CN101286824B - Pre-coding method and system in MIMO system with multiple users - Google Patents

Abstract

The embodiment of the invention discloses a precoding method which comprises the steps that: a first codebook and a second codebook are arranged at a base station and a user terminal, the second codebook is a set of unitary matrices generated by vectors in the first codebook and by making use of channel matrices, vectors that ensure that the user terminal obtains the maximum coding gain are respectively selected for every user terminal in the first codebook and corresponding CQI values are calculated; the base station determines the user terminal to be precoded and chooses a unitary matrix from the second codebook according to the vectors in the first codebook and corresponding CQI values corresponding to the user terminal to be precoded or chooses a unitary matrix from the second codebook and determines the user terminal to be precoded according to vectors in the first codebook and the corresponding CQI values corresponding to every user terminal; the unitary matrices or submatrices thereof are used as precoding matrices to carry out precoding towards the user terminal to be precoded. The embodiment of the invention also discloses a precoding system, and failure of matching among a plurality of user terminals can be avoided by the method and the system of the invention.

Description

Method for precoding and system in the multiple users mimo system

Technical field

The present invention relates to the communications field, particularly method for precoding and system in the multiple users mimo system.

Background technology

Along with going deep into of multi-antenna technology research, multiple-input and multiple-output (Multiple-InputMultiple-Output, MIMO) technology from point-to-point single subscriber terminal system extension to the multiple users mimo system of point-to-multipoint, be that some antennas are distinguished different user terminals by coded system, be used to transmit different data, to obtain the spatial reuse gain, obtain higher transmission rate, thereby enlarge the channel capacity zone of multiple users; And, be used to transmit identical data for some antennas with same-code, and to obtain diversity gain, effectively opposing decline, thus improve the errored bit performance.

In the multiple users mimo system, when a plurality of user terminals will be shared same time and frequency resource, the multiple users of introducing road altogether disturbs that (Co-Channel Interference CCI), adopts precoding technique to reduce interference between multiple users usually.Such as, MIMO technology based on the transmitting terminal precoding, this technology is utilized space division multiple access (Spatial Division Multiple Access, SDMA) can on identical time, frequency and sign indicating number territory resource, transmit the information of a plurality of user terminals, can improve its power system capacity and spectrum efficiency greatly than the single subscriber terminal mimo system, and this technology combines the advantage of space diversity and spatial reuse, various channel circumstances all there is adaptability preferably, can improve power system capacity by the space division multiple access access technology, the increase system is to the robustness of transmission environment.

From information theory view, according to dirty paper code (Dirty Paper Coding, DPC) theoretical proof, in having the system of interference, if transmitting terminal can be known interference signal exactly, by certain precoding processing, can make that the channel capacity of EVAC (Evacuation Network Computer Model) is identical with the channel capacity of Non-Interference System at transmitting terminal.Because DPC need be known definite interference signal, its theory is difficult to be applied in the real system, and some suboptimum precoding techniques arise at the historic moment.One class is based on the precoding technique that Real-time Channel is handled, need transmitting terminal to know all or part of channel condition information (Channel State Information, CSI), as channel counter-rotating (Channel Inversion), block of channels diagonalization (Block Digonolization, BD) etc., the feedback quantity of this class precoding technique channel information is bigger, is unfavorable for practical application.The another kind of precoding technique that is based on code book.Precoding technique based on code book is all known a fixed codebook in advance at transmitting terminal and receiving terminal, system is according to channel quality indication (CQI) information (the Channel Quality Information of feedback, CQI) come from fixed codebook, to select suitable precoding vector that transmitting terminal is carried out precoding, have higher practical value owing to its feedback quantity is little.

Following mask body is introduced a kind of method of carrying out precoding at present in the multiple users mimo system, this method is: by fast fourier transform (Fast Fourier Transform, FFT) produce a plurality of each pre-coding matrix that is square formation, wherein, precoding vector number in each pre-coding matrix and transmitting antenna number equate; With these pre-coding matrixes as the fixed codebook of knowing in advance at base station and user terminal; The base station selects suitable precoding vector that transmitting antenna is carried out precoding from this fixed codebook according to the CQI and the vectorial sequence number of each user terminal feedback, and user terminal is decoded according to the fixed codebook of knowing in advance, thus the data that obtain transmitting.Because each pre-coding matrix is produced by the FFT conversion, thereby the vector in each pre-coding matrix all is quadrature, thereby can reduce the interference between user terminal greatly.

The method of this precoding based on the FFT code book can obtain better performance under the unity feedback situation, each user terminal will feed back a plurality of CQI values during unity feedback, and the feedback quantity of this moment is very big, under the limited situation of capacity, is difficult for using in practice.And when non-unity feedback, such as under the low-feedback situation, if each user terminal can only feed back a small amount of CQI value, then can't select only precoding vector carries out precoding in the base station to transmitting antenna, thereby the precoding poor-performing.

In addition, because when carrying out first-time precoding, selected each precoding vector should belong to same pre-coding matrix, and for for the method for precoding of FFT code book, be to go out behind the precoding vector to base station feedback by user terminal selecting, this precoding vector is certain vector in certain pre-coding matrix in first code book, and when non-unity feedback, the precoding vector that a plurality of user terminal selecting go out does not often belong to same pre-coding matrix, thereby when user terminal being dispatched in the base station, can't carry out precoding simultaneously to these user terminals, or be called the pairing failure.

As seen, there is the problem that pairing is failed between a plurality of user terminals in the method for carrying out precoding at present in the multiple users mimo system.

Summary of the invention

In view of this, the embodiment of the invention provides a kind of method for precoding, and this method can be avoided the pairing failure between a plurality of user terminals.

The embodiment of the invention also provides a kind of pre-coding system, and this system can avoid the pairing failure between a plurality of user terminals.

For achieving the above object, the technical scheme of the embodiment of the invention specifically is achieved in that

A kind of method for precoding, this method comprises:

In base station and user terminal first code book and second code book are set, described second code book is the unitary matrice set that is generated by the vector in first code book,

Each user terminal utilizes channel matrix, selects one and make described user terminal obtain the vector of maximum precoding gain in first code book; According to the vector in first code book of selecting, calculate corresponding channel quality indication (CQI) information CQI value;

The user terminal of precoding is determined to treat in the base station, the corresponding CQI value of vector sum in described first code book that obtains according to the described user terminal for the treatment of precoding, go out a unitary matrice from second codebook selecting, perhaps the corresponding CQI value of vector sum in described first code book that obtains according to described each user terminal of base station goes out a unitary matrice and determines the user terminal for the treatment of precoding from second codebook selecting;

The base station as pre-coding matrix, uses described pre-coding matrix to carry out precoding to the described user terminal of precoding for the treatment of described unitary matrice or its submatrix.

A kind of pre-coding system, this system comprises: base station and at least one user terminal of first code book and second code book are set, and described second code book is the unitary matrice set that first code book generates;

Each described user terminal utilizes channel matrix, selects one and make described user terminal obtain the vector of maximum precoding gain in first code book; According to the vector in first code book of selecting, calculate corresponding C QI value;

The user terminal of precoding is determined to treat in described base station, the corresponding CQI value of vector sum in described first code book that obtains according to the described user terminal for the treatment of precoding, from second codebook selecting go out a unitary matrice or described first code book that obtains according to described each user terminal the corresponding CQI value of vector sum, go out a unitary matrice and determine the user terminal for the treatment of precoding from second codebook selecting; As pre-coding matrix, use described pre-coding matrix to carry out precoding described unitary matrice or its submatrix to the described user terminal of precoding for the treatment of.

As seen from the above technical solution, the embodiment of the invention provides a kind of method for precoding and system, in base station and user terminal first code book and second code book are set all, this second code book is the unitary matrice set that is generated by the vector in first code book, each user terminal is selected the CQI value that vector sum in first code book that can obtain maximum precoding gain should the vector correspondence earlier, utilize the corresponding CQI value of vector sum in this first code book from second code book, to select a unitary matrice by the base station, the user terminal for the treatment of precoding carries out precoding, and unitary matrice is meant each column vector or the vectorial pairwise orthogonal of row in the matrix here.Because the method for precoding that provides in the embodiment of the invention is to determine the pre-coding matrix that a plurality of user terminals use simultaneously by the base station, thereby can avoid in the prior art precoding vector owing to a plurality of user terminal selecting not belong to the problem of the pairing failure that same pre-coding matrix produces, therefore can avoid the pairing failure between a plurality of user terminals.

Description of drawings

Fig. 1 is the method for precoding simplified diagram of the embodiment of the invention;

Fig. 2 is the method for precoding schematic flow sheet of the embodiment of the invention one;

Fig. 3 is the pre-coding system structural representation of the embodiment of the invention two;

Fig. 4 is the pre-coding system structural representation of the embodiment of the invention three.

Embodiment

For making purpose of the present invention, technical scheme and advantage clearer, below with reference to the accompanying drawing embodiment that develops simultaneously, the present invention is described in further detail.

The method that provides in the present embodiment is: in base station and user terminal first code book and second code book are set, described second code book is the unitary matrice set that is generated by the vector in first code book, and this method also comprises:

Each user terminal utilizes channel matrix, selects one and make described user terminal obtain the vector of maximum precoding gain in first code book; According to the vector in first code book of selecting, calculate corresponding channel quality indication (CQI) information CQI value; The user terminal of precoding is determined to treat in the base station, the corresponding CQI value of vector sum in described first code book that obtains according to the described user terminal for the treatment of precoding, go out a unitary matrice from second codebook selecting, perhaps the corresponding CQI value of vector sum in described first code book that obtains according to described each user terminal of base station goes out a unitary matrice and determines the user terminal for the treatment of precoding from second codebook selecting;

The base station as pre-coding matrix, uses described pre-coding matrix to carry out precoding to the described user terminal of precoding for the treatment of described unitary matrice or its submatrix.

Specifically, the two identical cover code books of storing in advance with user terminal in the base station, the first cover code book is that Grassmanian code book or channel vector quantize code book, the second cover code book is the method for utilizing random search of the prior art or separating system of linear equations, by the unitary matrice that part or all of vector the generated set in the first cover code book, unitary matrice is meant each the vectorial pairwise orthogonal in the matrix here.Such as, when number of transmit antennas is M, then needs to generate M-1 vector and constitute the unitary matrice that dimension is M * M with a base vector.The generative process of unitary matrice is: earlier with a vector in first code book as base vector, according to the primary vector of base vector generation with this base vector quadrature, generate then and existing base vector and primary vector secondary vector of quadrature all, by that analogy, up to the vector that generates M-1 pairwise orthogonal, constitute unitary matrice.

With two transmitting antennas, each transmitting antenna is that example specifies at two kinds of alternative precoding vectors below, and form 1 is the first cover code book, is one 2 * 4 Grassmanian code book.

Index	0	1	2	3
					Column vector	-0.1612-0.7348j -0.5135-0.4128j	-0.1612-0.7348j -0.5135-0.4128j	-0.2399+0.5985j -0.7641-0.0212j	-0.9541 0.2996

Form 1

Form 2 is second cover code books, is 42 * 2 unitary matrice that adopt prior art to generate by 4 column vectors in the form 1, and promptly the size of unitary matrice is determined by transmitting antenna.Unitary matrice of each behavior in the form 2.

Index?0	-0.1612-0.7348j -0.5135-0.4128j	0.4094-0.5162j 0.0053+0.7523j
			Index?1	-0.1612-0.7348j -0.5135-0.4128j	0.6446-0.6902j 0.3132-0.1001j
Index?2	-0.2399+0.5985j -0.7641-0.0212j	-0.0568+0.7622j 0.6198-0.1777j
			Index?3	-0.9541 0.2996	0.2272+0.1954j 0.7233+0.6221j

Form 2

Following mask body describes for example, Fig. 1 is the method for precoding simplified diagram of the embodiment of the invention, as shown in Figure 1, if user terminal to be scheduled has N, each user terminal to be scheduled is selected one the code book from first cover and is made vector that user terminal precoding gain is maximum and calculate corresponding C QI value, sequence number and the corresponding C QI value of vector in the first cover code book of selecting sent to the base station, the base station recovers vector according to the sequence number and the first cover code book of vector in the first cover code book, and user terminal dispatched, from N user terminal, select K user terminal, from the second cover code book, select a unitary matrice, determine pre-coding matrix by unitary matrice, then K the user terminal of selecting carried out precoding simultaneously, and the sequence number of pre-coding matrix in the second cover code book of data after the precoding and employing sent to K user terminal.

Fig. 2 is the method for precoding schematic flow sheet of the embodiment of the invention.In the present embodiment, the transmitting antenna of supposing the base station is M, and according to the compromise of system feedback amount and performance, preestablishing the first cover code book is the matrix of M * L1, that is to say, comprises L1 column vector in the matrix of the first cover code book, and establishing the first cover code book is V=[v ₁v ₂... v _L1], wherein, v ₁First column vector in the expression first cover code book, v ₂Second column vector in the expression first cover code book, by that analogy.Can produce L2 unitary matrice by each column vector in the first cover code book, wherein, L2≤L1.Constitute the second cover code book W={w by these L2 unitary matrice ₁..., w _L2, w wherein _kUnitary matrice for M * M.When the number of unitary matrice in the vectorial number in the first cover code book and the second cover code book equated, the second cover code book had comprised the institute's directed quantity in the first cover code book fully, thereby logically, can be merged into a cover code book to two cover code books.As shown in the figure, specifically may further comprise the steps:

Step 201: each user terminal is selected the maximum column vector of the feasible gain of precoding separately from the first cover code book, and sequence number and the corresponding C QI value of this column vector in the first cover code book fed back to the base station.

In the present embodiment, each user terminal utilizes channel matrix, when selecting the vector of first cover in the code book as precoding vector, can obtain the column vector of maximum precoding gain, and calculate when adopting this column vector to carry out precoding, the CQI value that can produce, and with the value of this CQI with this vector is corresponding feeds back to the base station in first sequence number of overlapping column vector in the code book.Suppose that number of transmit antennas is M, the reception antenna number is N, then user terminal u _kThe mimo channel matrix H _kFor:

$H_k=\left[\begin{array}{cccc}{H}_{\mathrm{1,1}}& {\mathrm{<figure-callout\; id="2"\; label="H"\; filenames="S2008100568086E00011.png"\; state="\{\{state\}\}">H</figure-callout>}}_{\mathrm{2,1}}& ...& H_{1,\mathrm{<figure-callout\; id="2"\; label="M\; H"\; filenames="S2008100568086E00011.png"\; state="\{\{state\}\}">M</figure-callout>}}& \\ H_{}{}_{\mathrm{2,1}}& {\mathrm{<figure-callout\; id="2"\; label="H"\; filenames="S2008100568086E00011.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="H"\; filenames="S2008100568086E00011.png"\; state="\{\{state\}\}">H</figure-callout>\; </figure-callout>}}_{\mathrm{2,2}}& ...& {\mathrm{<figure-callout\; id="2"\; label="H"\; filenames="S2008100568086E00011.png"\; state="\{\{state\}\}">H</figure-callout>}}_{2,M}\\ ...& & & \\ H_{N,1}& H_{N,2}& ...& H_{N,M}\end{array}\right]---\left(1\right)$

Wherein, H _{I, j}Be the channel magnitude of j transmit antennas to i root reception antenna.For simplifying the analysis, present embodiment hypothesis channel is a flat fading channel, and those skilled in the art can draw method for precoding under other type fading channel according to present embodiment.

In the present embodiment, each column vector in each user terminal traversal first cover code book is therefrom selected the column vector that can obtain maximum precoding gain.For user terminal u _k, according to as formula (2), choose the column vector that can obtain maximum precoding gain:

$v_{i_{-}{u}_k}=\underset{v_i\&Element;V}{\arg \max }\left({\left|\right|H_k{v}_i\left|\right|}^2\right),i=1,...,L1---\left(2\right)$

Wherein, || H _kv _i|| ²Expression precoding gain, || x|| ²Be the norm of x, x=arg maxf (x) is the x value of correspondence when f (x) value is maximum, v _iExpression user terminal u _kVector in first code book of traversal.Selecting the column vector v that can obtain maximum precoding gain _{I_uk}After, calculate user terminal u according to formula (3) _kCQI, i.e. CQI _Uk:

$\mathrm{CQ}{I}_{u_k}={\left|\right|\frac{H_k{v}_{i_{-}{u}_k}}{{\mathrm{\&sigma;}}^2}\left|\right|}^2---\left(3\right)$

Here, σ ²Be noise variance.

Column vector v in selecting the first cover code book _{I_uk}And corresponding C QI _UkAfterwards, user terminal u _kWith v _{I_uk}Vectorial sequence number and CQI in the first cover code book _UkFeed back to the base station.

Step 202: the base station recovers first column vector of overlapping in the code book that each user terminal selecting goes out according to the column vector sequence number of each user terminal feedback.

Has received after the CQI value and column vector sequence number of each user terminal feedback the base station, recovers first column vector of overlapping in the code book that each user terminal selecting goes out.Such as, according to user terminal u _kThe column vector sequence number of feedback is 2, then recovers and makes user terminal u _kThe column vector that obtains maximum precoding gain is v ₂

Step 203: one group of user terminal and corresponding best unitary matrice thereof for the treatment of precoding is determined according to the column vector of first cover in the code book and the CQI value of each user terminal feedback that each user terminal selecting goes out in the base station from each user terminal.

In the present embodiment, the dispatching method of base station is the CQI by feedback _kAnd the vector v in the first cover code book that recovers _{I_uk}The scheduling cost function of being constructed is dispatched.Dispatch cost function as shown in Equation (4):

$f({\mathrm{CQI}}_{u_1},...,{\mathrm{CQI}}_{u_K};v_{i_{-}{u}_1},...,v_{i_{-}{u}_K};w_l)=\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}\left({\mathrm{CQI}}_{u_k}{v}_{i_{-}{u}_k}^H{\left[w_l\right]}_k\right),l=1,...,\mathrm{<figure-callout\; id="2"\; label="L"\; filenames="S2008100568086E00011.png"\; state="\{\{state\}\}">L</figure-callout>}2,---\left(4\right)$

Wherein, scheduling cost function physical significance is the total capacity value of system, particularly, and CQI _U1Be user terminal u ₁The CQI value of feedback, v _{I_uk}For according to user terminal u _kColumn vector in the first cover code book that the column vector SN recovery of feedback goes out, v _{I_uk} ^HBe v _{I_uk}The complex-conjugate transpose matrix, w _l∈ W, l=1 ..., L2 is a l unitary matrice among the second cover code book W, L2 is the unitary matrice sum among the second code book W, [w _l] _kExpression w _lK column vector.Dispatching principle is the user { u that chooses feasible scheduling cost function maximum ₁..., u _K} _SCHAs sending the user, choose a unitary matrice in the second cover code book simultaneously

Transmission pre-coding matrix as these users.As shown in Equation (5):

$\stackrel{\&OverBar;}{w},u_1,...,u_K=\arg \max \left(f\right)---\left(5\right)$

For the second cover code book that includes L2 unitary matrice in the present embodiment, travel through each user terminal at each unitary matrice respectively, find out the user terminal and the pre-coding matrix of feasible (5) formula maximum The unitary matrice size of second cover in the code book be M * M, then by in the definite pre-coding matrix of unitary matrice, supports at most that M is individual to treat that the user terminal of precoding carries out precoding simultaneously and sends data.When the number of the user terminal for the treatment of precoding equals the transmitting antenna number,

Be the unitary matrice of second cover in the code book, if the number of user terminal for the treatment of precoding is during less than the transmitting antenna number M,

Be a submatrix of a unitary matrice in the second cover code book, Comprised the part rows vector in the unitary matrice, its number equals to treat the number of the user terminal of precoding.Present embodiment for convenience of description, supposing need be at each user terminal u ₁..., u _KIn select the user terminal that M treats precoding.The method that travels through each user terminal at each unitary matrice has multiple, when realizing, can adopt some simple traversal methods according to complexity and the compromise of accuracy as a result, such as, comprised the unitary matrice w of M precoding vector for any _l=[t ₁, t ₂..., t _M] l=1 ..., L2 is at each user terminal u of traversal ₁' ..., u _N' time, can be as shown in table 3, can be earlier from t ₁Column vector begins, to each user terminal u _n' calculating CQI _{Un '}(v _{I_un '} ^Ht ₁), n=1,2......N writes down the row at result of calculation maximum place, and this row is and is adopting w _lIn one row t ₁During as a precoding vector, a user terminal u who treats precoding of the correspondence of selecting ₁, then as above-mentioned step process column vector t ₂, at this moment, no longer consider the user terminal for the treatment of precoding selected, can will treat all values zero setting of the pairing row of user terminal of precoding.By that analogy, up to finding out K different user terminal { u ₁..., u _K} _SCHObtain adopting this group user to send and adopt unitary matrice w the precoding user terminal substitution formula of selecting (4) for the treatment of _lPower system capacity during as pre-coding matrix.

Form 3

More than, all to travel through each user terminal, finally obtain corresponding power system capacity value, to code book W={w for each unitary matrice in the second cover code book for travel through the method for each user terminal at a unitary matrice ₁..., w _L2In L2 pre-coding matrix finish the traversal computing after, the pre-coding matrix that selects the power system capacity maximum with and one group of corresponding user terminal for the treatment of precoding.

Step 204: the one group user terminal { u that treat precoding of base station to determining ₁..., u _K} _SCHData carry out precoding, and will organize the pre-coding matrix of user terminal Sequence number informing user terminal in the second cover code book.

In the present embodiment, precoding is carried out to the data of these user terminals, just with pre-coding matrix in the base station behind the pre-coding matrix of selecting the user terminal for the treatment of precoding and employing thereof

In l column vector organize the precoding vector of l user terminal as this, each precoding vector respectively with this user terminal on data on the corresponding transmitting antenna multiply each other, row merges the back transmission during with the data after the precoding, and { u is also informed with the sequence number of pre-coding matrix in the second cover code book in the base station ₁..., u _K} _SCHIn each user terminal, detect use for each user terminal.Step 205 detects the process that receives data for user terminal.

Step 205: each treats the data after pre-coding matrix sequence number that precoding user terminal basis receives and the precoding that receives, and carries out MIMO and detects.

In the present embodiment, for user terminal u _k, the data y that it receives _kFor:

$y_k=H_k\stackrel{\&OverBar;}{w}s+n_k,k=1,...,K---\left(6\right)$

Wherein, $s={\left[\begin{array}{cccc}\sqrt{p_1}{s}_{u_1}& \sqrt{p_2}{s}_{{\mathrm{<figure-callout\; id="2"\; label="u"\; filenames="S2008100568086E00011.png"\; state="\{\{state\}\}">u</figure-callout>}}_2}& ...& \sqrt{p_k}{s}_{u_k}\end{array}\right]}^T,$ $\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}{p}_k=p_0$

N in the formula (6) _kBe user terminal u _kThe additive white Gaussian noise that receives, and obedience CN (0, N ₀) distribution.p _kIt is the transmitted power of k data flow; p ₀Transmission gross power for all data flow; S is a transmission signal vector, comprises the data flow of K user terminal;

Pre-coding matrix for M * K dimension comprises K precoding vector, [.] ^TExpression channel transposition.

{ u ₁..., u _K} _SCHIn the sequence number of pre-coding matrix in the second cover code book that adopted according to the transmission precoding of base station notice of each user terminal, obtain pre-coding matrix

, then user terminal obtains H in channel estimating _kAfter, can calculate the channel after the precoding

Present embodiment adopts MMSE criterion of the prior art, the linear transformation matrix that obtains user terminal uk as shown in Equation (8):

${\tilde{G}}_k={\tilde{h}}_k^H{({\tilde{H}}_k{\tilde{H}}_k^H+\frac{{\mathrm{KN}}_0}{p_0}{I}_N)}^{-1}---\left(7\right)$

Wherein, ${\tilde{h}}_k={\left[{\tilde{H}}_k\right]}_k={\left[H_k\stackrel{\&OverBar;}{w}\right]}_k$ K column vector of the channel after the expression precoding, I _NBe unit matrix.

If present embodiment adopts the LMMSE detection method, can obtain user terminal u _kDetected data ${\hat{s}}_k={\tilde{G}}_k{y}_k,k=1,...,K.$

If present embodiment is to estimate when according to the channel matrix of special pilot design after with precoding, then step 204 need not notified pre-coding matrix to user terminal by the base station Sequence number user terminal in the second cover code book.And step 205 changes into: each treat that the precoding user terminal is estimated the channel matrix after the precoding that according to special pilot tone and the precoding that receives after data, carry out MIMO and detect.

In the present embodiment, each user terminal only need feed back a CQI value and a sequence number value, and in the situation that the base station end can not exist pairing to lose efficacy, can obtain good performance simultaneously.

And if when TDD system lower channel has reciprocity, think that promptly each user's the channel information of up-downgoing is the same, can know the channel information that travelling carriage is descending at the base station end by the user uplink link, therefore the operation of user terminal in the step 201 can be placed on the base station carries out, be specially: the base station is according to the up channel of each user terminal, from the first cover code book, select the maximum column vector of the feasible gain of precoding separately and calculate its corresponding CQI value, thereby reduced this link of feedback.Then, do not need execution in step 202, and directly utilize this vector to find out optimum unitary matrice and treat one group of user of precoding in the second cover code book, thereby reduced vectorial recovery link, other steps are with identical in the present embodiment.

Certainly, present embodiment also is not limited to by dynamic dispatching, when determining the user terminal for the treatment of precoding, determine the best unitary matrice among the user to be encoded, it is pre-coding matrix, also can determine user terminal to be encoded earlier by other modes, vector sum corresponding C QI value in first code book that obtains according to user terminal to be encoded is determined best unitary matrice from second code book again.

Particularly, for the occasion that sets in advance the user terminal for the treatment of precoding, establish described preset user terminal u ₁..., u _NVector in described first code book that obtains is v _{I_u1}..., v _{I_uN}Corresponding C QI value is CQI _U1..., CQI _UN, then step 203 is: utilize the corresponding CQI value of vector sum in first code book that the preset user terminal obtains to construct the selection cost function, $f({\mathrm{CQI}}_{u_1},...,{\mathrm{CQI}}_{u_N};v_{i_{-}{u}_1},...,v_{i_{-}{u}_N};w_l)=\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}\left({\mathrm{CQI}}_{u_n}{v}_{i_{-}{u}_n}^H{\left[w_l\right]}_n\right),l=1,...,\mathrm{<figure-callout\; id="2"\; label="L"\; filenames="S2008100568086E00011.png"\; state="\{\{state\}\}">L</figure-callout>}2,$ Travel through each unitary matrice in second code book, from second code book, select to make a unitary matrice of described selection cost function maximum.The operation of other steps is same as the previously described embodiments.

And, determined the occasion of the user terminal for the treatment of precoding for before determining pre-coding matrix, establishing one group of user terminal for the treatment of precoding selecting in advance is { u ₁..., u _K} _SCH, then step 203 is: utilize the one group of user terminal { u that has selected ₁..., u _K} _SCHIn first code book that obtains of each user terminal in the corresponding CQI value of vector sum, structure scheduling cost function $f({\mathrm{CQI}}_{u_1},...,{\mathrm{CQI}}_{u_K};v_{i_{-}{u}_1},...,v_{i_{-}{u}_K};w_l)=\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}\left({\mathrm{CQI}}_{u_k}{v}_{i_{-}{u}_k}^H{\left[w_l\right]}_k\right),l=1,...,\mathrm{<figure-callout\; id="2"\; label="L"\; filenames="S2008100568086E00011.png"\; state="\{\{state\}\}">L</figure-callout>}2,$ Travel through each unitary matrice in second code book, from second code book, select to make a unitary matrice of described scheduling cost function maximum.The operation of other steps is same as the previously described embodiments.

In addition, the first cover code book and the second cover code book are that example describes with column vector as precoding vector all in the present embodiment, can certainly adopt the form of expression of row vector.Can adopt multiple traversal method in the prior art to each matrix in first each vector of cover code book, the second cover code book and to each traversal method of organizing user terminal.Each user terminal to the base station also can feed back a CQI value and characterize the vector of first cover in the code book out of Memory or should vector.

The pre-coding system structural representation that Fig. 3 adopts for the embodiment of the invention, this system comprises: base station 320 and at least one user terminal 310 that first code book and second code book are set;

Each user terminal 310 utilizes channel matrix, selects one and make described user terminal obtain the vector of maximum precoding gain in first code book; According to the vector in first code book of selecting, calculate corresponding C QI value, the vector in the described CQI value and first code book selected is sent to the base station.

The user terminal of precoding is determined to treat in base station 320, and the corresponding CQI value of vector sum in first code book that obtains according to the user terminal 310 for the treatment of precoding, go out a unitary matrice from second codebook selecting, perhaps the corresponding CQI value of vector sum in described first code book that obtains according to described each user terminal goes out a unitary matrice and determines the user terminal for the treatment of precoding from second codebook selecting; As pre-coding matrix, the user terminal for the treatment of precoding uses pre-coding matrix to carry out precoding with unitary matrice or its submatrix.

User terminal 310 comprises:

Vector is selected module 311, utilizes channel matrix, selects one and make described user terminal obtain the vector of maximum precoding gain in first code book.

CQI value computing module 312 is used for the vector according to first code book of selecting, and calculates corresponding C QI value.

Vector selects module 311 to comprise:

First computing unit 313 is used for user terminal u _kTravel through each vector in first code book, according to formula || H _kv _i|| ², i=1 ..., L1 calculates the precoding gain when adopting each described vector to carry out precoding, wherein, H _kBe user terminal u _kThe mimo channel matrix, v _iExpression user terminal u _kVector in first code book of traversal.

Selected cell 314 is used for the result of calculation that obtains according to described computing module, selects a vector that makes described result of calculation maximum.

CQI value computing module 312 comprises:

Acquiring unit 315 is used for obtaining the vector v that described vector is selected first code book that module 311 selects _{I_uk}

Second computing unit 316 is used for according to formula $\mathrm{CQ}{I}_{u_k}={\left|\right|\frac{H_k{v}_{i_{-}{u}_k}}{{\mathrm{\&sigma;}}^2}\left|\right|}^2$ Calculate the v that described acquisition module obtains _{I_uk}Corresponding C QI value, wherein, σ ²The expression noise variance.

Base station 320 comprises:

Coding pretreatment module 321, when pre-determining the user terminal for the treatment of precoding, be used to determine to treat the user terminal of precoding, and the corresponding CQI value of vector sum in described first code book that obtains according to the described user terminal for the treatment of precoding, go out a unitary matrice that makes the power system capacity maximum from second codebook selecting, when not determining to treat the user terminal of precoding, the corresponding CQI value of vector sum that is used for described first code book that obtains according to each user terminal goes out a unitary matrice that makes the power system capacity maximum and determines the user terminal for the treatment of precoding from second codebook selecting.

Precoding module 322 is used for unitary matrice that described coding pretreatment module 321 is obtained or its submatrix as pre-coding matrix, uses described pre-coding matrix to carry out precoding to each user terminal.

For the occasion of not using scheduling, coding preliminary treatment 321 comprises:

The first cost construction of function unit 323 is used to utilize each preset user terminal u ₁..., u _NVector v in described first code book that obtains _{I_u1}..., v _{I_uN}With corresponding CQI value CQI _U1..., CQI _UN, structure is selected cost function $f({\mathrm{CQI}}_{u_1},...,{\mathrm{CQI}}_{u_N};v_{i_{-}{u}_1},...,v_{i_{-}{u}_N};w_l)=\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}\left({\mathrm{CQI}}_{u_n}{v}_{i_{-}{u}_n}^H{\left[w_l\right]}_n\right),l=1,...,\mathrm{<figure-callout\; id="2"\; label="L"\; filenames="S2008100568086E00011.png"\; state="\{\{state\}\}">L</figure-callout>}2,$ Wherein, v _{I_un} ^HUser terminal u _nVector v in first code book of selecting _{I_un}The complex-conjugate transpose matrix, w _lBe l unitary matrice among the second code book W, L2 is the unitary matrice sum among the second code book W, [w _l] _nExpression w _lN-component column vector.

The first matrix selected cell 324 is used for traveling through each unitary matrice of second code book, selects to make a unitary matrice of the selection cost function maximum in the described first cost construction of function module 323 from second code book.

When pre-determining out the user terminal for the treatment of precoding, the coding pretreatment module comprises:

Scheduling unit is used for by dynamic dispatching, the one group of user terminal { u that treats precoding that selects in advance ₁..., u _K} _SCH

The second cost construction of function unit is used to utilize described scheduling unit to select one group of user terminal { u ₁..., u _K} _SCHIn each user terminal u ₁..., u _KVector v in described first code book that obtains _{I_u1}..., v _{I_uK}With corresponding CQI value CQI _U1..., CQI _UK, structure scheduling cost function $f({\mathrm{CQI}}_{u_1},...,{\mathrm{CQI}}_{u_K};v_{i_{-}{u}_1},...,v_{i_{-}{u}_K};w_l)=\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}\left({\mathrm{CQI}}_{u_k}{v}_{i_{-}{u}_k}^H{\left[w_l\right]}_k\right),l=1,...,\mathrm{<figure-callout\; id="2"\; label="L"\; filenames="S2008100568086E00011.png"\; state="\{\{state\}\}">L</figure-callout>}2,$ Wherein, v _{I_uk} ^HUser terminal u _kVector v in first code book of selecting _{I_uk}The complex-conjugate transpose matrix, w _lBe l unitary matrice among the second code book W, L2 is the unitary matrice sum among the second code book W, [w _l] _kExpression w _lK column vector;

The second matrix selected cell is used for traveling through each unitary matrice of second code book, selects to make a unitary matrice of the scheduling cost function maximum in the described second cost construction of function unit from second code book.

The second matrix selected cell is used for traveling through each unitary matrice of second code book, selects to make a unitary matrice of the scheduling cost function maximum in the described second cost construction of function unit from second code book.

For the occasion of determining pre-coding matrix in scheduling, the coding pretreatment module comprises:

Spider module is used for traveling through each unitary matrice of second code book, travels through each user terminal to be scheduled at each unitary matrice.

The 3rd matrix selected cell is used for the result that obtains according to described spider module, selects to make described scheduling cost function $f({\mathrm{CQI}}_{u_1},...,{\mathrm{CQI}}_{u_K};v_{i_{-}{u}_1},...,v_{i_{-}{u}_K};w_l)=\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}\left({\mathrm{CQI}}_{u_k}{v}_{i_{-}{u}_k}^H{\left[w_l\right]}_k\right),l=1,...,\mathrm{<figure-callout\; id="2"\; label="L"\; filenames="S2008100568086E00011.png"\; state="\{\{state\}\}">L</figure-callout>}2,$ One group of maximum user terminal { u that treats precoding ₁..., u _K} _SCHWith a corresponding unitary matrice, wherein, v _{I_uk} ^HUser terminal u _kVector v in first code book of selecting _{I_uk}The complex-conjugate transpose matrix, w _lBe l unitary matrice among the second code book W, L2 is the unitary matrice sum among the second code book W, [w _l] _kExpression w _lK column vector.

The base station can also comprise:

Sending module is used for unitary matrice or unitary matrice that described matrix selects module to select are sent to each user terminal in the sequence number of described second code book; Or estimate according to the channel matrix of special pilot design after with precoding.

If when TDD system lower channel has reciprocity, can be by the up channel of base station according to each user terminal, from the first cover code book, select the maximum column vector of the feasible gain of precoding separately and calculate its corresponding CQI value, at this moment vector selects module and CQI value computing module to be arranged in the base station, specifically as shown in Figure 4.Fig. 4 is the pre-coding system in the TDD system that has reciprocity at channel, and this system comprises: base station 400 and at least one user terminal 450, the second code book that first code book and second code book are set are the unitary matrice set that first code book generates.Wherein, base station 400 comprises:

Vector is selected module 410, utilizes MIMO up channel matrix, selects one and make described user terminal obtain the vector of maximum precoding gain in first code book.

CQI value computing module 420 is used for the vector according to first code book of selecting, and calculates corresponding C QI value.

Coding pretreatment module 430, when pre-determining the user terminal for the treatment of precoding, the corresponding C QI value of selecting the vector sum CQI value computing module 420 in first code book of the user terminal correspondence for the treatment of precoding that module 410 obtains to draw according to vector goes out a unitary matrice that makes the power system capacity maximum from second codebook selecting; When not pre-determining the user terminal for the treatment of precoding, the corresponding C QI value of selecting the vector sum CQI value computing module 420 in first code book of each user terminal correspondence that module 410 obtains to draw according to vector, go out a unitary matrice that makes the power system capacity maximum from second codebook selecting, determine the user terminal for the treatment of precoding simultaneously.

Precoding module 440 as pre-coding matrix, uses described pre-coding matrix to carry out precoding to each user terminal the coding unitary matrice that obtains of pretreatment module 430 or its submatrix.

Vector selects the internal structure of module 410 to select module 311 identical with the vector in Fig. 3 user terminal, also comprise first computing unit and selected cell, the internal structure of CQI value computing module 420 is also identical with CQI value computing module 312 in Fig. 3 user terminal, comprises second computing unit and acquiring unit.Coding pretreatment module 430 is also identical with coding pretreatment module 321 among Fig. 3, for the occasion of not using scheduling, for the occasion of in scheduling, determining pre-coding matrix or for the occasion that pre-determines out the user terminal for the treatment of precoding, have different internal structures, do not give unnecessary details one by one here.

Certainly, the user terminal that provides in the present embodiment and the structure of inside of base station are functional modules, when realizing, can merge or do further segmentation, and can be arranged in different physical structures.

It will be appreciated by those skilled in the art that accompanying drawing is the schematic diagram of a preferred embodiment, the module in the accompanying drawing might not be that enforcement the present invention is necessary.

The invention described above embodiment sequence number is not represented the quality of embodiment just to description.

The scheme of the content record of claim also is the protection range of the embodiment of the invention.

One of ordinary skill in the art will appreciate that all or part of processing in the foregoing description method is to instruct relevant hardware to finish by program, described program can be stored in a kind of computer-readable recording medium.

The above is preferred embodiment of the present invention only, is not to be used to limit protection scope of the present invention.All any modifications of being done within the spirit and principles in the present invention, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (23)

1. a method for precoding is characterized in that, in base station and user terminal first code book and second code book is set, and described second code book is the unitary matrice set that is generated by the vector in first code book, and this method also comprises:

Utilize channel matrix, in first code book, select the vector that makes this user terminal obtain maximum precoding gain respectively for each user terminal; According to the vector in first code book of selecting, calculate corresponding channel quality indication (CQI) information CQI value;

The user terminal of precoding is determined to treat in the base station, according to the corresponding CQI value of vector sum in described first code book of the described user terminal correspondence for the treatment of precoding, go out a unitary matrice from second codebook selecting, perhaps the base station goes out a unitary matrice and determines the user terminal for the treatment of precoding from second codebook selecting according to the corresponding CQI value of vector sum in described first code book of described each user terminal correspondence;

The base station as pre-coding matrix, uses described pre-coding matrix to carry out precoding to the described user terminal of precoding for the treatment of described unitary matrice or its submatrix.

2. method according to claim 1 is characterized in that, describedly comprises for each user terminal selecting goes out a vector that makes this user terminal obtain maximum precoding gain in first code book: travel through each vector in first code book, according to formula || H _kv _i|| ², i=1 ..., L1 calculates the precoding gain when adopting each described vector to carry out precoding, v _iRepresent i column vector in first code book; According to result of calculation, select a vector that makes described result of calculation maximum I=1 ..., L1 also writes down the sequence number i_u of this vector in first code book _k, wherein, H _kBe arbitrary user terminal u _kMIMO down channel matrix or be arbitrary user terminal u in the TDD of channel reciprocity system _kMIMO up channel matrix, L1 is the summation of the column vector number that comprises of first code book.

3. method according to claim 1 and 2 is characterized in that, establishes H _kBe arbitrary user terminal u _kMIMO down channel matrix or be user terminal u in the TDD of channel reciprocity system _kMIMO up channel matrix,

Vector in first code book that expression is selected, described calculating corresponding C QI value comprises: according to formula

Calculate

Corresponding C QI value, wherein, σ ²The expression noise variance.

4. method according to claim 1 is characterized in that, described base station determines that the user terminal for the treatment of precoding comprises: according to setting in advance or dynamic dispatching, determine the user terminal for the treatment of precoding;

Described according to the corresponding CQI value of vector sum in described first code book of the described user terminal correspondence for the treatment of precoding, selecting a unitary matrice from second code book comprises: utilize the vector sum corresponding C QI value in described first code book that the described user terminal for the treatment of precoding obtains, select a unitary matrice that makes the power system capacity maximum from second code book.

5. method according to claim 4 is characterized in that, described base station determines to treat that the user terminal of precoding is preset user terminal u ₁..., u _N, establish described preset user terminal u ₁..., u _NVector in described first code book that obtains is Corresponding C QI value is

Describedly select one make the unitary matrice of power system capacity maximum comprise from second code book: utilize the corresponding CQI value of vector sum in described first code book that the preset user terminal obtains, structure is selected cost function

L=1 ..., L2 travels through each unitary matrice in second code book, selects to make a unitary matrice of described selection cost function maximum from second code book, wherein,

Be user terminal u _nVector in first code book of selecting

The complex-conjugate transpose matrix, w _lBe l unitary matrice among the second code book W, L2 is the unitary matrice sum among the second code book W, [w _l] _nExpression w _lN-component column vector.

6. method according to claim 4 is characterized in that, described base station determines that the user terminal for the treatment of precoding is: by dynamic dispatching, select one group of user terminal { u that treats precoding in advance ₁..., u _K} _SCH,

Describedly from second code book, select one and make the unitary matrice of power system capacity maximum comprise: utilize the one group of user terminal { u that has selected ₁..., u _K} _SCHIn each user terminal u ₁..., u _KVector in described first code book that obtains With corresponding CQI value

Structure scheduling cost function L=1 ..., L2 travels through each unitary matrice in second code book, selects to make a unitary matrice of described scheduling cost function maximum from second code book, wherein,

Vector in first code book that user terminal uk selects The complex-conjugate transpose matrix, w _lBe l unitary matrice among the second code book W, L2 is the unitary matrice sum among the second code book W, [w _l] _kExpression w _lK column vector.

7. method according to claim 1, it is characterized in that, described base station is according to the corresponding CQI value of vector sum in described first code book of described each user terminal correspondence, from second code book, select a unitary matrice and determine the user terminal for the treatment of precoding and comprise: utilize the corresponding CQI value of vector sum in described first code book of described each user terminal correspondence, from second code book, select a unitary matrice that makes the power system capacity maximum, determine the user terminal for the treatment of precoding simultaneously.

8. method according to claim 7 is characterized in that,

Describedly from second code book, select a unitary matrice that makes the power system capacity maximum, determining the user terminal for the treatment of precoding simultaneously comprises: travel through each unitary matrice in second code book, travel through each user terminal at each unitary matrice, select and make described scheduling cost function

One group of maximum user terminal { u that treats precoding ₁..., u _K} _SCHWith a corresponding unitary matrice.

9. according to claim 4,5 or 6 described methods, it is characterized in that, further comprise before the vector in described first code book that described base station obtains according to the described user terminal for the treatment of precoding: receive the vector in first code book of the user terminal feedback for the treatment of precoding or receive vectorial sequence number in first code book of the user terminal feedback for the treatment of precoding, go out vector in first code book according to described vectorial SN recovery.

10. according to claim 1,2,3,4,5,6,7 or 8 described methods, it is characterized in that, after using described pre-coding matrix to carry out precoding to each user terminal, further comprise: the base station sends to the user terminal that each treats precoding with unitary matrice or the sequence number of unitary matrice in described second code book of selecting; Or estimate according to the channel matrix of special pilot design after with precoding.

11., it is characterized in that described first code book is that Jim Glassman (Grassmanian) code book or channel vector quantize code book according to claim 1,2,3,4,5,6,7 or 8 described methods.

12. a pre-coding system is characterized in that, this system comprises: base station and at least one user terminal of first code book and second code book are set, and described second code book is the unitary matrice set that first code book generates;

Each described user terminal utilizes MIMO down channel matrix, selects one and make described user terminal obtain the vector of maximum precoding gain in first code book; According to the vector in first code book of selecting, calculate corresponding C QI value;

The user terminal of precoding is determined to treat in described base station, the corresponding CQI value of vector sum in described first code book that obtains according to the described user terminal for the treatment of precoding, go out a unitary matrice from second codebook selecting, perhaps the corresponding CQI value of vector sum in described first code book that obtains according to described each user terminal goes out a unitary matrice and determines the user terminal for the treatment of precoding from second codebook selecting; As pre-coding matrix, use described pre-coding matrix to carry out precoding described unitary matrice or its submatrix to the described user terminal of precoding for the treatment of.

13. system according to claim 12 is characterized in that, described user terminal comprises:

Vector is selected module, is used to utilize channel matrix, selects one and make described user terminal obtain the vector of maximum precoding gain in first code book;

CQI value computing module is used for the vector according to first code book of selecting, and calculates corresponding C QI value.

14. system according to claim 13 is characterized in that, described vector selects module to comprise:

First computing unit is used for user terminal u _kTravel through each vector in first code book, according to formula || H _kv _i|| ², i=1 ..., L1 calculates the precoding gain when adopting each described vector to carry out precoding, wherein, H _kBe user terminal u _kMIMO down channel matrix, v _iExpression user terminal u _kVector in first code book of traversal, L1 are the summation of the column vector number that comprises of first code book;

Selected cell is used for the result of calculation that obtains according to described computing module, selects a vector that makes described result of calculation maximum.

15., it is characterized in that described CQI value computing module comprises according to claim 13 or 14 described systems:

Acquiring unit is used for obtaining the vector that described vector is selected first code book that module selects

Second computing unit is used for according to formula

Calculate that described acquisition module obtains

Corresponding C QI value, wherein, σ ²The expression noise variance.

16. system according to claim 12 is characterized in that, described base station comprises:

The coding pretreatment module, when pre-determining the user terminal for the treatment of precoding, the corresponding CQI value of vector sum that is used for described first code book that obtains according to the fixed described user terminal for the treatment of precoding goes out a unitary matrice that makes the power system capacity maximum from second codebook selecting; When not pre-determining the user terminal for the treatment of precoding, the corresponding CQI value of vector sum that is used for described first code book that obtains according to described each user terminal, go out a unitary matrice that makes the power system capacity maximum from second codebook selecting, determine the user terminal for the treatment of precoding simultaneously;

Precoding module is used for unitary matrice that described coding pretreatment module is obtained or its submatrix as pre-coding matrix, uses described pre-coding matrix to carry out precoding to each user terminal.

17. system according to claim 16 is characterized in that, when pre-determining the user terminal for the treatment of precoding, described coding pretreatment module comprises:

The first cost construction of function unit is used to utilize described each preset user terminal u ₁..., u _NVector in described first code book that obtains

With corresponding CQI value

Structure is selected cost function

L=1 ..., L2, wherein, User terminal u _nVector in first code book of selecting

The complex-conjugate transpose matrix, w _lBe l unitary matrice among the second code book W, L2 is the unitary matrice sum among the second code book W, [w _l] _nExpression w _lN-component column vector;

The first matrix selected cell is used for traveling through each unitary matrice of second code book, selects to make a unitary matrice of the selection cost function maximum in the described first cost construction of function unit from second code book.

18. system according to claim 16 is characterized in that, when pre-determining the user terminal for the treatment of precoding, described coding pretreatment module comprises:

Scheduling unit is used for by dynamic dispatching, the one group of user terminal { u that treats precoding that selects in advance ₁..., u _K} _SCH

The second cost construction of function unit is used to utilize described scheduling unit to select one group of user terminal { u ₁..., u _K} _SCHIn each user terminal u ₁..., u _KVector in described first code book that obtains

With corresponding CQI value

Structure scheduling cost function

L=1 ..., L2, wherein,

User terminal u _kVector in first code book of selecting

The complex-conjugate transpose matrix, w _lBe l unitary matrice among the second code book W, L2 is the unitary matrice sum among the second code book W, [w _l] _kExpression w _lK column vector;

The second matrix selected cell is used for traveling through each unitary matrice of second code book, selects to make a unitary matrice of the scheduling cost function maximum in the described second cost construction of function unit from second code book.

19. system according to claim 16 is characterized in that, when not pre-determining the user terminal for the treatment of precoding, described coding pretreatment module comprises:

Travel through the unit, be used for traveling through each unitary matrice of second code book, travel through each user terminal to be scheduled at each unitary matrice;

The 3rd matrix selected cell is used for the result that obtains according to described traversal unit, selects to make described scheduling cost function

L=1 ..., L2, one group of maximum user terminal { u that treats precoding ₁..., u _K} _SCHWith a corresponding unitary matrice, wherein,

Vector in first code book of selecting for user terminal uk

The complex-conjugate transpose matrix, w _lBe l unitary matrice among the second code book W, L2 is the unitary matrice sum among the second code book W, [w _l] _kExpression w _lK column vector.

20., it is characterized in that described base station further comprises according to claim 16,17,18 or 19 described systems:

Sending module, the unitary matrice or the unitary matrice that are used for described coding pretreatment module is selected send to the user terminal that each treats precoding in the sequence number of described second code book.

21. a pre-coding system is characterized in that, this system comprises: base station and at least one user terminal of first code book and second code book are set, and described second code book is the unitary matrice set that first code book generates; Described base station comprises:

Vector is selected module, is used to utilize MIMO up channel matrix, selects one and make described user terminal obtain the vector of maximum precoding gain in first code book;

CQI value computing module is used for the vector according to first code book of selecting, and calculates corresponding C QI value;

The coding pretreatment module, when pre-determining the user terminal for the treatment of precoding, the corresponding C QI value that the described CQI value of the vector sum computing module of described first code book of the described user terminal correspondence for the treatment of precoding that is used for selecting according to vector that module obtains draws goes out a unitary matrice that makes the power system capacity maximum from second codebook selecting; When not pre-determining the user terminal for the treatment of precoding, be used for the corresponding C QI value of selecting the described CQI value of the vector sum computing module of described first code book of described each user terminal correspondence that module obtains to draw according to vector, go out a unitary matrice that makes the power system capacity maximum from second codebook selecting, determine the user terminal for the treatment of precoding simultaneously;

Precoding module is used for unitary matrice that described coding pretreatment module is obtained or its submatrix as pre-coding matrix, uses described pre-coding matrix to carry out precoding to each user terminal.

22. system according to claim 21 is characterized in that, described vector selects module to comprise:

First computing unit is used for traveling through each vector of first code book, according to formula || and H _kv _i|| ², i=1 ..., L1 calculates the precoding gain when adopting each described vector to carry out precoding, wherein, H _kBe user terminal u _kMIMO up channel matrix, v _iExpression user terminal u _kVector in first code book of traversal, L1 are the summation of the column vector number that comprises of first code book;

Selected cell is used for the result of calculation that obtains according to described computing module, selects a vector that makes described result of calculation maximum.

23., it is characterized in that described CQI value computing module comprises according to claim 21 or 22 described systems:

Acquiring unit is used for obtaining the vector that described vector is selected first code book that module selects

Second computing unit is used for according to formula

Calculate that described acquisition module obtains

Corresponding C QI value, wherein, σ ²The expression noise variance.

Images