CN102594489B - Pre-coded vector determination method, device, base station and user terminal - Google Patents

Abstract

The invention discloses a pre-coded vector determination method, a pre-coded vector determination device, a base station and a user terminal, aiming at solving the problem that many computing resources are consumed by adopting the existing pre-coded vector determination method. The method comprises the steps of: determining equivalent channel matrixes Tj of all user terminals, and aiming at each user terminal, executing the following steps of: according to the Tj of other user terminals, determining an interference channel matrix H<->I and executing orthogonalization processing to all row vectors in H<->i to obtain a group of standard orthogonal bases Bi; determining an orthogonal projection matrix of a subspace which is formed from the equivalent channel matrix of the user terminal to H<->i according to the Bi; determining the quantity of receiving antennas of the user terminal, taking the orthogonal projection matrix as a matrix to be conducted with singular value decomposition (SVD), and according to a preset corresponding relation between the quantity of the receiving antennas and simplified SVD algorithms, determining a simplified SVD algorithm corresponding to the determined quantity of the receiving antennas; and according to a right singular vector corresponding to a singular value determined based on the determined simplified SVD algorithm, determining a pre-coded vector aiming at the user terminal.

Description

Definite method, device, base station and the user terminal of precoding vector

Technical field

The present invention relates to wireless communication technology field, relate in particular to a kind of definite method and device, base station and user terminal of precoding vector.

Background technology

Downlink transfer scheme based on precoding is one of core transmission plan of multiaerial system.So-called precoding is that information symbol to be passed is multiplied by a precoding vector by base station, and the element in this precoding vector and transmitting antenna are relations one to one.Precoding vector is normally based on singular value decomposition (SVD, SingularValue Decomposition) mode obtain, to descending multiple-input, multiple-output (MIMO, Multiple-Input Multiple-Out-put) channel matrix H carries out SVD decomposition, the right singular vector corresponding by front n the maximum singular value of H directly obtains precoding vector, and n is the information symbol number simultaneously transmitting here.

At time division duplex (TDD, Time Division Duplexing) in system, base station utilizes detecting pilot frequency signal (SRS) and the channel reciprocity that user terminal sends to know that base station arrives the down channel matrix H of this user terminal, thereby directly H is carried out to SVD decomposition by base station.And at Frequency Division Duplexing (FDD) (FDD, FrequencyDivision Duplexing) in system, the down public guide frequency signal that user terminal sends by base station is known down channel matrix H, thereby carry out the SVD to H by user terminal, and by the corresponding pre-encoding codeword index of right singular vector (PMI) of user terminal feedback H to base station, thereby carry out follow-up precoding to base station can be determined corresponding to the pre-encoding codeword of this PMI.

In single subscriber terminal transmission situation, base station fill order user's precoding; In multiple users transmission situation, multi-user pre-coding is carried out in base station.If multi-user pre-coding needs to eliminate the interference between user, need to carry out ZF processing.ZF is processed normally to adopt and is solved that the method for kernel realizes, (kernel claims again the orthogonal complement space first to ask for the kernel of interference channel (other user of simultaneously transmitting channel be referred to as interference channel) matrix, characterized by one group orthonormal basis, can adopt the method that solves system of linear equations to obtain orthonormal basis, interference channel matrix and its kernel are orthogonal), and then SVD decomposition is carried out in the projection in this kernel to subscriber channel matrix, and decompose by current SVD the right singular vector obtaining and carry out the final precoding vector of structuring user's.

In precoding technique being applied in enhancing MIMO and coordinate multipoint (CoMP) system brought performance boost, the predicament that has also brought Project Realization difficulty.This be because in the situation that transmitting terminal antenna number is more (for example, strengthening in MIMO is 8 transmitting antennas, in CoMP system, reach 24 transmitting antennas), the derivation amount of SVD decomposition and interference channel matrix kernel is huge, therefore needs to consume very many calculation resources.

Summary of the invention

The embodiment of the present invention provides a kind of definite method and device of precoding vector, can consume the problem of more calculation resources in order to solve the mode of definite precoding vector that prior art provides.

The embodiment of the present invention also provides a kind of base station and a kind of user terminal.

The embodiment of the present invention is by the following technical solutions:

A definite method for precoding vector, comprising:

Determine respectively equivalent channel matrix T corresponding to each user terminal _j, and for user terminal described in each, carry out following step:

According to the corresponding T of other user terminals except this user terminal _j, determine the interference channel matrix of this user terminal and right in all row vectors carry out orthogonalization process, obtain one group of orthonormal basis B of the subspace of being opened _i; And according to B _ithe equivalent channel matrix of determining this user terminal arrives the orthogonal intersection cast shadow matrix of the subspace of being opened and the quantity of definite this user terminal reception antenna, and with as the matrix of pending singular value decomposition SVD, carry out the corresponding relation of simplifying svd algorithm from the reception antenna quantity and the matrix for pending SVD that set in advance, determine the simplification svd algorithm corresponding with this quantity of determining; Determined according to the simplification svd algorithm based on definite right singular vector corresponding to singular value, determine for the precoding vector of this user terminal.

A definite method for precoding vector, is applied in the single subscriber terminal transmission plan of TDD system, comprising:

The down channel matrix H that described user terminal is corresponding is determined in base station _i, and the quantity of definite described user terminal reception antenna; And with H _ias the matrix of pending SVD, carry out the corresponding relation of simplifying svd algorithm from the reception antenna quantity and the matrix for pending SVD that set in advance, determine the simplification svd algorithm corresponding with this quantity of determining; And determine H according to definite simplification svd algorithm _iright singular vector corresponding to singular value, determine for the precoding vector of described user terminal.

A definite method for precoding vector, is applied in the single subscriber terminal transmission plan of FDD system, comprising:

Described user terminal is determined the down channel matrix H of self _i, and the quantity of definite described user terminal reception antenna; And with H _ias the matrix of pending SVD, carry out the corresponding relation of simplifying svd algorithm from the reception antenna quantity and the matrix for pending SVD that set in advance, determine the simplification svd algorithm corresponding with this quantity of determining; And determine H according to definite simplification svd algorithm _iright singular vector corresponding to singular value, determine and send corresponding PMI to base station according to described right singular vector, corresponding precoding vector is determined according to PMI in instruction base station.

A determining device for precoding vector, comprising: equivalent channel matrix determining unit, and for determining equivalent channel matrix T corresponding to each user terminal _j; Interference channel matrix determining unit, for for each user terminal, the corresponding equivalent channel matrix T of other user terminals except this user terminal determining according to equivalent channel matrix determining unit _j, determine the interference channel matrix of this user terminal orthogonalization process unit, for what determine interference channel matrix determining unit in all row vectors carry out orthogonalization process, obtain one group of orthonormal basis B of the subspace of being opened _i; Orthogonal intersection cast shadow matrix determining unit, for the B obtaining according to orthogonalization process unit _i, determine that the equivalent channel matrix of this user terminal arrives the orthogonal intersection cast shadow matrix of the subspace of being opened simplify svd algorithm determining unit, for according to the quantity of this user terminal reception antenna, determine with orthogonal intersection cast shadow matrix determining unit as the matrix of pending singular value decomposition SVD, carry out the corresponding relation of simplifying svd algorithm from the reception antenna quantity and the matrix for pending SVD that set in advance, determine the simplification svd algorithm corresponding with the quantity of this user terminal reception antenna; Precoding vector determining unit, determined based on simplifying the definite simplification svd algorithm of svd algorithm determining unit for basis right singular vector corresponding to singular value, determine for the precoding vector of this user terminal.

A kind of base station, is applied in the single subscriber terminal transmission plan of TDD system, comprising: matrix and quantity determining unit, and for determining down channel matrix H corresponding to described user terminal _iquantity with described user terminal reception antenna; Simplify svd algorithm determining unit, for the matrix using matrix and the definite Hi of quantity determining unit as pending SVD, carry out the corresponding relation of simplifying svd algorithm with the matrix for pending SVD from the reception antenna quantity setting in advance, determine the corresponding simplification svd algorithm of this quantity that matrix and quantity determining unit are determined; Precoding vector determining unit, for determining H according to simplifying the definite simplification svd algorithm of svd algorithm determining unit _iright singular vector corresponding to singular value, determine for the precoding vector of described user terminal.

A kind of user terminal, is applied in the single subscriber terminal transmission plan of FDD system, it is characterized in that, comprising: matrix and quantity determining unit, and for determining the down channel matrix H of described user terminal _i, and the quantity of definite described user terminal reception antenna; Simplify svd algorithm determining unit, for matrix and the definite H of quantity determining unit _ias the matrix of pending SVD, carry out the corresponding relation of simplifying svd algorithm from the reception antenna quantity and the matrix for pending SVD that set in advance, determine simplification svd algorithm corresponding to this quantity of determining with matrix and quantity determining unit; Transmitting element, for according to right singular vector corresponding to singular value of simplifying the definite determined Hi of simplification svd algorithm of svd algorithm determining unit, determines and sends corresponding PMI to base station, corresponding precoding vector is determined according to PMI in instruction base station.

The beneficial effect of the embodiment of the present invention is as follows:

The said method providing by the embodiment of the present invention, in multiple users transmission plan, without asking for again interference channel matrix kernel and subscriber channel matrix projection being arrived to this kernel, but directly by the subspace to interference channel matrix by the rectangular projection of subscriber channel matrix, thereby greatly save calculation resources.In addition, simplify the application of svd algorithm in this programme, also avoided utilizing svd algorithm that prior art provides need to carry out repeatedly the defect of iteration, realized the saving to calculation resources.

Brief description of the drawings

The idiographic flow schematic diagram of the definite method that is applied to the precoding vector in multiple users transmission plan that Fig. 1 provides for the embodiment of the present invention;

The idiographic flow schematic diagram of definite method of the precoding vector in the single subscriber terminal transmission plan of a kind of TDD of being applied to system that Fig. 2 provides for the embodiment of the present invention;

The idiographic flow schematic diagram of definite method of the precoding vector in the single subscriber terminal transmission plan of a kind of FDD of being applied to system that Fig. 3 provides for the embodiment of the present invention;

The concrete structure schematic diagram of the determining device that is applied to the precoding vector in multiple users transmission plan that Fig. 4 provides for the embodiment of the present invention;

The concrete structure schematic diagram of the base station in the single subscriber terminal transmission plan of a kind of TDD of being applied to system that Fig. 5 provides for the embodiment of the present invention;

The concrete structure schematic diagram of the user terminal in the single subscriber terminal transmission plan of a kind of FDD of being applied to system that Fig. 6 provides for the embodiment of the present invention.

Embodiment

The problem that can consume more calculation resources in order to solve the mode of definite precoding vector that prior art provides, definite mode of the precoding vector that inventor provides prior art has been carried out following analysis:

The precoding of decomposing based on SVD is optimum, but the operand of typical SVD derivation algorithm (conventionally needing iteration) increases along with the increase of matrix dimension.In the situation that the large but transmission fluxion of number of transmit antennas is less (in LTE Rel.8/9/10 specification, single current and stream transmission pattern are defined especially, and under multiple users transmission mode, each UE transmission fluxion is no more than 2), SVD decomposition there is no need completely, because now only need maximum right singular vector corresponding to n singular value.In enhancement mode multiaerial system, if user terminal reception antenna number is configured to 1 or 2, no matter transmitting terminal antenna number how, can there is closed expression in SVD, the right singular vector that is user's mimo channel matrix H can directly be provided by formula, and does not need iterative computation.Equally, for example, in the situation that transmission fluxion is less (single current or double fluid), there is accurate approximate solution in SVD decomposition equally.

Equally, the derivation amount of the kernel of interference channel also increases along with the increase of interference channel dimension, and this,, reaching in the CoMP system of 24 transmitting antennas, is to be difficult to realize.In fact, the ZF processing of multi-user pre-coding, its essence is to obtain the rectangular projection of subscriber channel matrix to interference channel matrix.Rectangular projection both can be by having asked for interference channel matrix kernel, then subscriber channel matrix projection is realized to this kernel, also can be directly by the rectangular projection of subscriber channel matrix be realized to the subspace of interference channel matrix, obviously, the latter is downlink transfer fluxion less (number of transmit antennas relatively) in the situation that, and operand is much lower compared with the former.

By above-mentioned analysis, inventor thinks, in fact, under some typical scene, for example user terminal only configures under the scene of 1 or 2 reception antenna, and SVD decomposes and can simplify, and has closed solutions, in single current or stream transmission and 4 reception antenna situations of user terminal configuration, there is approximate solution in SVD decomposition equally.And no matter adopt which kind of SVD decomposition method, it is all avoidable disturbing the kernel carrying out to solve for elimination in multi-user pre-coding.

Given this, receive the transmission situation that is no more than 2 streams for user terminal, the invention provides a kind of method for precoding of new descending multiaerial system, in the method, SVD decomposes directly by formula, the ZF of multi-user pre-coding is processed and is obtained by rectangular projection, without the kernel that solves interference channel.

Below in conjunction with accompanying drawing, the scheme that the embodiment of the present invention is provided is elaborated.

First the embodiment of the present invention provides a kind of definite method that is applied to the precoding vector in multiple users transmission plan, and the method is determined first respectively equivalent channel matrix T corresponding to each user terminal _j, then, for each user terminal, by carrying out following steps as shown in Figure 1, realize the precoding vector of determining for this user terminal:

Step 11, according to the corresponding T of other user terminals except this user terminal _j, determine the interference channel matrix of this user terminal

Step 12 is right in all row vectors carry out orthogonalization process, acquire one group of orthonormal basis Bi of the subspace of being opened, wherein, orthogonalization process can be, but not limited to adopt Si Mite algorithm (comprising improved Si Mite algorithm), ranks elementary transformation algorithm, Householder mapping algorithm and Givens rotation scheduling algorithm to realize here;

Step 13, according to B _ithe equivalent channel matrix of determining this user terminal arrives the orthogonal intersection cast shadow matrix of the subspace of being opened

Step 14, determines the quantity of this user terminal reception antenna, and with as the matrix of pending singular value decomposition SVD, carry out the corresponding relation of simplifying svd algorithm from the reception antenna quantity and the matrix for pending SVD that set in advance, determine the simplification svd algorithm corresponding with this quantity of determining;

Step 15, determines according to definite simplification svd algorithm right singular vector corresponding to singular value, determine for the precoding vector of this user terminal.In this step 15, according to right singular vector corresponding to singular value while determining the precoding vector for this user terminal, can be first to the transmission means of this user terminal for judging, in the time judging this transmission means and be single-stream transmission, general right singular vector corresponding to maximum singular value be defined as the precoding vector that this user terminal is corresponding, and in the time judging this transmission means and be stream transmission, according to maximum and second largest singular value respectively corresponding right singular vector determine this precoding vector.

The said method providing by the embodiment of the present invention, in multiple users transmission plan, without asking for again interference channel matrix kernel and subscriber channel matrix projection being arrived to this kernel, but directly by the subspace to interference channel matrix by the rectangular projection of subscriber channel matrix, thereby greatly save calculation resources.

In addition, the application of above-mentioned simplification svd algorithm, has also avoided utilizing the svd algorithm that prior art provides need to carry out repeatedly the defect of iteration, thereby has also realized the saving of calculation resources.Particularly, the above-mentioned reception antenna quantity setting in advance is specially with the corresponding relation of simplifying svd algorithm:

1,, in the time that the quantity of reception antenna is 1, simplifies accordingly svd algorithm and be: using | H| as the maximum singular value that the matrix H of pending SVD is carried out obtaining after SVD, and with H|H| ^-1as corresponding to | the right singular value vector of H|, wherein, | H| represents to calculate the 2-norm of H;

2, in the time that the quantity of reception antenna is 2, simplifies accordingly svd algorithm and be: with λ ₁ ²with respectively as H is carried out to the maximum singular value and the second largest singular value that obtain after SVD, and with V ₁and V ₂conduct is corresponding to λ respectively ₁ ²with right singular value vector, wherein, according to the following equation [1] calculate λ ₁ ²:

${\mathrm{\&lambda;}}_1^2=\frac{{\left|h_1\right|}^2{(\mathrm{\&beta;}+\sqrt{1-{\mathrm{\&beta;}}^2}|\mathrm{\&eta;}\left|\right)}^2+{\left|h_2\right|}^2{\left(\mathrm{\&beta;}\right|\mathrm{\&eta;}|+\sqrt{1-{\mathrm{\&beta;}}^2})}^2}{1+2\mathrm{\&beta;}\sqrt{1-{\mathrm{\&beta;}}^2}\left|\mathrm{\&eta;}\right|}---\left[1\right]$

Calculate [2] according to the following equation

${\mathrm{\&lambda;}}_2^2={\left|h_1\right|}^2+{\left|h_2\right|}^2-{\mathrm{\&lambda;}}_1^2---\left[2\right]$

Calculate according to the following equation V ₁:

$V_1=\frac{\mathrm{\&beta;}{h}_1{\left|h_1\right|}^{-1}+\sqrt{1-{\mathrm{\&beta;}}^2}\mathrm{\&eta;}{\left|\mathrm{\&eta;}\right|}^{-1}{h}_2{\left|h_2\right|}^{-1}}{\sqrt{1+2\mathrm{\&beta;}\sqrt{1-{\mathrm{\&beta;}}^2}\left|\mathrm{\&eta;}\right|}}---\left[3\right]$

Calculate according to the following equation V ₂:

$V_2=\frac{h_1-h_1{V}_1^H{V}_1}{|h_1-h_1{V}_1^H{V}_1|}---\left[4\right]$

Wherein, h ₁for first row vector of H, h ₂for second row vector of H, the η in above-mentioned formula and β be [5], [6] calculating according to the following equation respectively:

$\mathrm{\&eta;}=h_1{h}_2^H{\left|h_1\right|}^{-1}{\left|h_2\right|}^{-1}---\left[5\right]$

$\mathrm{\&beta;}=\sqrt{\frac{1}{2}\frac{2{\left|h_1\right|}^2({\left|h_1\right|}^2+{\left|h_2\right|}^2){\left|\mathrm{\&eta;}\right|}^2+{({\left|h_1\right|}^2-{\left|h_2\right|}^2)}^2+({\left|h_1\right|}^2-{\left|h_2\right|}^2)\sqrt{{({\left|h_1\right|}^2-{\left|h_2\right|}^2)}^2+4{\left|h_1\right|}^2{\left|h_2\right|}^2{\left|\mathrm{\&eta;}\right|}^2}}{{({\left|h_1\right|}^2-{\left|h_2\right|}^2)}^2+{({\left|h_1\right|}^2+{\left|h_2\right|}^2)}^2{\left|\mathrm{\&eta;}\right|}^2}}---\left[6\right]$

3, in the time that the quantity of reception antenna is 4, simplifies accordingly svd algorithm and be: with matrix as the matrix of pending SVD, according to simplification svd algorithm corresponding to the reception antenna quantity that is 2, determine respectively right singular vector corresponding to right singular vector, second largest singular value corresponding to maximum singular value, second largest singular value, maximum singular value, as the corresponding maximum singular value of H, second largest singular value, the right singular vector that maximum singular value is corresponding, right singular vector corresponding to second largest singular value; Wherein, λ _1,1, V _1,1, λ _2,1, V _2,1determine in the following manner:

H is divided and obtains H ₁and H ₂, wherein H ₁two row vectors by H form, H ₂two other row vector by H forms; With H ₁and H ₂as the matrix of pending SVD, according to reception antenna quantity be 2 o'clock corresponding simplification svd algorithms, respectively determine H ₁maximum singular value λ _1,1with corresponding right singular vector V _1,1, and H ₂maximum singular value λ _2,1with corresponding right singular vector V _2,1.

It should be noted that, when the method that the embodiment of the present invention provides is applied to respectively TDD system and FDD, the equivalent channel matrix T that each user terminal is corresponding _jdefinite mode on there are some differences.Concrete difference is as follows:

In the time that the method is applied to TDD system, determine the equivalent channel matrix T that each user terminal is corresponding _jmode be: first, the pilot signal SRS sending according to each user terminal determines the down channel matrix H that each user terminal is corresponding _j; Then, judge that respectively for the transmission means of each user terminal be single-stream transmission or stream transmission; And for judging each single-stream transmission user terminal execution that transmission means is single-stream transmission: determine the quantity of this single-stream transmission user terminal reception antenna, and with H _jas the matrix of pending SVD, in described corresponding relation, determine the simplification svd algorithm corresponding with the quantity of this single-stream transmission user terminal reception antenna from the above; Determine H according to definite simplification svd algorithm _jright singular vector corresponding to maximum singular value, and right singular vector corresponding this maximum singular value is defined as to the T of this single-stream transmission user terminal _j; And for judging each stream transmission user terminal that transmission means is stream transmission, directly by the H of this stream transmission user terminal _jbe defined as the T of this stream transmission user terminal _j.

In the time that the method is applied to FDD system, determine the equivalent channel matrix T that each user terminal is corresponding _jmode be: first, obtain the codewords indexes PMI of the down public guide frequency signal CRS feedback that each user terminal sends according to base station; Then, judge that respectively for the transmission means of each user terminal be single-stream transmission or stream transmission; And for judging each single-stream transmission user terminal that transmission means is single-stream transmission, the first row vector of the pre-encoding codeword corresponding PMI of this single-stream transmission user terminal feedback is defined as to the T of this single-stream transmission user terminal _j; And for judging each stream transmission user terminal that described transmission means is stream transmission, determine the T of this stream transmission user terminal according to the first two columns vector of pre-encoding codeword corresponding to the PMI of this stream transmission user terminal feedback _j.

In addition, the embodiment of the present invention also provides definite method of the precoding vector in a kind of single subscriber terminal transmission plan of the TDD of being applied to system, and the method comprises following step as shown in Figure 2:

Step 21, the down channel matrix H that user terminal is corresponding is determined in base station _i, and the quantity of definite this user terminal reception antenna;

Step 22, with H _ias the matrix of pending SVD, carry out the above-mentioned corresponding relation of simplifying svd algorithm from the reception antenna quantity and the matrix for pending SVD that set in advance, determine the simplification svd algorithm corresponding with this quantity of determining;

Step 23, determines H according to definite simplification svd algorithm _iright singular vector corresponding to singular value, determine for the precoding vector of this user terminal.Particularly, can first judge the transmission means of this user terminal, in the time judging this transmission means and be single-stream transmission, by H _iright singular vector corresponding to maximum singular value be defined as the precoding vector that this user terminal is corresponding, and in the time judging this transmission means and be stream transmission, according to H _imaximum and second largest singular value respectively corresponding right singular vector determine the precoding vector that this user terminal is corresponding.

From the above-mentioned definite method that is applied in the precoding vector in TDD system single subscriber terminal transmission plan, because the mode without by iteration is to H _icarry out SVD, but only need utilize default simplification svd algorithm to come easily to H _icarry out SVD, thereby greatly saved calculation resources.

In addition, the embodiment of the present invention also provides a kind of definite method of precoding vector, is applied in the single subscriber terminal transmission plan of FDD system, and the method comprises following step as shown in Figure 3:

Step 31, user terminal is determined the down channel matrix H of self _i, and determine the quantity of self reception antenna;

Step 32, with H _ias the matrix of pending SVD, carry out the above-mentioned corresponding relation of simplifying svd algorithm from the reception antenna quantity and the matrix for pending SVD that set in advance, determine the simplification svd algorithm corresponding with this quantity of determining;

Step 33, determines H according to definite simplification svd algorithm _iright singular vector corresponding to singular value, and determine and send corresponding PMI to base station according to right singular vector, corresponding precoding vector is determined according to PMI in instruction base station.Similarly, user terminal also can, in the time judging transmission means used and be single-stream transmission, from precoding codebook, be chosen corresponding to H _ithe code word of the corresponding right singular vector of maximum singular value, and determine PMI corresponding to this code word choosing with transmission; And in the time judging transmission means and be stream transmission, from precoding codebook, choose corresponding to H _imaximum and second largest singular value divide the code word of the right singular matrix that forms of corresponding right singular vector, and determine PMI corresponding to this code word choosing with transmission.

Based on said method, being described in detail as follows of the scheme that the embodiment of the present invention is provided application in practice:

Step 1: estimating user terminal down channel matrix H is also carried out SVD.

In TDD system, estimate H and H is carried out to SVD by base station end.The estimation of H and SVD mode are as follows:

1, user side sends detecting pilot frequency signal SRS;

2, base station end utilizes SRS and the channel reciprocity that user side sends to estimate H;

3, base station end is carried out SVD to H.

In FDD system, estimate H and H is carried out to SVD by user side.The estimation of H and SVD mode are as follows:

1, base station end sends down public guide frequency signal CRS;

2, user side utilizes the CRS that base station end sends to estimate H;

3, user side is carried out SVD to H and is obtained the right singular vector of H, and from precoding codebook, selects a code word according to right singular vector, and the index PMI that feeds back this code word is to base station end.

In embodiments of the present invention, the matrix that note need to be carried out SVD is H, and a kind of SVD decomposition method of simplification is as follows:

1) if H is 1 × N dimension matrix, its maximum singular value be λ=| H|, corresponding singular value vector is V=H|H| ^-1, wherein | H| represents to calculate the 2-norm of H, and N is base station end number of transmit antennas, and value is to be more than or equal to 2 and be not more than 24 even number;

2) if H is 2 × N dimension matrix, remember that the 1st row vector is h ₁, the 2nd row vector is h ₂, the maximum singular value of H is corresponding singular value vector is wherein and second Largest Eigenvalue is corresponding singular value vector is

3) if H is 4 × N dimension matrix, carry out following steps, the SVD that obtains H decomposes approximation:

First, H is divided into two 2 × N dimension matrix H ₁and H ₂, wherein H ₁two row vectors by H form, H ₂two other row vector by H forms;

Then, according to above-mentioned 2) mode, to H ₁and H ₂carry out SVD, obtain H ₁maximum singular value λ _1,1and corresponding right singular vector V _1,1, obtain H ₂maximum singular value λ _2,1and corresponding right singular vector V _2,1;

Finally, according to step above-mentioned 2) mode, to matrix carry out SVD, obtain maximum singular value λ ₁and corresponding right singular vector V ₁, and inferior maximum singular value λ ₂and corresponding right singular vector V ₂, λ ₁and λ ₂and corresponding V ₁and V ₂be the SVD approximate solution of H.

Step 2: the precoding vector that calculates user terminal i.

In multiple users transmission plan, the precoding vector process that user terminal i is calculated in base station is as follows:

First, the interference channel matrix of note user terminal i is wherein M is simultaneous interpretation user terminal number, T _jfor user terminal j (j=1,2 ..., M) equivalent channel matrix;

In TDD system, if user terminal j is single-stream transmission mode, T _j=V _{1, j}, V here _{1, j}for user terminal mimo channel matrix H _jright singular vector corresponding to maximum singular value, according to above-mentioned 2) mode to H _jcarry out and simplify SVD acquisition, if user terminal j is stream transmission mode, T _j=H _j.

In FDD system, if user j terminal is single-stream transmission mode, T _jfor the first row vector of pre-encoding codeword corresponding to the PMI of user terminal j feedback, if user terminal j is stream transmission mode, T _jfor the first two columns vector of pre-encoding codeword corresponding to the PMI of user terminal j feedback.

Then, right in all row vectors carry out orthogonalization methods, obtain one group of orthonormal basis of the subspace of being opened wherein L is matrix row vector number;

Then, calculate user terminal i equivalent channel matrix T _iarrive the rectangular projection of the subspace of being opened ?

Finally, according to determine the precoding vector of user terminal i.

If what user terminal i adopted is single-stream transmission mode, the precoding vector of user terminal i is wherein V _{1, i}for the right singular vector that maximum singular value is corresponding, this right singular vector by adopt above-mentioned 2) mode pair carry out and simplify SVD acquisition; If what family terminal i adopted is stream transmission mode, the precoding vector of user terminal i is wherein V _{1, i}and V _{2, i}be respectively the right singular vector that maximum and second largest singular value is corresponding, this right singular vector by adopt above-mentioned 2) mode pair carry out and simplify SVD acquisition.

In embodiments of the present invention, for the scheme of single subscriber terminal transmission, the precoding vector process that user terminal i is calculated in base station is as follows:

In TDD system, if the single-stream transmission mode that user terminal i adopts, the precoding vector of user terminal i is wherein V _{1, i}for the channel matrix H of user i _ithe right singular vector that maximum singular value is corresponding, this right singular vector by adopt above-mentioned 2) mode to H _icarry out and simplify SVD acquisition; If stream transmission mode, the precoding vector of user terminal i is wherein V _{1, i}and V _{2, i}be respectively the channel matrix H of user terminal i _ithe right singular vector that maximum and second largest singular value is corresponding, this right singular vector by adopt above-mentioned 2) mode to H _icarry out and simplify SVD acquisition.

In FDD system, if user terminal i adopt single-stream transmission mode, user terminal i by adopt above-mentioned 2) in mode to H _icarry out and simplify right singular vector V corresponding to SVD acquisition maximum singular value _{1, i}, and according to V _{1, i}from precoding codebook, select a code word, and the index PMI that feeds back this code word is to base station, the precoding vector of user terminal i is the pre-encoding codeword corresponding to PMI of user i feedback; If stream transmission, user terminal i by the way 2) to H _icarry out and simplify SVD acquisition maximum singular value and right singular matrix corresponding to second largest singular value according to from precoding codebook, select a code word, and the index PMI that feeds back this code word is to base station, the precoding vector of user terminal i is the pre-encoding codeword corresponding to PMI of user terminal i feedback.

Definite method of the precoding vector providing corresponding to the embodiment of the present invention, the embodiment of the present invention also provides a kind of determining device of precoding vector, and the concrete structure schematic diagram of this device as shown in Figure 4, comprises following functions unit:

Equivalent channel matrix determining unit 41, for determining equivalent channel matrix T corresponding to each user terminal _j;

Interference channel matrix determining unit 42, for for each user terminal, the corresponding equivalent channel matrix T of other user terminals except this user terminal determining according to equivalent channel matrix determining unit 41 _j, determine the interference channel matrix of this user terminal

Orthogonalization process unit 43, for what determine interference channel matrix determining unit 42 in all row vectors carry out orthogonalization process, obtain one group of orthonormal basis B of the subspace of being opened _i;

Orthogonal intersection cast shadow matrix determining unit 44, for the B obtaining according to orthogonalization process unit 43 _i, determine that the equivalent channel matrix of this user terminal arrives the orthogonal intersection cast shadow matrix of the subspace of being opened

Simplify svd algorithm determining unit 45, for according to the quantity of this user terminal reception antenna, determine with orthogonal intersection cast shadow matrix determining unit 44 as the matrix of pending singular value decomposition SVD, carry out the corresponding relation of simplifying svd algorithm from the reception antenna quantity and the matrix for pending SVD that set in advance, determine the simplification svd algorithm corresponding with the quantity of this user terminal reception antenna;

Precoding vector determining unit 46, determined based on simplifying the definite simplification svd algorithm of svd algorithm determining unit 45 for basis right singular vector corresponding to singular value, determine for the precoding vector of this user terminal.

When this application of installation is during in TDD system, above-mentioned equivalent channel matrix determining unit 41 specifically can be divided into following functions module:

Down channel matrix determination module, for the pilot signal SRS sending according to each user terminal, determines the down channel matrix H that each user terminal is corresponding _j; The first judge module, for judging that respectively for the transmission means of each user terminal be single-stream transmission or stream transmission; The first Executive Module, for judging for the first judge module each single-stream transmission user terminal that transmission means is single-stream transmission, carry out: determine the quantity of this single-stream transmission user terminal reception antenna, and the H of this single-stream transmission user terminal of determining with descending channel matrix determination module _jas the matrix of pending SVD, from described corresponding relation, determine the simplification svd algorithm corresponding with the quantity of this single-stream transmission user terminal reception antenna; Determine the H of this single-stream transmission user terminal according to definite simplification svd algorithm _jright singular vector corresponding to maximum singular value, and right singular vector corresponding described maximum singular value is defined as to the T of this single-stream transmission user terminal _j; The second Executive Module, for judge each stream transmission user terminal that transmission means is stream transmission for the first judge module, carries out: by the H of this stream transmission user terminal _jbe defined as the T of this stream transmission user terminal _j.

When this application of installation is during in FDD system, above-mentioned equivalent channel matrix determining unit 41 specifically can be divided into following functions module:

Codewords indexes obtains module, for obtaining each user terminal according to the codewords indexes PMI of the down public guide frequency signal CRS feedback of base station transmission;

The first judge module, for judging that respectively for the transmission means of each user terminal be single-stream transmission or stream transmission;

The first Executive Module, for judging for the first judge module each single-stream transmission user terminal that transmission means is single-stream transmission, carry out: the T that the first row vector of the pre-encoding codeword corresponding PMI of this single-stream transmission user terminal feedback is defined as to this single-stream transmission user terminal _j;

The second Executive Module, for judging for the first judge module each stream transmission user terminal that transmission means is stream transmission, carry out: the T that determines this stream transmission user terminal according to the first two columns vector of pre-encoding codeword corresponding to the PMI of this stream transmission user terminal feedback _j.

In a preferred embodiment, may be that single-stream transmission also may be for stream transmission in the transmission means of user terminal, above-mentioned precoding vector determining unit 46 specifically can be divided into following functions module:

The second judge module, for judging that for the transmission means of this user terminal be single-stream transmission or stream transmission; Determination module, while being single-stream transmission for judging described transmission means when the second judge module, will right singular vector corresponding to maximum singular value be defined as precoding vector, in the time that the second judge module is judged described transmission means and is stream transmission, according to maximum and second largest singular value respectively corresponding right singular vector determine precoding vector.

Definite method of the precoding vector in the single subscriber terminal transmission plan of a kind of TDD of the being applied to system providing corresponding to the embodiment of the present invention, the embodiment of the present invention also provides a kind of base station, the concrete structure schematic diagram of this base station as shown in Figure 5, comprises following functions unit:

Matrix and quantity determining unit 51, for determining down channel matrix H corresponding to user terminal _iquantity with user terminal reception antenna;

Simplify svd algorithm determining unit 52, for matrix and the definite H of quantity determining unit 51 _ias the matrix of pending SVD, carry out the corresponding relation of simplifying svd algorithm from the reception antenna quantity and the matrix for pending SVD that set in advance, determine the corresponding simplification svd algorithm of this quantity that matrix and quantity determining unit are determined;

Precoding vector determining unit 53, for determining H according to simplifying svd algorithm determining unit 52 definite simplification svd algorithms _iright singular vector corresponding to singular value, determine for the precoding vector of described user terminal.

May be that single-stream transmission also may be for stream transmission in the transmission means of user terminal, above-mentioned precoding vector determining unit 53 specifically can be divided into following functions module:

Judge module, for judging that for the transmission means of described user terminal be single-stream transmission or stream transmission; Precoding vector determination module, while being single-stream transmission for judge transmission means at judge module, by H _iright singular vector corresponding to maximum singular value be defined as precoding vector, in the time that judge module is judged transmission means and is stream transmission, according to H _imaximum and second largest singular value respectively corresponding right singular vector determine precoding vector.

Definite method of the precoding vector in the single subscriber terminal transmission plan of a kind of FDD of the being applied to system providing corresponding to the embodiment of the present invention, the embodiment of the present invention also provides a kind of user terminal, the concrete structure schematic diagram of this user terminal as shown in Figure 6, comprises following functions unit:

Matrix and quantity determining unit 61, for determining the down channel matrix H of user terminal _i, and the quantity of definite user terminal reception antenna;

Simplify svd algorithm determining unit 62, for matrix and the definite H of quantity determining unit 61 _ias the matrix of pending SVD, carry out the corresponding relation of simplifying svd algorithm from the reception antenna quantity and the matrix for pending SVD that set in advance, determine simplification svd algorithm corresponding to this quantity of determining with matrix and quantity determining unit 61;

Transmitting element 63, for the simplification svd algorithm determined H definite according to simplification svd algorithm determining unit 62 _iright singular vector corresponding to singular value, determine and send corresponding PMI to base station, corresponding precoding vector is determined according to PMI in instruction base station.

In a preferred embodiment, above-mentioned transmitting element 63 specifically can be divided into following functions module, comprising:

Judge module, for judging that for the transmission means of user terminal be single-stream transmission or stream transmission;

Choose module, while being single-stream transmission for judge transmission means when judge module, from precoding codebook, choose corresponding to H _ithe code word of the corresponding right singular vector of maximum singular value in the time that disconnected module is judged transmission means and is stream transmission, from precoding codebook, choose corresponding to H _imaximum and second largest singular value divide the code word of the right singular matrix that forms of corresponding right singular vector;

Determination module, chooses for determining PMI corresponding to code word that module is chosen;

Sending module, for send the definite PMI of determination module to base station, the PMI that instruction base station sends according to sending module determines corresponding precoding vector.

The said apparatus, base station and the user terminal that provide for the embodiment of the present invention, the above-mentioned reception antenna quantity setting in advance with simplify svd algorithm corresponding relation can but be not limited to following form:

Be that 1 simplification svd algorithm corresponding to reception antenna quantity is: using | H| as the maximum singular value that the matrix H of pending SVD is carried out obtaining after SVD, and with H|H| ^-1as corresponding to | the right singular value vector of H|, wherein, | H| represents to calculate the 2-norm of H;

Be that 2 simplification svd algorithm corresponding to reception antenna quantity is: with λ ₁ ²with respectively as H is carried out to the maximum singular value and the second largest singular value that obtain after SVD, and with V ₁and V ₂conduct is corresponding to λ respectively ₁ ²with right singular value vector, wherein, λ ₁ ², v ₁, V ₂calculate according to the following equation:

${\mathrm{\&lambda;}}_1^2=\frac{{\left|h_1\right|}^2{(\mathrm{\&beta;}+\sqrt{1-{\mathrm{\&beta;}}^2}|\mathrm{\&eta;}\left|\right)}^2+{\left|h_2\right|}^2{\left(\mathrm{\&beta;}\right|\mathrm{\&eta;}|+\sqrt{1-{\mathrm{\&beta;}}^2})}^2}{1+2\mathrm{\&beta;}\sqrt{1-{\mathrm{\&beta;}}^2}\left|\mathrm{\&eta;}\right|}$

${\mathrm{\&lambda;}}_2^2={\left|h_1\right|}^2+{\left|h_2\right|}^2-{\mathrm{\&lambda;}}_1^2$

$V_1=\frac{\mathrm{\&beta;}{h}_1{\left|h_1\right|}^{-1}+\sqrt{1-{\mathrm{\&beta;}}^2}\mathrm{\&eta;}{\left|\mathrm{\&eta;}\right|}^{-1}{h}_2{\left|h_2\right|}^{-1}}{\sqrt{1+2\mathrm{\&beta;}\sqrt{1-{\mathrm{\&beta;}}^2}\left|\mathrm{\&eta;}\right|}}$

$V_2=\frac{h_1-h_1{V}_1^H{V}_1}{|h_1-h_1{V}_1^H{V}_1|}$

Wherein, h ₁for first row vector of H, h ₂for second row vector of H, η and β calculate according to the following equation:

$\mathrm{\&eta;}=h_1{h}_2^H{\left|h_1\right|}^{-1}{\left|h_2\right|}^{-1}$

$\mathrm{\&beta;}=\sqrt{\frac{1}{2}\frac{2{\left|h_1\right|}^2({\left|h_1\right|}^2+{\left|h_2\right|}^2){\left|\mathrm{\&eta;}\right|}^2+{({\left|h_1\right|}^2-{\left|h_2\right|}^2)}^2+({\left|h_1\right|}^2-{\left|h_2\right|}^2)\sqrt{{({\left|h_1\right|}^2-{\left|h_2\right|}^2)}^2+4{\left|h_1\right|}^2{\left|h_2\right|}^2{\left|\mathrm{\&eta;}\right|}^2}}{{({\left|h_1\right|}^2-{\left|h_2\right|}^2)}^2+{({\left|h_1\right|}^2+{\left|h_2\right|}^2)}^2{\left|\mathrm{\&eta;}\right|}^2}}$

；

Be that 4 simplification svd algorithm corresponding to reception antenna quantity is:

With matrix as the matrix of pending SVD, according to simplification svd algorithm corresponding to the reception antenna quantity that is 2, determine respectively right singular vector corresponding to right singular vector, second largest singular value corresponding to maximum singular value, second largest singular value, maximum singular value, as the corresponding maximum singular value of H, second largest singular value, the right singular vector that maximum singular value is corresponding, right singular vector corresponding to second largest singular value; Wherein, λ _1,1, V _1,1, λ _2,1, V _2,1determine in the following manner:

H is divided and obtains H ₁and H ₂, wherein H ₁two row vectors by H form, H ₂two other row vector by H forms; With H ₁and H ₂as the matrix of pending SVD, according to simplification svd algorithm corresponding to the reception antenna quantity that is 2, determine respectively H ₁maximum singular value λ _1,1with corresponding right singular vector V _1,1, and H ₂maximum singular value λ _2,1with corresponding right singular vector V _2,1.

Obviously, those skilled in the art can carry out various changes and modification and not depart from the spirit and scope of the present invention the present invention.Like this, if these amendments of the present invention and within modification belongs to the scope of the claims in the present invention and equivalent technologies thereof, the present invention is also intended to comprise these changes and modification interior.

Claims (18)

1. a definite method for precoding vector, is characterized in that, comprising:

Determine respectively equivalent channel matrix T corresponding to each user terminal _j, and for user terminal described in each, carry out following step:

According to the corresponding T of other user terminals except this user terminal _j, determine the interference channel matrix of this user terminal and

Right in all row vectors carry out orthogonalization process, obtain one group of orthonormal basis B of the subspace of being opened _i; And

According to B _ithe equivalent channel matrix of determining this user terminal arrives the orthogonal intersection cast shadow matrix of the subspace of being opened ; And

Determine the quantity of this user terminal reception antenna, and with as the matrix of pending singular value decomposition SVD, carry out the corresponding relation of simplifying svd algorithm from the reception antenna quantity and the matrix for pending SVD that set in advance, determine the simplification svd algorithm corresponding with this quantity of determining;

Determined according to the simplification svd algorithm based on definite right singular vector corresponding to singular value, determine for the precoding vector of this user terminal.

2. the method for claim 1, is characterized in that, determines the equivalent channel matrix T that each user terminal is corresponding _j, specifically comprise:

The pilot signal SRS sending according to each user terminal, determines the down channel matrix H that each user terminal is corresponding _j; And

Judge that respectively for the transmission means of each user terminal be single-stream transmission or stream transmission;

For judging each single-stream transmission user terminal that described transmission means is single-stream transmission, carry out: determine the quantity of this single-stream transmission user terminal reception antenna, and with the H of this single-stream transmission user terminal _jas the matrix of pending SVD, from described corresponding relation, determine the simplification svd algorithm corresponding with the quantity of this single-stream transmission user terminal reception antenna; Determine the H of this single-stream transmission user terminal according to definite simplification svd algorithm _jright singular vector corresponding to maximum singular value, and right singular vector corresponding described maximum singular value is defined as to the T of this single-stream transmission user terminal _j;

For judging each stream transmission user terminal that described transmission means is stream transmission, carry out: by the H of this stream transmission user terminal _jbe defined as the T of this stream transmission user terminal _j.

3. the method for claim 1, is characterized in that, determines the equivalent channel matrix T that each user terminal is corresponding _j, specifically comprise:

Obtain each user terminal according to the codewords indexes PMI of the down public guide frequency signal CRS feedback of base station transmission; And

Judge that respectively for the transmission means of each user terminal be single-stream transmission or stream transmission;

For judging each single-stream transmission user terminal that described transmission means is single-stream transmission, carry out: the T that the first row vector of the pre-encoding codeword corresponding PMI of this single-stream transmission user terminal feedback is defined as to this single-stream transmission user terminal _j;

For judging each stream transmission user terminal that described transmission means is stream transmission, carry out: the T that determines this stream transmission user terminal according to the first two columns vector of pre-encoding codeword corresponding to the PMI of this stream transmission user terminal feedback _j.

4. the method as described in as arbitrary in claim 1～3, is characterized in that, described in the reception antenna quantity setting in advance and the corresponding relation of simplifying svd algorithm be specially:

Be that 1 simplification svd algorithm corresponding to reception antenna quantity is: using | H| as the maximum singular value that the matrix H of pending SVD is carried out obtaining after SVD, and with H|H| ^-1as corresponding to | the right singular value vector of H|, wherein, | H| represents to calculate the 2-norm of H;

Be that 2 simplification svd algorithm corresponding to reception antenna quantity is: with λ ₁ ²with respectively as H is carried out to the maximum singular value and the second largest singular value that obtain after SVD, and with V ₁and V ₂conduct is corresponding to λ respectively ₁ ²with right singular value vector, wherein, λ ₁ ², v ₁, V ₂calculate according to the following equation:

${{\mathrm{\&lambda;}}_1}^2=\frac{{\left|h_1\right|}^2{(\mathrm{\&beta;}+\sqrt{1-{\mathrm{\&beta;}}^2}|\mathrm{\&eta;}\left|\right)}^2+{\left|h_2\right|}^2{\left(\mathrm{\&beta;}\right|\mathrm{\&eta;}|+\sqrt{1-{\mathrm{\&beta;}}^2})}^2}{1+2\mathrm{\&beta;}\sqrt{1-{\mathrm{\&beta;}}^2}\left|\mathrm{\&eta;}\right|}$

${\mathrm{\&lambda;}}_2^2={\left|h_1\right|}^2+{\left|h_2\right|}^2-{{\mathrm{\&lambda;}}_1}^2$

$V_1=\frac{\mathrm{\&beta;}{h}_1{\left|h_1\right|}^{-1}+\sqrt{1-{\mathrm{\&beta;}}^2}\mathrm{\&eta;}{\left|\mathrm{\&eta;}\right|}^{-1}{h}_2{\left|h_2\right|}^{-1}}{\sqrt{1+2\mathrm{\&beta;}\sqrt{1-{\mathrm{\&beta;}}^2}\left|\mathrm{\&eta;}\right|}}$

$V_2=\frac{h_1-h_1{V_1}^H{V}_1}{|h_1-h_1{V_1}^H{V}_1|}$

Wherein, h ₁for first row vector of H, h ₂for second row vector of H, η and β calculate according to the following equation:

$\mathrm{\&eta;}=h_1{h}_2^H{\left|h_1\right|}^{-1}{\left|h_2\right|}^{-1}$

$\mathrm{\&beta;}=\sqrt{\frac{1}{2}\frac{2{\left|h_1\right|}^2({\left|h_1\right|}^2+{\left|h_2\right|}^2){\left|\mathrm{\&eta;}\right|}^2+{({\left|h_1\right|}^2-{{|h}_2|}^2)}^2+({\left|h_1\right|}^2-{\left|h_2\right|}^2)\sqrt{{({\left|h_1\right|}^2-{\left|h_2\right|}^2)}^2+4{\left|h_1\right|}^2{\left|h_2\right|}^2{\left|\mathrm{\&eta;}\right|}^2}}{{({\left|h_1\right|}^2-{\left|h_2\right|}^2)}^2+{({\left|h_1\right|}^2+{\left|h_2\right|}^2)}^2{\left|\mathrm{\&eta;}\right|}^2}};$

Be that 4 simplification svd algorithm corresponding to reception antenna quantity is:

With matrix as the matrix of pending SVD, according to simplification svd algorithm corresponding to the reception antenna quantity that is 2, determine respectively right singular vector corresponding to right singular vector, second largest singular value corresponding to maximum singular value, second largest singular value, maximum singular value, as the corresponding maximum singular value of H, second largest singular value, the right singular vector that maximum singular value is corresponding, right singular vector corresponding to second largest singular value; Wherein, λ _1,1, V _1,1, λ _2,1, V _2,1determine in the following manner:

H is divided and obtains H ₁and H ₂, wherein H ₁two row vectors by H form, H ₂two other row vector by H forms; With H ₁and H ₂as the matrix of pending SVD, according to simplification svd algorithm corresponding to the reception antenna quantity that is 2, determine respectively H ₁maximum singular value λ _1,1with corresponding right singular vector V _1,1, and H ₂maximum singular value λ _2,1with corresponding right singular vector V _2,1.

5. the method as described in as arbitrary in claim 1～3, is characterized in that, according to right singular vector corresponding to singular value determine the precoding vector for this user terminal, specifically comprise:

Judge that for the transmission means of this user terminal be single-stream transmission or stream transmission;

In the time judging described transmission means and be single-stream transmission, will right singular vector corresponding to maximum singular value be defined as described precoding vector;

In the time judging described transmission means and be stream transmission, according to maximum and second largest singular value respectively corresponding right singular vector determine described precoding vector.

6. a definite method for precoding vector, is applied in the single subscriber terminal transmission plan of TDD system, it is characterized in that, comprising:

The down channel matrix H that described user terminal is corresponding is determined in base station _i, and the quantity of definite described user terminal reception antenna; And

With H _ias the matrix of pending SVD, carry out the corresponding relation of simplifying svd algorithm from the reception antenna quantity and the matrix for pending SVD that set in advance, determine the simplification svd algorithm corresponding with this quantity of determining; And

According to the determined H of simplification svd algorithm based on definite _iright singular vector corresponding to singular value, determine for the precoding vector of described user terminal;

According to H _iright singular vector corresponding to singular value determine the precoding vector for described user terminal, specifically comprise:

Judge that for the transmission means of described user terminal be single-stream transmission or stream transmission;

In the time judging described transmission means and be single-stream transmission, by H _iright singular vector corresponding to maximum singular value be defined as described precoding vector;

In the time judging described transmission means and be stream transmission, according to H _imaximum and second largest singular value respectively corresponding right singular vector determine described precoding vector.

7. method as claimed in claim 6, is characterized in that, described in the reception antenna quantity that sets in advance be specially with the corresponding relation of simplifying svd algorithm:

Be that 1 simplification svd algorithm corresponding to reception antenna quantity is: using | H| as the maximum singular value that the matrix H of pending SVD is carried out obtaining after SVD, and with H|H| ^-1as corresponding to | the right singular value vector of H|, wherein, | H| represents to calculate the 2-norm of H;

Be that 2 simplification svd algorithm corresponding to reception antenna quantity is: with λ ₁ ²with respectively as H is carried out to the maximum singular value and the second largest singular value that obtain after SVD, and with V ₁and V ₂conduct is corresponding to λ respectively ₁ ²with right singular value vector, wherein, λ ₁ ², v ₁, V ₂calculate according to the following equation:

${{\mathrm{\&lambda;}}_1}^2=\frac{{\left|h_1\right|}^2{(\mathrm{\&beta;}+\sqrt{1-{\mathrm{\&beta;}}^2}|\mathrm{\&eta;}\left|\right)}^2+{\left|h_2\right|}^2{\left(\mathrm{\&beta;}\right|\mathrm{\&eta;}|+\sqrt{1-{\mathrm{\&beta;}}^2})}^2}{1+2\mathrm{\&beta;}\sqrt{1-{\mathrm{\&beta;}}^2}\left|\mathrm{\&eta;}\right|}$

${\mathrm{\&lambda;}}_2^2={\left|h_1\right|}^2+{\left|h_2\right|}^2-{{\mathrm{\&lambda;}}_1}^2$

$V_1=\frac{\mathrm{\&beta;}{h}_1{\left|h_1\right|}^{-1}+\sqrt{1-{\mathrm{\&beta;}}^2}\mathrm{\&eta;}{\left|\mathrm{\&eta;}\right|}^{-1}{h}_2{\left|h_2\right|}^{-1}}{\sqrt{1+2\mathrm{\&beta;}\sqrt{1-{\mathrm{\&beta;}}^2}\left|\mathrm{\&eta;}\right|}}$

$V_2=\frac{h_1-h_1{V_1}^H{V}_1}{|h_1-h_1{V_1}^H{V}_1|}$

Wherein, h ₁for first row vector of H, h ₂for second row vector of H, η and β calculate according to the following equation:

$\mathrm{\&eta;}=h_1{h}_2^H{\left|h_1\right|}^{-1}{\left|h_2\right|}^{-1}$

$\mathrm{\&beta;}=\sqrt{\frac{1}{2}\frac{2{\left|h_1\right|}^2({\left|h_1\right|}^2+{\left|h_2\right|}^2){\left|\mathrm{\&eta;}\right|}^2+{({\left|h_1\right|}^2-{{|h}_2|}^2)}^2+({\left|h_1\right|}^2-{\left|h_2\right|}^2)\sqrt{{({\left|h_1\right|}^2-{\left|h_2\right|}^2)}^2+4{\left|h_1\right|}^2{\left|h_2\right|}^2{\left|\mathrm{\&eta;}\right|}^2}}{{({\left|h_1\right|}^2-{\left|h_2\right|}^2)}^2+{({\left|h_1\right|}^2+{\left|h_2\right|}^2)}^2{\left|\mathrm{\&eta;}\right|}^2}};$

Be that 4 simplification svd algorithm corresponding to reception antenna quantity is:

With matrix as the matrix of pending SVD, according to simplification svd algorithm corresponding to the reception antenna quantity that is 2, determine respectively right singular vector corresponding to right singular vector, second largest singular value corresponding to maximum singular value, second largest singular value, maximum singular value, as the corresponding maximum singular value of H, second largest singular value, the right singular vector that maximum singular value is corresponding, right singular vector corresponding to second largest singular value; Wherein, λ _1,1, V _1,1, λ _2,1, V _2,1determine in the following manner:

H is divided and obtains H ₁and H ₂, wherein H ₁two row vectors by H form, H ₂two other row vector by H forms; With H ₁and H ₂as the matrix of pending SVD, according to simplification svd algorithm corresponding to the reception antenna quantity that is 2, determine respectively H ₁maximum singular value λ _1,1with corresponding right singular vector V _1,1, and H ₂maximum singular value λ _2,1with corresponding right singular vector V _2,1.

8. a definite method for precoding vector, is applied in the single subscriber terminal transmission plan of FDD system, it is characterized in that, comprising:

Described user terminal is determined the down channel matrix H of self _i, and the quantity of definite described user terminal reception antenna; And

With H _ias the matrix of pending SVD, carry out the corresponding relation of simplifying svd algorithm from the reception antenna quantity and the matrix for pending SVD that set in advance, determine the simplification svd algorithm corresponding with this quantity of determining; And

According to the determined H of simplification svd algorithm based on definite _iright singular vector corresponding to singular value, determine and send corresponding PMI to base station, corresponding precoding vector is determined according to PMI in instruction base station;

Determine PMI according to described right singular vector, specifically comprise:

Described user terminal judges that for the transmission means of described user terminal be single-stream transmission or stream transmission;

In the time judging described transmission means and be single-stream transmission, from precoding codebook, choose corresponding to H _ithe code word of the corresponding right singular vector of maximum singular value, and determine PMI corresponding to this code word choosing;

In the time judging described transmission means and be stream transmission, from precoding codebook, choose corresponding to H _imaximum and second largest singular value divide the code word of the right singular matrix that forms of corresponding right singular vector, and determine PMI corresponding to this code word choosing.

9. method as claimed in claim 8, is characterized in that, described in the reception antenna quantity that sets in advance be specially with the corresponding relation of simplifying svd algorithm:

Be that 1 simplification svd algorithm corresponding to reception antenna quantity is: using | H| as the maximum singular value that the matrix H of pending SVD is carried out obtaining after SVD, and with H|H| ^-1as corresponding to | the right singular value vector of H|, wherein, | H| represents to calculate the 2-norm of H;

Be that 2 simplification svd algorithm corresponding to reception antenna quantity is: with λ ₁ ²with respectively as H is carried out to the maximum singular value and the second largest singular value that obtain after SVD, and with V ₁and V ₂conduct is corresponding to λ respectively ₁ ²with right singular value vector, wherein, λ ₁ ², v ₁, V ₂calculate according to the following equation:

${{\mathrm{\&lambda;}}_1}^2=\frac{{\left|h_1\right|}^2{(\mathrm{\&beta;}+\sqrt{1-{\mathrm{\&beta;}}^2}|\mathrm{\&eta;}\left|\right)}^2+{\left|h_2\right|}^2{\left(\mathrm{\&beta;}\right|\mathrm{\&eta;}|+\sqrt{1-{\mathrm{\&beta;}}^2})}^2}{1+2\mathrm{\&beta;}\sqrt{1-{\mathrm{\&beta;}}^2}\left|\mathrm{\&eta;}\right|}$

${\mathrm{\&lambda;}}_2^2={\left|h_1\right|}^2+{\left|h_2\right|}^2-{{\mathrm{\&lambda;}}_1}^2$

$V_1=\frac{\mathrm{\&beta;}{h}_1{\left|h_1\right|}^{-1}+\sqrt{1-{\mathrm{\&beta;}}^2}\mathrm{\&eta;}{\left|\mathrm{\&eta;}\right|}^{-1}{h}_2{\left|h_2\right|}^{-1}}{\sqrt{1+2\mathrm{\&beta;}\sqrt{1-{\mathrm{\&beta;}}^2}\left|\mathrm{\&eta;}\right|}}$

$V_2=\frac{h_1-h_1{V_1}^H{V}_1}{|h_1-h_1{V_1}^H{V}_1|}$

Wherein, h ₁for first row vector of H, h ₂for second row vector of H, η and β calculate according to the following equation:

$\mathrm{\&eta;}=h_1{h}_2^H{\left|h_1\right|}^{-1}{\left|h_2\right|}^{-1}$

$\mathrm{\&beta;}=\sqrt{\frac{1}{2}\frac{2{\left|h_1\right|}^2({\left|h_1\right|}^2+{\left|h_2\right|}^2){\left|\mathrm{\&eta;}\right|}^2+{({\left|h_1\right|}^2-{{|h}_2|}^2)}^2+({\left|h_1\right|}^2-{\left|h_2\right|}^2)\sqrt{{({\left|h_1\right|}^2-{\left|h_2\right|}^2)}^2+4{\left|h_1\right|}^2{\left|h_2\right|}^2{\left|\mathrm{\&eta;}\right|}^2}}{{({\left|h_1\right|}^2-{\left|h_2\right|}^2)}^2+{({\left|h_1\right|}^2+{\left|h_2\right|}^2)}^2{\left|\mathrm{\&eta;}\right|}^2}};$

Be that 4 simplification svd algorithm corresponding to reception antenna quantity is:

With matrix as the matrix of pending SVD, according to simplification svd algorithm corresponding to the reception antenna quantity that is 2, determine respectively right singular vector corresponding to right singular vector, second largest singular value corresponding to maximum singular value, second largest singular value, maximum singular value, as the corresponding maximum singular value of H, second largest singular value, the right singular vector that maximum singular value is corresponding, right singular vector corresponding to second largest singular value; Wherein, λ _1,1, V _1,1, λ _2,1, V _2,1determine in the following manner:

H is divided and obtains H ₁and H ₂, wherein H ₁two row vectors by H form, H ₂two other row vector by H forms; With H ₁and H ₂as the matrix of pending SVD, according to simplification svd algorithm corresponding to the reception antenna quantity that is 2, determine respectively H ₁maximum singular value λ _1,1with corresponding right singular vector V _1,1, and H ₂maximum singular value λ _2,1with corresponding right singular vector V _2,1.

10. a determining device for precoding vector, is characterized in that, comprising:

Equivalent channel matrix determining unit, for determining equivalent channel matrix T corresponding to each user terminal _j;

Interference channel matrix determining unit, for for each user terminal, the corresponding equivalent channel matrix T of other user terminals except this user terminal determining according to equivalent channel matrix determining unit _j, determine the interference channel matrix of this user terminal

Orthogonalization process unit, for what determine interference channel matrix determining unit in all row vectors carry out orthogonalization process, obtain one group of orthonormal basis B of the subspace of being opened _i;

Orthogonal intersection cast shadow matrix determining unit, for the B obtaining according to orthogonalization process unit _i, determine that the equivalent channel matrix of this user terminal arrives the orthogonal intersection cast shadow matrix of the subspace of being opened ;

Simplify svd algorithm determining unit, for according to the quantity of this user terminal reception antenna, determine with orthogonal intersection cast shadow matrix determining unit as the matrix of pending singular value decomposition SVD, carry out the corresponding relation of simplifying svd algorithm from the reception antenna quantity and the matrix for pending SVD that set in advance, determine the simplification svd algorithm corresponding with the quantity of this user terminal reception antenna;

Precoding vector determining unit, determined based on simplifying the definite simplification svd algorithm of svd algorithm determining unit for basis right singular vector corresponding to singular value, determine for the precoding vector of this user terminal.

11. devices as claimed in claim 10, is characterized in that, described equivalent channel matrix determining unit specifically comprises:

Down channel matrix determination module, for the pilot signal SRS sending according to each user terminal, determines the down channel matrix H that each user terminal is corresponding _j;

The first judge module, for judging that respectively for the transmission means of each user terminal be single-stream transmission or stream transmission;

The first Executive Module, for judging for the first judge module each single-stream transmission user terminal that described transmission means is single-stream transmission, carry out: determine the quantity of this single-stream transmission user terminal reception antenna, and the H of this single-stream transmission user terminal of determining with descending channel matrix determination module _jas the matrix of pending SVD, from described corresponding relation, determine the simplification svd algorithm corresponding with the quantity of this single-stream transmission user terminal reception antenna; Determine the H of this single-stream transmission user terminal according to definite simplification svd algorithm _jright singular vector corresponding to maximum singular value, and right singular vector corresponding described maximum singular value is defined as to the T of this single-stream transmission user terminal _j;

The second Executive Module, for judge each stream transmission user terminal that described transmission means is stream transmission for the first judge module, carries out: by the H of this stream transmission user terminal _jbe defined as the T of this stream transmission user terminal _j.

12. devices as claimed in claim 10, is characterized in that, described equivalent channel matrix determining unit specifically comprises:

Codewords indexes obtains module, for obtaining each user terminal according to the codewords indexes PMI of the down public guide frequency signal CRS feedback of base station transmission;

The first judge module, for judging that respectively for the transmission means of each user terminal be single-stream transmission or stream transmission;

The first Executive Module, for judging for the first judge module each single-stream transmission user terminal that described transmission means is single-stream transmission, carry out: the T that the first row vector of the pre-encoding codeword corresponding PMI of this single-stream transmission user terminal feedback is defined as to this single-stream transmission user terminal _j;

The second Executive Module, for judging for the first judge module each stream transmission user terminal that described transmission means is stream transmission, carry out: the T that determines this stream transmission user terminal according to the first two columns vector of pre-encoding codeword corresponding to the PMI of this stream transmission user terminal feedback _j.

13. devices as described in as arbitrary in claim 10～12, is characterized in that, described in the reception antenna quantity setting in advance and the corresponding relation of simplifying svd algorithm be specially:

Be that 1 simplification svd algorithm corresponding to reception antenna quantity is: using | H| as the maximum singular value that the matrix H of pending SVD is carried out obtaining after SVD, and with H|H| ^-1as corresponding to | the right singular value vector of H|, wherein, | H| represents to calculate the 2-norm of H;

Be that 2 simplification svd algorithm corresponding to reception antenna quantity is: with λ ₁ ²with respectively as H is carried out to the maximum singular value and the second largest singular value that obtain after SVD, and with V ₁and V ₂conduct is corresponding to λ respectively ₁ ²with right singular value vector, wherein, λ ₁ ², v ₁, V ₂calculate according to the following equation:

${{\mathrm{\&lambda;}}_1}^2=\frac{{\left|h_1\right|}^2{(\mathrm{\&beta;}+\sqrt{1-{\mathrm{\&beta;}}^2}|\mathrm{\&eta;}\left|\right)}^2+{\left|h_2\right|}^2{\left(\mathrm{\&beta;}\right|\mathrm{\&eta;}|+\sqrt{1-{\mathrm{\&beta;}}^2})}^2}{1+2\mathrm{\&beta;}\sqrt{1-{\mathrm{\&beta;}}^2}\left|\mathrm{\&eta;}\right|}$

${\mathrm{\&lambda;}}_2^2={\left|h_1\right|}^2+{\left|h_2\right|}^2-{{\mathrm{\&lambda;}}_1}^2$

$V_1=\frac{\mathrm{\&beta;}{h}_1{\left|h_1\right|}^{-1}+\sqrt{1-{\mathrm{\&beta;}}^2}\mathrm{\&eta;}{\left|\mathrm{\&eta;}\right|}^{-1}{h}_2{\left|h_2\right|}^{-1}}{\sqrt{1+2\mathrm{\&beta;}\sqrt{1-{\mathrm{\&beta;}}^2}\left|\mathrm{\&eta;}\right|}}$

$V_2=\frac{h_1-h_1{V_1}^H{V}_1}{|h_1-h_1{V_1}^H{V}_1|}$

Wherein, h ₁for first row vector of H, h ₂for second row vector of H, η and β calculate according to the following equation:

$\mathrm{\&eta;}=h_1{h}_2^H{\left|h_1\right|}^{-1}{\left|h_2\right|}^{-1}$

$\mathrm{\&beta;}=\sqrt{\frac{1}{2}\frac{2{\left|h_1\right|}^2({\left|h_1\right|}^2+{\left|h_2\right|}^2){\left|\mathrm{\&eta;}\right|}^2+{({\left|h_1\right|}^2-{{|h}_2|}^2)}^2+({\left|h_1\right|}^2-{\left|h_2\right|}^2)\sqrt{{({\left|h_1\right|}^2-{\left|h_2\right|}^2)}^2+4{\left|h_1\right|}^2{\left|h_2\right|}^2{\left|\mathrm{\&eta;}\right|}^2}}{{({\left|h_1\right|}^2-{\left|h_2\right|}^2)}^2+{({\left|h_1\right|}^2+{\left|h_2\right|}^2)}^2{\left|\mathrm{\&eta;}\right|}^2}};$

Be that 4 simplification svd algorithm corresponding to reception antenna quantity is:

With matrix as the matrix of pending SVD, according to simplification svd algorithm corresponding to the reception antenna quantity that is 2, determine respectively right singular vector corresponding to right singular vector, second largest singular value corresponding to maximum singular value, second largest singular value, maximum singular value, as the corresponding maximum singular value of H, second largest singular value, the right singular vector that maximum singular value is corresponding, right singular vector corresponding to second largest singular value; Wherein, λ _1,1, V _1,1, λ _2,1, V _2,1determine in the following manner:

H is divided and obtains H ₁and H ₂, wherein H ₁two row vectors by H form, H ₂two other row vector by H forms; With H ₁and H ₂as the matrix of pending SVD, according to simplification svd algorithm corresponding to the reception antenna quantity that is 2, determine respectively H ₁maximum singular value λ _1,1with corresponding right singular vector V _1,1, and H ₂maximum singular value λ _2,1with corresponding right singular vector V _2,1.

14. devices as described in as arbitrary in claim 10～12, is characterized in that, described precoding vector determining unit specifically comprises:

The second judge module, for judging that for the transmission means of this user terminal be single-stream transmission or stream transmission;

Determination module, while being single-stream transmission for judging described transmission means when the second judge module, will right singular vector corresponding to maximum singular value be defined as described precoding vector, in the time that the second judge module is judged described transmission means and is stream transmission, according to maximum and second largest singular value respectively corresponding right singular vector determine described precoding vector.

15. 1 kinds of base stations, are applied in the single subscriber terminal transmission plan of TDD system, it is characterized in that, comprising:

Matrix and quantity determining unit, for determining down channel matrix H corresponding to described user terminal _iquantity with described user terminal reception antenna;

Simplify svd algorithm determining unit, for matrix and the definite H of quantity determining unit _ias the matrix of pending SVD, carry out the corresponding relation of simplifying svd algorithm from the reception antenna quantity and the matrix for pending SVD that set in advance, determine the corresponding simplification svd algorithm of this quantity that matrix and quantity determining unit are determined;

Precoding vector determining unit, for according to simplifying svd algorithm determining unit definite simplification H that svd algorithm is determined _iright singular vector corresponding to singular value, determine for the precoding vector of described user terminal;

Described precoding vector determining unit specifically comprises:

Judge module, for judging that for the transmission means of described user terminal be single-stream transmission or stream transmission;

Precoding vector determination module, while being single-stream transmission for judge described transmission means at judge module, by H _iright singular vector corresponding to maximum singular value be defined as described precoding vector, in the time that judge module is judged described transmission means and is stream transmission, according to H _imaximum and second largest singular value respectively corresponding right singular vector determine described precoding vector.

16. base stations as claimed in claim 15, is characterized in that, described in the reception antenna quantity that sets in advance be specially with the corresponding relation of simplifying svd algorithm:

Be that 1 simplification svd algorithm corresponding to reception antenna quantity is: using | H| as the maximum singular value that the matrix H of pending SVD is carried out obtaining after SVD, and with H|H| ^-1as corresponding to | the right singular value vector of H|, wherein, | H| represents to calculate the 2-norm of H;

Be that 2 simplification svd algorithm corresponding to reception antenna quantity is: with λ ₁ ²with respectively as H is carried out to the maximum singular value and the second largest singular value that obtain after SVD, and with V ₁and V ₂conduct is corresponding to λ respectively ₁ ²with right singular value vector, wherein, λ ₁ ², v ₁, V ₂calculate according to the following equation:

${{\mathrm{\&lambda;}}_1}^2=\frac{{\left|h_1\right|}^2{(\mathrm{\&beta;}+\sqrt{1-{\mathrm{\&beta;}}^2}|\mathrm{\&eta;}\left|\right)}^2+{\left|h_2\right|}^2{\left(\mathrm{\&beta;}\right|\mathrm{\&eta;}|+\sqrt{1-{\mathrm{\&beta;}}^2})}^2}{1+2\mathrm{\&beta;}\sqrt{1-{\mathrm{\&beta;}}^2}\left|\mathrm{\&eta;}\right|}$

${\mathrm{\&lambda;}}_2^2={\left|h_1\right|}^2+{\left|h_2\right|}^2-{{\mathrm{\&lambda;}}_1}^2$

$V_1=\frac{\mathrm{\&beta;}{h}_1{\left|h_1\right|}^{-1}+\sqrt{1-{\mathrm{\&beta;}}^2}\mathrm{\&eta;}{\left|\mathrm{\&eta;}\right|}^{-1}{h}_2{\left|h_2\right|}^{-1}}{\sqrt{1+2\mathrm{\&beta;}\sqrt{1-{\mathrm{\&beta;}}^2}\left|\mathrm{\&eta;}\right|}}$

$V_2=\frac{h_1-h_1{V_1}^H{V}_1}{|h_1-h_1{V_1}^H{V}_1|}$

Wherein, h ₁for first row vector of H, h ₂for second row vector of H, η and β calculate according to the following equation:

$\mathrm{\&eta;}=h_1{h}_2^H{\left|h_1\right|}^{-1}{\left|h_2\right|}^{-1}$

$\mathrm{\&beta;}=\sqrt{\frac{1}{2}\frac{2{\left|h_1\right|}^2({\left|h_1\right|}^2+{\left|h_2\right|}^2){\left|\mathrm{\&eta;}\right|}^2+{({\left|h_1\right|}^2-{{|h}_2|}^2)}^2+({\left|h_1\right|}^2-{\left|h_2\right|}^2)\sqrt{{({\left|h_1\right|}^2-{\left|h_2\right|}^2)}^2+4{\left|h_1\right|}^2{\left|h_2\right|}^2{\left|\mathrm{\&eta;}\right|}^2}}{{({\left|h_1\right|}^2-{\left|h_2\right|}^2)}^2+{({\left|h_1\right|}^2+{\left|h_2\right|}^2)}^2{\left|\mathrm{\&eta;}\right|}^2}};$

Be that 4 simplification svd algorithm corresponding to reception antenna quantity is:

With matrix as the matrix of pending SVD, according to simplification svd algorithm corresponding to the reception antenna quantity that is 2, determine respectively right singular vector corresponding to right singular vector, second largest singular value corresponding to maximum singular value, second largest singular value, maximum singular value, as the corresponding maximum singular value of H, second largest singular value, the right singular vector that maximum singular value is corresponding, right singular vector corresponding to second largest singular value; Wherein, λ _1,1, V _1,1, λ _2,1, V _2,1determine in the following manner:

H is divided and obtains H ₁and H ₂, wherein H ₁two row vectors by H form, H ₂two other row vector by H forms; With H ₁and H ₂as the matrix of pending SVD, according to simplification svd algorithm corresponding to the reception antenna quantity that is 2, determine respectively H ₁maximum singular value λ _1,1with corresponding right singular vector V _1,1, and H ₂maximum singular value λ _2,1with corresponding right singular vector V _2,1.

17. 1 kinds of user terminals, are applied in the single subscriber terminal transmission plan of FDD system, it is characterized in that, comprising:

Matrix and quantity determining unit, for determining the down channel matrix H of described user terminal _i, and the quantity of definite described user terminal reception antenna;

Simplify svd algorithm determining unit, for matrix and the definite H of quantity determining unit _ias the matrix of pending SVD, carry out the corresponding relation of simplifying svd algorithm from the reception antenna quantity and the matrix for pending SVD that set in advance, determine simplification svd algorithm corresponding to this quantity of determining with matrix and quantity determining unit;

Transmitting element, for according to right singular vector corresponding to singular value of simplifying the definite determined Hi of simplification svd algorithm of svd algorithm determining unit, determines and sends corresponding PMI to base station, corresponding precoding vector is determined according to PMI in instruction base station;

Described transmitting element specifically comprises:

Judge module, for judging that for the transmission means of described user terminal be single-stream transmission or stream transmission;

Choose module, while being single-stream transmission for judge described transmission means when judge module, from precoding codebook, choose corresponding to H _ithe code word of the corresponding right singular vector of maximum singular value in the time that disconnected module is judged described transmission means and is stream transmission, from precoding codebook, choose corresponding to H _imaximum and second largest singular value divide the code word of the right singular matrix that forms of corresponding right singular vector;

Determination module, chooses for determining PMI corresponding to code word that module is chosen;

Sending module, for send the definite PMI of determination module to base station, the PMI that instruction base station sends according to sending module determines corresponding precoding vector.

18. user terminals as claimed in claim 17, is characterized in that, described in the reception antenna quantity that sets in advance be specially with the corresponding relation of simplifying svd algorithm:

Be that 1 simplification svd algorithm corresponding to reception antenna quantity is: using | H| as the maximum singular value that the matrix H of pending SVD is carried out obtaining after SVD, and with H|H| ^-1as corresponding to | the right singular value vector of H|, wherein, | H| represents to calculate the 2-norm of H;

Be that 2 simplification svd algorithm corresponding to reception antenna quantity is: with λ ₁ ²with respectively as H is carried out to the maximum singular value and the second largest singular value that obtain after SVD, and with V ₁and V ₂conduct is corresponding to λ respectively ₁ ²with right singular value vector, wherein, λ ₁ ², v ₁, V ₂calculate according to the following equation:

${{\mathrm{\&lambda;}}_1}^2=\frac{{\left|h_1\right|}^2{(\mathrm{\&beta;}+\sqrt{1-{\mathrm{\&beta;}}^2}|\mathrm{\&eta;}\left|\right)}^2+{\left|h_2\right|}^2{\left(\mathrm{\&beta;}\right|\mathrm{\&eta;}|+\sqrt{1-{\mathrm{\&beta;}}^2})}^2}{1+2\mathrm{\&beta;}\sqrt{1-{\mathrm{\&beta;}}^2}\left|\mathrm{\&eta;}\right|}$

${\mathrm{\&lambda;}}_2^2={\left|h_1\right|}^2+{\left|h_2\right|}^2-{{\mathrm{\&lambda;}}_1}^2$

$V_1=\frac{\mathrm{\&beta;}{h}_1{\left|h_1\right|}^{-1}+\sqrt{1-{\mathrm{\&beta;}}^2}\mathrm{\&eta;}{\left|\mathrm{\&eta;}\right|}^{-1}{h}_2{\left|h_2\right|}^{-1}}{\sqrt{1+2\mathrm{\&beta;}\sqrt{1-{\mathrm{\&beta;}}^2}\left|\mathrm{\&eta;}\right|}}$

$V_2=\frac{h_1-h_1{V_1}^H{V}_1}{|h_1-h_1{V_1}^H{V}_1|}$

Wherein, h ₁for first row vector of H, h ₂for second row vector of H, η and β calculate according to the following equation:

$\mathrm{\&eta;}=h_1{h}_2^H{\left|h_1\right|}^{-1}{\left|h_2\right|}^{-1}$

$\mathrm{\&beta;}=\sqrt{\frac{1}{2}\frac{2{\left|h_1\right|}^2({\left|h_1\right|}^2+{\left|h_2\right|}^2){\left|\mathrm{\&eta;}\right|}^2+{({\left|h_1\right|}^2-{{|h}_2|}^2)}^2+({\left|h_1\right|}^2-{\left|h_2\right|}^2)\sqrt{{({\left|h_1\right|}^2-{\left|h_2\right|}^2)}^2+4{\left|h_1\right|}^2{\left|h_2\right|}^2{\left|\mathrm{\&eta;}\right|}^2}}{{({\left|h_1\right|}^2-{\left|h_2\right|}^2)}^2+{({\left|h_1\right|}^2+{\left|h_2\right|}^2)}^2{\left|\mathrm{\&eta;}\right|}^2}};$

Be that 4 simplification svd algorithm corresponding to reception antenna quantity is:

With matrix as the matrix of pending SVD, according to simplification svd algorithm corresponding to the reception antenna quantity that is 2, determine respectively right singular vector corresponding to right singular vector, second largest singular value corresponding to maximum singular value, second largest singular value, maximum singular value, as the corresponding maximum singular value of H, second largest singular value, the right singular vector that maximum singular value is corresponding, right singular vector corresponding to second largest singular value; Wherein, λ _1,1, V _1,1, λ _2,1, V _2,1determine in the following manner:

H is divided and obtains H ₁and H ₂, wherein H ₁two row vectors by H form, H ₂two other row vector by H forms; With H ₁and H ₂as the matrix of pending SVD, according to simplification svd algorithm corresponding to the reception antenna quantity that is 2, determine respectively H ₁maximum singular value λ _1,1with corresponding right singular vector V _1,1, and H ₂maximum singular value λ _2,1with corresponding right singular vector V _2,1.