CN103378945A - Pre-coding method for multi-user multi-input multi-output system and apparatus thereof - Google Patents

Abstract

The invention discloses a pre-coding method for a multi-user multi-input multi-output system and an apparatus thereof. The method comprises the steps of acquiring channel gains of a plurality of user terminals; initializing an autocorrelation matrix of receiving weights of each user terminal; calculating a pre-coding vector of each user terminal based on channel gain matrixes and the autocorrelation matrixes; calculating updated autocorrelation matrixes based on the pre-coding vectors; calculating a updated pre-coding vector of each user terminal; executing the steps of calculating the updated autocorrelation matrixes and calculating the updated pre-coding vector of each user terminal in a cyclic manner; and adjusting power and phase of a signal transmitted to each user terminal based on the calculated pre-coding vectors of each user terminal by a preset number of times. The method or the apparatus can be used for removing the limitation that the number of antennas transmitted by the base station must not be less than the total number of antennas received by all the other users thereby reducing a bit error rate of the system.

Description

The method for precoding of multiuser mimo system and device

Technical field

The present invention relates to the communications field, particularly relate to a kind of method for precoding and device of multiuser mimo system.

Background technology

Multi-user's multiple-input and multiple-output (Multi-User Multiple Input Multiple Output, MU-MIMO) system broadcast channel refers to that transmit leg (for example base station) and recipient's (for example travelling carriage) all are furnished with the down link of a plurality of antennas.In the MU-MIMO system, owing to carry out the transmitting-receiving of a plurality of subscriber signals simultaneously, so there is common-channel interference.In the prior art, mainly solve the problem of common-channel interference by the method for the output signal of transmit leg being carried out precoding.

Fig. 1 is the mimo system structural representation under the multi-user environment of down link list residential quarter.As shown in Figure 1, base station while and K telex network.Base station and user adopt many antenna configurations, and i user's reception antenna number is M _i, the base station transmit antennas number is N.I user's channel matrix is H _i, precoding is weighted to W _i

The method for precoding of a kind of mimo system of the prior art, its basic thought are based on block diagonalization (BD) algorithm of zero interference conditions to release.

The BD algorithm is exactly to find other users' kernel vector as active user's precoding vector.In order to eliminate the interference between the user fully, will satisfy:

$\left\{\begin{array}{c}{H}_1\underset{k=1,k\≠1}{\overset{K}{\mathrm{\Σ}}}{w}_k{s}_k\left(n\right)=0\\ H_2\underset{k=1,k\≠2}{\overset{K}{\mathrm{\Σ}}}{w}_k{s}_k\left(n\right)=0\\ .\\ .\\ .\\ H_K\underset{k=1,k\≠K}{\overset{K}{\mathrm{\Σ}}}{w}_k{s}_k\left(n\right)=0\end{array}\right.$

Namely all users' precoding vector will satisfy:

H _kw _j＝0(k≠j)

To the total channel matrix except k user's channel matrix Carry out SVD (singular value decomposition) and decompose, obtain the kernel Vector Groups, that is:

${\tilde{H}}_k=\left[\begin{array}{c}{H}_1\\ .\\ .\\ .\\ H_{k-1}\\ H_{k+1}\\ .\\ .\\ .\\ H_K\end{array}\right]=\left[{\tilde{U}}_k^{\left(1\right)}{\tilde{U}}_k^{\left(0\right)}\right]\begin{array}{}\end{array}\left[\begin{array}{cc}\mathrm{\Λ}& 0\\ 0& 0\end{array}\right]{\left[{\tilde{V}}_k^{\left(1\right)}{\tilde{V}}_k^{\left(0\right)}\right]}^{*}(\underset{j\≠k}{\mathrm{\Σ}}{M}_j\×N)$

Matrix

Order be L _kSo, Just consisted of L _kIndividual right singular vector,

Dimension be N * (N-L _k), namely by (N-L _k) individual vector formed channel matrix The kernel Vector Groups, satisfy

${\tilde{H}}_k\×{\tilde{V}}_k^{\left(0\right)}=0$

Namely

k＝{1，...，K}，i≠k。

But the method for precoding of this mimo system of the prior art has a constraints: concerning user i, the base station transmit antennas number can not be less than every other user's reception antenna sum.This is a very large restriction for the multi-user MIMO system of reality.

Summary of the invention

The method for precoding and the device that the purpose of this invention is to provide a kind of multiuser mimo system, the restriction that can remove the base station transmit antennas number, and the elimination of the interference between the realization multi-user effectively reduce error rate of system.

For achieving the above object, the invention provides following scheme:

A kind of method for precoding of multiuser mimo system comprises:

Obtain the channel gain of a plurality of user terminals; Each user terminal has a plurality of channels, and the channel gain of a plurality of channels of each user terminal consists of channel gain matrix;

Autocorrelation matrix to the reception weights of each user terminal carries out initialization, is set to unit matrix;

According to described channel gain matrix and described autocorrelation matrix, calculate the precoding vector of each user terminal;

According to the autocorrelation matrix after the described precoding vector calculating renewal;

According to the autocorrelation matrix after described channel gain matrix and the described renewal, calculate the precoding vector after each user terminal upgrades;

The step of the precoding vector after the described autocorrelation matrix after the described calculating of circulation execution is upgraded and each user terminal of described calculating upgrade;

When carrying out number of times and reach preset times, the precoding vector of each user terminal that calculates according to described preset times is adjusted power and the phase place of the signal that sends to each user terminal.

Optionally, described according to the autocorrelation matrix after the described precoding vector calculating renewal, comprising:

According to expression formula

Calculate the autocorrelation matrix after upgrading;

Wherein, R _iAutocorrelation matrix after expression is upgraded, H _jBe the channel gain matrix of j user terminal,

Be the conjugate transpose of the channel gain matrix of j user terminal, The precoding vector of i user terminal when expression execution number of times is l,

The conjugate transpose of the precoding vector of i user terminal when expression execution number of times is l.

Optionally, the precoding vector of described each user terminal of calculating comprises:

For i user terminal, according to expression formula

Calculate the equivalent channel correlation matrix sum of other user terminals; Wherein, Q _iExpression does not comprise i user terminal at the equivalent channel correlation matrix sum of other interior user terminals, H _jBe the channel gain matrix of j user terminal, Be the conjugate transpose of the channel gain matrix of j user terminal, R _jIt is the described autocorrelation matrix of j user terminal;

Judge Q _iWhether be non-singular matrix, obtain the first judged result;

When described the first judged result represents Q _iDuring for non-singular matrix, calculate the first eigenvector matrix of i user terminal, from described first eigenvector matrix, select maximal eigenvector corresponding to eigenvalue of maximum, described maximal eigenvector is carried out normalized, obtain the precoding vector of i user terminal;

When described the first judged result represents Q _iWhen being not non-singular matrix, to Q _iDo singular value decomposition, utilize right singular vector corresponding to non-zero singular value wherein, calculate the Second Characteristic vector matrix of i user terminal, from described Second Characteristic vector matrix, select maximal eigenvector corresponding to eigenvalue of maximum, described maximal eigenvector is carried out normalized, obtain the precoding vector of i user terminal.

Optionally, the first eigenvector matrix of i user terminal of described calculating comprises:

According to expression formula Calculate the first eigenvector matrix of i user terminal;

Wherein, A ₁The first eigenvector matrix that represents i user terminal,

The inverse matrix that represents the equivalent channel correlation matrix sum of other user terminals, H _iBe the channel gain matrix of i user terminal,

It is the conjugate transpose of the channel gain matrix of i user terminal.

Optionally, the Second Characteristic vector matrix of i user terminal of described calculating comprises:

According to formula $Q_i=U\left(\begin{array}{cc}\mathrm{\Σ}& \\ & 0\end{array}\right){\left[V^{\left(1\right)}{V}^{\left(0\right)}\right]}^{*}$ To Q _iCarry out singular value decomposition;

Wherein, U is left singular vector, V ⁽¹⁾The corresponding right singular vector of non-zero singular value, V ⁽⁰⁾Be zero right singular vector corresponding to singular value;

According to expression formula A ₂=H _iV ⁽⁰⁾Calculate the Second Characteristic vector matrix of i user terminal;

Wherein, A ₂The Second Characteristic vector matrix that represents i user terminal, H _iIt is the channel gain matrix of i user terminal.

Optionally, described described maximal eigenvector is carried out normalized, comprising:

According to formula Described maximal eigenvector is carried out normalized;

Wherein,

The described maximal eigenvector of i user terminal when expression execution number of times is l,

The described maximal eigenvector after the normalized is carried out in expression.

A kind of pre-coding apparatus of multiuser mimo system comprises:

The channel gain acquisition module is for the channel gain that obtains a plurality of users; Each user has a plurality of channels, and the channel gain of a plurality of channels of each user consists of channel gain matrix;

The autocorrelation matrix initialization module is used for the autocorrelation matrix of each user's reception weights is carried out initialization, is set to unit matrix;

The precoding vector computing module is used for according to described channel gain matrix and described autocorrelation matrix, calculates each user's precoding vector;

The autocorrelation matrix update module is used for according to the autocorrelation matrix after the described precoding vector calculating renewal;

The precoding vector update module is used for according to the autocorrelation matrix after described channel gain matrix and the described renewal, calculates the precoding vector after each user upgrades;

Control module is upgraded in circulation, is used for controlling the step that the precoding vector after described autocorrelation matrix after described calculating is upgraded and each user of described calculating upgrade is carried out in described autocorrelation matrix update module and the circulation of described precoding vector update module;

The transmitted signal adjusting module is used for when carrying out number of times and reach preset times, and each user's who calculates according to described preset times precoding vector is adjusted to power and the phase place of the signal of each user's transmission.

Optionally, described autocorrelation matrix update module comprises:

The autocorrelation matrix updating block is used for according to expression formula

Calculate the autocorrelation matrix after upgrading;

Wherein, R _iAutocorrelation matrix after expression is upgraded, H _jBe j user's channel gain matrix, Be the conjugate transpose of j user's channel gain matrix, I user's precoding vector when expression execution number of times is l, The conjugate transpose of i user's precoding vector when expression execution number of times is l.

Optionally, described precoding vector computing module comprises:

Other user's equivalent channel correlation matrix and computing units are used for for i user, according to expression formula

Calculate other users' equivalent channel correlation matrix sum; Wherein, Q _iExpression does not comprise i user at other interior users equivalent channel correlation matrix sum, H _jBe j user's channel gain matrix,

Be the conjugate transpose of j user's channel gain matrix, R _jIt is j user's described autocorrelation matrix;

The first judging unit is used for judging Q _iWhether be non-singular matrix, obtain the first judged result;

The first precoding vector computing unit is used for representing Q when described the first judged result _iDuring for non-singular matrix, calculate i user's first eigenvector matrix, from described first eigenvector matrix, select maximal eigenvector corresponding to eigenvalue of maximum, described maximal eigenvector is carried out normalized, obtain i user's precoding vector;

The second precoding vector computing unit is used for representing Q when described the first judged result _iWhen being not non-singular matrix, to Q _iDo singular value decomposition, utilize right singular vector corresponding to non-zero singular value wherein, calculate i user's Second Characteristic vector matrix, from described Second Characteristic vector matrix, select maximal eigenvector corresponding to eigenvalue of maximum, described maximal eigenvector is carried out normalized, obtain i user's precoding vector.

Optionally, described the first precoding vector computing unit comprises:

First eigenvector matrix computations subelement is used for according to expression formula A ₁=H _iV ⁽⁰⁾Calculate i user's first eigenvector matrix;

Wherein, A ₁The first eigenvector matrix that represents i user,

The inverse matrix that represents other users' equivalent channel correlation matrix sum, H _jBe j user's channel gain matrix,

It is the conjugate transpose of j user's channel gain matrix.

Optionally, described the second precoding vector computing unit comprises:

The singular value decomposition subelement is used for according to formula $Q_i=U\left(\begin{array}{cc}\mathrm{\Σ}& \\ & 0\end{array}\right){\left[V^{\left(1\right)}{V}^{\left(0\right)}\right]}^{*}$ To Q _iCarry out singular value decomposition;

Wherein, V ⁽⁰⁾Be right singular vector corresponding to non-zero singular value;

Second Characteristic vector matrix computation subunit is used for according to expression formula A ₂=H _iV ⁽⁰⁾Calculate i user's Second Characteristic vector matrix;

Wherein, A ₂The Second Characteristic vector matrix that represents i user, H _iIt is i user's channel gain matrix.

Optionally, described the first precoding vector computing unit and/or described the second precoding vector computing unit comprise:

The normalized subelement is used for according to formula

Described maximal eigenvector is carried out normalized;

Wherein,

I user's described maximal eigenvector when expression execution number of times is l,

The described maximal eigenvector after the normalized is carried out in expression.

According to specific embodiment provided by the invention, the invention discloses following technique effect:

The method for precoding of multiuser mimo system of the present invention and device after the weights of the precoding vector of transmitting terminal are determined, utilize the weights of the precoding vector of this transmitting terminal of determining to remove to upgrade the autocorrelation matrix of receiving terminal again.And then with the autocorrelation matrix after upgrading as the input of calculating precoding vector next time, calculate new precoding vector.By iteration can approximation theory on the optimal value of user terminal precoding vector, thereby can obtain to be better than the gain of existing algorithm, realize that the interference between the multiple users is eliminated, effectively reduce error rate of system.And, owing to according to described channel gain matrix and described autocorrelation matrix, calculate the precoding vector of each user terminal, so can remove the restriction that the base station transmit antennas number must be not less than every other user's reception antenna sum.

Description of drawings

In order to be illustrated more clearly in the embodiment of the invention or technical scheme of the prior art, the below will do to introduce simply to the accompanying drawing of required use among the embodiment, apparently, accompanying drawing in the following describes only is some embodiments of the present invention, for those of ordinary skills, under the prerequisite of not paying creative work, can also obtain according to these accompanying drawings other accompanying drawing.

Fig. 1 is the flow chart of the method for precoding embodiment one of multiuser mimo system of the present invention;

Fig. 2 is the flow chart of the method for precoding embodiment two of multiuser mimo system of the present invention;

Fig. 3 is the flow chart of the pre-coding apparatus embodiment one of multiuser mimo system of the present invention;

Fig. 4 is the flow chart of the pre-coding apparatus embodiment two of multiuser mimo system of the present invention.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the invention, the technical scheme in the embodiment of the invention is clearly and completely described, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, those of ordinary skills belong to the scope of protection of the invention not making the every other embodiment that obtains under the creative work prerequisite.

For above-mentioned purpose of the present invention, feature and advantage can be become apparent more, the present invention is further detailed explanation below in conjunction with the drawings and specific embodiments.

Fig. 1 is the flow chart of the method for precoding embodiment one of multiuser mimo system of the present invention.As shown in Figure 1, the method can comprise:

Step S101: the channel gain that obtains a plurality of user terminals;

Concrete, each user terminal has a plurality of channels, and the channel gain of a plurality of channels of each user terminal consists of channel gain matrix.

In the present specification, also the equipment such as base station can be called transmitting terminal, user terminal is called receiving terminal.When obtaining the channel gain of user terminal, can calculate and each channel yield value initiatively is emitted to transmitting terminal by user terminal self.Also can be calculated voluntarily by transmitting terminal each channel gain of this user terminal according to the user terminal signal intensity that receives.

Step S102: the autocorrelation matrix to each user's reception weights carries out initialization, is set to unit matrix;

It all is 1 that unit matrix refers on the diagonal, and all the other elements are all 0 matrix.The line number of unit matrix or columns are identical with the channel number of user terminal.

Step S103: according to described channel gain matrix and described autocorrelation matrix, calculate the precoding vector of each user terminal;

Concrete, the precoding vector that calculates each user terminal can be in the following ways:

Suppose to have K user terminal, then for i user terminal, according to expression formula

Calculate the equivalent channel correlation matrix sum of other user terminals; Wherein, Q _iExpression does not comprise i user terminal at the equivalent channel correlation matrix sum of other interior user terminals, H _jBe the channel gain matrix of j user terminal,

Be the conjugate transpose of the channel gain matrix of j user terminal, R _jIt is the autocorrelation matrix of j user terminal;

Judge Q _iWhether be non-singular matrix, obtain the first judged result;

When described the first judged result represents Q _iDuring for non-singular matrix, calculate the first eigenvector matrix of i user terminal, from described first eigenvector matrix, select maximal eigenvector corresponding to eigenvalue of maximum, described maximal eigenvector is carried out normalized, obtain the precoding vector of i user terminal;

When calculating the first eigenvector matrix of i user terminal, can be according to expression formula Calculate the first eigenvector matrix of i user terminal;

Wherein, A ₁The first eigenvector matrix that represents i user terminal,

The inverse matrix that represents the equivalent channel correlation matrix sum of other user terminals, H _iBe the channel gain matrix of i user terminal,

It is the conjugate transpose of the channel gain matrix of i user terminal.Need to prove, because Q _iBe non-singular matrix, so can invert.

When described the first judged result represents Q _iWhen being not non-singular matrix, to Q _iDo singular value decomposition, utilize right singular vector corresponding to non-zero singular value wherein, calculate the Second Characteristic vector matrix of i user terminal, from described Second Characteristic vector matrix, select maximal eigenvector corresponding to eigenvalue of maximum, described maximal eigenvector is carried out normalized, obtain the precoding vector of i user terminal.

Because the method for precoding of the application's multiuser mimo system is at Q _iWhen being not non-singular matrix, when for example the base station transmit antennas number is less than the reception antenna sum of every other user terminal, also can calculate the precoding vector of i user terminal, so can remove the restriction of base station transmit antennas number.

When calculating the Second Characteristic vector matrix of i user terminal, can be according to formula $Q_i=U\left(\begin{array}{cc}\mathrm{\Σ}& \\ & 0\end{array}\right){\left[V^{\left(1\right)}{V}^{\left(0\right)}\right]}^{*}$ To Q _iCarry out singular value decomposition;

Wherein, U is left singular vector, V ⁽¹⁾The corresponding right singular vector of non-zero singular value, V ⁽⁰⁾Be zero right singular vector corresponding to singular value;

According to expression formula A ₂=H _iV ⁽⁰⁾Calculate the Second Characteristic vector matrix of i user terminal;

Wherein, A ₂The Second Characteristic vector matrix that represents i user terminal, H _iIt is the channel gain matrix of i user terminal.

Step S104: according to the autocorrelation matrix after the described precoding vector calculating renewal;

Can be according to expression formula

Calculate the autocorrelation matrix after upgrading;

Wherein, R _iAutocorrelation matrix after expression is upgraded, H _jBe the channel gain matrix of j user terminal,

Be the conjugate transpose of the channel gain matrix of j user terminal,

The precoding vector of i user terminal when expression execution number of times is l,

The conjugate transpose of the precoding vector of i user terminal when expression execution number of times is l.

Because the weights of the precoding vector of transmitting terminal are relevant with the detection receiving algorithm that receiving terminal adopts in theory, so the present invention passes through iteration, after the weights of the precoding vector of transmitting terminal are determined, utilize the weights of the precoding vector of this transmitting terminal of determining to remove again to upgrade the autocorrelation matrix of receiving terminal.And then with the autocorrelation matrix after upgrading as the input of calculating precoding vector next time, calculate new precoding vector.

And prior art is not considered the receiving terminal algorithm or only suppose the detection receiving algorithm (its detection algorithm does not upgrade with the right value update of precoding vector) that certain is fixing, therefore the application's scheme can the approximation theory optimal value by iteration, thereby obtains to be better than having now the gain of algorithm.Realize the interference elimination between the multiple users, effectively reduce error rate of system.

Step S105: according to the autocorrelation matrix after described channel gain matrix and the described renewal, calculate the precoding vector after each user terminal upgrades;

Step and the method among the step S103 of the precoding vector after calculate upgrading are basic identical, are autocorrelation matrix is replaced with autocorrelation matrix after the renewal.

Step S106: the step of the precoding vector after the described autocorrelation matrix after the described calculating of circulation execution is upgraded and each user terminal of described calculating upgrade;

This step namely circulate execution in step S104 and step S105.Reach preset times until carry out number of times.Need to prove that the number of times of execution is more, namely iterations is more, and the precoding vector that then obtains is got over the optimal value on the approximation theory.

Step S107: when carrying out number of times and reach preset times, the precoding vector of each user terminal that calculates according to described preset times is adjusted power and the phase place of the signal that sends to each user terminal.

Concrete, the precoding vector that finally calculates has size and Orientation.The power of the signal that sends to each user terminal can be adjusted into the equal and opposite in direction with the precoding vector that finally calculates, the phase place of the signal that sends to each user terminal according to the direction adjustment of the precoding vector that finally calculates.

In sum, the method for precoding of the multiuser mimo system of present embodiment after the weights of the precoding vector of transmitting terminal are determined, utilizes the weights of the precoding vector of this transmitting terminal of determining to remove to upgrade the autocorrelation matrix of receiving terminal again.And then with the autocorrelation matrix after upgrading as the input of calculating precoding vector next time, calculate new precoding vector.By iteration can approximation theory on the optimal value of user terminal precoding vector, thereby can obtain to be better than the gain of existing algorithm, realize that the interference between the multiple users is eliminated, effectively reduce error rate of system.And, owing to according to described channel gain matrix and described autocorrelation matrix, calculate the precoding vector of each user terminal, so can remove the restriction of base station transmit antennas number.

Fig. 2 is the flow chart of the method for precoding embodiment two of multiuser mimo system of the present invention.As shown in Figure 2, the method can comprise:

Step S201: the channel gain that obtains a plurality of user terminals;

Step S202: the autocorrelation matrix to each user's reception weights carries out initialization, is set to unit matrix;

Step S203: for i user terminal, according to expression formula

Calculate the equivalent channel correlation matrix sum of other user terminals;

Wherein, Q _iExpression does not comprise i user terminal at the equivalent channel correlation matrix sum of other interior user terminals, H _jBe the channel gain matrix of j user terminal,

Be the conjugate transpose of the channel gain matrix of j user terminal, R _jIt is the autocorrelation matrix of j user terminal

Step S204: judge Q _iWhether be non-singular matrix, obtain the first judged result;

Step S205: when described the first judged result represents Q _iDuring for non-singular matrix, according to expression formula Calculate the first eigenvector matrix of i user terminal;

Wherein, A ₁The first eigenvector matrix that represents i user terminal,

The inverse matrix that represents the equivalent channel correlation matrix sum of other user terminals, H _iBe the channel gain matrix of i user terminal,

It is the conjugate transpose of the channel gain matrix of i user terminal.

Step S206: when described the first judged result represents Q _iWhen being not non-singular matrix, according to formula $Q_i=U\left(\begin{array}{cc}\mathrm{\Σ}& \\ & 0\end{array}\right){\left[V^{\left(1\right)}{V}^{\left(0\right)}\right]}^{*}$ To Q _iCarry out singular value decomposition; According to expression formula A ₂=H _iV ⁽⁰⁾Calculate the Second Characteristic vector matrix of i user terminal;

Wherein, U is left singular vector, V ⁽¹⁾The corresponding right singular vector of non-zero singular value, V ⁽⁰⁾Be zero right singular vector corresponding to singular value; A ₂The Second Characteristic vector matrix that represents i user terminal, H _iIt is the channel gain matrix of i user terminal.

Step S207: from described first eigenvector matrix or described Second Characteristic vector matrix, select maximal eigenvector corresponding to eigenvalue of maximum, according to formula

Described maximal eigenvector is carried out normalized;

Wherein,

The described maximal eigenvector of i user terminal when expression execution number of times is l,

The described maximal eigenvector after the normalized is carried out in expression.

Step S208: according to expression formula

Calculate the autocorrelation matrix after upgrading;

Wherein, R _iAutocorrelation matrix after expression is upgraded, H _jBe the channel gain matrix of j user terminal,

Be the conjugate transpose of the channel gain matrix of j user terminal,

The precoding vector of i user terminal when expression execution number of times is l,

The conjugate transpose of the precoding vector of i user terminal when expression execution number of times is l.

Step S209: according to the autocorrelation matrix after described channel gain matrix and the described renewal, calculate the precoding vector after each user terminal upgrades;

Step S210: the step of the precoding vector after the described autocorrelation matrix after the described calculating of circulation execution is upgraded and each user terminal of described calculating upgrade;

Step S211: when carrying out number of times and reach preset times, the precoding vector of each user terminal that calculates according to described preset times is adjusted power and the phase place of the signal that sends to each user terminal.

The invention also discloses a kind of pre-coding apparatus of multiuser mimo system.Fig. 3 is the flow chart of the pre-coding apparatus embodiment one of multiuser mimo system of the present invention.As shown in Figure 3, this device can comprise:

Channel gain acquisition module 301 is for the channel gain that obtains a plurality of users; Each user has a plurality of channels, and the channel gain of a plurality of channels of each user consists of channel gain matrix;

Autocorrelation matrix initialization module 302 is used for the autocorrelation matrix of each user's reception weights is carried out initialization, is set to unit matrix;

Precoding vector computing module 303 is used for according to described channel gain matrix and described autocorrelation matrix, calculates each user's precoding vector;

Autocorrelation matrix update module 304 is used for according to the autocorrelation matrix after the described precoding vector calculating renewal;

Precoding vector update module 305 is used for according to the autocorrelation matrix after described channel gain matrix and the described renewal, calculates the precoding vector after each user upgrades;

Control module 306 is upgraded in circulation, is used for controlling the step that the precoding vector after described autocorrelation matrix after described calculating is upgraded and each user of described calculating upgrade is carried out in described autocorrelation matrix update module and the circulation of described precoding vector update module;

Transmitted signal adjusting module 307 is used for when carrying out number of times and reach preset times, and each user's who calculates according to described preset times precoding vector is adjusted to power and the phase place of the signal of each user's transmission.

Fig. 4 is the flow chart of the pre-coding apparatus embodiment two of multiuser mimo system of the present invention.As shown in Figure 4, this device can comprise:

Channel gain acquisition module 301 is for the channel gain that obtains a plurality of users; Each user has a plurality of channels, and the channel gain of a plurality of channels of each user consists of channel gain matrix;

Autocorrelation matrix initialization module 302 is used for the autocorrelation matrix of each user's reception weights is carried out initialization, is set to unit matrix;

Other user's equivalent channel correlation matrix and computing unit 3031 are used for for i user, according to expression formula

Calculate other users' equivalent channel correlation matrix sum; Wherein, Q _iExpression does not comprise i user at other interior users equivalent channel correlation matrix sum, H _jBe j user's channel gain matrix,

Be the conjugate transpose of j user's channel gain matrix, R _jIt is j user's described autocorrelation matrix;

The first judging unit 3032 is used for judging Q _iWhether be non-singular matrix, obtain the first judged result;

The first precoding vector computing unit 3033 is used for representing Q when described the first judged result _iDuring for non-singular matrix, calculate i user's first eigenvector matrix, from described first eigenvector matrix, select maximal eigenvector corresponding to eigenvalue of maximum, described maximal eigenvector is carried out normalized, obtain i user's precoding vector;

The second precoding vector computing unit 3034 is used for representing Q when described the first judged result _iWhen being not non-singular matrix, to Q _iDo singular value decomposition, utilize right singular vector corresponding to non-zero singular value wherein, calculate i user's Second Characteristic vector matrix, from described Second Characteristic vector matrix, select maximal eigenvector corresponding to eigenvalue of maximum, described maximal eigenvector is carried out normalized, obtain i user's precoding vector.

Need to prove, above-mentioned other user's equivalent channel correlation matrix and computing unit 3031, the first judging unit 3032, the first precoding vector computing units 3033 and the second precoding vector computing unit 3034 can be contained in the precoding vector computing module 303.

And described the first precoding vector computing unit 3033 can comprise:

First eigenvector matrix computations subelement is used for according to expression formula A ₁=H _iV ⁽⁰⁾Calculate i user's first eigenvector matrix;

Wherein, A ₁The first eigenvector matrix that represents i user,

The inverse matrix that represents other users' equivalent channel correlation matrix sum, H _jBe j user's channel gain matrix,

It is the conjugate transpose of j user's channel gain matrix.

Described the second precoding vector computing unit 3034 can comprise:

The singular value decomposition subelement is used for according to formula $Q_i=U\left(\begin{array}{cc}\mathrm{\Σ}& \\ & 0\end{array}\right){\left[V^{\left(1\right)}{V}^{\left(0\right)}\right]}^{*}$ To Q _iCarry out singular value decomposition;

Wherein, V ⁽⁰⁾Be right singular vector corresponding to non-zero singular value;

Second Characteristic vector matrix computation subunit is used for according to expression formula A ₂=H _iV ⁽⁰⁾Calculate i user's Second Characteristic vector matrix;

Wherein, A ₂The Second Characteristic vector matrix that represents i user, H _iIt is i user's channel gain matrix.

Described the first precoding vector computing unit 3033 and/or described the second precoding vector computing unit 3034 can also comprise:

The normalized subelement is used for according to formula

Described maximal eigenvector is carried out normalized;

Wherein,

I user's described maximal eigenvector when expression execution number of times is l,

The described maximal eigenvector after the normalized is carried out in expression.

Described pre-coding apparatus also comprises:

Autocorrelation matrix updating block 3041 is used for according to expression formula

Calculate the autocorrelation matrix after upgrading;

Wherein, R _iAutocorrelation matrix after expression is upgraded, H _jBe j user's channel gain matrix,

Be the conjugate transpose of j user's channel gain matrix,

I user's precoding vector when expression execution number of times is l,

The conjugate transpose of i user's precoding vector when expression execution number of times is l.

Described autocorrelation matrix updating block 3041 can be contained in autocorrelation matrix update module 304.

Described pre-coding apparatus also comprises:

Precoding vector update module 305 is used for according to the autocorrelation matrix after described channel gain matrix and the described renewal, calculates the precoding vector after each user upgrades;

Control module 306 is upgraded in circulation, is used for controlling the step that the precoding vector after described autocorrelation matrix after described calculating is upgraded and each user of described calculating upgrade is carried out in described autocorrelation matrix update module and the circulation of described precoding vector update module;

Transmitted signal adjusting module 307 is used for when carrying out number of times and reach preset times, and each user's who calculates according to described preset times precoding vector is adjusted to power and the phase place of the signal of each user's transmission.

In sum, the pre-coding apparatus of multiuser mimo system of the present invention after the weights of the precoding vector of transmitting terminal are determined, utilizes the weights of the precoding vector of this transmitting terminal of determining to remove to upgrade the autocorrelation matrix of receiving terminal again.And then with the autocorrelation matrix after upgrading as the input of calculating precoding vector next time, calculate new precoding vector.By iteration can approximation theory on the optimal value of user terminal precoding vector, thereby can obtain to be better than the gain of existing algorithm, realize that the interference between the multiple users is eliminated, effectively reduce error rate of system.And, owing to according to described channel gain matrix and described autocorrelation matrix, calculate the precoding vector of each user terminal, so can remove the restriction of base station transmit antennas number.

Each embodiment adopts the mode of going forward one by one to describe in this specification, and what each embodiment stressed is and the difference of other embodiment that identical similar part is mutually referring to getting final product between each embodiment.For the disclosed device of embodiment, because it is corresponding with the disclosed method of embodiment, so description is fairly simple, relevant part partly illustrates referring to method and gets final product.

Used specific case herein principle of the present invention and execution mode are set forth, the explanation of above embodiment just is used for helping to understand method of the present invention and core concept thereof; Simultaneously, for one of ordinary skill in the art, according to thought of the present invention, all will change in specific embodiments and applications.In sum, this description should not be construed as limitation of the present invention.

Claims (12)

1. the method for precoding of a multiuser mimo system is characterized in that, comprising:

Obtain the channel gain of a plurality of user terminals; Each user terminal has a plurality of channels, and the channel gain of a plurality of channels of each user terminal consists of channel gain matrix;

Autocorrelation matrix to the reception weights of each user terminal carries out initialization, is set to unit matrix;

According to described channel gain matrix and described autocorrelation matrix, calculate the precoding vector of each user terminal;

According to the autocorrelation matrix after the described precoding vector calculating renewal;

According to the autocorrelation matrix after described channel gain matrix and the described renewal, calculate the precoding vector after each user terminal upgrades;

The step of the precoding vector after the described autocorrelation matrix after the described calculating of circulation execution is upgraded and each user terminal of described calculating upgrade;

When carrying out number of times and reach preset times, the precoding vector of each user terminal that calculates according to described preset times is adjusted power and the phase place of the signal that sends to each user terminal.

2. method according to claim 1 is characterized in that, and is described according to the autocorrelation matrix after the described precoding vector calculating renewal, comprising:

According to expression formula

Calculate the autocorrelation matrix after upgrading;

Wherein, R _iAutocorrelation matrix after expression is upgraded, H _jBe the channel gain matrix of j user terminal,

Be the conjugate transpose of the channel gain matrix of j user terminal,

The precoding vector of i user terminal when expression execution number of times is l,

The conjugate transpose of the precoding vector of i user terminal when expression execution number of times is l.

3. method according to claim 1 is characterized in that, the precoding vector of described each user terminal of calculating comprises:

For i user terminal, according to expression formula

Calculate the equivalent channel correlation matrix sum of other user terminals; Wherein, Q _iExpression does not comprise i user terminal at the equivalent channel correlation matrix sum of other interior user terminals, H _jBe the channel gain matrix of j user terminal,

Be the conjugate transpose of the channel gain matrix of j user terminal, R _jIt is the described autocorrelation matrix of j user terminal;

Judge Q _iWhether be non-singular matrix, obtain the first judged result;

When described the first judged result represents Q _iDuring for non-singular matrix, calculate the first eigenvector matrix of i user terminal, from described first eigenvector matrix, select maximal eigenvector corresponding to eigenvalue of maximum, described maximal eigenvector is carried out normalized, obtain the precoding vector of i user terminal;

When described the first judged result represents Q _iWhen being not non-singular matrix, to Q _iDo singular value decomposition, utilize right singular vector corresponding to non-zero singular value wherein, calculate the Second Characteristic vector matrix of i user terminal, from described Second Characteristic vector matrix, select maximal eigenvector corresponding to eigenvalue of maximum, described maximal eigenvector is carried out normalized, obtain the precoding vector of i user terminal.

4. method according to claim 3 is characterized in that, the first eigenvector matrix of i user terminal of described calculating comprises:

According to expression formula

Calculate the first eigenvector matrix of i user terminal;

Wherein, A ₁The first eigenvector matrix that represents i user terminal,

The inverse matrix that represents the equivalent channel correlation matrix sum of other user terminals, H _iBe the channel gain matrix of i user terminal,

It is the conjugate transpose of the channel gain matrix of i user terminal.

5. method according to claim 3 is characterized in that, the Second Characteristic vector matrix of i user terminal of described calculating comprises:

According to formula $Q_i=U\left[\begin{array}{cc}\mathrm{\Σ}& \\ & 0\end{array}\right]{\left[V^{\left(1\right)}{V}^{\left(0\right)}\right]}^{*}$ To Q _iCarry out singular value decomposition;

Wherein, U is left singular vector, V ⁽¹⁾The corresponding right singular vector of non-zero singular value, V ⁽⁰⁾Be zero right singular vector corresponding to singular value;

According to expression formula A ₂=H _iV ⁽⁰⁾Calculate the Second Characteristic vector matrix of i user terminal;

Wherein, A ₂The Second Characteristic vector matrix that represents i user terminal, H _iIt is the channel gain matrix of i user terminal.

6. method according to claim 3 is characterized in that, described described maximal eigenvector is carried out normalized, comprising:

According to formula

Described maximal eigenvector is carried out normalized;

Wherein,

The described maximal eigenvector of i user terminal when expression execution number of times is l,

The described maximal eigenvector after the normalized is carried out in expression.

7. the pre-coding apparatus of a multiuser mimo system is characterized in that, comprising:

The channel gain acquisition module is for the channel gain that obtains a plurality of users; Each user has a plurality of channels, and the channel gain of a plurality of channels of each user consists of channel gain matrix;

The autocorrelation matrix initialization module is used for the autocorrelation matrix of each user's reception weights is carried out initialization, is set to unit matrix;

The precoding vector computing module is used for according to described channel gain matrix and described autocorrelation matrix, calculates each user's precoding vector;

The autocorrelation matrix update module is used for according to the autocorrelation matrix after the described precoding vector calculating renewal;

The precoding vector update module is used for according to the autocorrelation matrix after described channel gain matrix and the described renewal, calculates the precoding vector after each user upgrades;

Control module is upgraded in circulation, is used for controlling the step that the precoding vector after described autocorrelation matrix after described calculating is upgraded and each user of described calculating upgrade is carried out in described autocorrelation matrix update module and the circulation of described precoding vector update module;

The transmitted signal adjusting module is used for when carrying out number of times and reach preset times, and each user's who calculates according to described preset times precoding vector is adjusted to power and the phase place of the signal of each user's transmission.

8. pre-coding apparatus according to claim 7 is characterized in that, described autocorrelation matrix update module comprises:

The autocorrelation matrix updating block is used for according to expression formula

Calculate the autocorrelation matrix after upgrading;

Wherein, R _iAutocorrelation matrix after expression is upgraded, H _jBe j user's channel gain matrix,

Be the conjugate transpose of j user's channel gain matrix,

I user's precoding vector when expression execution number of times is l,

The conjugate transpose of i user's precoding vector when expression execution number of times is l.

9. pre-coding apparatus according to claim 7 is characterized in that, described precoding vector computing module comprises:

Other user's equivalent channel correlation matrix and computing units are used for for i user, according to expression formula

Calculate other users' equivalent channel correlation matrix sum; Wherein, Q _iExpression does not comprise i user at other interior users equivalent channel correlation matrix sum, H _jBe j user's channel gain matrix, Be the conjugate transpose of j user's channel gain matrix, R _jIt is j user's described autocorrelation matrix;

The first judging unit is used for judging Q _iWhether be non-singular matrix, obtain the first judged result;

The first precoding vector computing unit is used for representing Q when described the first judged result _iDuring for non-singular matrix, calculate i user's first eigenvector matrix, from described first eigenvector matrix, select maximal eigenvector corresponding to eigenvalue of maximum, described maximal eigenvector is carried out normalized, obtain i user's precoding vector;

The second precoding vector computing unit is used for representing Q when described the first judged result _iWhen being not non-singular matrix, to Q _iDo singular value decomposition, utilize right singular vector corresponding to non-zero singular value wherein, calculate i user's Second Characteristic vector matrix, from described Second Characteristic vector matrix, select maximal eigenvector corresponding to eigenvalue of maximum, described maximal eigenvector is carried out normalized, obtain i user's precoding vector.

10. pre-coding apparatus according to claim 9 is characterized in that, described the first precoding vector computing unit comprises:

First eigenvector matrix computations subelement is used for according to expression formula A ₁=H _iV ⁽⁰⁾Calculate i user's first eigenvector matrix;

Wherein, A ₁The first eigenvector matrix that represents i user, The inverse matrix that represents other users' equivalent channel correlation matrix sum, H _jBe j user's channel gain matrix,

It is the conjugate transpose of j user's channel gain matrix.

11. pre-coding apparatus according to claim 9 is characterized in that, described the second precoding vector computing unit comprises:

The singular value decomposition subelement is used for according to formula $Q_i=U\left[\begin{array}{cc}\mathrm{\Σ}& \\ & 0\end{array}\right]{\left[V^{\left(1\right)}{V}^{\left(0\right)}\right]}^{*}$ To Q _iCarry out singular value decomposition;

Wherein, V ⁽⁰⁾Be right singular vector corresponding to non-zero singular value;

Second Characteristic vector matrix computation subunit is used for according to expression formula A ₂=H _iV ⁽⁰⁾Calculate i user's Second Characteristic vector matrix;

Wherein, A ₂The Second Characteristic vector matrix that represents i user, H _iIt is i user's channel gain matrix.

12. pre-coding apparatus according to claim 9 is characterized in that, described the first precoding vector computing unit and/or described the second precoding vector computing unit comprise:

The normalized subelement is used for according to formula

Described maximal eigenvector is carried out normalized;

Wherein,

I user's described maximal eigenvector when expression execution number of times is l,

The described maximal eigenvector after the normalized is carried out in expression.

Images