WO2008083576A1

WO2008083576A1 - Communication method, transmission method, reception method and device thereof

Info

Publication number: WO2008083576A1
Application number: PCT/CN2007/071287
Authority: WO
Inventors: Jianguo Liu; Lvxi Yang; Lixia Xue; Bin Li
Original assignee: Huawei Technologies Co., Ltd.
Priority date: 2006-12-31
Filing date: 2007-12-20
Publication date: 2008-07-17
Also published as: CN101212281A; CN101212281B

Abstract

A communication method, transmission method, reception method and device thereof. A receiving end selects an optimum pre-coding matrix according to current channel state, and feeds back information delegating the optimum pre-coding matrix to a sending end; the sending end performs orthogonal space-time coding for the information needing to be transmitted, and sends it after multiplying by the optimum pre-coding matrix delegated by the information which is fed back by the receiving end; the receiving end multiplies the received signal by the conjugate transposition of the pre-coding matrix used by the sending end, and then obtains the information needing to be transmitted by the sending end after performing orthogonal space-time decoding.

Description

Communication method, transmission method, reception method and device

This application claims priority to Chinese Patent Application No. 200610063878.5, entitled "Communication Method and Equipment Based on Multiple Input Multiple Output System", filed on December 31, 2006, the entire contents of which are incorporated by reference. In this application.

Technical field

The present invention relates to the field of wireless communications, and in particular, to a communication method, a transmitting method, a receiving method, a transmitting device, and a receiving device.

Background technique

Multiple Input Multiple Output (MIMO) technology achieves high data rates, high system capacity, and high transmission in the third generation of mobile communications (The Third Generation, referred to as "3G") and future mobile communication systems. An important way of quality. At present, space-time transmit diversity technology has become one of the key technologies of 3G (Beyond 3G, referred to as "B3G") / 4G. Transmit diversity can improve system capacity and organically combine code modulation diversity technology.

In order to improve the performance of MIMO systems, there are currently two technical solutions.

One scheme uses orthogonal space-time coding, and orthogonal space-time coding is an open-loop transmit diversity technique in MIMO systems. Because it can effectively resist fading and has very low decoding complexity, it has gained widespread attention. One of the most practical space-time transmit diversity techniques. However, as an open-loop transmit diversity technique, its performance is degraded due to the lack of channel state information at low speeds.

Another solution uses closed-loop diversity techniques. In the Frequency Division Duplex (FDD) mode, if all users feed back the Channel State Information (CSI) to the base station, this mechanism is too expensive when the number of users is large. It is also not realistic in wireless communication. Orthogonal space-time coding systems often use linear precoding to obtain feedback gain. In the actual point-to-point MIMO system precoding design, in order to reduce the amount of feedback information, the codebook method is often used. Based on the codebook design scheme of the orthogonal space-time coding system, a larger array gain is obtained with less feedback information. The scheme maintains an identical precoding codebook (Codebook) at both the transmitting end and the receiving end, based on The instantaneous channel state receiving end selects an optimal codeword matrix from the finite codebook set, and sends the codeword sequence number to the transmitting end, and the transmitting end uses the codeword matrix of the sequence number as the transmitting precoding matrix, and the orthogonal space time The coded codeword matrix is multiplied and sent out.

The problem with closed-loop diversity techniques is poor performance at high speeds. On the one hand, the feedback information itself cannot be guaranteed to be correctly transmitted to the transmitting end. On the other hand, the channel change is too fast, so that the feedback information is invalidated, thereby degrading the high-speed performance.

Another problem with closed-loop diversity techniques is how to design a precoding codebook and how to choose the optimal codeword matrix from the codebook. There is no better method in the current publication.

Summary of the invention

Embodiments of the present invention provide a communication method, a transmission method, a reception method, and a device, so that a MIMO system can have satisfactory performance at both high speed and low speed.

An embodiment of the present invention discloses a communication method, in which a receiving end selects an optimal precoding matrix according to a current channel state, and feeds back information about the optimal precoding matrix to a transmitting end; the transmitting end performs information to be transmitted. Orthogonal space-time coding, and then multiplying by the optimal precoding matrix represented by the information fed back by the receiving end, and transmitting; the receiving end multiplies the received signal by the optimal precoding matrix used by the transmitting end. After the transposition is performed, the information to be transmitted by the transmitting end is obtained after performing orthogonal space-time decoding.

The embodiment of the present invention further discloses a receiving device, including: selecting a unit of an optimal precoding matrix according to a current channel state; and feeding back information of the optimal precoding matrix to a sender a unit that multiplies a signal received from the transmitting end by a unit of a common precoding matrix of the optimal precoding matrix used by the transmitting end; and a transmutation transpose of the optimal precoding matrix used by the transmitting end The subsequent signal is subjected to orthogonal space-time decoding. The embodiment of the present invention further discloses a transmitting device, including: obtaining, by a receiving end, a unit that represents optimal precoding matrix information; and performing orthogonal space time coding on the information that needs to be transmitted; The time-coded signal is multiplied by the unit transmitted after the optimal precoding matrix represented by the information fed back by the receiving end. An embodiment of the present invention further discloses a receiving method, selecting an optimal precoding matrix according to a current channel state, and feeding back information of the optimal precoding matrix to a transmitting end; multiplying a signal received from the transmitting end Performing a common 扼 transposition of the optimal precoding matrix used by the transmitting end; performing orthogonal space-time decoding on the conjugated transposed signal multiplied by the optimal precoding matrix used by the transmitting end to obtain the transmitting The information that needs to be transmitted at the end.

The embodiment of the present invention further discloses a sending method, which obtains information representing the optimal precoding matrix fed back by the receiving end, performs orthogonal space time coding on the information to be transmitted, and multiplies the signal encoded by the orthogonal space time. The optimal precoding matrix represented by the information fed back by the receiving end is sent. The embodiment of the present invention proposes a closed-loop orthogonal space-time coding scheme, that is, the receiving end selects a precoding matrix according to the current channel state and feeds relevant information back to the transmitting end, and the transmitting end multiplies the signal encoded by the orthogonal space-time multiplication by the receiving end. The selected precoding matrix is sent after. Robust for channel fading conditions. At low speeds, because of the closed-loop CSI feedback, the gain of the system diversity is improved while the transmit diversity gain is obtained with less feedback information. Therefore, the MIMO system has Better transmission performance; At high speed, although the correctness of the feedback information cannot be guaranteed, because orthogonal space-time coding is used, the system maintains the open-loop orthogonal space-time coded transmit diversity gain, so the MIMO system remains at high speed. Have better performance. DRAWINGS

1 is a flow chart of a communication method based on a MIMO system according to a preferred embodiment of the present invention; FIG. 2 is a schematic diagram of a MIMO system model according to a preferred embodiment of the present invention; FIG. 3 is a case where the codebook length and the moving speed are different. Simulation results of embodiments of the present invention; FIG. 4 is a simulation result of an embodiment of the present invention in the case where the codebook lengths are different.

detailed description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail with reference to the accompanying drawings. In a preferred embodiment of the present invention, a communication method based on a MIMO system is shown in FIG. In step 110, the receiving end selects an optimal precoding matrix according to the current channel state. When receiving the signal, the receiving end can estimate the current channel state according to the receiving condition of the signal.

The same codebook can be saved on both the receiving end and the transmitting end, and the precoding matrix is included, and the receiving end selects the precoding matrix that best matches the current channel state from the codebook as the optimal precoding matrix.

The optimal precoding matrix is a precoding matrix with a maximum of |H in the codebook, where H is a channel matrix, F is a precoding matrix in the codebook, and | | represents a 2 norm. The specific selection method of the optimal precoding matrix and the method of generating the codebook will be described in detail later.

Thereafter, proceeding to step 120, the receiving end feeds back information representing the optimal precoding matrix to the transmitting end. In this embodiment, the information representing the precoding matrix is the sequence number of the precoding matrix in the codebook. By using a codebook, the amount of information that needs to be fed back can be greatly reduced.

Of course, the present invention may also use the codebook without using the codebook, but adopt other methods to select the optimal precoding matrix and feed back related information, as long as the receiving end can uniquely determine the precoding matrix consistent with the transmitting end. For example, you can feed back all the information of the compressed optimal precoding matrix, or find it in the codebook first. The codeword closest to the optimal precoding matrix, the sequence number of the codeword and the difference between the optimal precoding matrix and the codeword are fed back in a compressed form.

Thereafter, proceeding to step 130, the transmitting end performs orthogonal space time coding on the information to be transmitted.

Thereafter, the process proceeds to step 140, where the transmitting end multiplies the signal encoded by the orthogonal space-time encoding by the precoding matrix represented by the information fed back by the receiving end, and then transmits. The sending device saves the same codebook as the receiving end, and the transmitting end finds the corresponding precoding matrix in the codebook according to the sequence number fed back by the receiving end.

Thereafter, proceeding to step 150, the receiving end multiplies the received signal by the common transposition of the precoding matrix used by the transmitting end. The precoding matrix used by the transmitting end here is the optimal precoding matrix that the receiving end feeds back to the transmitting end in step 120.

Thereafter, the process proceeds to step 160 where the receiving end performs orthogonal space-time decoding.

It should be noted that steps 110, 120 and steps 150, 160 may be triggered by the same signal. After receiving the signal, the receiving end decodes the signal through steps 150 and 160, and on the other hand, feedbacks to the transmitting end through steps 110 and 120, and the feedback information is used by the transmitting end for subsequent information transmission.

In Fig. 2, a system model of the above flow is illustrated by taking a base station transmitting information to a mobile station as an example.

Considering the downlink of the MIMO system using orthogonal space-time coding, the base station is configured with a root transmit antenna, and the UE is configured with an N antenna. It is assumed that there are abundant scatterers around the transmitting antenna and the receiving antenna, and the channel coefficient _m (the response from the second transmitting antenna to the channel between the "root receiving antennas") obeys the Gaussian distribution CN(o, i). The quasi-static is maintained for r time slot intervals. Therefore, the user's equivalent baseband received signal model in the r time slots can be expressed as:

Where y is the received signal matrix of N _r x T dimension, the total power of the transmitting antenna of the user, NN, the channel matrix H of the dimension is composed of complex Gaussian random variables, which is a Gaussian white noise matrix of N _f xr dimension, whose elements are independent The same distribution of (0,1) distributed random variables, 7^ is the precoding matrix of the dimension; (: is the transmission code symbol matrix of the user's transmitted symbols after orthogonal space-time coding, g from the constellation S Independent modulation symbols = , ... composition, each symbol obeys the power constraint ^ sV ] = l, and

CC ^H = I _M , where is the expected operator, which is a conjugate transpose operation. M is the data stream

The number.

Under the condition that the base station already knows the user channel state information, the precoding matrix F in equation (1) can be designed and obtained by appropriate optimization criteria, but when the system adopts the Frequency Division Duplex (FDD) mode. The user needs to feed back all the CSI to the base station. When the number of system users is relatively large, the mechanism is too expensive, and in the wireless communication system, it is not realistic for the base station to acquire all the CSIs of all users. In view of the above situation, in order to reduce the amount of feedback information, in the actual point-to-point MIMO system, the codebook is often used for precoding design.

How to select the optimal precoding codeword is explained in detail below.

It is assumed that the error-containing codeword matrix received by the user of the MIMO wireless communication system based on orthogonal space-time coding is that the channel matrix is H, and the precoding matrix is the transmitted codeword matrix (:, under the condition of orthogonal space time) The conditional error probability upper bound of the maximum likelihood estimation sequence codeword of the coding system can be expressed as:

Where | | represents 2 norm, y is a function related to the data stream size M, the average signal-to-noise ratio of the receiving antenna per user, and the constellation S, which remains unchanged during a transmission, so orthogonal space The conditional error probability of the coded codeword is related to the transmit diversity gain Hif, and the conditional error probability of the codeword decreases exponentially with the increase of Hi3⁄4. In order to minimize the conditional error probability of the codeword in one transmission process, the coding matrix F should be designed to maximize Hi3⁄4, so the optimal precoding matrix is determined by the following constraints:

/(H) = argm _aX |Hi3⁄4 (3)

F

Under the condition that the base station knows the user channel state information, assuming H = UV ^H , the front queue of Ϋ ,, the matrix optimal precoding matrix is F _opt = V. In practical systems, the codebook is often used to design the precoding matrix. The optimal precoding matrix F selects, from the given codebook Ω, the codeword satisfying the condition (3) as its precoding matrix, and feeds back the sequence number of the selected codeword in the codebook to the base station, and the base station The codeword matrix of the sequence number is transmitted as a transmit precoding matrix, multiplied by the code matrix of the orthogonal space-time coded, and Τ^Ω Α, ..., ^ satisfies ^^=^, which is the codebook length.

The following example shows how to construct a pre-coded codebook for orthogonal space-time coding:

Define the codebook space Ω element F _t , Fj chord distance d person Fi'Fj) is:

-F _jF fl = _t r(l _M -F^ ) (4) The design method of precoding codebook based on orthogonal analysis of orthogonal space-time coding is as follows:

1) Let = 0, given the initialization codebook ie{l,...,N}, where Θ, =£> ( ',... ), φ _η ,φ _ί2 are 0 to 2; The random variable, F _DFT is the Discrete Fourier Transformation ("DFT") matrix of N _r xM. Define Ω^^,···,^}, and calculate the minimum chord distance of the codebook min_d(A:)= min d(F _t , FX , .e Ω ^Λ

― !</'< j≤N ^JJ

2) Randomly generate ^ 维各各各正交 , , , , , , , , , , , The calendar should make ^ big enough.

3) i§ik = k + Use the cluster analysis method to divide ί/ into N classes, so that ., = 1,..., ^ are all members of each class:

R _t = {U, F _t )< d _c (U,Fj), j≠ i,F _j ,F _i e Ω"}

, = 1,···, N (5) Here is a brief example of the meaning of cluster analysis:

Initializing a codebook can be called an old codebook, and this codebook is composed of 16 codewords. Randomly generate 10000 sample sample code words, each sample sample code word and 16 code words respectively calculate the distance to obtain the code word corresponding to the minimum distance, then the code sample word and the minimum distance correspond to the code word is divided into one class This can be classified into 16 categories, which is the method of cluster analysis.

4) Find the optimal codeword matrix for each class:

1 1

F° ^pt = arg min - d _c U„, F = arg max tr F ^H ,i = l,---,N (6)

U„eR

5) Definition = 丄XV _n V _n ^H , eigenvalue decomposition = P i ^ff , so the optimal codeword matrix should be the eigenvector corresponding to the first M largest features of the eigenvector, define Ω ^ορ ' ={F ₁ ^Opt , -- ^,, F° ^! "} , and calculate the minimum chord distance e Ω of the codebook a ^opt , where N is the total number of members in the class,

i=l,2,...N

6) If min - d(t) ≥ min - d^ - 1) then = Ω. ^Ρί , otherwise ί^=Ω". Continue back to step (2) until max e does not increase after multiple iterations.

It can be seen from the above description that the minimum distance of the codebook space based on cluster analysis can be maximized by iterative method, so that the codebooks are distributed in a given space. The optimal precoding matrix based on orthogonal space-time coding is composed of the right eigenvector corresponding to the first M singular values of the channel matrix, so it is specific to Channel conditions, such as correlation channel, Rice channel, etc., can perform Monte-Carlo simulation on the channel, and optimally pre-code the channel pair to generate N, xM-dimensional states go through matrix ί/, and then cluster analysis to generate codebook . The method generates a codebook that is more in line with the actual precoding statistical properties and can be used in an adaptive codebook design.

The codebook design can be used for any antenna configuration and arbitrary codebook lengths. Several simulation results are exemplified below.

The simulation experiment of an embodiment of the present invention considers a multi-user chirp system using orthogonal space-time coding with limited feedback precoding. It is assumed that the base station is configured with four antennas, one antenna is configured for each user, the base station transmits two data streams to each user, and the Alamouti code of two transmit antennas is used for the transmitted symbols.

Figure 3 shows the proposed closed-loop orthogonal space-time coding scheme, the IEEE802.20 Rate-One scheme of the antenna configured by the base station, and the system error symbol rate of the open-loop Alamouti coding scheme in the case where the codebook length and the moving speed are different. The performance varies with the total power transmitted by the base station antenna. It can be seen that the symbol error rate of the embodiment of the present invention is lower than that of IEEE802.20 at the same codebook size and moving rate.

Figure 4 shows the system error rate performance of the proposed codebook and IEEE802.20 codebook (Codebook index = 2) with the total power of the base station antenna when the rate is 3km/h and the codebook length is different. curve. It can be seen that the symbol error rate of the embodiment of the present invention is lower than that of IEEE802.20 under the same codebook size.

The simulation results in Figure 3 and Figure 4 use the orthogonal transform phase key ( Quadrature Phasesshift

Keying, referred to as "QPSK").

The receiving device and the transmitting device in the comparative embodiment of the present invention will be described below. The receiving device contains:

Selecting a unit of an optimal precoding matrix according to a current channel state;

Feeding information representing the precoding matrix to the unit at the transmitting end;

Multiplying the received signal by the conjugated transposed unit of the precoding matrix used by the transmitting end;

A unit that performs orthogonal space-time decoding on a signal that is multiplied by the pre-transformed matrix.

The receiving device stores the same codebook as the transmitting end, where the codebook includes at least one precoding matrix; the receiving device selects a precoding matrix that best matches the current channel state from the codebook as the optimal precoding matrix; The information of the matrix is the sequence number of the precoding matrix in the codebook.

The sending device contains:

a unit that obtains feedback information from the receiving end;

A unit for orthogonal space time coding of information to be transmitted;

The orthogonally space-time encoded signal is multiplied by the pre-coding matrix represented by the information fed back by the receiving end.

The sending device stores the same codebook as the receiving end, and the codebook includes at least one precoding matrix;

The information representing the precoding matrix fed back by the receiving end is the sequence number of the precoding matrix in the codebook. It should be noted that the various units mentioned in the embodiments of the present invention are logical concepts, physically implemented in different devices, or implemented in the same device. These units may have various names, but the effects of the present invention can be achieved as long as they have the above functions, and are within the protection scope of the present invention.

In summary, in various embodiments of the present invention, the channel fading condition is robust (Robust), and at low speed, the MIMO system has better transmission performance because closed-loop feedback is used; At high speeds, MIMO systems still have better performance because of the use of orthogonal space-time coding. In general, after using the technical solution of the present invention, the performance of the MIMO system is relatively balanced, and the system performance does not fluctuate greatly due to the change of the moving speed of the terminal.

Secondly, the same codebook is set at the transmitting end and the receiving end, and the receiving end selects the optimal precoding matrix from the codebook, and only feeds back the sequence number of the optimal precoding matrix in the codebook to the transmitting end, thereby reducing the feedback. Resource consumption on.

The optimal precoding matrix selected by the receiving end is a precoding matrix with a maximum of |H in the codebook, where H is the channel matrix, F is the precoding matrix in the codebook, and | represents the 2 norm. Thereby solving the problem that there is no better precoding matrix selection method in the current publication.

The optimized search method for codebook generation proposed by the embodiment of the present invention can maximize the minimum distance of the codewords in the codebook in the Grassmanian space, so that the codebooks are distributed in a given space. .

Although the invention has been illustrated and described with reference to the preferred embodiments of the present invention, it will be understood The spirit and scope of the invention.

Claims

Rights request

A communication method, comprising the steps of:

The receiving end selects an optimal precoding matrix according to the current channel state, and feeds back information representing the optimal precoding matrix to the transmitting end;

The transmitting end performs orthogonal space time coding on the information to be transmitted, and then multiplies the optimal precoding matrix represented by the information fed back by the receiving end, and then sends the information;

The receiving end multiplies the received signal by the common transposition of the optimal precoding matrix used by the transmitting end, and performs orthogonal space-time decoding to obtain information that needs to be transmitted by the transmitting end.

The communication method according to claim 1, wherein the same codebook is stored at both the receiving end and the transmitting end, and at least one precoding matrix is included;

The selecting, by the receiving end, the optimal precoding matrix according to the current channel state, includes: the receiving end selecting, from the codebook, a precoding matrix that best matches a current channel state as an optimal precoding matrix;

The information representing the optimal precoding matrix is a sequence number of the optimal precoding matrix in the codebook.

The communication method according to claim 2, wherein the optimal precoding matrix is a precoding matrix that maximizes |H 3⁄4 in the codebook, where H is a channel matrix, and F is the code In the precoding matrix of this, I represents 2 norm.

4. The communication method according to claim 2, wherein the codebook is obtained by the following steps:

The optimized search finds N N, xM-dimensional states that are orthogonal to the unitary matrix U as the codeword of the codebook, and the objective function of the optimized search is the minimum distance of the codewords in the Grassmanian space, so that the objective function The maximum number of N Us is the codeword of the codebook; where N is the number of receiving antennas, M For the number of data streams, N is the codebook length.

5. The communication method according to claim 4, wherein the optimized search is implemented by the following steps:

Generate an initial codebook;

Generate ^11;

Divide U into N categories, members R _t {U,F _t )< d _c (U,Fj), j≠ i,F _j ,F _i e

Ω"}; Find the optimal codeword matrix for each class = argmin 丄 V d _c (U _n , F ), i = l, ---, N to form the candidate codebook with the optimal codeword matrix of each class Ω. ^ρί , and calculate the minimum chord of the candidate codebook from e il ^opt ;

If the minimum chord distance of the candidate codebook is greater than or equal to the minimum chord distance of the current codebook, the candidate codebook Ω will be used. ^Ίί as the new current codebook Ω;

Repeating the above steps until the minimum chord distance of the candidate codebook is less than or equal to the minimum number of consecutive chord distances of the current codebook reaches a predetermined threshold;

Where is a positive integer and N>N, representing the chord distance of the two matrices.

6. The communication method according to claim 5, wherein the initial codebook

F _t ^k = Q _t F _DFT , iG{l,-,N] , where Θ,. ) ,

₁ , a random variable uniformly distributed between 0 and 2; τ, F _DPT is a discrete Fourier transform DFT matrix of N _r xM.

7. A receiving device, characterized in that

Feeding information representing the optimal precoding matrix to the unit at the transmitting end; Multiplying a signal received from the transmitting end by a unit of a common transposition of the optimal precoding matrix used by the transmitting end;

A unit for orthogonal space-time decoding of a signal obtained by multiplying the conjugated transposed signal of the optimal precoding matrix used by the transmitting end.

The receiving device according to claim 7, wherein the receiving device has the same codebook as the transmitting end, and the codebook includes at least one precoding matrix;

The optimal precoding matrix is a precoding matrix that best matches a current channel state in the codebook; the information representing the optimal precoding matrix is a sequence number of the optimal precoding matrix in the codebook.

9. The receiving device according to claim 8, wherein the optimal precoding matrix is a precoding matrix that maximizes |H 3⁄4 in the codebook, where H is a channel matrix and F is the code In the precoding matrix of this, I represents 2 norm.

10. A transmitting device, characterized in that

Obtaining, by the receiving end, a unit representing the optimal precoding matrix information;

A unit for orthogonal space time coding of information to be transmitted;

The orthogonally space-time encoded signal is multiplied by the unit transmitted after the optimal precoding matrix represented by the information fed back by the receiving end.

The transmitting device according to claim 10, wherein the transmitting device stores the same codebook as the receiving end, and the codebook includes at least one precoding matrix;

The information representative of the optimal precoding matrix fed back by the receiving end is the sequence number of the precoding matrix in the codebook.

12. A receiving method, characterized in that:

Selecting an optimal precoding matrix according to the current channel state; Feeding information representing the optimal precoding matrix to the transmitting end;

Multiplying a signal received from the transmitting end by a common transposition of an optimal precoding matrix used by the transmitting end;

Performing orthogonal space-time decoding on the signal after multiplication and transposition of the optimal precoding matrix used by the transmitting end to obtain information to be transmitted by the transmitting end.

The receiving method according to claim 12, further comprising: storing the same codebook as the transmitting end, where at least one precoding matrix is included;

The optimal precoding matrix is a precoding matrix that maximizes |HF| in the codebook, where H is a channel matrix, ^ is a precoding matrix in the codebook, and ll l represents a 2 norm.

14. A method of transmitting, characterized by:

Obtaining information representative of the optimal precoding matrix fed back by the receiving end;

Perform orthogonal space time coding on the information that needs to be transmitted;

The orthogonally space-time encoded signal is multiplied by the optimal precoding matrix represented by the information fed back by the receiving end, and then transmitted.