CN105163329A - Single stream and double stream switching method based on effective speed for use in beam forming - Google Patents

Abstract

The invention discloses a single stream and double stream switching method based on an effective speed for use in beam forming, relates to a single stream and double stream switching method, and aims to solve the problems that a data transmission speed in single-stream or double-stream beam forming is low, the fairness among users cannot be ensured, and transmission speeds of specific users cannot reach a target speed. The method is implemented by the following steps: I, acquiring a channel matrix defined in the description of all receiving antennas when a user k is on RBn through feedback information of the user k; II, selecting a number defined in the description of the receiving antennas actually used by the user k on an nth resource block, and determining the actually-used receiving antennas according to the number defined in the description; III, establishing a corresponding relation rho<k, n> between the user k of the actually-used receiving antennas and the RBn; and IV, calculating an effective speed defined in the description by the corresponding relation rho<k, n> between the user k of the actually-used receiving antennas and the RBn. The method is applied to the field of single stream and double stream switching.

Description

Based on the double-current changing method of the list of effective speed in a kind of wave beam forming

Technical field

The present invention relates to single double-current changing method, particularly in a kind of wave beam forming based on the double-current changing method of list of effective speed.

Background technology

In recent years, under the background of LTE, Development of Wireless Communications is advanced by leaps and bounds.Along with the surge of smart phone user, radio communication continues to increase for the demand of multimedia service.In order to provide better experience to user, MIMO technology is introduced into.MIMO technology is one of core technology of proposing the earliest of LTE.It adopts many antennas on Receiver And Transmitter, and combines with the signal processing technology of some advanced persons.MIMO technology can be used for obtaining the systematic function improved, and comprises and improves power system capacity (single community support more users) and Extended Cell coverage, and improve provided business, such as, improve single user data rate etc.Multiple antennas MIMO technology the earliest by E.Telatar and G.J.Foshini etc. by theoretical proof draw this technology the single antenna transmissions capacity limitation that provides of obtainable channel capacity far super mountain agriculture, therefore MIMO technology is obtained for extensive use in LTE and LTE-A.

At present, the transmission mode of MIMO is mainly divided three classes: transmit diversity, spatial reuse and wave beam forming.Beamforming technique, as an important core technology, is just applied to some extent in China Mobile's TD-SCDMA network.In the LTE technical specification R8 version of 3GPP, introduce single current beamforming technique (TM7), this technology can significantly promote cell edge throughput and reduce presence of intercell interference.In March, 2009, the dual-stream beamforming technology jointly promoted by Datang Mobile and China Mobile is set up the project.In March, 2010, its standardization effort completes smoothly, and is written in 3GPPLTE technical specification R9 version.Dual-stream beamforming technology (TM8) defines new dual-port dedicated pilot and corresponding control, feedback mechanism, can think that the one of intelligent antenna technology and space multiplexing technique combines.Under the dual-mode of fdd mode, the frequency spectrum of uplink downlink is not shared, and its fading characteristic is incoherent, often needs reporting of user PMI, then selects limited several beamforming scheme in the codebook to carry out.And have benefited from the characteristic that up-downgoing channel frequency is shared under TDD dual-mode, its channel has up-downgoing reciprocity property, base station can utilize the channel information of up estimation when downlink transfer, the expense of such channel estimating will reduce greatly, the acquisition of channel information becomes very convenient, this also makes the non-code book wave beam forming changing wave beam forming strategy based on channel-quality feedback in real time be achieved, and this is also the wave beam forming form that the present invention mainly studies.Non-code book wave beam forming is through years development, through various test and checking, the performance advantage of 8 antenna dual-stream beamforming technology obtains sufficient proof, and increasing operator tends to the 8 antenna beam figuration technology selecting to have advantage in outdoor macro base station at present.

TM7 and TM8 comparatively speaking, the data rate of the single current beamforming transmission that TM7 supports is lower, be applicable to the scene that user is not high to rate requirement, the dual-stream beamforming that TM8 supports has benefited from each user can transmit two different data flow, so unique user can obtain higher data rate, but there is interference between stream between two streams, so total data rate can be caused slightly lower than two single current user sums, be therefore applicable to the occasion that subscriber channel condition is good and higher to data rate requirement.

The sky line options of current multiuser MIMO and dispatching algorithm are mainly based on single current user, and the scene research mixed with multithread single current is less.For the feature of the transmission mode of different fluxion, existing a part of algorithm gives the method that fluxion is selected.Document [1] " Capacitymaximizationforzero-forcingMIMO-OFDMAdownlinksys temswithmultiuserdiversity " (comprising the maximum capacity algorithm based on ZF in the MIMO-OFDM down link of diversity) is for the purpose of maximum system throughput, give zero-forcing beamforming condition place an order double fluid choice criteria, by choosing for different user combination the fluxion that suitable antenna combination decides each user, such fluxion and antenna combination are easy to reach by making the best water line of water-filling algorithm, thus significant increase throughput of system.But the problem of this algorithm is the fairness can not guaranteeing user, the poor user of channel condition will be difficult to obtain service opportunity, and the good user of channel condition can obtain excess resource, make Resourse Distribute unreasonable, the user of a lot of bad channel conditions is difficult to reach targeted rate owing to can not get resource simultaneously, and this is that impact is great for real time business.The algorithm that document [2] " Rankadaptivetransmissiontoimprovethedetectionperformance oftheBLASTinspatiallycorrelatedMIMOchannel " (for improvement of the rank adaptation transmission algorithm of detection perform in BLAST space encoder correlated MIMO channel) provides improves to some extent in fairness, but still do not guarantee that the transmission rate of particular user can reach targeted rate, this algorithm can accept for non-real-time service, but the transmission of business will be caused to occur great problems for targeted rate cannot be reached real time business.

Summary of the invention

The object of the invention is to solve that the data rate of prior art to single current or dual-stream beamforming transmission is lower, in fairness that user can not be guaranteed and a kind of wave beam forming targeted rate that the transmission rate of particular user reaches can not guaranteed and propose based on the double-current changing method of list of effective speed.

Above-mentioned goal of the invention is achieved through the following technical solutions:

Step one, suppose LTE mobile cellular communication system single community in total K user, each user is equipped with N _rroot reception antenna, cell base station is equipped with N _ttransmit antennas; By the feedback information of user k obtain user k on RBn time all reception antennas channel matrix

Wherein, RBn is the n-th Resource Block, n=1,2,3 ...., N; N is total number of resource blocks; the transmission vector of representative of consumer k i-th reception antenna on the n-th Resource Block; K=1,2,3 ..., K;

Step 2, utilization calculate channel norm relatively channel norm with threshold value H ₀determine at all N _rselect user k at the reception antenna number of the actual use of the n-th Resource Block in root reception antenna according to determine the reception antenna of actual use;

Step 3, the user k setting up the actual reception antenna used and RBn corresponding relation ρ _k,n;

Step 4, the user k utilizing the actual reception antenna used and RBn corresponding relation ρ _k,ncalculate effective speed

Invention effect

The algorithm of the single current wave beam forming of the present invention program and traditional fixing fluxion, dual-stream beamforming and document " Capacitymaximizationforzero-forcingMIMO-OFDMAdownlinksys temswithmultiuserdiversity " contrasts, thus Integrated comparative performance.

Emulate the lte-a system based on 5M band width configuration, channel model takes TU3 model; Fig. 1 describes the contrast situation of several algorithm effective speed.When number of users is less, because RB resource is very sufficient, nearly all user can meet oneself targeted rate, now system effective speed and to close speed be almost equal, so to maximize the highest effective speed of algorithm realization in document [1] " Capacitymaximizationforzero-forcingMIMO-OFDMAdownlinksys temswithmultiuserdiversity " that speed is target.And start to increase along with number of users, resource starts the effective speed that can not meet all users gradually, and now the advantage of our algorithm starts to embody gradually.When number of users is greater than 15, algorithm of the present invention remains the highest effective speed.For double-current algorithm, because each user transmits two data flow, itself allow the number of users supported just to be less than single current algorithm, and the speed that there is no this moment of user of resource in double-current algorithm is 0, is difficult to make Mean Speed reach targeted rate.And increase along with number of users, the scarcity of resource is more and more serious, there will be a lot of user and is difficult to reach targeted rate, so its performance is worse than algorithm of the present invention, and this gap can widen further along with user Su Zengchang.For single current algorithm, each user only distributes a data flow, when number of users is more, although there will not be a lot of user rate be 0 situation, but because its fluxion is fixed, can not according to the demand dynamic conditioning of user and targeted rate, so its effect is still worse than algorithm of the present invention.

Fig. 2 further describes algorithm and meets ability to user's QoS demand.If the Mean Speed of user reaches targeted rate, then think that this user is a satisfied user.The satisfied users number of algorithms of different under different excited users said conditions that what Fig. 2 provided be.When user is less, resource is sufficient, and the rate requirement of nearly all user can be satisfied, therefore as K=10, therefore four kinds of algorithms almost do not have difference.And increase along with number of users, when RB starts to meet all users gradually, algorithm of the present invention can realize the highest meeting number of users.Document [1] " Capacitymaximizationforzero-forcingMIMO-OFDMAdownlinksys temswithmultiuserdiversity " algorithm distributes ample resources owing to giving the good user of channel, and the user of bad channel almost cannot obtain resource, so cause only having a few users to meet rate requirement, speed is lower all the time cannot meet the demands for other users.Traditional fluxion fixes single current or double-current algorithm owing to cannot dynamically mate according to user's request, also will lower than our algorithm so meet number of users.

Fairness is then another evaluation index of algorithm.At this, we carry out evaluation algorithms fairness with Jain ' the s index of weighting.Can see that our algorithm is owing to considering the demand of user, make each user close with the targeted rate of oneself as far as possible, thus achieve good fairness.The algorithm of single current wave beam forming only transmits stream data due to each user, so the targeted rate that each user has more chance to come close to oneself, therefore also has good fairness, but lower than our algorithm.The fairness of double fluid algorithm obviously declines, this is owing to increasing along with user, the throughput that the user that can be assigned to resource will obtain close to single current user twice, and unallocated to resource and user can not obtain any speed, and along with the aggravation of scarcity of resources, this imbalance is more and more obvious, so its fairness index decreased is very fast.For document [1] " Capacitymaximizationforzero-forcingMIMO-OFDMAdownlinksys temswithmultiuserdiversity ", fairness is ignored completely, so remain the poorest fairness because this algorithm is started with from throughput.

Generally speaking, algorithm of the present invention in most of the cases can keep meeting user's effective speed, remains the highest and meets number of users and fairness, and keep the highest effective speed when user is not little in systems in which always.In practical communication, in order to expand capacity, the user often accessed in system is more, has more highlighted the advantage of algorithm of the present invention in such a scenario.

Accompanying drawing explanation

Fig. 1 is the effective speed of embodiment one proposition and the schematic diagram of excited user number;

Fig. 2 be embodiment one propose meet number of users and excited user number schematic diagram.

Embodiment

Embodiment one: based on the double-current changing method of list of effective speed in a kind of wave beam forming of present embodiment, specifically prepare according to following steps:

For single current multithread mixed transport scene under forming multiuser wave beam, be that each user under this scene selects suitable transmission fluxion according to the channel quality of user and rate requirement, and the fluxion of foundation different user and channel condition, complete the distribution of multi-user on RB, the single double fluid achieved based on effective speed switches;

Step one, utilize system channel to feed back to obtain the channel information of user; Suppose total K user in single community of LTE mobile cellular communication system, each user is equipped with N _rroot reception antenna, cell base station is equipped with N _ttransmit antennas; The calculating of matrix RBn depends on the channel matrix of user, and in mimo systems, a user may be equipped with many reception antennas, therefore first by the feedback information of user k obtain user k on RBn time all reception antennas channel matrix

(in step 2, calculate channel norm compare threshold and selection antenna time all need the channel matrix using step one, obtain in step one row inside matrix; And calculate in step 3 used also from step one obtain, be exactly a part (namely wherein row)):

Wherein, RBn is the n-th Resource Block, n=1,2,3 ...., N; N is total number of resource blocks; the transmission vector of representative of consumer k i-th reception antenna on the n-th Resource Block; K=1,2,3 ..., K;

Step 2, complete sky line options to user; Utilize calculate channel norm relatively channel norm with threshold value H ₀determine at all N _rselect user k at the reception antenna number of the actual use of the n-th Resource Block in root reception antenna according to determine the reception antenna of actual use;

Step 3, the user k setting up the actual reception antenna used and RBn corresponding relation ρ _k,n, completing user antenna is to the distribution of Resource Block;

Step 4, next introduce the concept of effective speed; Because the otherness of transport service between user causes the demand of different user to speed to be different, we often do not need to make the overall transmission rate of system very large, but need the targeted rate λ that allows each user reach oneself as far as possible _k; Conveniently system optimization, utilizes user k and the RBn corresponding relation ρ of the actual reception antenna used _k,ncalculate effective speed

Present embodiment effect:

In TD-LTE system, wave beam forming carries out different weightings by the different antenna element at transmitting terminal, thus forms the Signal to Interference plus Noise Ratio that wave beam effectively improves user.Wave beam forming can be divided into single current and multithread, and current multithread main-lobe shaping is mainly double-current, the TM8 transmission mode in corresponding 3GPP standard.Double fluid user exists and disturbs with between indoor stream, but thus can obtain larger speed comparatively greatly due to total data fluxion; Disturb although single current user does not exist between stream, total speed is less.Thus dual-stream beamforming is suitable for the user that channel condition is good, rate requirement is high.Be suitable for scene and user's own situation for single double fluid, the double-current changing method of list based on effective speed can the fluxion select permeability of user in resolution system effectively, realizes single dual-stream beamforming self adaptation and switches.

Embodiment two: present embodiment and embodiment one unlike: utilize in step 2 calculate channel norm relatively channel norm with threshold value H ₀determine at all N _rselect user k at the reception antenna number of the actual use of the n-th Resource Block in root reception antenna according to determine that the reception antenna detailed process of actual use is:

1), except the channel information of user, rate requirement is also a major criterion determining fluxion; Utilize r_remain _k,nt () carrys out the rate requirement of characterizing consumer:

$r\_{\mathrm{remain}}_{k,n}\left(t\right)=t\&times;{\mathrm{\&lambda;}}_k-(t-1)\&times;R_k(t-1)-\underset{b=1,b\&NotEqual;n}{\overset{N}{\&Sigma;}}{\mathrm{\&rho;}}_{k,b}{r}_{k,b}\left(t\right)---\left(3\right)$

Wherein, R _k(t-1) be the Mean Speed that user k obtains in a front t-1 subframe, r _k,bt () is the speed that user k obtains in t subframe on RBb, real-time update; λ _kfor the target transmission speed of user k; RBb represents b RB; ρ _k,bfor user k and the RBb corresponding relation of the reception antenna of reality use;

T × λ _kfor wishing the target accumulated speed reached, (t-1) × R to user k during t subframe _k(t-1) speed obtained in t-1 subframe before t is illustrated, for the speed obtained at present t subframe, and r_remain _k,nt () is the current speed also differed with target accumulated speed;

2), according in step one obtain calculate the Frobenius norm of channel (Frobenius norm be exactly with this name Frobenius norm be a mathematical concept, belong to the known concept in matrix analysis opinion, this mathematical concept be widely used in communication theory characterize channel quality)

3), in order to determine whether dual-stream beamforming can be used, the rate requirement setting for channel norm and user is needed threshold value H ₀with the threshold value r of residue speed ₀, when two indexs of certain user exceed threshold value simultaneously, the correlation between its antenna can be judged further, if correlation is enough good, adopt dual-stream beamforming;

(1) if or r_remain _k,n(t) <r ₀, then user k adopts single-stream transmission on RBn, makes the data fluxion that user k transmits on RBn meanwhile, be l that user k selective channel norm is maximum ^*the reception antenna that root antenna uses as reality, namely $l^{*}=\arg \underset{l\&Element;\{1,2,...N_r\}}{max}\left(\right|\left|h_{k,n}^l\right|{|}_F);$ Then $H_{k,n}=h_{k,n}^{l^{*}};$

Do not considering under extreme transmission environment, think user reception antenna between be separate, namely user's transmission matrix is full rank every root reception antenna is by data flow different for correspondence one, and total data fluxion equals the antenna number selected, and the data fluxion of K user is equal with the order of user's transmission matrix simultaneously, namely therefore, determine to transmit the antenna number that fluxion not only will determine selection, also should determine to select which root antenna actually;

Wherein, it is user k actual number of antennas used on RBn; the transmission vector of representative of consumer k l root reception antenna on the n-th Resource Block, H _k,nfor by the user k transmission matrix that the actual antenna selected forms on RBn; H _k,nbe a submatrix, representative of consumer k is l on the n-th Resource Block ^*the transmission vector of root reception antenna;

(2) if user k and r_remain _k,n(t)>=r ₀, then a pair antenna (i selecting relative coefficient minimum from the antenna of user k ^*, j ^*), utilize weigh relative coefficient, after obtaining relative coefficient, select $(i^{*},j^{*})=\arg \underset{i,j\&Element;\{1,2,...N_r\}}{min}\left({\mathrm{\&eta;}}_{k_i,k_j}^n\right)$ As the reception antenna that reality uses;

Wherein, k _irepresent i-th antenna of user k; k _jrepresent the jth root antenna of user k; represent i-th antenna of user k and the space correlation property coefficient of jth root antenna on RBn of user k; computing formula is as follows:

${\mathrm{\&eta;}}_{k_i,k_j}^n=\frac{\left|h_{k,n}^i{\left(h_{k,n}^j\right)}^H\right|}{\left|\right|h_{k,n}^i\left|{|}_F\right|\left|h_{k,n}^j\right|{|}_F};$

(3), setting threshold η ₀if the antenna selected in (2) is to (i ^*, j ^*) relative coefficient meet then user k adopts stream transmission on RBn, order and make if otherwise user k adopts single-stream transmission on RBn, and is the maximum l of user k selective channel norm ^*the reception antenna that root antenna uses as reality, namely now $H_{k,n}=h_{k,n}^{l^{*}};$

Wherein, representative of consumer k is on the n-th Resource Block i-th ^*the transmission vector of root reception antenna, representative of consumer k is jth on the n-th Resource Block ^*the transmission vector of root reception antenna;

By said process, the antenna of the fluxion that all users are transmitted on all RBn and use is decided; This process synthesis considers channel quality and the rate requirement of user, and can according to user's real time rate dynamic conditioning, and the self adaptation achieving single dual-stream beamforming switches;

But in general, the data fluxion of transmission is probably less than the reception antenna sum of user, under the condition not considering diversity reception, may only need to select fewer antenna; Therefore by all N _rselect user k at the reception antenna number (number of user k reality of the transmission vector of i-th reception antenna on the n-th Resource Block) of the actual use of the n-th Resource Block in root reception antenna by the reception antenna of the actual use of root the transmission matrix that composition is actual matrix for submatrix;

Fluxion determination main purpose is to determine the data fluxion that user is transmitted; Can transmit two data flow due to dual-stream beamforming simultaneously and provide higher and speed, but the interference between two streams can produce certain influence to speed, so be applicable to the user that channel condition is better, rate requirement is higher; According to subscriber channel situation and demand rate determination fluxion;

The channel conditions of user is decomposed into channel quality and the antenna spatial correlation Two Variables of user; The channel quality of user is user's fading profiles on transport channels; The concrete computational process of the fading profiles of user on channel is: the signal strength signal intensity that the channel that the channel norm calculating user carrys out characterizing consumer receives.Other step and parameter identical with embodiment one.

Embodiment three: present embodiment and embodiment one or two are unlike the user k and the RBn corresponding relation ρ that set up the actual reception antenna used in step 3 _k,ndetailed process:

Step 2 has completed user's fluxion and has determined and the determination of antenna space, but the determination of this fluxion is only indicate the transmission mode being about to when user k is assigned on RBn adopt, and in fact whether user k can be assigned on RBn is uncertain, in user k selection course, solve the user's combinatorial problem on different RB, thus complete result that in integrating step two, fluxion is determined and carried out a relatively complete scheduling process; In order to choose the less user's combination of inter-user interference, in user selects, we will continue the concept of usage space relative coefficient, and difference is here be for assessment of user between interference size; Except inter-user interference, this channel condition on RB of user is also one of our factor of considering,

(1), the Customs Assigned Number that all RB distribute is put into user and gather Ui;

(2), on RBn, select that there is minimum value r _k(t)/λ _kuser, be assigned on RBn; The user that user k adds RBn is gathered U _nin, and by the reception antenna set A of user k in step 2 in the actual use of RBn _nin, make ρ _k,nthe user of the actual reception antenna used of=1 expression and RBn corresponding relation _x

Wherein, r _kt () is for user k is in the transmission rate of user's original transmitted data of t subframe; λ _kfor targeted rate; ρ _k,nfor distributing designator;

(3) method, calculating the mean space relative coefficient of user k between the reception antenna and the antenna not being assigned to the user on RBn of the actual use of RBn on RBn is as follows:

$\stackrel{\&OverBar;}{{\mathrm{\&eta;}}_m^n}=\frac{\underset{k_i\&Element;A_n}{\&Sigma;}\underset{m_j\&Element;T_{m,n}}{\&Sigma;}{\mathrm{\&eta;}}_{k_i,m_j}^n}{card\left(A_n\right)\&times;card\left(T_{m,n}\right)},\&ForAll;m\&Element;Ui$

Wherein, k _ifor i-th antenna of user k; m _jfor the jth root antenna of user m; T _m,nwhat represent is the antenna set of user m on RBn, and card represents the element number of set; what represent is calculating for spatial coherence;

${\mathrm{\&eta;}}_{k_i,m_j}^n=\frac{\left|h_{k,n}^i{\left(h_{m,n}^j\right)}^H\right|}{\left|\right|h_{k,n}^i\left|{|}_F\right|\left|h_{m,n}^j\right|{|}_F}---\left(7\right)$

Wherein, represent i-th antenna transmission vector of user k, represent the jth root antenna transmission vector of user m on RBn, for i-th antenna of user k and the space correlation property coefficient of jth root antenna on RBn of user m, the interference size between the data flow that can be used for assessment two antenna transmission; value larger, can think that the interference between i-th antenna of user k and the jth root antenna of user m is larger;

(4), from the user of use reception antenna actual on RBn, L is chosen be worth minimum user, from L user, select user f, the alternative condition of user f is numerical value maximum;

Wherein, r _k,nrepresent the speed that this user k obtains under present sub-frame on RBn; Here L is the customer parameter set up in order to overall balance user channel quality and inter-user interference, in the present invention, and L value min{card (Ui), N _t/ N _r; representative is all on RBn has been assigned to antenna on RBn and unallocated mean space relative coefficient between the antenna on RBn;

(5) if after user f adds RBn, all users of RBn comprising f and speed can not be allowed to promote (because adding of f introduces new interference to some extent, now necessarily not the improving with speed of all users on this RB), then user f is not assigned on RBn, the all users simultaneously stopped on this RB share out the work, and no longer continue as this RB and select user; If RBn comprises the speed sum of all users of f improve, then user f to be assigned on RBn and (that is: the user that the numbering of user f adds RBn to be gathered U _n, U _n=U _nand make ρ U{f}), _f,n=1, upgrade user and gather U _nwith antenna set A _nrepeat above-mentioned (3) and (4) process until card (A _n)=N _tnamely the user's selection course on RBn is completed; ρ _f,nrepresent the distribution designator of user f; K=1,2,3 ..., f ..., K

(6), after completing the user selected on RBn, the average transmission rate R of user k in T subframe is calculated _kif, R _k>=λ _k, then user k is shifted out set Ui; K=1,2,3 ..., K; If R _k< λ _k, then user k is retained in set Ui, completes the user selected on RBn;

(7) make n=n+1, repeat step (2) to (6), until namely n=N completes the user k and RBn corresponding relation ρ that set up the actual reception antenna used _k,n.Other step and parameter identical with embodiment one or two.

Embodiment four: one of present embodiment and embodiment one to three unlike: calculate the average transmission rate R of user k in T subframe _kdetailed process is:

In base station end, signal is transmitted into after transmission channel through serioparallel exchange and through precoding beam figuration and arrives user's receiving terminal; When power averaging distributes, the signal y that the user k of receiving terminal receives _k,nbe expressed as:

Wherein, the original transmitted data of user k, n _kfor white Gaussian noise; W _j,nfor the wave beam formed matrix of user j on RBn; x _j,nbe the original transmitted data of user j, d is S _k,nin the number of data in each data flow; x _{k, n, S}the initial data of the S stream of user k on RBn; S=1,2 ..., S _k,n; The data fluxion of each user's transmission is S; The wave beam formed matrix W of user k on RBn (ResouceBlock, Resource Block) _k,n(base station is according to the channel matrix H of user _k,n, calculate wave beam formed matrix W by precoding _k,n; Utilize the channel matrix H of user _k,nwith wave beam formed matrix W _k,n, its Signal to Interference plus Noise Ratio γ can be calculated _k,nas follows:

${\mathrm{\&gamma;}}_{k,n}=\frac{\left|\right|H_{k,n}{W}_{k,n}|{|}_F^2}{N_R^{k,n}{\mathrm{\&sigma;}}_{k,n}^2+{\mathrm{\&Sigma;}}_{j=1,j\&NotEqual;k}^K\left|\right|H_{j,n}{W}_{k,n}|{|}_F^2}$

Wherein, for the noise power of user k on RBn, LTE system adopts Adaptive Modulation and Coding strategy, and after calculating Signal to Interference plus Noise Ratio, LTE system will according to γ _k,nnumerical value be that user k selects suitable modulation, coded system and bit rate, calculate the speed r of user k at RBn at receiving terminal according to the actual data volume received of user _k,n; High γ _k,nvalue is corresponding higher order of modulation and bit rate usually, thus can make the r that user's receiving terminal records _k,nnumerical value is comparatively large, r _k,nwith γ _k,nbecome positive correlation; H _j,nfor the transmission matrix of user j on RBn;

For the wave beam formed matrix W of user k on RBn (ResouceBlock, Resource Block) _k,nbe expressed as:

W _{k, n, S}the emission value of s flow data at transmitting terminal of a kth user n-th Resource Block; S _k,nthe transmitting data stream number of user k on RBn; K=1,2,3 ..., K;

Because diversity technique is mainly used in the scene of user channel quality extreme difference, and as a rule beamforming technique can not be applied to this kind of user, so under not considering the scene of multi-antenna diversity, in (3) formula represent from the interference of other users, the Signal to Interference plus Noise Ratio of user becomes positive correlation with received signal power, and the Signal to Interference plus Noise Ratio of user, inter-user interference and white noise power become inverse correlation;

1) due to the employing of Adaptive Modulation and Coding technology in LTE system, the height of Signal to Interference plus Noise Ratio SINR directly determines the size of transmission rate; The bit rate r of the transmission data of user k in present sub-frame _kbe expressed from the next:

$r_k=\underset{n=1}{\overset{N}{\&Sigma;}}{\mathrm{\&rho;}}_{k,n}{r}_{k,n}---\left(4\right)$

Wherein, r _k,nrepresent that user k obtains speed, if ρ on RBn _k,n=1 represents that user k is distributed on RBn, otherwise user k is not distributed on RBn;

2) according to r _kcalculate R _k:

$R_k={\mathrm{\&Sigma;}}_{t=1}^T{r}_k\left(t\right)/T$

R _krepresent the average transmission rate of user k in T subframe; Work as R _kbe less than targeted rate λ _k, the qos requirement of this user is not satisfied; According to result of calculation, the performance of assessment algorithm, and determine whether to adjust system.Other step and parameter identical with one of embodiment one to three.

Embodiment five: one of present embodiment and embodiment one to four are unlike effective speed in step 4 concept definition is as follows:

${\tilde{R}}_k=\left\{\begin{array}{c}{R}_k,if{R}_k\&GreaterEqual;{\mathrm{\&lambda;}}_k\\ 0,if{R}_k<{\mathrm{\&lambda;}}_k\end{array}\right.---\left(5\right)$

Wherein, represent the average transmission rate of user k in T subframe, r _kt () represents r _kthe speed that middle user k obtains when current t subframe; r _kfor the bit rate of the transmission data of user k in present sub-frame; And the size of T depends on QoS (QualityofService, the service quality) requirement of user; Work as R _kbe more than or equal to targeted rate λ _k, user k meets qos requirement; Work as R _kbe less than targeted rate λ _k, the qos requirement of this user is not satisfied; According to result of calculation, the performance of assessment algorithm, and determine whether to adjust system;

So be different from the way of traditional optimization problem, optimization aim be maximization system and effective speed, that is:

$\underset{{\mathrm{\&rho;}}_{k,n},H_{k,n}}{max}\underset{k=1}{\overset{K}{\&Sigma;}}{\tilde{R}}_k---\left(6\right)$

ρ _k,nand H _k,nnot only need to solve the distribution of user on RB, prior, be the fluxion switching problem solving user;

Obtain ρ _k,nand H _k,noptimal solution, direct method takes traversal search, by the fluxion situation of all users and user combination all attempt, select optimal solution; But the method amount of calculation is excessive, cannot apply completely in systems in practice; In order to this problem solving can be realized under feasible complexity, choose the fluxion changing method of rank adaptation using indexs such as the channel quality of user, rate requirement and Antenna Correlation as considering foundation, judge single current wave beam forming or the dual-stream beamforming transmission mode of user, realize the self adaptation fluxion adjustment of real-time antenna number and rate requirement coupling; Wherein, the fluxion changing method of rank adaptation comprises fluxion and determines to select two major parts with user

Selected by such user, we not only ensure that the user channel quality on each RB is better, also ensure that the low interference between user to greatest extent, avoid the high complexity that traversal search brings simultaneously.Other step and parameter identical with one of embodiment one to four.

Following examples are adopted to verify beneficial effect of the present invention:

Embodiment one:

Based on the double-current changing method of list of effective speed in a kind of wave beam forming of the present embodiment, specifically prepare according to following steps:

Principle: total K user in supposing the system, every user is equipped with N _rroot reception antenna, base station is equipped with N _ttransmit antennas, the data fluxion of sole user's transmission is S.For user k, its wave beam formed matrix on RBn can be expressed as:

Wherein w _k,sthe emission value of s flow data at transmitting terminal of a kth user.S _k,nthe transmitting data stream number of user k on RBn.The whole transmission matrixs that accept antenna of user k on RBn can be expressed as:

Wherein the transmission matrix of i-th reception antenna of representative of consumer k.But in general, the data fluxion of our transmission is probably less than the reception antenna sum of user, under the condition not considering diversity reception, may only need to select fewer antenna.Therefore we will from all N _rselect in root reception antenna root as the reception antenna of reality, thus forms actual transmission matrix this matrix is submatrix.At this, we do not consider extreme transmission environment, think user reception antenna between be separate, namely user's transmission matrix is full rank now every root reception antenna is by data flow different for correspondence one, and total data fluxion equals the antenna number selected, simultaneously also equal with the order of transmission matrix, namely therefore, determine to transmit the antenna number that fluxion not only will determine selection, also should determine to select which root antenna actually.

In base station end, signal is through serioparallel exchange and launch through precoding beam figuration, arrives receiving terminal after experience transmission channel.When power averaging distributes, the signal that the user k of receiving terminal receives can be expressed as:

the original transmitted data of user k, n _kfor white Gaussian noise.Because diversity technique is mainly used in the scene of user channel quality extreme difference, and as a rule beamforming technique can not be applied to this kind of user, so we do not consider the scene of multi-antenna diversity herein.(3) in formula, Section 2 represents the interference from other users, the Signal to Interference plus Noise Ratio of user becomes positive correlation with received signal power, inverse correlation is become with inter-user interference and white noise power, due to the employing of Adaptive Modulation and Coding technology in LTE system, the height of Signal to Interference plus Noise Ratio SINR directly determines the size of transmission rate.The calculating of SINR and the transmission matrix of user, wave beam formed matrix and resource distribution mode have relation, value according to SINR can according to CQI and the SINR table of comparisons in LTE system, determine the modulation system of being correlated with and bit rate, and then the transmission rate r of user k on RBn _k,n, and the transmission rate r that user is total _kcan be expressed from the next:

$r_k=\underset{n=1}{\overset{N}{\&Sigma;}}{\mathrm{\&rho;}}_{k,n}{r}_{k,n}---\left(4\right)$

Wherein, r _k,nrepresent that user k obtains speed, ρ on RBn _k,nfor distributing designator, if ρ _k,n=1 represents that user k is distributed on RBn, otherwise user k is not distributed on RBn.

Next the concept of effective speed is introduced.Because the otherness of transport service between user causes the demand of different user to speed to be different, we often do not need to make the overall transmission rate of system very large, but need the targeted rate λ that allows each user reach oneself as far as possible _k.Conveniently system optimization, it is as follows that we introduce effective speed concept definition:

${\tilde{R}}_k=\left\{\begin{array}{c}{R}_k,if{R}_k\&GreaterEqual;{\mathrm{\&lambda;}}_k\\ 0,if{R}_k<{\mathrm{\&lambda;}}_k\end{array}\right.---\left(5\right)$

Wherein, represent the average transmission rate of user k in T subframe, and the size of T depends on QoS (QualityofService, the service quality) requirement of user; Work as R _kbe less than targeted rate λ _k, the qos requirement of this user is not satisfied.

$\underset{{\mathrm{\&rho;}}_{k,n},H_{k,n}}{max}\underset{k=1}{\overset{K}{\&Sigma;}}{\tilde{R}}_k---\left(6\right)$

This problem not only needs to solve the distribution of user on RB, prior, is the fluxion switching problem solving user.

Want to obtain ρ _k,nand H _k,nthe optimal solution of optimal solution, direct method takes traversal search, the fluxion situation of all users and user's combination all attempted, select optimal solution.But the method amount of calculation is excessive, cannot apply completely in systems in practice.In order to this problem solving can be realized under feasible complexity, we have proposed a kind of fluxion changing method of rank adaptation, the method is using indexs such as the channel quality of user, rate requirement, Antenna Correlation as considering foundation, synthetically judge single current or the stream transmission pattern of user, realize the self adaptation fluxion adjustment of real-time matching.The method can be divided into two major parts: fluxion is determined to select with user.

1. fluxion is determined

The main purpose that fluxion is determined is to determine the data fluxion that user is transmitted.Can transmit two data flow due to dual-stream beamforming simultaneously and provide higher and speed, but the interference between two streams can produce certain influence to speed, so be applicable to the user that channel condition is better, rate requirement is higher.For this feature, we go out to send from subscriber channel situation and rate requirement two angles and determine fluxion.

Here, the channel conditions of user is decomposed into channel quality and the antenna spatial correlation Two Variables of user by us.The former represents user's fading profiles on transport channels, and latter can interference size between characterizing consumer different antennae thus the size of interference between stream in estimating user.In order to embody the fading profiles of user on channel, the signal strength signal intensity that the channel norm that we calculate user receives to characterize it.For the calculating of spatial coherence, we introduce variable be defined as follows:

${\mathrm{\&eta;}}_{k_i,m_j}^n=\frac{\left|h_{k,n}^i{\left(h_{m,n}^j\right)}^H\right|}{\left|\right|h_{k,n}^i\left|{|}_F\right|\left|h_{m,n}^j\right|{|}_F}---\left(7\right)$

Wherein k _irepresent that i-th of user k with antenna, mj represents the jth root antenna of user m, represent the i root antenna of user k and the space correlation property coefficient of jth root antenna on RBn of user m, the interference size between the data flow that can be used for assessment two antenna transmission. value larger, can think k _iwith m _jbetween interference larger.

Except the channel information of user, rate requirement is also a major criterion determining fluxion.The difference r_remain of the speed that we have been obtained at present by user and its targeted rate _k,nt () carrys out the rate requirement of characterizing consumer:

$r\_{\mathrm{remain}}_{k,n}\left(t\right)=t\&times;{\mathrm{\&lambda;}}_k-(t-1)\&times;R_k(t-1)-\underset{i=1,i\&NotEqual;n}{\overset{N}{\&Sigma;}}{\mathrm{\&rho;}}_{k,b}{r}_{k,b}\left(t\right)---\left(8\right)$

R _k(t-1) be the Mean Speed that user k has obtained in a front t-1 subframe, r _k,bt () is the speed that user k obtains in current TTI (TransmissionTimeInterval) on RBb, real-time update.

In order to determine whether dual-stream beamforming can be used, we need for channel norm and rate requirement setting threshold values H ₀, r ₀, when two indexs of certain user exceed threshold values simultaneously, the correlation between its antenna can be judged further, if correlation is enough good, adopt dual-stream beamforming.Concrete step is as follows:

(1) all users are gathered to the user in U, calculate its channel norm on t Transmission Time Interval (TransmissionTimeInterval, TTI) with residue rate requirement r_remain _k,n(t), if or r_remain _k,n(t) <r ₀, then this user adopts single-stream transmission on RBn, order meanwhile, be an antenna l that this user's selective channel norm is maximum ^*as reception antenna, namely

(2) for meeting and r_remain _k,n(t)>=r ₀user, a pair antenna selecting relative coefficient minimum from all antennas of this user, namely $(i^{*},j^{*})=\arg \underset{i,j\&Element;\{1,2,...N_r\}}{min}\left({\mathrm{\&eta;}}_{k_i,k_j}^n\right).$

(3) if then user k adopts stream transmission on RBn, order and make otherwise this user still adopts single-stream transmission on RBn, and find out a maximum antenna of channel norm for it $l^{*}=\underset{l\&Element;\{1,2,\mathrm{...}{N}_r\}}{max}\left(\right|\left|h_{k,n}^l\right|{|}_F).$

By said process, the antenna of the fluxion that all users are transmitted on all RBn and use is decided.This process synthesis considers channel quality and the rate requirement of user, and can according to user's real time rate dynamic conditioning, and the self adaptation achieving single dual-stream beamforming switches.

2. user selects

In last process, we have completed user's fluxion and have determined and the determination of antenna space, but the determination of this fluxion is only indicate the transmission mode being about to when user k is assigned on RBn adopt, and in fact whether user k can be assigned on RBn is uncertain.In user's selection course, we will solve the user's combinatorial problem on different RB just, thus complete the result determined in conjunction with fluxion in previous step and carried out a relatively complete scheduling process.In order to choose the less user's combination of inter-user interference, in user selects, we will continue the concept of usage space relative coefficient, and difference is here be for assessment of user between interference size.Except inter-user interference, this channel condition on RB of user is also one of our factor of considering.The process that whole user selects is as follows:

(1) all users wanting RB to distribute that still needs are put into available subscribers set Ui.

(2) on RBn, select that there is minimum value r _k(t)/λ _kuser, be assigned on RBn.The user that user k adds RBn is gathered U _n, and the antenna set of k is assigned to the antenna set A of n _nin, order

(3) on RBn, calculate all mean space relative coefficients be assigned on n between antenna and the antenna of other users, its computational methods are as follows:

$\stackrel{\&OverBar;}{{\mathrm{\&eta;}}_m^n}=\frac{\underset{k_i\&Element;A_n}{\&Sigma;}\underset{m_j\&Element;T_{m,n}}{\&Sigma;}{\mathrm{\&eta;}}_{k_i,m_j}^n}{card\left(A_n\right)\&times;card\left(T_{m,n}\right)},\&ForAll;m\&Element;Ui---\left(9\right)$

(4) L is therefrom chosen be worth minimum user, choose from this L user maximum user; Here r _k,nrepresent the speed that this user obtains under current TTI on RBn.Here L is the customer parameter set up in order to overall balance user channel quality and inter-user interference, in the present invention, and L value min{card (Ui), N _t/ N _r.

(5) if the user selected can promote whole RB's and speed, be then assigned on this RB, and repeat above-mentioned (3), (4) process until complete selects the user of this RB.Otherwise do not distribute, the user simultaneously terminated on this RB selects.

(6) after completing the distribution of a RB, what calculate all users reaches speed, if meet the demands, is then shifted out available subscribers set Ui.

(7) above-mentioned steps (2) is repeated to (6), until all RB all completing user selections.

Selected by such user, we not only ensure that the user channel quality on each RB is better, also ensure that the low interference between user to greatest extent, avoid the high complexity that traversal search brings simultaneously.

The present invention also can have other various embodiments; when not deviating from the present invention's spirit and essence thereof; those skilled in the art are when making various corresponding change and distortion according to the present invention, but these change accordingly and are out of shape the protection range that all should belong to the claim appended by the present invention.

Claims (5)

1. in wave beam forming based on the double-current changing method of the list of effective speed, it is characterized in that: specifically carry out according to following steps based on the double-current changing method of the list of effective speed in a kind of wave beam forming:

Step one, suppose LTE mobile cellular communication system single community in total K user, each user is equipped with N _rroot reception antenna, cell base station is equipped with N _ttransmit antennas; By the feedback information of user k obtain user k on RBn time all reception antennas channel matrix

Wherein, RBn is the n-th Resource Block, n=1,2,3 ...., N; N is total number of resource blocks; the transmission vector of representative of consumer k i-th reception antenna on the n-th Resource Block; K=1,2,3 ..., K;

Step 2, utilization calculate channel norm relatively channel norm with threshold value H ₀determine at all N _rselect user k at the reception antenna number of the actual use of the n-th Resource Block in root reception antenna according to determine the reception antenna of actual use;

Step 3, the user k setting up the actual reception antenna used and RBn corresponding relation ρ _k,n;

Step 4, the user k utilizing the actual reception antenna used and RBn corresponding relation ρ _k,ncalculate effective speed

2. according to claim 1 in a kind of wave beam forming based on the double-current changing method of list of effective speed, it is characterized in that: utilize in step 2 calculate channel norm relatively channel norm with threshold value H ₀determine at all N _rselect user k at the reception antenna number of the actual use of the n-th Resource Block in root reception antenna according to determine that the reception antenna detailed process of actual use is:

1), r_remain is utilized _k,nt () carrys out the rate requirement of characterizing consumer:

$r\_{\mathrm{remain}}_{k,n}\left(t\right)=t\&times;{\mathrm{\&lambda;}}_k-(t-1)\&times;R_k(t-1)-\underset{b=1,b\&NotEqual;n}{\overset{N}{\&Sigma;}}{\mathrm{\&rho;}}_{k,b}{r}_{k,b}\left(t\right)---\left(3\right)$

Wherein, R _k(t-1) be the Mean Speed that user k obtains in a front t-1 subframe, r _k,bt () is the speed that user k obtains in t subframe on RBb; λ _kfor the target transmission speed of user k; RBb represents b RB; ρ _k,bfor user k and the RBb corresponding relation of the reception antenna of reality use;

2), according in step one obtain calculate the Frobenius norm of channel

3), set threshold value H ₀with the threshold value r of residue speed ₀;

(1) if or r_remain _k,n(t) <r ₀, then user k adopts single-stream transmission on RBn, makes the data fluxion that user k transmits on RBn meanwhile, be l that user k selective channel norm is maximum ^*the reception antenna that root antenna uses as reality, namely $l^{*}=\arg \underset{l\&Element;\{1,2,...N_r\}}{max}\left(\right|\left|h_{k,n}^l\right|{|}_F);$ Then $H_{k,n}=k_{k,n}^{l^{*}};$

Wherein, it is user k actual number of antennas used on RBn; the transmission vector of representative of consumer k l root reception antenna on the n-th Resource Block, H _k,nfor by the user k transmission matrix that the actual antenna selected forms on RBn; H _k,nbe a submatrix, representative of consumer k is l on the n-th Resource Block ^*the transmission vector of root reception antenna;

(2) if user k and r_remain _k,n(t)>=r ₀, then a pair antenna (i selecting relative coefficient minimum from the antenna of user k ^*, j ^*), utilize weigh relative coefficient, after obtaining relative coefficient, select $(i^{*},j^{*})=\arg \underset{i,j\&Element;\{1,2,...N_r\}}{min}\left({\mathrm{\&eta;}}_{k_i,k_j}^n\right)$ As the reception antenna that reality uses;

Wherein, k _irepresent i-th antenna of user k; k _jrepresent the jth root antenna of user k; represent i-th antenna of user k and the space correlation property coefficient of jth root antenna on RBn of user k; computing formula is as follows:

${\mathrm{\&eta;}}_{k_i,k_j}^n=\frac{\left|h_{k,n}^i{\left(h_{k,n}^j\right)}^H\right|}{\left|\right|h_{k,n}^i\left|{|}_F\right|\left|h_{k,n}^j\right|{|}_F};$

(3), setting threshold η ₀if the antenna selected in (2) is to (i ^*, j ^*) relative coefficient meet then user k adopts stream transmission on RBn, order and make if otherwise user k adopts single-stream transmission on RBn, and is the maximum l of user k selective channel norm ^*the reception antenna that root antenna uses as reality, namely $l^{*}=\arg \underset{l\&Element;\{1,2,...N_r\}}{max}\left(\right|\left|h_{k,n}^l\right|{|}_F);$ Now $H_{k,n}=h_{k,n}^{l^{*}};$

Wherein, representative of consumer k is on the n-th Resource Block i-th ^*the transmission vector of root reception antenna, representative of consumer k is jth on the n-th Resource Block ^*the transmission vector of root reception antenna.

3. according to claim 2 in a kind of wave beam forming based on the double-current changing method of list of effective speed, it is characterized in that: the user k and the RBn corresponding relation ρ that set up the actual reception antenna used in step 3 _k,ndetailed process:

(1), the Customs Assigned Number that all RB distribute is put into user and gather Ui;

(2), on RBn, select that there is minimum value r _k(t)/λ _kuser, be assigned on RBn; The user that user k adds RBn is gathered U _nin, and by the reception antenna set A of user k in step 2 in the actual use of RBn _nin, make ρ _k,nthe user of the actual reception antenna used of=1 expression and RBn corresponding relation;

Wherein, r _kt () is for user k is in the transmission rate of user's original transmitted data of t subframe; λ _kfor targeted rate; ρ _k,nfor distributing designator;

(3) method, calculating the mean space relative coefficient of user k between the reception antenna and the antenna not being assigned to the user on RBn of the actual use of RBn on RBn is as follows:

$\stackrel{\&OverBar;}{{\mathrm{\&eta;}}_m^n}=\frac{\underset{k_i\&Element;A_n}{\&Sigma;}\underset{m_j\&Element;T_{m,n}}{\&Sigma;}{\mathrm{\&eta;}}_{k_i,m_j}^n}{card\left(A_n\right)\&times;card\left(T_{m,n}\right)},\&ForAll;m\&Element;Ui$

Wherein, k _ifor i-th antenna of user k; m _jfor the jth root antenna of user m; T _m,nwhat represent is the antenna set of user m on RBn, and card represents the element number of set; what represent is calculating for spatial coherence;

${\mathrm{\&eta;}}_{k_i,m_j}^n=\frac{\left|h_{k,n}^i{\left(h_{m,n}^j\right)}^H\right|}{\left|\right|h_{k,n}^i\left|{|}_F\right|\left|h_{m,n}^j\right|{|}_F}---\left(7\right)$

Wherein, represent i-th antenna transmission vector of user k, represent the jth root antenna transmission vector of user m on RBn, for i-th antenna of user k and the space correlation property coefficient of jth root antenna on RBn of user m;

(4), from the user of use reception antenna actual on RBn, L is chosen be worth minimum user, from L user, select user f, the alternative condition of user f is numerical value maximum;

Wherein, r _k,nrepresent the speed that this user k obtains under present sub-frame on RBn; L value min{card (Ui), N _t/ N _r; representative is all on RBn has been assigned to antenna on RBn and unallocated mean space relative coefficient between the antenna on RBn;

(5) if after user f adds RBn, if RBn comprises the speed sum of all users of f improve, then the user numbering of user f being added RBn gathers U _n, U _n=U _nu{f} also makes ρ _f,n=1, upgrade user and gather U _nwith antenna set A _nrepeat above-mentioned (3) and (4) process until card (A _n)=N _tnamely the user's selection course on RBn is completed; ρ _f,nrepresent the distribution designator of user f; K=1,2,3 ..., f ..., K;

(6), after completing the user selected on RBn, the average transmission rate R of user k in T subframe is calculated _kif, R _k>=λ _k, then user k is shifted out set Ui; K=1,2,3 ..., K; If R _k< λ _k, then user k is retained in set Ui, completes the user selected on RBn;

(7) make n=n+1, repeat step (2) to (6), until namely n=N completes the user k and RBn corresponding relation ρ that set up the actual reception antenna used _k,n.

4. according to claim 3 in a kind of wave beam forming based on the double-current changing method of list of effective speed, it is characterized in that: calculate the average transmission rate R of user k in T subframe _kdetailed process is:

1) the bit rate r of the transmission data of user k in present sub-frame _kbe expressed from the next:

$r_k=\underset{n=1}{\overset{N}{\&Sigma;}}{\mathrm{\&rho;}}_{k,n}{r}_{k,n}---\left(4\right)$

Wherein, r _k,nrepresent that user k obtains speed, if ρ on RBn _k,n=1 represents that user k is distributed on RBn;

2) according to r _kcalculate R _k:

$R_k={\mathrm{\&Sigma;}}_{t=1}^T{r}_k\left(t\right)/T.$

5. according to claim 4 in a kind of wave beam forming based on the double-current changing method of list of effective speed, it is characterized in that: effective speed in step 4 concept definition is as follows:

${\tilde{R}}_k=\left\{\begin{array}{ccc}{R}_k,& if& R_k\&GreaterEqual;{\mathrm{\&lambda;}}_k\\ 0,& if& R_k<{\mathrm{\&lambda;}}_k\end{array}\right.---\left(5\right)$

Wherein, represent the average transmission rate of user k in T subframe, r _kt () represents r _kthe speed that middle user k obtains when current t subframe; r _kfor the bit rate of the transmission data of user k in present sub-frame.