount of amount and/or each user i is directly proportional.

Described this base station gulps down in the estimation of each not scheduling resource block k according to each user i in j Transmission Time IntervalThe residue token quantity of the amount of telling and described user i, and the buffer memory remaining data amount of each user i, determine dispatching priority because ofAttached bag is drawn together:

According to each user i in j Transmission Time Interval the estimation handling capacity of each not scheduling resource block k and described inThe residue token quantity of user i, and average throughput and the each user i of each user i in this j Transmission Time IntervalBuffer memory remaining data amount, determine one or more scheduling priority factor.

Described this base station gulps down in the estimation of each not scheduling resource block k according to each user i in j Transmission Time IntervalThe residue token quantity of the amount of telling and described user i, and the buffer memory remaining data amount of each user i, determine dispatching priority because ofAttached bag is drawn together: according to formula $P_{i,j,k}=\frac{min({\mathrm{Token}}_{i,j},{\mathrm{Buffer}}_{i,j})\&CenterDot;r_{i,j,k}}{B_{Bucket\_size}\&CenterDot;R_{i,j}},k\&Element;U$ Determine equitable proportion factor P_i,j,kAsDescribed scheduling priority factor;

Wherein, Token_i,jFor the residue token quantity of each user i; Buffer_i,jFor the buffer memory remainder of each user iAccording to amount; B_{Bucket_size}For token bucket size; r_i,j,kBe in j Transmission Time Interval each user i at each not scheduling resourceThe estimation handling capacity of piece k; R_i,jIt is the average throughput of each user i in j Transmission Time Interval; U is scheduling resource block notSet.

Described this base station gulps down in the estimation of each not scheduling resource block k according to each user i in j Transmission Time IntervalThe residue token quantity of the amount of telling and described user i, and the buffer memory remaining data amount of each user i, determine dispatching priority because ofSon further comprises: determine described scheduling priority factor according to token bucket size.

The method further comprises:

From one or more scheduling priority factor of determining, find maximum scheduling priority factor;

Judge whether this maximum scheduling priority factor is greater than 0, if be greater than 0, by this maximum scheduling priority factor pairThe first resource piece k ' answering distributes to first user i ' corresponding to this maximum scheduling priority factor, to remaining of this first user i 'Remaining token quantity is upgraded, and scheduling resource set of blocks is not upgraded, and described first resource piece k ' is never dispatched to moneyIn the set of blocks of source, delete.

The method further comprises:

Described in judging, whether scheduling resource set of blocks is not empty, if be not empty, according to the residue token quantity after upgradingRedefine scheduling priority factor, and according to the one or more scheduling priority factor that redefine out by upgrade after notThe not scheduling resource block k comprising in scheduling resource set of blocks distributes to corresponding user.

The method further comprises:

First resource piece k ' corresponding maximum scheduling priority factor is distributed to this maximum scheduling priority factor correspondenceFirst user i ' time, the buffer memory remaining data amount of this first user i ' is upgraded;

Redefine scheduling priority factor according to the buffer memory remaining data amount after upgrading.

The method further comprises:

Be not more than at 0 o'clock in described maximum scheduling priority factor, be dispatching algorithm according to max carrier to interference algorithm or wheelDescribed first resource piece k ' is distributed to corresponding user;

Or, according to formulaDetermine the second equitable proportion factor P '_i,j, and according to the second ratio of determiningFairness index P '_i,jDescribed first resource piece k ' is distributed to corresponding user; Wherein, described r_i,jBe between j transmission timeEvery in each user i in the estimation handling capacity of first resource piece k '; R_i,jBe the flat of each user i in j Transmission Time IntervalAll handling capacities.

The method further comprises:

In the time that described not scheduling resource set of blocks is sky, determine that in j Transmission Time Interval, the actual of each user i gulps downThe amount of telling, and determine the average throughput of each user i in j+1 Transmission Time Interval.

Described base station determines that the residue token quantity of the each user i administering comprises: open at j Transmission Time IntervalWhen beginning, upgrade the residue token quantity of the each user i administering this base station;

The method further comprises:

After the goodput of determining each user i in j Transmission Time Interval and between this j transmission timeBefore finishing, upgrade the residue token quantity of described user i according to the goodput of described each user i.

Adopt base station provided by the invention and method, in the time that the priority of allocation of radio resources is calculated, consider residue token numberAmount, thus realize the quality of service guarantee based on token, and make resource be distributed in to a certain extent more optimization.

Brief description of the drawings

Fig. 1 is a rough schematic of token bucket algorithm realization flow;

Fig. 2 is the realization flow figure of scheduling of resource in concrete example of the present invention;

Fig. 3 is the relatively signal that in dispatching method of the present invention and conventional method, satisfied users ratio increases with number of usersFigure;

Fig. 4 is the comparison schematic diagram that in dispatching method of the present invention and conventional method, handling capacity increases with number of users.

Detailed description of the invention

For making object of the present invention, technical scheme and advantage clearer, referring to the accompanying drawing embodiment that develops simultaneously, rightThe present invention further describes.

The invention provides the resource regulating method in a kind of radio communication, comprising:

The residue token quantity of the each user i administering is determined in base station;

Scheduling priority factor is determined according to the residue token quantity of described user i in this base station, and according to determine oneDescribed not scheduling resource block k is distributed to corresponding user by individual or multiple scheduling priority factor.

Fig. 2 is the realization flow figure of scheduling of resource in concrete example of the present invention, comprising:

Step 201: j TTI starts, each user i's that base station is administered it according to new token arrival situationToken bucket upgrades, and determines the residue token quantity Token of each user i_i,j。

When specific implementation, this step can be by logic link control (LogicLinkControl, the LLC) layer in base stationCarry out, or carried out by media Access Control (MediaAccessControl, the MAC) layer in base station. It may be noted that LLCLayer and MAC layer are arranged in base station two independently logic modules, and therefore LLC layer can be described as again logic link control mouldPiece, MAC layer can be described as again media Access Control module.

Have, the renewal described in step 201 can be that original token quantity is added to newly-increased token number measures certain use againThe current token quantity at family.

Step 202: the Resource Block S set of having dispatched is set to sky (null), arranges the Resource Block set U of not schedulingFor comprising all Resource Block. This step can be completed by the media Access Control module in base station.

It may be noted that step 201-202 is the initial setting up of base station before carrying out scheduling of resource.

Step 203: determine in this j TTI that user i is at the scheduling priority factor P of scheduling resource block k not_i,j,k. OneIn individual specific implementation, can be according to formula $P_{i,j,k}=\frac{min({\mathrm{Token}}_{i,j},{\mathrm{Buffer}}_{i,j})\&CenterDot;r_{i,j,k}}{B_{Bucket\_size}\&CenterDot;R_{i,j}},k\&Element;U$ Determine described P_i,j,k, now shouldP_i,j,kA kind of equitable proportion (PF) factor, but different from traditional PF factor.

Wherein, Token_i,jCurrent residue token quantity in token bucket for user i; Buffer_i,jFor user i is at buffer memoryIn current remaining data amount; r_i,j,kBe the estimation handling capacity of user i on scheduling resource block k not in j TTI; R_i,jBeThe average throughput of user i in j TTI, this parameter after the scheduling of resource of j-1 TTI completes by media Access Control mouldPiece calculates; B_{Bucket_size}For token bucket size, it is generally constant. It may be noted that all users under normal circumstancesB_{Bucket_size}Be the same, but also have an exception, such as for 2G mobile phone and 4G cell phone, due to mobile phone capacity differenceGreatly, so B_{Bucket_size}Value can be different. In addition, at different TTI, R_i,j、r_i,j,kValue may be different.

Step 204: any one from any one user i among scheduling resource set of blocks U not be not on scheduling resource block kThe one or more P that determine_i,j,kThe scheduling priority factor of middle selection value maximum, according to formulaDetermine maximum scheduling priority factor P_i′,j,k′. Wherein, i ' represents and maximum schedulingThe specific user that priority factors is corresponding, k ' represents a specific not scheduling money corresponding with maximum scheduling priority factorSource piece.

In this step, select the scheduling priority factor of value maximum can adopt conventional method to realize, such as adopting one dimensionThe modes such as search, two-dimensional search repeat no more herein.

Step 205: judge P_i′,j,k′Whether be greater than 0. If be greater than 0, by this not scheduling resource block k ' distribute to userI ', and according to this user i ' data transmission capabilities on scheduling resource block k ' not to Token_i',jAnd/or Buffer_i',jCarry outUpgrade, such as at Token_i',jAnd/or Buffer_i',jIn reduce estimate transmission data volume. Like this, enter into this TTI underOne take turns when scheduling (in step 207, U does not return to execution step 203 when empty), by the Token utilizing after upgrading_i',jAnd/orBuffer_i',jDetermine the scheduling priority factor of user i '.

Further, if be not more than for 0 (being less than or equal to 0), adopt traditional dispatching method, such as according to considerationThe max carrier to interference algorithm of buffer carries out the not distribution of scheduling resource block k ', or adopts RR algorithm, or utilizes traditionalThe PF factor () etc.

In a specific implementation of this step, can only consider to process being greater than 0 situation, be not more than and ignore0 situation, does not carry out any processing for the situation that is not more than 0, directly enters step 206.

Step 206: S is updated to S+k ', U is updated to U-k '. , Resource Block k ' is added to scheduling resource block collectionClose in S, and Resource Block k ' is never deleted in scheduling resource set of blocks U.

Step 207: judge that whether U is empty, if U is not sky, returns to execution step 203. Further, if U is emptyPerform step 208.

It may be noted that step 203-207 is the main body process that scheduling of resource is carried out in base station, above-mentioned steps is by the matchmaker in base stationBody control AM access module completes.

Step 208: the goodput of determining the one or more users that administer this base station.

Step 209: upgrade the token bucket of this user i according to the goodput of each user i, so far j TTI knotBundle.

Particularly, the renewal described in step 209 can be that the current token quantity of user i is deducted to goodput institute is rightThe token quantity of answering, this step can be completed by the logic link control module in base station.

It may be noted that any one when j TTI that shown in Fig. 2, flow process is mentioned can be understood as base station and normally workTTI, any one user or each user that user i administers for base station, scheduling resource block k is not for giving for base station assignsAny one Resource Block or any one unallocated Resource Block of going out that user uses. There is again parameter Token_i,jIn differenceTTI can dynamically update (seeing step 201 and step 209), and the execution along with scheduling of resource also can be dynamic in a TTIUpgrade (seeing step 205).

In the time of specific implementation, in step 203, be various for the mode of determining scheduling priority factor, such as:

1, according to residue token quantity Token_i,j, estimate handling capacity r_i,j, average throughput R_i,j, buffer memory remaining data amountBuffer_i,j, token bucket size B_{Bucket_size}Determine one or more scheduling priority factor. Following formula (A)-be only (F)Determine the concrete example of scheduling priority factor according to which, following formula is carried out to other distortion, such as index is becomeCoefficient and/or product become add and etc., be equally also applicable to resource regulating method of the present invention.

(A) $P_{i,j}=\frac{\min ({\mathrm{Token}}_{i,j},{\mathrm{Buffer}}_{i,j})\&CenterDot;r_{i,j}}{B_{Bucket\_size}\&CenterDot;R_{i,j}}$

(B) $P_{i,j}=\frac{{\&lsqb;\min ({\mathrm{Token}}_{i,j},{\mathrm{Buffer}}_{i,j})\&rsqb;}^{\mathrm{\&alpha;}}\&CenterDot;{r_{i,j}}^{\mathrm{\&beta;}}}{{B_{Bucket\_size}}^{\mathrm{\&theta;}}{R_{i,j}}^{\mathrm{\&gamma;}}}$

(C) $P_{i,j}=\frac{{{\mathrm{Token}}_{i,j}}^{\mathrm{\&alpha;}}\&CenterDot;{{\mathrm{Buffer}}_{i,j}}^{\mathrm{\&delta;}}\&CenterDot;{r_{i,j}}^{\mathrm{\&beta;}}}{{B_{Bucket\_size}}^{\mathrm{\&theta;}}\&CenterDot;{R_{i,j}}^{\mathrm{\&gamma;}}}$

(D) $P_{i,j}=\frac{{\&lsqb;\min ({\mathrm{Token}}_{i,j},{\mathrm{Buffer}}_{i,j},r_{i,j})\&rsqb;}^{\mathrm{\&alpha;}}}{{B_{Bucket\_size}}^{\mathrm{\&theta;}}\&CenterDot;{R_{i,j}}^{\mathrm{\&gamma;}}}$

(E) $P_{i,j}=\frac{{{\mathrm{Token}}_{i,j}}^{\mathrm{\&alpha;}}\&CenterDot;\min {\&lsqb;{\mathrm{Buffer}}_{i,j},r_{i,j}\&rsqb;}^{\mathrm{\&beta;}}}{{B_{Bucket\_size}}^{\mathrm{\&theta;}}\&CenterDot;{R_{i,j}}^{\mathrm{\&gamma;}}}$

(F) $P_{i,j}=\frac{{{\mathrm{Buffer}}_{i,j}}^{\mathrm{\&alpha;}}\&CenterDot;\min {\&lsqb;{\mathrm{Token}}_{i,j},r_{i,j}\&rsqb;}^{\mathrm{\&beta;}}}{{B_{Bucket\_size}}^{\mathrm{\&theta;}}\&CenterDot;{R_{i,j}}^{\mathrm{\&gamma;}}}$

2, according to residue token quantity Token_i,j, estimate handling capacity r_i,jDetermine one or more dispatching priorities because ofSon.

3, according to residue token quantity Token_i,j, estimate handling capacity r_i,j, average throughput R_i,jDetermine one or moreScheduling priority factor.

4, according to residue token quantity Token_i,j, estimate handling capacity r_i,j, buffer memory remaining data amount Buffer_i,jDetermineOne or more scheduling priority factor.

5, according to residue token quantity Token_i,j, estimate handling capacity r_i,j, average throughput R_i,j, token bucket sizeB_{Bucket_size}Determine one or more scheduling priority factor.

6, according to residue token quantity Token_i,j, estimate handling capacity r_i,j, buffer memory remaining data amount Buffer_i,j, token bucketSize B_{Bucket_size}Determine one or more scheduling priority factor.

7, according to residue token quantity Token_i,jDetermine one or more scheduling priority factor, such as P_i,j＝Token_i,j。

In other words, while determining scheduling priority factor in step 203, considering residue token quantity Token_i,jThis ginsengOn the basis of number, can further consider and estimate handling capacity r_i,j, average throughput R_i,j, buffer memory remaining data amount Buffer_i,j、Token bucket size B_{Bucket_size}In one or more parameter. And, can determine above-mentioned ginseng according to actual application scenariosThe operational form of number, such as adopting and formula (A)-(F) similarly operational form or further distortion etc.

In addition the above-mentioned residue token quantity Token mentioning,_i,jCan determine according to guaranteed rate token mode, alsoCan determine according to maximum rate token mode. For residue token quantity Token_i,jFor the situation of maximum rate token,The scheduling priority factor of determining according to said method is carried out scheduling of resource, shows that system tends to make all users all to approachMaximum rate, the User Priority far away apart from maximum rate is higher. Like this, considering the high speed data transfer clothes of fairnessIn business, all users transfer of data under the prerequisite that meets fairness can approach maximum rate as much as possible.

Certainly, for residue token quantity Token_i,jFor the situation of maximum rate token, also can according to formula (G)-(J) or its distortion determine scheduling priority factor. In this case, it is quick that system tends to make to approach the user of maximum rateReach maximum rate, the User Priority nearer apart from maximum rate is higher, is mainly used in not considering the high speed of fairnessIn data transport service, thereby wish to utilize limited resource to meet the maximum rate transmission requirement of more users.

(G) $P_{i,j}=\frac{{{\mathrm{Buffer}}_{i,j}}^{\mathrm{\&delta;}}\&CenterDot;{r_{i,j}}^{\mathrm{\&beta;}}}{{{\mathrm{Token}}_{i,j}}^{\mathrm{\&alpha;}}\&CenterDot;{B_{Bucket\_size}}^{\mathrm{\&theta;}}\&CenterDot;{R_{i,j}}^{\mathrm{\&gamma;}}}$

(H) $P_{i,j}=\frac{\min {\&lsqb;{\mathrm{Buffer}}_{i,j},r_{i,j}\&rsqb;}^{\mathrm{\&beta;}}}{{{\mathrm{Token}}_{i,j}}^{\mathrm{\&alpha;}}\&CenterDot;{B_{Bucket\_size}}^{\mathrm{\&theta;}}\&CenterDot;{R_{i,j}}^{\mathrm{\&gamma;}}}$

(I) $P_{i,j}=\frac{{(B_{Bucket\_size}-{\mathrm{Token}}_{i,j})}^{\mathrm{\&alpha;}}\&CenterDot;{{\mathrm{Buffer}}_{i,j}}^{\mathrm{\&delta;}}\&CenterDot;{r_{i,j}}^{\mathrm{\&beta;}}}{{B_{Buc\mathrm{ke}t\_size}}^{\mathrm{\&theta;}}\&CenterDot;{R_{i,j}}^{\mathrm{\&gamma;}}}$

(J) $P_{i,j}=\frac{{(B_{Bucket\_size}-{\mathrm{Token}}_{i,j})}^{\mathrm{\&alpha;}}\&CenterDot;\min {\&lsqb;{\mathrm{Buffer}}_{i,j},r_{i,j}\&rsqb;}^{\mathrm{\&beta;}}}{{B_{Bucket\_size}}^{\mathrm{\&theta;}}\&CenterDot;{R_{i,j}}^{\mathrm{\&gamma;}}}$

It should be noted that, because a TTI is the minimum interval of scheduling, therefore resource regulating method of the present inventionCan carry out by each TTI. Certainly, also can just dispatch once by multiple TTI, such as 2 TTI dispatch one inferior. Correspondingly, do not haveThe TTI that has scheduling to indicate can continue to use nearest one the scheduling result of dispatching the TTI indicating.

The present invention also provides the base station in a kind of wireless communication system, comprising: logic link control module and media accessControl module.

Wherein, logic link control module, will for upgrading the residue token quantity of the each user i administering this base stationThe residue token quantity of described user i offers media Access Control module.

Media Access Control module is for determining scheduling priority factor according to the residue token quantity of described user i, andAccording to one or more scheduling priority factor of determining, described not scheduling resource block k is distributed to corresponding user.

In concrete example of the present invention, described media Access Control module is used for:

Estimation handling capacity and institute according to each user i in j Transmission Time Interval at each not scheduling resource block kState the residue token quantity of user i, determine one or more scheduling priority factor. Wherein, according to estimating handling capacity and surplusRemaining token quantity determines that the operational form of scheduling priority factor can be various.

In concrete example of the present invention, described media Access Control module is used for:

According to each user i in j Transmission Time Interval the estimation handling capacity of each not scheduling resource block k and described inThe residue token quantity of user i, and the average throughput of each user i in this j Transmission Time Interval, determine oneOr multiple scheduling priority factor. Wherein, determine dispatching priority according to estimation handling capacity, residue token quantity, average throughputThe operational form of the level factor can be various.

In another concrete example of the present invention, described media Access Control module is used for:

According to each user i in j Transmission Time Interval the estimation handling capacity of each not scheduling resource block k and described inThe residue token quantity of user i, and the buffer memory remaining data amount of each user i, determine one or more dispatching prioritiesThe factor. Wherein, determine the computing of scheduling priority factor according to estimation handling capacity, residue token quantity, buffer memory remaining data amountForm can be various.

In another concrete example of the present invention, described media Access Control module is used for:

According to each user i in j Transmission Time Interval the estimation handling capacity of each not scheduling resource block k and described inThe residue token quantity of user i, and average throughput and the each user i of each user i in this j Transmission Time IntervalBuffer memory remaining data amount, determine one or more scheduling priority factor. Wherein, according to estimating handling capacity, residue tokenQuantity, average throughput, buffer memory remaining data amount determine that the operational form of scheduling priority factor can be various, such as adoptingWith with similarly operational form of formula (A)-(J), just by the token bucket size B in above-mentioned formula_{Bucket_size}Remove.

Particularly, described media Access Control module is used for: according to formula $P_{i,j,k}=\frac{\min ({\mathrm{Token}}_{i,j},{\mathrm{Buffer}}_{i,j})\&CenterDot;r_{i,j,k}}{B_{Bucket\_size}\&CenterDot;R_{i,j}},k\&Element;U$ Determine equitable proportion factor P_i,j,kAs described scheduling priority factor. Wherein, Token_i,jFor the residue order of each user iBoard quantity; Buffer_i,jFor the buffer memory remaining data amount of each user i; B_{Bucket_size}For token bucket size; r_i,j,kIt is jIn Transmission Time Interval, each user i is in the estimation handling capacity of each not scheduling resource block k; R_i,jBe j Transmission Time IntervalIn the average throughput of each user i; U is scheduling resource set of blocks not.

Certainly, described media Access Control module is further used for: determine described dispatching priority according to token bucket sizeThe factor. ,, in the time determining scheduling priority factor, further consider this parameter of token bucket size.

Further, described media Access Control module is used for:

From one or more scheduling priority factor of determining, find maximum scheduling priority factor, judge this maximumWhether scheduling priority factor is greater than 0, if be greater than 0, by first resource piece k ' corresponding this maximum scheduling priority factor pointThe first user i ' that this maximum scheduling priority factor of dispensing is corresponding, carries out more the residue token quantity of this first user i 'Newly, and scheduling resource set of blocks is not upgraded, described first resource piece k ' is never deleted in scheduling resource set of blocks.

Described media Access Control module is further used for:

Described in judging, whether scheduling resource set of blocks is not empty, if be not empty, according to the residue token quantity after upgradingRedefine scheduling priority factor, and according to the one or more scheduling priority factor that redefine out by upgrade after notThe not scheduling resource block k comprising in scheduling resource set of blocks distributes to corresponding user.

Described media Access Control module is further used for:

First resource piece k ' corresponding maximum scheduling priority factor is distributed to this maximum scheduling priority factor correspondenceFirst user i ' time, the buffer memory remaining data amount of this first user i ' is upgraded, and according to the buffer memory residue after upgradingData volume redefines scheduling priority factor.

Described media Access Control module is further used for:

Be not more than at 0 o'clock in described maximum scheduling priority factor, be dispatching algorithm according to max carrier to interference algorithm or wheelDescribed first resource piece k ' is distributed to corresponding user; Or, according to formulaDetermine the second equitable proportion factorP′_i,j, and according to the second equitable proportion factor P ' determining_i,jDescribed first resource piece k ' is distributed to corresponding user;

Wherein, described r_i,jBe in j Transmission Time Interval each user i in the estimation handling capacity of first resource piece k ';R_i,jIt is the average throughput of each user i in j Transmission Time Interval.

Described media Access Control module is further used for:

In the time that described not scheduling resource set of blocks is sky, determine that in j Transmission Time Interval, the actual of each user i gulps downThe amount of telling, provides it to described logic link control module, and determines each user i's in j+1 Transmission Time IntervalAverage throughput.

Described logic link control module is further used for: in the time that j Transmission Time Interval starts, upgrade this base station instituteThe residue token quantity of each user i of administration;

After the goodput that receives each user i and before this j Transmission Time Interval finish, according to described everyThe goodput of individual user i upgrades the residue token quantity of described user i.

It may be noted that token belongs to QoS carries relevant data, is according to upper strata (EvolvedPacket by base stationSystem, EPS) qos parameter that provides while setting up QoS carrying (bearer) (comprises the maximum rate of this carrying and ensures speedRate etc.) produce. And base station is according to user's that these parameter maintenances are administered token bucket. Particularly, in base station, be responsible for maintenanceToken be logical link control layer, LLC layer is mainly responsible for the QoS associative operation such as flow-control, irrelevant with physical media.

Although and dispatching method of the present invention is also completed by base station, scheduling of resource belongs to the media access of base stationThe operation of key-course, its master is to be processed is the control relevant to physical media, as problems such as the distribution of solution physical resource. PassIn system method, the media access control layer of base station does not use token, and therefore the resource regulating method in the present invention needs baseStand the parameter relevant to token passed to media access control layer from logical link control layer, need to carry out inside of base station mouldThe cross-layer parameter of interblock is transmitted.

Fig. 3 illustrates the situation that satisfied users ratio increases with number of users, and Fig. 4 illustrates the feelings that handling capacity increases with number of usersCondition, the simulation parameter adopting is as shown in table 1. Can find out from Fig. 3 and Fig. 4, compared with conventional method, the money that the present invention proposesSource dispatching method and for the more optimally Resources allocation of base station of scheduling of resource, to meet user's rate requirement. CorrespondingGround, base station of the present invention and method can retain more resource and carry out max carrier to interference scheduling, thereby aspect handling capacity alsoThere is advantage.

Table 1

The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, all in essence of the present inventionWithin god and principle, any amendment of making, be equal to replacement, improvement etc., within the scope of protection of the invention all should be included in.

Claims (20)

1. the base station in wireless communication system, is characterized in that, comprising:

Logic link control module, for upgrading the residue token quantity of the each user i administering this base station, by described userThe residue token quantity of i offers media Access Control module;

Described media Access Control module, for according to j the each user i of Transmission Time Interval at each not scheduling resourceThe residue token quantity of the estimation handling capacity of piece k and described user i, and the buffer memory remaining data amount of each user i, determine and adjustDegree priority factors, and according to one or more scheduling priority factor of determining, described not scheduling resource block k is distributed toCorresponding user, wherein, in described scheduling priority factor and j Transmission Time Interval, each user i is in each not schedulingThe buffer memory remaining data amount of the estimation handling capacity of Resource Block k and/or each user i is directly proportional.

2. base station according to claim 1, is characterized in that, described media Access Control module is used for:

Estimation handling capacity and described user according to each user i in j Transmission Time Interval at each not scheduling resource block kThe residue token quantity of i, and the average throughput of each user i and delaying of each user i in this j Transmission Time IntervalDeposit remaining data amount, determine one or more scheduling priority factor.

3. base station according to claim 2, is characterized in that, described media Access Control module is used for: according to formulaDetermine equitable proportion factor P_i,j,kAs described dispatching priority because ofSon;

Wherein, Token_i,jFor the residue token quantity of each user i; Buffer_i,jFor the buffer memory remaining data of each user iAmount; B_{Bucket_size}For token bucket size; r_i,j,kBe in j Transmission Time Interval each user i at each not scheduling resource blockThe estimation handling capacity of k; R_i,jIt is the average throughput of each user i in j Transmission Time Interval; U is scheduling resource block collection notClose.

4. according to the base station described in claim 1-3 any one, it is characterized in that, described media Access Control module is further usedIn: determine described scheduling priority factor according to token bucket size.

5. according to the base station described in claim 1-3 any one, it is characterized in that, described media Access Control module is further usedIn:

From one or more scheduling priority factor of determining, find maximum scheduling priority factor, judge this maximum schedulingWhether priority factors is greater than 0, if be greater than 0, first resource piece k ' corresponding this maximum scheduling priority factor is distributed toThe first user i ' that this maximum scheduling priority factor is corresponding, upgrades the residue token quantity of this first user i ', andScheduling resource set of blocks is not upgraded, described first resource piece k ' is never deleted in scheduling resource set of blocks.

6. base station according to claim 5, is characterized in that, described media Access Control module is further used for:

Described in judging, whether scheduling resource set of blocks is not empty, if be not empty, according to the residue token quantity after upgrading againDetermine scheduling priority factor, and according to the one or more scheduling priority factor that redefine out by upgrade after not schedulingThe not scheduling resource block k comprising in Resource Block set distributes to corresponding user.

7. base station according to claim 6, is characterized in that, described media Access Control module is further used for:

By first resource piece k ' corresponding maximum scheduling priority factor distribute to this maximum scheduling priority factor correspondingWhen one user i ', the buffer memory remaining data amount of this first user i ' is upgraded, and according to the buffer memory remaining data after upgradingAmount redefines scheduling priority factor.

8. base station according to claim 5, is characterized in that, described media Access Control module is further used for:

Be not more than at 0 o'clock in described maximum scheduling priority factor, cry dispatching algorithm by institute according to max carrier to interference algorithm or wheelState first resource piece k ' and distribute to corresponding user; Or, according to formulaDetermine the second equitable proportion factor P '_i,j，And according to the second equitable proportion factor P ' determining_i,jDescribed first resource piece k ' is distributed to corresponding user;

Wherein, described r_i,jBe in j Transmission Time Interval each user i in the estimation handling capacity of first resource piece k '; R_i,jIt is the average throughput of each user i in j Transmission Time Interval.

9. base station according to claim 5, is characterized in that, described media Access Control module is further used for:

In the time that described not scheduling resource set of blocks is sky, determine that in j Transmission Time Interval, the actual of each user i handles upAmount, provides it to described logic link control module, and determines putting down of each user i in j+1 Transmission Time IntervalAll handling capacities.

10. base station according to claim 9, is characterized in that, described logic link control module is further used for:J Transmission Time Interval upgrades the residue token quantity of the each user i administering this base station while beginning;

After the goodput that receives each user i and before this j Transmission Time Interval finish, according to described each useThe goodput of family i upgrades the residue token quantity of described user i.

Resource regulating method in 11. 1 kinds of radio communications, is characterized in that, comprising:

The residue token quantity of the each user i administering is determined in base station;

This base station is estimation handling capacity and the institute at each not scheduling resource block k according to each user i in j Transmission Time IntervalState the residue token quantity of user i, and the buffer memory remaining data amount of each user i, determine scheduling priority factor, and according toDescribed not scheduling resource block k is distributed to corresponding user by one or more scheduling priority factor of determining, wherein, described inIn scheduling priority factor and j Transmission Time Interval, each user i is in the estimation handling capacity of each not scheduling resource block kAnd/or the buffer memory remaining data amount of each user i is directly proportional.

12. methods according to claim 11, is characterized in that, described this base station is according in j Transmission Time IntervalEach user i is in the estimation handling capacity of each not scheduling resource block k and the residue token quantity of described user i, and each useThe buffer memory remaining data amount of family i, determine that scheduling priority factor comprises:

Estimation handling capacity and described user according to each user i in j Transmission Time Interval at each not scheduling resource block kThe residue token quantity of i, and the average throughput of each user i and delaying of each user i in this j Transmission Time IntervalDeposit remaining data amount, determine one or more scheduling priority factor.

13. methods according to claim 12, is characterized in that, described this base station is according in j Transmission Time IntervalEach user i is in the estimation handling capacity of each not scheduling resource block k and the residue token quantity of described user i, and each useThe buffer memory remaining data amount of family i, determines that scheduling priority factor comprises: according to formulaDetermine equitable proportion factor P_i,j,kAs described dispatching priority because ofSon;

Wherein, Token_i,jFor the residue token quantity of each user i; Buffer_i,jFor the buffer memory remaining data of each user iAmount; B_{Bucket_size}For token bucket size; r_i,j,kBe in j Transmission Time Interval each user i at each not scheduling resource blockThe estimation handling capacity of k; R_i,jIt is the average throughput of each user i in j Transmission Time Interval; U is scheduling resource block collection notClose.

14. according to the method described in claim 11-13 any one, it is characterized in that, when transmit according to j described this base stationBetween in interval each user i in the estimation handling capacity of each not scheduling resource block k and the residue token quantity of described user i, withAnd the buffer memory remaining data amount of each user i, determine that scheduling priority factor further comprises: determine institute according to token bucket sizeState scheduling priority factor.

15. according to the method described in claim 11-13 any one, it is characterized in that, further comprises:

From one or more scheduling priority factor of determining, find maximum scheduling priority factor;

Judge whether this maximum scheduling priority factor is greater than 0, if be greater than 0, by corresponding this maximum scheduling priority factorFirst resource piece k ' distributes to first user i ' corresponding to this maximum scheduling priority factor, to the residue order of this first user i 'Board quantity is upgraded, and scheduling resource set of blocks is not upgraded, by never scheduling resource block of described first resource piece k 'In set, delete.

16. methods according to claim 15, is characterized in that, further comprise:

Described in judging, whether scheduling resource set of blocks is not empty, if be not empty, according to the residue token quantity after upgrading againDetermine scheduling priority factor, and according to the one or more scheduling priority factor that redefine out by upgrade after not schedulingThe not scheduling resource block k comprising in Resource Block set distributes to corresponding user.

17. methods according to claim 16, is characterized in that, further comprise:

By first resource piece k ' corresponding maximum scheduling priority factor distribute to this maximum scheduling priority factor correspondingWhen one user i ', the buffer memory remaining data amount of this first user i ' is upgraded;

Redefine scheduling priority factor according to the buffer memory remaining data amount after upgrading.

18. methods according to claim 15, is characterized in that, further comprise:

Be not more than at 0 o'clock in described maximum scheduling priority factor, cry dispatching algorithm by institute according to max carrier to interference algorithm or wheelState first resource piece k ' and distribute to corresponding user;

Or, according to formulaDetermine the second equitable proportion factor P '_i,j, and according to the second equitable proportion of determining because ofSub-P '_i,jDescribed first resource piece k ' is distributed to corresponding user; Wherein, described r_i,jBe every in j Transmission Time IntervalIndividual user i is in the estimation handling capacity of first resource piece k '; R_i,jBe on average the handling up of each user i in j Transmission Time IntervalAmount.

19. methods according to claim 15, is characterized in that, further comprise:

In the time that described not scheduling resource set of blocks is sky, determine that in j Transmission Time Interval, the actual of each user i handles upMeasure, and determine the average throughput of each user i in j+1 Transmission Time Interval.

20. methods according to claim 19, is characterized in that, the definite each user i's administering in described base station is surplusRemaining token quantity comprises: the residue token that upgrades the each user i administering this base station in the time that j Transmission Time Interval startsQuantity;

The method further comprises:

After the goodput of determining each user i in j Transmission Time Interval and this j Transmission Time Interval knotShu Qian, upgrades the residue token quantity of described user i according to the goodput of described each user i.