CN1838559A - Receiver accessorial beam forming method - Google Patents

Abstract

This invention relates to an assistant wave beam shaped method of a receiving machine, the character of which is that the base station and user end are stored with a wave beam code with pluralities of wave beam matrixes given by the theory analysis or the computer simulation production; the channel state information of the user reviving machine is used to demodulate data and choose the wave beam matrix and wave beam vector with the maximum velocity; and the user end sends the wave beam matrix sequence with maximum velocity to the base station through the feedback circuit; and the base station computes each wave beam matrix's ejecting rate according to the users' feedback information, then it chooses wave beam matrix with maximum ejecting rate to multiple with the signal vector; then the base station uses amplitude amplifying method according to the users' data to allot total power to the used wave beam. Comparing to present algorithm, this invention uses minimum feedback quantum and least complexity of modulation algorithum to obtain higher ejecting rate.

Description

The beam forming method that a kind of receiver is auxiliary

Technical field:

The invention belongs to the single sub-district of mobile communication multi-antenna broadcast channel capacity technical field, particularly relate in the multi-antenna broadcast channel base station and dispatch and pretreated beam forming method according to user's feedback information.

Background technology:

Multi-user diversity is the effective ways of the single sub-district of present mobile communication multi-antenna broadcast channel capacity technical field broad research, raising wireless system spectrum efficiency.

" international electronics and The Institution of Electrical Engineers's information journal " (" On the achievable throughput of amultiantenna Gaussian broadcast channel ", IEEE Transactions on Information Theory, Vol.49No.7, July 2003) introduced and a kind ofly many antennas have been arranged at the base station end, user side has in the broadcast channel of single antenna and utilizes dpc techniques to reach the method for downlink broadcast channel and capacity, but because dirty paper code is non-linear, and need know other user's interfere information, make this method complexity too high, and the base station end needs channel condition information completely, and this is difficult to realize in practice.Zero beam forming method also introduced in this literary composition, because it must reverse to subscriber channel at the base station end, make this method need the base station end to have channel condition information completely, and its optimum user scheduling is owing to need search for optimum user's combination in all possible user's combination, computation complexity can be exponential increase along with the increase of number of users.

" international electronics and The Institution of Electrical Engineers's information journal " (" On the capacity of MIMO broadcast channelwith partial side information ", IEEE Transactions on Information Theory,) introduced a kind of random beam forming method that only needs partial channel knowledge that realizes multi-user diversity, but because the information of user feedback very little, the channel condition information of receiver is not fully utilized, and the throughput loss of this method is too big.

Technology contents:

The present invention proposes the auxiliary beam forming method of a kind of receiver, can make full use of the channel condition information of user side receiver, reduce the loss of throughput, and the base station end only need the part channel condition information, its optimum user dispatches the increase to linear growth of computation complexity with number of users.

The beam forming method that receiver of the present invention is auxiliary will send signal phasor in the base station and beamforming matrix multiplies each other, and described beamforming matrix is a unitary matrix; Each user feeds back to the base station by based on feedback link with the Signal Interference and Noise Ratio of maximum in its all possible Signal Interference and Noise Ratio that obtains and the sequence number of the beam vector of correspondence; Each beam vector is distributed to the user who obtains the peak signal interference-to-noise ratio on that beam vector according to feedback information in the base station; Receiver user is oppositely found the solution acquisition channel condition information at training stage pilot transmitted signal and this receiver in the received signal of training stage according to the base station;

It is characterized in that:

Store a wave beam code book that generates or provide by theory analysis, comprise a plurality of beamforming matrixs by Computer Simulation respectively in base station and user side; The channel condition information of receiver user also is used for selecting to make this user obtain the beamforming matrix and the beam vector of maximum rate at the wave beam code book except being used for data demodulates; User side sends the beamforming matrix sequence number that obtains maximum rate in the wave beam code book to base station by based on feedback link; The base station is calculated the throughput that each beamforming matrix can reach earlier according to user's feedback information, the beamforming matrix of therefrom selecting the throughput maximum then as with send the beamforming matrix that signal phasor multiplies each other; The mode that the base station enlarges by the amplitude of carrying out to the user's data in the validated user set with the gross power mean allocation to the wave beam of using.

Below pass through to analyze the principle of the inventive method institute foundation, and advantage of the present invention is described compared with prior art.

Comprise in single cell system that the present invention is suitable for a base station and K user (be numbered U_1 ..., U_K), the base station is equipped with M transmitting antenna, and each is user assembled a reception antenna, and K＞M, and the input/output relation of this system can be described as:

Y＝HSX+N (F1)

Wherein, Y is user's received signal vector of K * 1, and H is the channel matrix of K * M, and S is the preconditioning matrix of M * M, and X is the user data of M * 1, and N is the additive white Gaussian noise vector that the user of K * 1 receives.

In order to obtain higher throughput with less feedback quantity, each different to the preconditioning matrix S that subscriber channel reverses at base station end generation preconditioning matrix S or close-to zero beam forming randomly with existing random beam forming technology at the base station end, preconditioning matrix S among the present invention is to be that target selection is come out with the maximum throughput the wave beam code book of M * M beamforming matrix independently from the L that contains that is stored in base station and user side, wherein each beamforming matrix S _l, l=1 ..., L is a unitary matrix, and satisfies

E{|s _l，m| ²}＝1，l＝1，…，L；m＝1，…，M，

S wherein _LmBe beamforming matrix S _lM row.Do not carry out the channel reverse turn operation at the base station end like this, avoided since the channel reverse turn operation bring need the shortcoming of complete channel knowledge feedback in the basic beginning of a war; The result who selects beamforming matrix by the user in the wave beam code book simultaneously and will select feeds back to the mode of base station, and the base station obtains more channel knowledge than random beam forming method, thereby has improved throughput.

The user needs to obtain channel condition information in order to select optimum beamforming matrix in the wave beam code book, and this is oppositely to find the solution in the received signal of training stage with this receiver by the pilot signal of utilizing the base station end to send at user side to obtain.The operation of base station is divided into two stages: training stage and transfer phase.In the training stage, the base station does not send data at each antenna on pilot signal transmitted; At transfer phase, the base station only sends the dateout of pretreatment module on each antenna, do not send pilot tone.The dateout of pilot tone and pretreatment module is carried out time division multiplexing by multiplexer.

In the training stage of base station, the base station sends mutually orthogonal pilot tone on each antenna, and each user estimates channel gain vector h from M transmitting antenna of base station to this user's reception antenna by channel estimation module _i=[h _I1H _IM] ^T, wherein i is user's a sequence number.The user is with the channel gain vector h that estimates then _i=[h _I1H _IM] ^TBe input to signal noise interference ratio SINR computing module.SINR computing module user according to h _i=[H _I1H _IM] ^TIn conjunction with the beamforming matrix in the wave beam code book with following formula calculate its L * M the SINR that might reach:

${\mathrm{SINR}}_{i,l,m}=\frac{{\left|h_i{s}_{l,m}\right|}^2}{1/\mathrm{SNR}+\underset{j\≠i}{\mathrm{\Σ}}{\left|h_i{s}_{l,j}\right|}^2},m=1,\·\·\·,M;l=,\·\·\·,L---\left(F2\right)$

Wherein SNR is each user's a received signal noise ratio; Total transmitted power of supposing the base station is ρ, and the power on each wave beam is the same, so at this time total signal noise ratio TSNR=ρ/σ ²And each user's SNR=ρ/(M σ ²), σ wherein ²Variance for additive white Gaussian noise in user's the received signal.

After calculating all possible SINR, user i can obtain a beamforming matrix sequence number l _i ^*And the sequence number m of certain beam vector in the beamforming matrix _i ^*Make the SINR of their correspondences _{I, l*, m*} ^*Be maximum, that is:

$l_i^{*}=\underset{1\&le;l\&le;L}{\mathrm{<figure-callout\; id="1"\; label="arg"\; filenames="A20051003858600111.png,A20051003858600112.png"\; state="\{\{state\}\}">arg</figure-callout>}}\left\{\underset{1\&le;l\&le;L}{\mathrm{<figure-callout\; id="1"\; label="max"\; filenames="A20051003858600111.png,A20051003858600112.png"\; state="\{\{state\}\}">max</figure-callout>}}\right\{\max \left\{\underset{1\&le;m\&le;M}{{\mathrm{SINR}}_{i,l,\mathrm{<figure-callout\; id="1"\; label="m"\; filenames="A20051003858600111.png,A20051003858600112.png"\; state="\{\{state\}\}">m</figure-callout>}}}\right\}\left\}\right\}$

$m_i^{*}=\underset{1\&le;m\&le;M}{\mathrm{<figure-callout\; id="1"\; label="arg"\; filenames="A20051003858600111.png,A20051003858600112.png"\; state="\{\{state\}\}">arg</figure-callout>}}\left\{\underset{1\&le;l\&le;L}{\max }\right\{\max \left\{\underset{1\&le;m\&le;M}{{\mathrm{SINR}}_{i,l,\mathrm{<figure-callout\; id="1"\; label="m"\; filenames="A20051003858600111.png,A20051003858600112.png"\; state="\{\{state\}\}">m</figure-callout>}}}\right\}\left\}\right\}---\left(F3\right)$

Each user is with SINR then _{I, l*, m*} ^*, m _i ^*With the distinctive beamforming matrix sequence number of the present invention l _i ^*Form feedback information, send the base station to by based on feedback link, each user's feedback quantity is a real number and two integers like this, thereby has avoided in existing dirty paper code and the zero beam forming method and need have the huge feedback overhead that the method for whole channel condition informations is brought at the base station end.For the convenience of representing, definition:

${\mathrm{\&gamma;}}_{i,l,m}=\left\{\begin{array}{cc}{\mathrm{SINR}}_{i,l^{*},m^{*}}^{*},& l=l^{*},m=m^{*}\\ 0,& \mathrm{else}\end{array}\right.,i=1,\&CenterDot;\&CenterDot;\&CenterDot;,K---\left(F4\right)$

In the present invention, owing to introduced the wave beam code book, thus the feedback information that makes the base station obtain by based on feedback link is the matrix of a three-dimensional ${\&aleph;}_{K\&times;L\&times;M}=\left\{{\mathrm{\&gamma;}}_{i,l,m}\right\}.$ Base station end among the present invention is obtaining feedback information ${\&aleph;}_{K\&times;L\&times;M}=\left\{{\mathrm{\&gamma;}}_{i,l,m}\right\}$ Following step is followed in operation afterwards:

The first step to each beamforming matrix in the wave beam code book and each beam vector, is obtained and is had maximum γ _{I, l, m}That user, promptly obtain $i_{l,m}^{*}=\underset{1\&le;i\&le;K}{\mathrm{<figure-callout\; id="1"\; label="arg"\; filenames="A20051003858600111.png,A20051003858600112.png"\; state="\{\{state\}\}">arg</figure-callout>}}\left\{\underset{1\&le;i\&le;K}{\mathrm{<figure-callout\; id="1"\; label="max"\; filenames="A20051003858600111.png,A20051003858600112.png"\; state="\{\{state\}\}">max</figure-callout>}}\right\{{\mathrm{\&gamma;}}_{i,l,m}\left\}\right\},l=1,\&CenterDot;\&CenterDot;\&CenterDot;,L;m=1,\&CenterDot;\&CenterDot;\&CenterDot;,M;$

Second step, each beamforming matrix in the wave beam code book, calculate the throughput that it can reach with following formula:

$R_l=\underset{m=1}{\overset{M}{\mathrm{\&Sigma;}}}\left(\log (1+\underset{1\&le;i\&le;K}{\mathrm{<figure-callout\; id="1"\; label="max"\; filenames="A20051003858600111.png,A20051003858600112.png"\; state="\{\{state\}\}">max</figure-callout>}}\{{\mathrm{\&gamma;}}_{i,l,m}\}\right),l=1,\&CenterDot;\&CenterDot;\&CenterDot;,L---\left(F5\right)$

The 3rd goes on foot, and obtains to have the beamforming matrix S that maximum can reach throughput _L*, that is to say and obtain

$l^{*}=\underset{1\&le;l\&le;L}{\mathrm{<figure-callout\; id="1"\; label="arg"\; filenames="A20051003858600111.png,A20051003858600112.png"\; state="\{\{state\}\}">arg</figure-callout>}}\underset{1\&le;l\&le;L}{\max }\left\{R_i\right\}---\left(F6\right)$

In the 4th step, the user who finally chooses set can be expressed as $A=\left\{k\left(m\right)\right|k\left(m\right)=i_{l^{*},m}^{*},m=1,\&CenterDot;\&CenterDot;\&CenterDot;,M\},$ The user's who chooses M * 1 data vector X=[x _{K (1)}..., x _{K (M)}] ^TAmong M the user who supposes to comprise among the A, satisfy condition ${\mathrm{\&gamma;}}_{k\left(m\right),l^{*},m}\&NotEqual;0$ Number of users be M ', obvious M '≤M, so to each m=1 ..., M, carry out following operation: if ${\mathrm{\&gamma;}}_{k\left(m\right),l^{*},m}=0,$ Then the m circuit-switched data of data vector X multiply by 0, otherwise multiply by Obtain after the operation data output M * 1 data vector X of scheduler '.

The 5th step is with the dateout X ' input pretreatment module of scheduler; With l ^*Be input to wave beam code book module, obtain required beamforming matrix S by the beamforming matrix of storing in the index wave beam code book _L*In pretreatment module with X ' premultiplication beamforming matrix S _L*Obtain the dateout of watermark pre-processor.Go out by transmission antennas transmit then.

The set of definition validated user $A^{\&prime;}=\left\{k\right|k=i_{i^{*},m}^{*},\mathrm{and}{\mathrm{\&gamma;}}_{{k,l}^{*},m}\&NotEqual;0,m=1,\&CenterDot;\&CenterDot;\&CenterDot;,M\}.$ In the present invention, the base station is by enlarging the amplitude of carrying out of the user data in the validated user set in scheduler, to distributing power ρ/M ' on each m ' ∈ B beam vector.Like this, to each validated user k ∈ A ', its actual SINR can be expressed as:

${\mathrm{SINR}}_{{k,l}^{*},m^{\&prime;}}^{\&prime;}=\frac{{\left|h_k{s}_{m^{\&prime;}}\right|}^2}{1/S{\mathrm{NR}}^{\&prime;}+\underset{i{\&NotEqual;m}^{\&prime;},j\&Element;B}{\mathrm{\&Sigma;}}{\left|h_k{s}_j\right|}^2},k\&Element;A^{\&prime;}---\left(F7\right)$

The received signal (9) of user k ∈ A ' is:

$y_k=h_k{s}_{l^{*},m^{\&prime;}}{x}_k+\underset{j{\&NotEqual;m}^{\&prime;},j\&Element;B}{\mathrm{\&Sigma;}}{h}_i{s}_{i^{*},j}{x}_j+n_k,k{\&Element;A}^{\&prime;}---\left(F8\right)$

The user is with channel gain vector h _k=[h _K1H _KM] ^TAnd s _{L*, m '}Be input to data demodulation module and carry out coherent demodulation, the i.e. data message that can obtain to send to received signal.

The throughput that utilizes the present invention to reach is:

$R_{\mathrm{sum}}=\underset{k\&Element;A,m^{\&prime;}\&Element;B}{\mathrm{\&Sigma;}}\log (1+{\mathrm{SINR}}_{{k,l}^{*},m^{\&prime;}}^{\&prime;})---\left(F9\right)$

With each all at random the random wave bundle matrixes of generation of existing random beam forming technology, the SINR of each user feedback maximum compares with the sequence number of the beam vector of correspondence then, because the present invention introduces the wave beam code book at base station end, user side, the beamforming matrix that mates most with this subscriber channel by selection in the wave beam code book at the receiver user place, just can obtain the beamforming matrix of maximum SINR, then the sequence number of this SINR, corresponding beamforming matrix and the sequence number of beam vector be fed back to the base station.

Feeding back completely channel condition information with existing dirty paper code and close-to zero beam forming Technology Need is that M plural number is different, owing to need not carry out the channel counter-rotating at the base station end, the present invention only need feed back a real number and two integers, has saved feedback quantity effectively.Compare with existing random beam forming, the present invention is owing to carried out the operation that beamforming matrix is selected user and base station, the beamforming matrix feedback information that the base station provides at receiver down auxiliary, can obtain more subscriber channel state information, thereby the channel condition information that more effectively utilizes receiver to have has improved throughput significantly under the cost of the feedback quantity that only increases an integer.

With dirty paper code need be in all user arranges the optimum user of search arrange different and close-to zero beam forming need be in all user's combinations the optimum user of search make up different, because the dispatching algorithm among the present invention is in each beamforming matrix each beam vector to be carried out peaked search in all users, do not relate to and arranging and combinatorial operation, and the number of wave beam code book medium wave beam matrix and the number of beam vector fix, so its computation complexity is linear growth along with the growth of number of users K.

The auxiliary beam forming method of the receiver that the present invention proposes selects to have obtained gain by introducing the wave beam code book to the beamforming matrix of the used pretreatment module of each channel realization, more effectively utilized the channel condition information of receiver, thereby and utilized this state information to carry out more rational scheduling and improved throughput with the cost that increases very little feedback overhead.

Description of drawings:

Fig. 1 is the auxiliary beam forming method system schematic of receiver of the present invention.

Fig. 2 is that the throughput of auxiliary beam forming method of receiver and random beam forming method compares schematic diagram.

Embodiment:

Embodiment below in conjunction with description of drawings this method.

Embodiment 1:

Present embodiment has with the base station end that in M=3 transmitting antenna, the wave beam code book multi-antenna broadcast channel of L=8 beamforming matrix to be arranged be example, and the enforcement of the beam forming method that receiver is auxiliary is described.

Realization block diagram of the present invention is as shown in Figure 1:

At the base station end, K independent user data 1 are input to scheduler module 2, scheduler module is dispatched (wherein feedback information 14 comprises the sequence number of each user's maximum SINR, corresponding wave beam code book and the sequence number of beam vector) according to the user by the feedback information 14 that based on feedback link 15 feeds back to the base station end, the sequence number of obtaining the beamforming matrix of maximum throughput is delivered to the beamforming matrix that the memory index of storage wave beam code book 3 goes out to need, and selects the effective user data line amplitude of going forward side by side to enlarge the input data that obtain pretreatment module 4 simultaneously from the user data of K road.Also import from the beamforming matrix of wave beam code book 3 output and to lead pretreatment module 4, at pretreatment module 4 obtains the input data premultiplication of this module from wave beam code book 3 beamforming matrix, the data that obtain finally will send are launched by transmitting antenna 8 then.

At user's receiving terminal, the user obtains to receive data from reception antenna 9, and the reception data are input to channel estimation module 10 and data demodulation module 11 is carried out channel estimating and data demodulates respectively.At channel estimation module, the pilot signal of utilizing the base station end to send is oppositely found the solution in the received signal of training stage with this receiver and can be obtained channel condition information, and outputs to SINR computing module 12.The another one of SINR computing module 12 is input as all beamforming matrixs of wave beam code book 3, and the wave beam code book of user side is identical with the base station end.At SINR computing module 12, the user is according to obtaining feedback information 14 from the channel condition information of channel estimation module 10 inputs and the beamforming matrix of wave beam code book 3 inputs.Data demodulation module 11 is according to received signal, utilizes the beam vector in the wave beam code book 3 and the channel condition information of channel estimation module 10 to carry out the user data 13 that coherent demodulation finally recovers the base station end.Storing 8 beamforming matrixs that produce by Computer Simulation in the wave beam code book 3 in advance, as follows:

${\mathrm{<figure-callout\; id="1"\; label="S"\; filenames="A20051003858600111.png,A20051003858600112.png"\; state="\{\{state\}\}">S</figure-callout>}}_1=\left[\begin{array}{ccc}-0.6354+0.1489i& 0.6109-0.3518i& 0.1551-0.2304i\\ -0.5178+0.0370i& -0.3681+0.4141i& -0.4566-0.4637i\\ -0.5217-0.1801i& -0.3615-0.2556i& -0.0768+0.7025i\end{array}\right];$

${\mathrm{<figure-callout\; id="2"\; label="S"\; filenames="A20051003858600111.png,A20051003858600112.png"\; state="\{\{state\}\}">S</figure-callout>}}_2=\left[\begin{array}{ccc}-0.4581+0.7896i& 0.0871+0.0675i& -0.1352-0.3690i\\ -0.3689+0.1672i& 0.0442-0.2056i& -0.2082+0.8651i\\ 0.0060+0.0497i& -0.5302+0.8139i& -0.1495+0.1775i\end{array}\right];$

${\mathrm{<figure-callout\; id="3"\; label="S"\; filenames="A20051003858600111.png"\; state="\{\{state\}\}">S</figure-callout>}}_3=\left[\begin{array}{ccc}-0.1386-0.3426i& 0.3754+0.1864i& -0.8135-0.1612i\\ -0.0215+0.4983i& 0.7844-0.1230i& 0.1859-0.2937i\\ -0.4606+0.6345i& -0.4400-0.0136i& -0.3404-0.2749i\end{array}\right];$

$S_4=\left[\begin{array}{ccc}-0.3996+0.6219i& 0.5637-0.0228i& 0.2849-0.2326i\\ -0.1784+0.4119i& -0.4282-0.3232i& 0.2255+0.6781i\\ 0.2698+0.4234i& 0.1367+0.6126i& -0.4800+0.3517i\end{array}\right];$

${\mathrm{<figure-callout\; id="5"\; label="S"\; filenames="A20051003858600111.png,A20051003858600112.png"\; state="\{\{state\}\}">S</figure-callout>}}_5=\left[\begin{array}{ccc}-0.0081-0.12671i& 0.2317+0.9412i& -0.1810-0.1080i\\ 0.0524+0.58831i& 0.1007-0.0789i& -0.2427-0.7589i\\ -0.7952-0.0523i& -0.1997-0.0655i& -0.5626+0.0648i\end{array}\right];$

${\mathrm{<figure-callout\; id="6"\; label="S"\; filenames="A20051003858600111.png,A20051003858600112.png"\; state="\{\{state\}\}">S</figure-callout>}}_6=\left[\begin{array}{ccc}-0.4201-0.2545i& 0.2526+0.5874i& -0.3823-0.4513i\\ 0.1120-0.6523i& -0.2445-0.1475i& -0.5596+0.4089i\\ -0.4024-0.398i& 0.2110-0.6820i& 0.2143-0.3519i\end{array}\right];$

${\mathrm{<figure-callout\; id="7"\; label="S"\; filenames="A20051003858600111.png"\; state="\{\{state\}\}">S</figure-callout>}}_7=\left[\begin{array}{ccc}-0.1363+0.2811i& -0.3004+0.2544i& 0.6667+0.5505i\\ 0.2678+0.0007i& -0.4843+0.7329i& -0.3905-0.0639i\\ -0.3126+0.8561i& 0.2476+0.1100i& -0.3032-0.0631i\end{array}\right];$

${\mathrm{<figure-callout\; id="8"\; label="S"\; filenames="A20051003858600111.png"\; state="\{\{state\}\}">S</figure-callout>}}_8=\left[\begin{array}{ccc}-0.0001+0.3122& 0.6704-0.5923i& 0.2478+0.2021i\\ -0.8298+0.04444i& -0.1951-0.3457i& -0.3877+0.0400i\\ 0.4436-0.1234i& 0.0303-0.2030i& -0.7544+0.4205i\end{array}\right]$

S ₁..., S ₈All be unitary matrix, and satisfy E{|s _{L, m}| ²}=1, l=1 ..., 8; M=1,2,3, s wherein _{L, m}Be beamforming matrix S _lM row.

The operation of base station is divided into two stages: training stage and transfer phase.In the training stage, the base station does not send data at each antenna on pilot signal transmitted; At transfer phase, the base station only sends the dateout of pretreatment module on each antenna, do not send pilot tone.The dateout of pilot tone and pretreatment module is carried out time division multiplexing by multiplexer.

The operation of base station is divided into two stages: training stage and transfer phase.In the training stage, the base station only sends pilot tone 5 on each antenna, does not send the dateout 6 of pretreatment module 3, and wherein pilot tone is to be used for estimating channel condition information at user's receiver; At transfer phase, the base station only sends the dateout of pretreatment module on each antenna, do not send pilot tone.The dateout of pilot tone and pretreatment module is carried out time division multiplexing by multiplexer.

In the training stage of base station, the base station sends mutually orthogonal pilot tone on each antenna, and each user estimates channel gain vector h from M transmitting antenna of base station to this user's reception antenna by channel estimation module _i=[h _I1h _I2h _I3] ^T, wherein i is user's a sequence number.The user is with the channel gain vector h that estimates then _i=[h _I1h _I2h _I3] ^TBe input to signal noise interference ratio SINR computing module 13.At SINR computing module 13, the user is according to h _i=[h _I1h _I2h _I3] ^TIn conjunction with the beamforming matrix S in the wave beam code book 3 ₁..., S ₈Calculate the SINR that institute might reach with formula F 2, wherein to follow the wave beam code book of base station end be identical to the wave beam code book 3 of user side.

After calculating all possible SINR, user i obtains a beamforming matrix sequence number l according to formula F 3 _i ^*And the sequence number m of certain beam vector in the beamforming matrix _i ^*Make the SINR of their correspondences _{I, l*, m*} ^*Be maximum, each user is with SINR then _{I, l*, m*} ^*, n _i ^*With the distinctive beamforming matrix sequence number of the present invention l _i ^* Form feedback information 14, send the base station with 14 to by based on feedback link 15 at last.The base station obtains each user's feedback information by based on feedback link 15, forms the matrix of a three-dimensional ${\&aleph;}_{K\&times;8\&times;3}=\left\{{\mathrm{\&gamma;}}_{i,l,m}\right\},$ γ wherein _{I, l, m}As formula F 4 definition.Following operation is carried out in the base station then:

The first step:, obtain and have maximum γ to each beamforming matrix in the wave beam code book and each beam vector _{I, l, m}That user, that is to say and obtain $i_{l,m}^{*}=\underset{1\&le;i\&le;K}{\mathrm{<figure-callout\; id="1"\; label="arg"\; filenames="A20051003858600111.png,A20051003858600112.png"\; state="\{\{state\}\}">arg</figure-callout>}}\left\{\underset{1\&le;i\&le;K}{\mathrm{<figure-callout\; id="1"\; label="max"\; filenames="A20051003858600111.png,A20051003858600112.png"\; state="\{\{state\}\}">max</figure-callout>}}\right\{{\mathrm{\&gamma;}}_{i,l,m}\left\}\right\},l=1,\&CenterDot;\&CenterDot;\&CenterDot;,8;m=\mathrm{1,2,3};$

Second step:, calculate its throughput R that can reach with formula F 5 to each beamforming matrix in the wave beam code book _l, l=1 ..., 8;

The 3rd step: obtain according to formula F 6 and to have the beamforming matrix S that maximum can reach throughput _L*

In the 4th step, the user who finally chooses set can be expressed as $A=\left\{k\left(m\right)\right|k\left(m\right)=i_{l^{*},m}^{*},m=1,\&CenterDot;\&CenterDot;\&CenterDot;,3\},$ 3 * 1 data vector X=[x of the user who chooses _{K (1)}, x _{K (2)}, x _{K (3)}] ^TSatisfy condition ${\mathrm{\&gamma;}}_{k\left(m\right),l^{*},m}\&NotEqual;0$ Number of users be M ', obvious M '≤3, so to each m=1 ..., 3, carry out following operation: if ${\mathrm{\&gamma;}}_{k\left(m\right),l^{*},m}=0,$ Then the m circuit-switched data of data vector X multiply by 0, otherwise multiply by

Obtain after the operation data output 3 * 1 data vector X of scheduler '.

The 5th step is with the dateout X ' input pretreatment module of scheduler; With l ^*Be input to wave beam code book module, obtain required beamforming matrix S by the beamforming matrix of storing in the index wave beam code book _L*In pretreatment module with X ' premultiplication beamforming matrix S _L*Obtain the dateout of watermark pre-processor.Go out by transmission antennas transmit then.

The validated user set is $A^{\&prime;}=\left\{k\right|k=i_{l^{*},m}^{*},\mathrm{and}{\mathrm{\&gamma;}}_{k,l^{*},m}\&NotEqual;0,m=1,\&CenterDot;\&CenterDot;\&CenterDot;,3\},$ The received signal 9 of user k ∈ A ' can be represented with formula F 8.The user is with channel gain vector h _i=[h _I1h _I2h _I3] ^TAnd s _{L*, m '}Be input to data demodulation module 11 and carry out coherent demodulation to received signal, the data message 13 that promptly can obtain to send.

It is under the uncorrelated channel in Ruili at channel H that Fig. 2 has provided present embodiment, supposes the user in the independent identically distributed system, the beam forming that receiver is auxiliary and the throughput comparison diagram of random beam forming.Abscissa is a number of users among the figure, and it is that step-length increases progressively with 4 that number of users gets from 4 to 128, the system throughput of ordinate for representing with every hertz of Nat's per second.Curve A has provided the throughput of random beam forming when total signal to noise ratio TSNR is 0dB; Curve B has provided the throughput of the beam forming that receiver is assisted when total signal to noise ratio TSNR is 0dB; Curve C has provided the throughput of random beam forming when total signal to noise ratio TSNR is 5dB; Curve D has provided the throughput of the beam forming that receiver is assisted when total signal to noise ratio TSNR is 5dB; Curve E has provided the throughput of random beam forming when total signal to noise ratio TSNR is 10dB; Curve F has provided the throughput of the beam forming that receiver is assisted when total signal to noise ratio TSNR is 10dB.As seen from Figure 2, under the situation of number of users from 4 to 128, be 0,5 at total signal to noise ratio TSNR, under the situation of 10dB, the auxiliary throughput that beam forming method reached of receiver is than throughput height that random beam forming reached.And signal to noise ratio is high more the time, and this raising is big more.At number of users is 60, and when total signal to noise ratio was 10dB, the auxiliary throughput that beam forming method reached of receiver was than throughput height that random beam forming reached nearly 20%.

The receiver secondary beam forming method that the present invention proposes does not need channel condition information completely at the base station end, each user only need feed back a real number and two integers, than random beam forming, the feedback information (being used to refer to the sequence number of selected beamforming matrix in the wave beam code book) that has only increased by 3 bits improves system throughput effectively; The computation complexity of dispatching algorithm is with user's growth linear growth; Only having increased under the cost of 3 bit feedback amounts, obtained higher system throughput than random beam forming method.The auxiliary beam forming method of the receiver that the present invention proposes selects to have obtained gain by introducing the wave beam code book to the beamforming matrix of the used pretreatment module of each channel realization, more effectively utilized the channel condition information of receiver, thereby and utilized this state information to carry out more rational scheduling and improved throughput.

Claims (1)

1, the auxiliary beam forming method of a kind of receiver will send signal phasor in the base station and beamforming matrix multiplies each other, and described beamforming matrix is a unitary matrix; Each user feeds back to the base station by based on feedback link with the Signal Interference and Noise Ratio of maximum in its all possible Signal Interference and Noise Ratio that obtains and the sequence number of the beam vector of correspondence; Each beam vector is distributed to the user who obtains the peak signal interference-to-noise ratio on that beam vector according to feedback information in the base station; Receiver user is oppositely found the solution acquisition channel condition information at training stage pilot transmitted signal and this receiver in the received signal of training stage according to the base station;

It is characterized in that:

Store a wave beam code book that generates or provide by theory analysis, comprise a plurality of beamforming matrixs by Computer Simulation respectively in base station and user side; The channel condition information of receiver user also is used for selecting to make this user obtain the beamforming matrix and the beam vector of maximum rate at the wave beam code book except being used for data demodulates; User side sends the beamforming matrix sequence number that obtains maximum rate in the wave beam code book to base station by based on feedback link; The base station is calculated the throughput that each beamforming matrix can reach earlier according to user's feedback information, the beamforming matrix of therefrom selecting the throughput maximum then as with send the beamforming matrix that signal phasor multiplies each other; The mode that the base station enlarges by the amplitude of carrying out to the user's data in the validated user set with the gross power mean allocation to the wave beam of using.

Images