CN101873190B - Precoding method and device - Google Patents

Abstract

A precoding method and device are disclosed. The precoding method includes: the transmitting end obtains the channel information of its multiple antennas to two users respectively, and obtains the channel correlation matrix according to the channel information, the transmitting end obtains the generalized eigenvector of the channel correlation matrix, and the transmitting end determines the two channels according to the generalized eigenvector transmit precoding vectors of two users, obtain the receiving vectors of two users according to the transmitting precoding vectors, and obtain the equivalent channel matrices corresponding to the two users respectively according to the receiving vectors and the channel information, and the transmitting end obtains the corresponding equivalent channel matrices according to the said equal channel information The effective channel matrix obtains the power scaling matrix, feedback matrix and feedforward matrix respectively corresponding to two users, and performs THP for the two users according to the receiving vector, power scaling matrix, feedback matrix and feedforward matrix. The above technical solution can make the number of antennas configured by two users unlimited.

Description

Precoding method and device

technical field

The invention relates to the technical field of network communication, in particular to THP technology in a Multi-Input-Multi-Output (MIMO) system.

Background technique

Under MIMO Broadcast Channel, Dirty Paper Coding (DPC) is an optimized technical solution. Tomplinson Harashima Precoding (THP) can not only implement the DPC technical solution, but also THP technology has the characteristics of simple implementation and good performance.

The system model of THP is shown in Figure 1.

The relationship between the sending and receiving of the system model in Figure 1 can be expressed as: y=GHFB ^-1 v+Gn, where G is the power scaling matrix, H is the channel information from the transmitting antennas of the transmitting end to the user, and F is the front Feed matrix, B is the feedback matrix, v is the transmit signal vector, n is the noise vector on the user's receive antenna.

The realization principle of THP will be described below with reference to FIG. 1 .

Usually, under the multi-user MIMO broadcast channel, after the transmitted signal passes through the channel H, the signals sent to multiple users will interfere with each other. In order to eliminate the mutual interference of signals between users, the THP system first performs linear precoding at the transmitter, that is, Perform linear precoding on information flow a according to channel information. Since the channel is decomposed into an orthogonal triangular matrix, the channel can be equivalent to the form of an orthogonal matrix channel concatenated with a lower triangular matrix channel. After setting, the channel can be equivalent to the channel of a lower triangular matrix. At this time, the interference characteristics between the transmitted signals of the layers of the triangular matrix can be used, through the formula Perform linear precoding, where b _{k, l} are elements in the feedback matrix B, and k is the number of users. After the linearly precoded transmission signal x passes through F and the channel H, the interference between the layers of the transmission signal x is eliminated, so that each user can independently decode their own data. Since linear precoding will lead to an increase in power, the THP system introduces a MOD (modulo) operation to reduce the transmission power of the transmitted signal. The modulo operation turns the original linear precoding into a nonlinear precoding.

The inventors have found that at least the following problems exist in the above-mentioned prior art: In MIMO Broadcast Downlink (multiple-input multiple-output downlink broadcasting), and each user receives a single stream, since THP is limited by the channel space, the number of receiving users cannot exceed The number of transmitting antennas; that is, the existing THP can only be applied to the application scenario where the number of users does not exceed the number of transmitting antennas and each user is equipped with a single receiving antenna. It cannot solve the problem of how to perform THP when the number of receiving antennas exceeds the number of transmitting antennas.

Contents of the invention

In the precoding method and device provided by the embodiments of the present invention, when multiple antennas are set at the transmitting end, the number of antennas configured by two users may not be limited.

The precoding method provided by the embodiment of the present invention includes:

The transmitting end obtains channel information from its multiple antennas to two users respectively, and obtains a channel correlation matrix according to the channel information;

The transmitting end obtains the generalized eigenvector of the channel correlation matrix;

The transmitting end determines the transmitting precoding vectors of the two users according to the generalized eigenvectors, obtains the receiving vectors of the two users according to the transmitting precoding vectors of the two users, and obtains the receiving vectors of the two users according to the receiving vectors of the two users The vector and the channel information obtain equivalent channel matrices respectively corresponding to the two users;

The transmitting end obtains the power scaling matrix, feedback matrix and feedforward matrix corresponding to the two users respectively according to the equivalent channel matrices corresponding to the two users, and according to the receiving vector, power scaling matrix, feedback matrix and forward The feed matrix performs Tomlinson-Harashima precoding THP for the system composed of the two users.

The precoding device provided in the embodiment of the present invention includes:

A channel correlation matrix module, configured to obtain channel information from multiple antennas at the transmitting end to two users respectively, and obtain a channel correlation matrix according to the channel information;

An eigenvector module, configured to obtain a generalized eigenvector of the channel correlation matrix;

An equivalent channel matrix module, configured to obtain transmit precoding vectors of the two users according to the generalized eigenvectors, obtain receive vectors of the two users according to the transmit precoding vectors of the two users, and obtain the receive vectors of the two users according to the The receiving vectors of the two users and the channel information obtain the equivalent channel matrices respectively corresponding to the two users;

A precoding module, configured to obtain power scaling matrices, feedback matrices, and feedforward matrices respectively corresponding to the two users according to the equivalent channel matrix, and obtain the power scaling matrix, feedback matrix, and feedforward matrix according to the receiving vector, power scaling matrix, feedback matrix, and feedforward matrix THP is performed for the two-user system.

From the description of the above technical solution, it can be seen that when the transmitting end is equipped with multiple antennas, the transmitting end obtains the channel correlation matrix by using the channel information of the multiple antennas to two users respectively, and obtains the generalized feature by using the channel correlation matrix vector, and then further use the generalized eigenvectors to obtain the receiving vectors of the two users. Since the receiving vectors are used in the process of determining the equivalent channel matrix, the order of the equivalent channel matrices of the two users is reduced, so that the equal The order of the effective channel matrix is only related to the number of antennas at the transmitting end and the number of users, but has nothing to do with the number of receiving antennas of the users, so that the complexity of THP has nothing to do with the number of receiving antennas of the users, and finally the number of receiving antennas of the users can be unlimited .

Description of drawings

Fig. 1 is the system model schematic diagram of THP of prior art;

FIG. 2 is a flowchart of a precoding method according to Embodiment 1 of the present invention;

3 is a schematic diagram of a MIMO system model according to Embodiment 1 of the present invention;

Fig. 4 is a schematic diagram of the BER performance comparison between Embodiment 1 of the present invention and the prior art;

FIG. 5 is a schematic diagram of system capacity comparison between Embodiment 1 of the present invention and the prior art;

FIG. 6 is a structural diagram of a precoding device according to Embodiment 2 of the present invention.

Detailed ways

Embodiment 1 of the present invention provides a precoding method. In this embodiment and the following embodiments, description is made by taking the transmitting end provided with two transmitting antennas as an example.

In the MIMO system of this embodiment, the transmitting end is provided with two antennas, that is, the number of transmitting antennas N _t is 2, and the transmitting end can know all channel state information; the number of users K in the MIMO system is 2, and the receiving antennas of the users The number N _r is not limited and N _r can be greater than 2, and each user corresponds to a data stream. Under the above circumstances, the flow of the method is shown in FIG. 2 .

In FIG. 2, S200. The transmitting end such as the base station obtains channel information from two transmitting antennas to two users respectively, and obtains a channel correlation matrix according to the channel information.

其中，

为H₁的共轭转置矩阵，为H₂的共轭转置矩阵。If the channel information obtained by the transmitting end from the two transmitting antennas to the two users is denoted as H ₁ and H ₂ , then the channel correlation matrix obtained according to H ₁ and H ₂ is and

in,

is the conjugate transpose matrix of H ₁ , is the conjugate transpose matrix of _H2 .

S210. The transmitting end acquires the generalized eigenvector of the channel correlation matrix.

即根据公式

和计算并确定t₁和t₂。t₁和t₂具有对称性。Set the generalized eigenvectors as t ₁ and t ₂ , then t ₁ and t ₂ should satisfy respectively: and

i.e. according to the formula

and Calculate and determine t ₁ and t ₂ . t ₁ and t ₂ have symmetry.

S220. The transmitting end acquires the transmitting precoding vectors of the two users according to the generalized eigenvectors determined by the above calculation.

和

当两个用户的接收机采用

形式的接收向量进行接收时，可以实现两个用户间信号的相互干扰为0。If the transmitter regards t ₁ and t ₂ as the linear precoding vectors of two users, according to the characteristics of the generalized eigenvector, that is, at this time

and

When two user receivers employ

When the reception vector of the form is received, the mutual interference of signals between two users can be realized to be 0.

Considering the problem of system capacity, the transmitting end in this embodiment may adopt the system capacity maximization criterion to obtain the receiving vector and transmitting precoding vector of two users. For example, the transmitter selects the optimal transmit precoding vectors t _1,opt and t _2,opt for user 1 and user 2 respectively according to the following formula (1), so as to maximize the system capacity.

$\{t_{1,\mathrm{opt}},{\mathrm{<figure-callout\; id="2"\; label="t"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">t</figure-callout>}}_{2,\mathrm{opt}}\}=\arg \underset{t,\stackrel{\&OverBar;}{t}\&Element;{\mathrm{<figure-callout\; id="1"\; label="t"\; filenames="H2009101309948E00022.png"\; state="\{\{state\}\}">t</figure-callout>}}_1,{\mathrm{<figure-callout\; id="2"\; label="t"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">t</figure-callout>}}_2,t\&NotEqual;\stackrel{\&OverBar;}{t}}{\max }\{{\log }_2(1+\frac{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">P</figure-callout>}}{2{\mathrm{\&sigma;}}_1^2}{\left|h_{1}t\right|}^2)+{\log }_2(1+\frac{P}{2{\mathrm{\&sigma;}}_2^2}{\left|h_2\stackrel{\&OverBar;}{t}\right|}^2)\},\mathrm{the\; s}.t,\left|\right|t_1\left|\right|=1,\left|\right|t_2\left|\right|=1$ Formula 1)

为另一个用户的广义特征向量。In formula (1): P is the signal transmission power, H ₁ and H ₂ are the channel information from the two antennas to two users respectively, t is the generalized eigenvector of one of the users, and σ _i is the i-th user’s receiver noise variance,

is the generalized feature vector of another user.

S230. The transmitting end obtains receiving vectors of the two users according to the determined transmitting precoding vectors of the two users, and obtains equivalent channel matrices corresponding to the two users respectively according to the receiving vectors of the two users and the above channel information.

和

的特性，因而可消除用户间的干扰。In this embodiment, the reception matrix containing the reception vectors of two users is W, and the use of this reception matrix makes the whole system equivalent to the generalized eigenvector

and

characteristics, thus eliminating interference between users.

The reception matrix containing the reception vectors of the two users $W=\left(\begin{array}{cc}{\mathrm{<figure-callout\; id="1"\; label="w"\; filenames="H2009101309948E00022.png"\; state="\{\{state\}\}">w</figure-callout>}}_1& 0\\ 0& {\mathrm{<figure-callout\; id="2"\; label="w"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">w</figure-callout>}}_2\end{array}\right);$

Among them, w ₁ is the receiving vector of the first user, and w ₂ is the receiving vector of the second user.

The transmitter can obtain normalized receiving vectors w ₁ and w 2 by using t _1,opt and t _{2 ,opt} , for example, the receiving vector $w_{}$${}_1={\left(h_1{\mathrm{<figure-callout\; id="1"\; label="t"\; filenames="H2009101309948E00022.png"\; state="\{\{state\}\}">t</figure-callout>}}_{1,\mathrm{opt}}\right)}^h/{\left|\right|h_1{\mathrm{<figure-callout\; id="1"\; label="t"\; filenames="H2009101309948E00022.png"\; state="\{\{state\}\}">t</figure-callout>}}_{1,\mathrm{opt}}\left|\right|}_2^2,$ receive vector $w_{}$${}_2={\left(h_2{\mathrm{<figure-callout\; id="2"\; label="t"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">t</figure-callout>}}_{2,\mathrm{opt}}\right)}^h/{\left|\right|h_2{\mathrm{<figure-callout\; id="2"\; label="t"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">t</figure-callout>}}_{2,\mathrm{opt}}\left|\right|}_2^2;$

Among them, t _1,opt and t _2,opt are transmit precoding vectors of two users, H ₁ and H ₂ are channel information to two users respectively through two antennas at the transmitting end.

和

The transmitter can calculate the equivalent channel matrix of two users by using w ₁ and w ₂ according to the formula (2)

and

${\tilde{h}}_1=w_1{\mathrm{<figure-callout\; id="1"\; label="h"\; filenames="H2009101309948E00022.png"\; state="\{\{state\}\}">h</figure-callout>}}_1,$ ${\tilde{h}}_2=w_2{\mathrm{<figure-callout\; id="2"\; label="h"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">h</figure-callout>}}_2$ Formula (2)

The equivalent channel matrix of the above two users can be expressed as $h_{\mathrm{eff}}=\left[\begin{array}{c}{\tilde{h}}_1\\ {\tilde{h}}_2\end{array}\right].$

It can be seen from the above description that the order of the equivalent channel matrix of the two users is reduced due to the use of the receiving vector in the process of determining the equivalent channel matrix, and the order of the equivalent channel matrix is only equal to the number of antennas at the transmitting end It is related to the number of users, but has nothing to do with the number of receiving antennas of users, so that the complexity of THP has nothing to do with the number of receiving antennas of users, and finally the number of receiving antennas of users can be unlimited.

S240. The transmitting end obtains the power scaling matrix G, the feedback matrix B, and the feedforward matrix F corresponding to the two users respectively according to the equivalent channel matrices corresponding to the two users, that is, using the equivalent channel $h_{\mathrm{eff}}=\left[\begin{array}{c}{\tilde{h}}_1\\ {\tilde{h}}_2\end{array}\right]$ Calculate F, B, and G corresponding to two users respectively.

The process of obtaining F, B and G by using the equivalent channel matrix is: $h_{\mathrm{eff}}=\left[\begin{array}{c}{\tilde{h}}_1\\ {\tilde{h}}_2\end{array}\right]$ Orthogonal triangular decomposition such as QR decomposition of the conjugate transpose of is obtained: Therefore, F can be obtained, that is, F is a unitary matrix obtained by performing orthogonal triangular decomposition on the conjugate transposition of the equivalent channel matrix. The above ^SH is a triangular matrix obtained by performing a conjugate transpose operation on the triangular matrix S obtained by performing orthogonal triangular decomposition on the conjugate transpose of the equivalent channel matrix.

获得，其中，s₁₁,...，s_KK为等效信道矩阵 $H_{\mathrm{eff}}=\left[\begin{array}{c}{\tilde{H}}_1\\ {\tilde{H}}_2\end{array}\right]$ 进行正交三角分解后得到的三角矩阵S的对角线上的元素，K为用户数。The power scaling matrix can be obtained by the formula

Obtained, where, s ₁₁ ,..., s _KK is the equivalent channel matrix $h_{\mathrm{eff}}=\left[\begin{array}{c}{\tilde{h}}_1\\ {\tilde{h}}_2\end{array}\right]$ Elements on the diagonal of the triangular matrix S obtained after orthogonal triangular decomposition, K is the number of users.

The feedback matrix B can be obtained by the formula B=GS, where G is a power scaling matrix, and S is a triangular matrix obtained by performing orthogonal triangular decomposition on the equivalent channel matrix.

It should be noted that, according to different orthogonal triangular decomposition forms used for the equivalent channel matrix, the above triangular matrix S may be an upper triangular matrix or a lower triangular matrix.

S250. The transmitting end performs THP for a system composed of two users according to the above reception vector, the power scaling matrix G, the feedback matrix B, and the feedforward matrix F corresponding to the two users respectively.

其中的a为发射信号向量，d为扰动向量，d使得B^-1(a+d)等价为B^-1a的求模运算。The transmitter can perform THP for a system composed of two users according to the system model shown in Figure 3, and the

Wherein, a is a transmitting signal vector, d is a disturbance vector, and d makes B ^-1 (a+d) equivalent to a modulo operation of B ^-1 a.

The relationship between the sending and receiving of the system model shown in Figure 3 can be expressed as: y=GWHFB ^-1 v+GWn, that is, the transmitting end performs THP for the system composed of two users according to the formula y=GWHFB ^-1 v+GWn;

E[]表示期望，

表示高斯白噪声的方差，I表示单位阵。Among them, y is the transmit vector, G is the power scaling matrix, W is the user’s receive vector matrix, H is the user’s channel information, v is the transmit signal vector, n is the noise vector on the user’s receive antenna, each component of n independent of each other, and

E[] means expectation,

Represents the variance of Gaussian white noise, and I represents the identity matrix.

The system model shown in FIG. 3 can be designed using Zero-Forcing Criterion (ZFC), that is, GWHFB ⁻¹ =I. I stands for identity matrix.

It should be emphasized that this embodiment is not limited to using the ZFC criterion in THP, and this embodiment can support THP based on other criteria, for example, this embodiment can support THP under the MMSE (Minimum Mean Square Error) criterion.

It can be obtained through experimental data: when the number of user receiving antennas is different, the BER (Bit Error Rate) performance obtained by adopting the technical solution of Embodiment 1 is shown in FIG. 4 .

In Fig. 4, the horizontal coordinate represents SNR (signal-to-noise ratio), and the vertical coordinate represents the bit error rate. The circled curve is the BER performance curve when a user has one receiving antenna, and the crossed curve is the BER performance curve when a user has two receiving antennas.

It can be seen from the curve shown in FIG. 4 that the bit error rate can be greatly improved by adopting the first embodiment, and the diversity gain in multi-antenna reception can be obtained.

In addition, it can be obtained through experimental data: the comparison between the technical solution of the first embodiment and the existing DPC technical solution in terms of system capacity is shown in Fig. 5 .

In Fig. 5, the horizontal coordinate represents the transmission power, and the vertical coordinate represents the sum rate. The curve with circles is the system capacity curve of the DPC technical solution, and the curve with crosses is the system capacity curve of the technical solution of the first embodiment above.

It can be seen from the two curves shown in FIG. 5 that the system capacity of the first embodiment is very close to the system capacity using the DPC technology.

Embodiment 2 of the present invention provides a precoding device. The transmitting end where the device is located is provided with two antennas. The structure of this device is shown in accompanying drawing 6.

The apparatus 600 in FIG. 6 includes: a channel correlation matrix module 601 , an eigenvector module 602 , an equivalent channel matrix module 603 and a precoding module 604 .

The channel correlation matrix module 301 is used to obtain channel information from two antennas of the transmitting end to two users respectively, and obtain a channel correlation matrix according to the channel information.

和

其中，为H₁的共轭转置矩阵，

为H₂的共轭转置矩阵。If the channel information obtained by the channel correlation matrix module 301 from the two transmitting antennas to the two users is denoted as H ₁ and H ₂ , the channel correlation matrix obtained by the channel correlation matrix module 301 according to H ₁ and H ₂ is

and

in, is the conjugate transpose matrix of H ₁ ,

is the conjugate transpose matrix of _H2 .

The eigenvector module 602 is used to obtain the generalized eigenvector of the channel correlation matrix.

Set the generalized eigenvectors obtained by the eigenvector module 602 as t ₁ and t ₂ , then t ₁ and t ₂ should satisfy respectively: $t_2^h{h}_1^h{h}_1{\mathrm{<figure-callout\; id="1"\; label="t"\; filenames="H2009101309948E00022.png"\; state="\{\{state\}\}">t</figure-callout>}}_1=0$ and $t_2^h{h}_2^h{h}_2{\mathrm{<figure-callout\; id="1"\; label="t"\; filenames="H2009101309948E00022.png"\; state="\{\{state\}\}">t</figure-callout>}}_1=0,$ That is, the feature vector module 602 according to the formula $t_2^h{h}_1^h{h}_1{\mathrm{<figure-callout\; id="1"\; label="t"\; filenames="H2009101309948E00022.png"\; state="\{\{state\}\}">t</figure-callout>}}_1=0$ and $t_2^h{h}_2^h{h}_2{\mathrm{<figure-callout\; id="1"\; label="t"\; filenames="H2009101309948E00022.png"\; state="\{\{state\}\}">t</figure-callout>}}_1=0$ Calculate and determine t ₁ and t ₂ , t ₁ and t ₂ are symmetrical.

The equivalent channel matrix module 603 is used to obtain the transmit precoding vectors of the two users according to the above-mentioned generalized eigenvectors, obtain the receiving vectors of the two users according to the transmitting precoding vectors of the two users, and obtain the receiving vectors of the two users according to the receiving vectors of the two users and the channel The information acquires equivalent channel matrices corresponding to the two users respectively.

Considering the problem of system capacity, the equivalent channel matrix module 603 in this embodiment may adopt the system capacity maximization criterion in the process of acquiring the transmit precoding vectors of two users. For example, the equivalent channel matrix module 603 selects optimal transmit precoding vectors t _1,opt and t _2,opt for user 1 and user 2 respectively according to the following formula (1), so as to maximize the system capacity.

$\{{\mathrm{<figure-callout\; id="1"\; label="t"\; filenames="H2009101309948E00022.png"\; state="\{\{state\}\}">t</figure-callout>}}_{1,\mathrm{opt}},{\mathrm{<figure-callout\; id="2"\; label="t"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">t</figure-callout>}}_{2,\mathrm{opt}}\}=\arg \underset{t,\stackrel{\&OverBar;}{t}\&Element;{\mathrm{<figure-callout\; id="1"\; label="t"\; filenames="H2009101309948E00022.png"\; state="\{\{state\}\}">t</figure-callout>}}_1,{\mathrm{<figure-callout\; id="2"\; label="t"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">t</figure-callout>}}_2,t\&NotEqual;\stackrel{\&OverBar;}{t}}{\max }\{{\log }_2(1+\frac{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">P</figure-callout>}}{2{\mathrm{\&sigma;}}_1^2}{\left|h_{1}t\right|}^2)+{\log }_2(1+\frac{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">P</figure-callout>}}{2{\mathrm{\&sigma;}}_2^2}{\left|h_2\stackrel{\&OverBar;}{t}\right|}^2)\},\mathrm{the\; s}.t,\left|\right|{\mathrm{<figure-callout\; id="1"\; label="t"\; filenames="H2009101309948E00022.png"\; state="\{\{state\}\}">t</figure-callout>}}_1\left|\right|=1,\left|\right|{\mathrm{<figure-callout\; id="2"\; label="t"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">t</figure-callout>}}_2\left|\right|=1$ Formula 1)

In formula (1): P is the signal transmission power, H ₁ and H ₂ are the channel information from the two antennas to two users respectively, t is the generalized eigentransmission precoding vector of the first user, σ _i is the receiver noise variance of the i-th user, The generalized feature precoding vector for the transmission of another second user, st (subject to, constrained by), expresses the formula

$\{{\mathrm{<figure-callout\; id="1"\; label="t"\; filenames="H2009101309948E00022.png"\; state="\{\{state\}\}">t</figure-callout>}}_{1,\mathrm{opt}},{\mathrm{<figure-callout\; id="2"\; label="t"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">t</figure-callout>}}_{2,\mathrm{opt}}\}=\arg \underset{t,\stackrel{\&OverBar;}{t}\&Element;{\mathrm{<figure-callout\; id="1"\; label="t"\; filenames="H2009101309948E00022.png"\; state="\{\{state\}\}">t</figure-callout>}}_1,{\mathrm{<figure-callout\; id="2"\; label="t"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">t</figure-callout>}}_2,t\&NotEqual;\stackrel{\&OverBar;}{t}}{\max }\{{\log }_2(1+\frac{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">P</figure-callout>}}{2{\mathrm{\&sigma;}}_1^2}{\left|h_{1}t\right|}^2)+{\log }_2(1+\frac{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">P</figure-callout>}}{2{\mathrm{\&sigma;}}_2^2}{\left|h_2\stackrel{\&OverBar;}{t}\right|}^2)\}$ Subject to ∥t ₁ ∥=1 and ∥t ₂ ∥=1.

The receiving matrix obtained by the equivalent channel matrix module 603 and composed of receiving vectors of two users can be expressed as:

$\mathrm{<span\; class="notranslate">\; W</span>}<span\; class="notranslate">\; =</span>\left(\begin{array}{cc}{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="1"\; label="w"\; filenames="H2009101309948E00022.png"\; state="\{\{state\}\}">w</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& {\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="w"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">w</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}\end{array}\right)<span\; class="notranslate">\; ;</span>$

Among them, w ₁ is the receiving vector of the first user, and w ₂ is the receiving vector of the second user.

The equivalent channel matrix module 603 can obtain normalized reception vectors w ₁ and w 2 by using t _1,opt and t _{2 ,opt} , for example, the reception vector $w_{}$${}_1={\left(h_1{\mathrm{<figure-callout\; id="1"\; label="t"\; filenames="H2009101309948E00022.png"\; state="\{\{state\}\}">t</figure-callout>}}_{1,\mathrm{opt}}\right)}^h/{\left|\right|h_1{\mathrm{<figure-callout\; id="1"\; label="t"\; filenames="H2009101309948E00022.png"\; state="\{\{state\}\}">t</figure-callout>}}_{1,\mathrm{opt}}\left|\right|}_2^2,$ receive vector $w_{}$${}_2={\left(h_2{\mathrm{<figure-callout\; id="2"\; label="t"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">t</figure-callout>}}_{2,\mathrm{opt}}\right)}^h/{\left|\right|h_2{\mathrm{<figure-callout\; id="2"\; label="t"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">t</figure-callout>}}_{2,\mathrm{opt}}\left|\right|}_2^2;$

Among them, t _1,opt and t _2,opt are transmit precoding vectors of two users, and H ₁ and H ₂ are channel information from two antennas at the transmitting end to two users respectively.

和 The equivalent channel matrix module 603 can calculate the equivalent channel matrix of two users by using w ₁ and w ₂ according to formula (2)

and

${\tilde{h}}_1=w_1{\mathrm{<figure-callout\; id="1"\; label="h"\; filenames="H2009101309948E00022.png"\; state="\{\{state\}\}">h</figure-callout>}}_1,$ ${\tilde{h}}_2=w_2{\mathrm{<figure-callout\; id="2"\; label="h"\; filenames="H2009101309948E00022.png,H2009101309948E00031.png"\; state="\{\{state\}\}">h</figure-callout>}}_2$ Formula (2)

The equivalent channel matrix of the above two users can be expressed as $h_{\mathrm{eff}}=\left[\begin{array}{c}{\tilde{h}}_1\\ {\tilde{h}}_2\end{array}\right].$

The precoding module 604 is used to obtain the power scaling matrix G, the feedback matrix B, and the feedforward matrix F corresponding to the two users respectively according to the equivalent channel matrices corresponding to the two users, and according to the above-mentioned reception vector, the corresponding The power scaling matrix G, the feedback matrix B and the feedforward matrix F perform THP for a system composed of two users.

The process of the precoding module 604 using the equivalent channel matrix to obtain F, B and G can be:

从而预编码模块604可以获得F，即F为对等效信道矩阵进行正交三角分解后获得的酉矩阵。上述S^H为对等效信道矩阵进行正交三角分解后获得的三角矩阵S进行共轭转置运算后获得的三角矩阵。The precoding module 604 pairs $h_{\mathrm{eff}}=\left[\begin{array}{c}{\tilde{h}}_1\\ {\tilde{h}}_2\end{array}\right]$ Orthogonal triangular decomposition of the conjugate transpose of to get:

Therefore, the precoding module 604 can obtain F, that is, F is a unitary matrix obtained by performing orthogonal triangular decomposition on the equivalent channel matrix. The above ^SH is a triangular matrix obtained by performing a conjugate transpose operation on the triangular matrix S obtained by performing orthogonal triangular decomposition on the equivalent channel matrix.

获得功率缩放矩阵，其中，s₁₁,...，s_KK为等效信道矩阵 $H_{\mathrm{eff}}=\left[\begin{array}{c}{\tilde{H}}_1\\ {\tilde{H}}_2\end{array}\right]$ 进行正交三角分解后得到的三角矩阵S的对角线上的元素，K为用户数。The precoding module 604 can use the formula

Obtain the power scaling matrix, where, s ₁₁ ,..., s _KK is the equivalent channel matrix $h_{\mathrm{eff}}=\left[\begin{array}{c}{\tilde{h}}_1\\ {\tilde{h}}_2\end{array}\right]$ Elements on the diagonal of the triangular matrix S obtained after orthogonal triangular decomposition, K is the number of users.

The precoding module 604 can use the formula B=GS to obtain the feedback matrix B, where G is the power scaling matrix, and S is the triangular matrix obtained by performing orthogonal triangular decomposition on the equivalent channel matrix.

It should be noted that, according to the different orthogonal triangular decomposition forms used by the precoding module 604 for the equivalent channel matrix, the above triangular matrix S may be an upper triangular matrix or a lower triangular matrix.

其中的a为发射信号向量，d为扰动向量，d使得B^-1(a+d)等价为B^-1a的求模运算。After the precoding module 604 obtains the power scaling matrix G, the feedback matrix B and the feedforward matrix, it can perform THP for the system composed of two users according to the system model shown in FIG. 3 .

Wherein, a is a transmitting signal vector, d is a disturbance vector, and d makes B ^-1 (a+d) equivalent to a modulo operation of B ^-1 a.

Since the interrelationship between the transceivers of the system model shown in Figure 3 can be expressed as the following form:

y=GWHFB ^-1 v+GWn;

E[ ]表示期望，

表示高斯白噪声的方差，I表示单位阵。Among them, y is the transmit vector, G is the power scaling matrix, W is the user’s receive vector matrix, H is the user’s channel information, v is the transmit signal vector, n is the noise vector on the user’s receive antenna, each component of n independent of each other, and

E[ ] means expectation,

Represents the variance of Gaussian white noise, and I represents the identity matrix.

Therefore, the precoding module 604 can perform THP for the system composed of two users according to the formula y=GWHFB ^-1 v+GWn.

It can be seen from the description of the second embodiment above that, since the equivalent channel matrix module 603 utilizes the receiving vector in the process of determining the equivalent channel matrix, the orders of the equivalent channel matrices of the two users are reduced, and the equivalent channel The order of the matrix is only related to the number of antennas at the transmitting end and the number of users, but has nothing to do with the number of receiving antennas of the users. Therefore, the complexity of the THP operation performed by the precoding module 604 has nothing to do with the number of receiving antennas of the users. Finally, the receiving antennas of the users The quantity can be unlimited.

Through the description of the above embodiments, those skilled in the art can clearly understand that the present invention can be realized by means of software plus a necessary hardware platform, and of course all can be implemented by hardware, but in many cases the former is better implementation. Based on this understanding, all or part of the contribution made by the technical solution of the present invention to the background technology can be embodied in the form of software products, and the computer software products can be stored in storage media, such as ROM/RAM, magnetic disks, optical disks, etc. , including several instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute the methods described in various embodiments or some parts of the embodiments of the present invention.

Although the present invention has been described by way of example, those of ordinary skill in the art know that there are many variations and changes in the present invention without departing from the spirit of the invention, and the claims of the application document of the present invention include these variations and changes.

Claims (7)

1. A precoding method, comprising:

a transmitting terminal acquires channel information of a plurality of antennas of the transmitting terminal from two users respectively, and acquires a channel correlation matrix according to the channel information;

the transmitting terminal obtains generalized eigenvectors of the channel correlation matrix, wherein the generalized eigenvectors are t respectively₁And t₂And t is₁And t₂Respectively satisfy: $t_2^H{H}_1^H{H}_1{t}_1=0$ and $t_2^H{H}_2^H{H}_2{t}_1=0;$ wherein,

H₁and

H₂is a channel correlation matrix, H₁And H₂Channel information to two users for the multiple antennas respectively,

is H₁The conjugate transpose matrix of (a) is,

is H₂The conjugate transpose matrix of (a);

the transmitting terminal determines the transmitting pre-coding vectors t of the two users according to the generalized characteristic vector and by utilizing the system capacity maximization criterion_1,optAnd t_2,optAcquiring normalized receiving vectors of the two users according to the transmitting precoding vectors of the two users, and multiplying the normalized receiving vectors of the two users with the channel information respectively to acquire equivalent channel matrixes corresponding to the two users respectively; wherein the transmitting end obtains t according to the following formula_1,optAnd t_2,opt：

$\{t_{1,\mathrm{opt}},t_{2,\mathrm{opt}}\}=\arg \underset{t,\stackrel{-}{t}\&Element;t_1,t_2,t\&NotEqual;\stackrel{-}{t}}{\max }\{{\log }_2(1+\frac{P}{{2\mathrm{\&sigma;}}_1^2}{\left|H_{1}t\right|}^2)+{\log }_2(1+\frac{P}{{2\mathrm{\&sigma;}}_2^2}{\left|H_2\stackrel{-}{t}\right|}^2)\},s.t\left|\right|t_1|=1,|\left|t_2\right||=1;$

Wherein: p is signal transmitting power, t is generalized eigenvector of 1 user in the signal transmitting power, and sigma_iFor the receiver noise variance of the ith user,

a generalized eigenvector for another user;

and the transmitting terminal acquires a power scaling matrix, a feedback matrix and a feedforward matrix respectively corresponding to the two users according to the equivalent channel matrixes respectively corresponding to the two users, and performs Thomlinson-Harashima precoding (THP) on a system consisting of the two users according to the receiving vector, the power scaling matrix, the feedback matrix and the feedforward matrix.

2. The method of claim 1, wherein a receiving matrix composed of the receiving vectors of the two users is W, and the W is obtained by the following formula:

$W=\left(\begin{array}{cc}{w}_1& 0\\ 0& w_2\end{array}\right);$

wherein, the 1 st user normalized receiving vector

2 nd user normalized received vector $w_2={\left(H_2{t}_{2,\mathrm{opt}}\right)}^H/{\left|\right|H_2{t}_{2,\mathrm{opt}}\left|\right|}_2^2.$

3. The method of claim 1, wherein the multiplying the received vectors of the two users by the channel information to obtain equivalent channel matrices corresponding to the two users respectively comprises:

according to the formula $H_{\mathrm{eff}}=\left[\begin{array}{c}{\tilde{H}}_1\\ {\tilde{H}}_2\end{array}\right]$ Acquiring equivalent channel matrixes corresponding to two users;

wherein,

w₁received vector for the 1 st user, w₂Received vector for the 2 nd user, H₁And H₂Channel information to two users for the plurality of antennas respectively.

4. The method of claim 1, wherein the power scaling matrix G is formulated by the formula

Obtaining wherein s₁₁,...,s_KKFor the equivalent channel matrix $H_{\mathrm{eff}}=\left[\begin{array}{c}{\tilde{H}}_1\\ {\tilde{H}}_2\end{array}\right]$ Go on toThe element on the diagonal line of the triangular matrix S obtained after the triangular decomposition, K is the number of users,w₁received vector for the 1 st user, w₂Received vector for the 2 nd user, H₁And H₂Channel information to two users for the plurality of antennas respectively.

5. The method of claim 1, wherein the feedback matrix B is obtained by a formula B ═ GS, where G is the power scaling matrix and S is a triangular matrix obtained by orthogonal triangular decomposition of an equivalent channel matrix; or

The feedforward matrix F is a unitary matrix obtained by performing orthogonal triangular decomposition on the equivalent channel matrix.

6. The method of any one of claims 1 to 5, wherein said performing THP for a system of two users according to said receive vector, power scaling matrix, feedback matrix and feedforward matrix comprises:

according to y ═ GWFFB^-1v + Gwn is THP carried out by a system consisting of two users;

y is a transmitting vector, G is a power scaling matrix, W is a receiving vector matrix of a user, H is channel information of the user, F is a feedforward matrix, B is a feedback matrix, v is a transmitting signal vector, and n is a noise vector on a receiving antenna of the user.

7. A precoding apparatus having a plurality of antennas provided in a transmitting end, the apparatus comprising:

the channel correlation matrix module is used for acquiring channel information from the plurality of antennas of the transmitting end to two users respectively and acquiring a channel correlation matrix according to the channel information;

an eigenvector module for obtaining generalized eigenvectors of the channel correlation matrixThe generalized eigenvectors are respectively t₁And t₂And t is₁And t₂Respectively satisfy: $t_2^H{H}_1^H{H}_1{t}_1=0$ and $t_2^H{H}_2^H{H}_2{t}_1=0;$ wherein,

H₁and

H₂is a channel correlation matrix, H₁And H₂Channel information to two users for the multiple antennas respectively,

is H₁The conjugate transpose matrix of (a) is,

is H₂The conjugate transpose matrix of (a);

an equivalent channel matrix module for obtaining the transmitting pre-coding vectors t of the two users according to the generalized eigenvector and by using the system capacity maximization criterion_1,optAnd t_2,optObtaining normalized receiving vectors of the two users according to the transmitting pre-coding vectors of the two users, and respectively combining the two usersMultiplying the receiving vector of user normalization by the channel information to obtain equivalent channel matrixes respectively corresponding to the two users; wherein the transmitting end obtains t according to the following formula_1,optAnd t_2,opt：

$\{t_{1,\mathrm{opt}},t_{2,\mathrm{opt}}\}=\arg \underset{t,\stackrel{-}{t}\&Element;t_1,t_2,t\&NotEqual;\stackrel{-}{t}}{\max }\{{\log }_2(1+\frac{P}{{2\mathrm{\&sigma;}}_1^2}{\left|H_{1}t\right|}^2)+{\log }_2(1+\frac{P}{{2\mathrm{\&sigma;}}_2^2}{\left|H_2\stackrel{-}{t}\right|}^2)\},s.t\left|\right|t_1|=1,|\left|t_2\right||=1;$

Wherein: p is signal transmitting power, t is generalized eigenvector of 1 user in the signal transmitting power, and sigma_iFor the receiver noise variance of the ith user,

a generalized eigenvector for another user;

and the precoding module is used for acquiring a power scaling matrix, a feedback matrix and a feedforward matrix which respectively correspond to the two users according to the equivalent channel matrix, and carrying out THP (total harmonic distortion) on a system consisting of the two users according to the receiving vector, the power scaling matrix, the feedback matrix and the feedforward matrix.

Images