KR20090101804A - Open-loop spatial multiplexing for 4tx system with rank adaptation - Google Patents

Abstract

PURPOSE: An open-loop spatial multiplexing method for a 4Tx system with rank adaptation is provided to reduce the complexity of a wireless mobile communication system according to the number of pre-coding matrixes based on phase transition. CONSTITUTION: An open-loop spatial multiplexing method for a 4Tx system with rank adaptation performs pre-coding for subcarriers or virtual resources by using a phase transition based matrix. The phase transition based matrix is configured in unitary sparse matrix, and a code book configured in pre-coding matrix has 4 pre-coding matrixes W2(i) for rank 2, wherein i=0,1,2,3. The code book configured in pre-coding matrix has 4 pre-coding matrixes W3(i) for rank 3, wherein i=0,1,2,3. The code book configured in pre-coding matrix has 4 pre-coding matrixes W4 for rank 4, wherein i=0,1,2,3.

Description

Open-Loop Spatial Multiplexing for 4Tx System with Rank Adaptation}

The present invention relates to a data transmission / reception method using a phase shift based precoding in a multi-antenna system using a plurality of subcarriers and a transceiver supporting the same.

In order to cope with the recently increasing demand for wireless communication services, the capacity of a communication system must be increased. In order to increase the communication capacity in a wireless communication environment, a method for increasing the efficiency of limited resources can be considered. At this time, a multi-antenna transmission and reception technology for taking a space diversity gain or increasing a transmission capacity by mounting a plurality of antennas on a transceiver has been recently actively received with great attention.

Referring to FIG. 1, the general structure of a multiple-input multiple-output (MIMO) system using Orthogonal Frequency Division Multiplexing (OFDM) among multiple antenna transmission / reception techniques is as follows.

In the transmitting end, the channel encoder 101 attaches redundant bits to the transmitted data bits to reduce the influence of the channel or noise, and the mapper 103 converts the data bit information into data symbol information, and the serial-parallel converter. 105 parallelizes the data symbols onto a plurality of subcarriers, and the multi-antenna encoder 107 converts the parallelized data symbols into space-time signals. The multiple antenna decoder 109, the parallel-to-serial converter 111, the de-mapper 113, and the channel decoder 115 at the receiving end are the multiple antenna encoder 107, the serial-to-parallel converter 105, the mapper ( 103 and the reverse function of the channel encoder 101, respectively.

In a multi-antenna OFDM system, a scheme for increasing spatial diversity gain includes a space-time code (STC), a cyclic delay diversity (CDD), and a signal to noise ratio. Techniques for increasing the ratio (SNR) include beamforming (BF) and precoding. Here, space-time code and cyclic delay diversity are mainly used to increase transmission reliability of an open loop system in which feedback information is not available at a transmitter, and beamforming and precoding are applicable in a closed loop system in which feedback information is available at a transmitter. It is used to maximize the signal to noise ratio through the feedback information.

Among the above-described techniques, a technique for increasing the spatial diversity gain and a technique for increasing the signal-to-noise ratio, in particular, look at cyclic delay diversity and precoding as follows.

In the cyclic delay diversity scheme, in the OFDM signal transmission system, all the antennas transmit signals with different delays or different sizes to obtain frequency diversity gains at the receiver. 2 illustrates a configuration of a transmitter of a multiple antenna system using a cyclic delay diversity scheme.

OFDM symbols are separately transmitted to each antenna through a serial-to-parallel converter and a multi-antenna encoder, and then a Cyclic Prefix (CP) is attached to the receiver to prevent interference between channels. In this case, the data sequence transmitted to the first antenna is transmitted to the receiver as it is, but the data sequence transmitted to the next antenna is cyclically delayed by a certain sample compared to the antenna of the previous sequence.

On the other hand, if the cyclic delay diversity scheme is implemented in the frequency domain, the cyclic delay can be expressed as a product of a phase sequence. That is, as shown in FIG. 3, the data sequence in the frequency domain may be multiplied by a predetermined phase sequence (phase sequence 1 to phase sequence M) set differently for each antenna, and then may be transmitted to a receiver by performing a fast inverse Fourier transform (IFFT). This is called a phase shift diversity technique.

Using the phase shift diversity technique, a flat fading channel can be transformed into a frequency selective channel, and frequency diversity gain can be obtained through channel codes or multi-user diversity gain can be obtained through frequency selective scheduling. have.

Precoding schemes include a codebook based precoding scheme used when the feedback information is finite in a closed loop system, and a method of quantizing and feeding back channel information. The codebook-based precoding is a method of obtaining a signal-to-noise ratio (SNR) gain by feeding back an index of a precoding matrix known to the transmitter / receiver to the transmitter.

4 illustrates a configuration of a transmitting and receiving end of a multiple antenna system using the codebook based precoding. Here, each of the transmitter and the receiver has finite precoding matrices ( P ₁ to P _L ), and the receiver feeds back the optimal precoding matrix index ( l ) to the transmitter using channel information. Apply the precoding matrix corresponding to the transmission data (χ ₁ ~ χ _Mt) .

The above-described phase shift diversity technique has attracted much attention because of the advantages of gaining frequency selective diversity gain in open loops and gaining frequency scheduling gain in closed loops in addition to the above-mentioned advantages. The gains are difficult to achieve if the rate cannot be expected and the resource allocation is fixed.

In addition, the above-described codebook based precoding scheme can use a high spatial multiplexing rate while requiring a small amount of feedback information (index information), so that efficient data transmission is possible, but a stable channel must be secured for feedback. It is not suitable for severe mobile environments and is especially applicable only in closed loop systems.

In the codebook based precoding scheme, there is a problem that the complexity of the system increases as the number of precoding matrices increases.

An object of the present invention is to reduce the complexity of a wireless mobile communication system by limiting the number of phase shift based precoding matrices according to rank.

According to an aspect of the present invention, a method of performing precoding in a multi-antenna system using a plurality of subcarriers and using a phase shift based matrix is provided by using the phase shift based matrix. Performing a precoding on a subcarrier or a virtual resource, wherein the phase shift based matrix is composed of a product of a precoding matrix, a diagonal matrix for phase shifting, and a unitary matrix, and the precoding matrix A codebook consisting of has four precoding matrices W2 (i) for rank 2 (where i = 0,1,2,3) and four precoding matrices W3 (i) for rank 3 , i = 0,1,2,3) and one precoding matrix W4 (i) for rank 4, i = 0. At this time, of the above W2 (i), W3 (i), W4 (i), W2 (0), W3 (0), and W4 (0) are generated from the vector [1 -1 -1 1] ^T , W2 (1) and W3 (1) are generated from the vector [1 -1 1 -1] ^T , W2 (2) and W3 (2) are generated from the vector [1 1 -1 -1] ^T , and W2 ( 3) and W3 (3) can be generated from the vector [1 1 1 1] ^T.

According to another aspect of the present invention, a method of performing precoding in a multi-antenna system using a plurality of subcarriers and using a phase shift based matrix, precoding a subcarrier or a virtual resource using the above phase shift based matrix Wherein the phase shift based matrix comprises a product of a precoding matrix, a diagonal matrix for phase shifting, and a unitary matrix, and a codebook consisting of the precoding matrix is a rank value. The larger this value is, the smaller the number of precoding matrices defined for the rank is. In this case, the above codebook has four precoding matrices W2 (i) for rank 2, i = 0,1,2,3, and three precoding matrices W3 (i) for rank 3 , i = 0,1,2, and two precoding matrices W4 (i) (for i = 0,1) for rank 4. At this time, of the above W2 (i), W3 (i), W4 (i), W2 (0), W3 (0), and W4 (0) are generated from the vector [1 -1 -1 1] ^T , W2 (1), W3 (1), W4 (1) are generated from the vector [1 -1 1 -1] ^T , and W2 (2) and W3 (2) are generated from the vector [1 1 -1 -1] ^T W2 (3) can be generated from the vector [1 1 1 1] ^T.

Or, according to another aspect of the present invention, in the method of performing precoding in a multi-antenna system using a plurality of subcarriers and using a phase shift based matrix, the codebook includes four pre-orders for rank 2. Has a coding matrix W2 (i) (where i = 0,1,2,3), has two precoding matrices W3 (i) (where i = 0,1) for rank 3, and rank 4 One may have one precoding matrix W4 (i) (where i = 0). At this time, of the above W2 (i), W3 (i), W4 (i), W2 (0), W3 (0), and W4 (0) are generated from the vector [1 -1 -1 1] ^T , W2 (1) and W3 (1) are generated from the vector [1 -1 1 -1] ^T , W2 (2) is generated from the vector [1 1 -1 -1] ^T , and W2 (3) is generated from the vector [ 1 1 1 1] Method of performing precoding, generated from ^T.

According to another aspect of the present invention, a method of performing precoding in a multi-antenna system using a plurality of subcarriers and using a phase shift based matrix, precoding a subcarrier or a virtual resource using the above phase shift based matrix Wherein the phase shift based matrix comprises a product of a precoding matrix, a diagonal matrix for phase shifting, and a unitary matrix, and the codebook consisting of the precoding matrix is rank 2 We have four precoding matrices W2 (i) for i, i = 0,1,2,3, and four precoding matrices W3 (i) for rank 3, i = 0,1,2 , 3) and four precoding matrices W4 (i) for rank 4, i = 0,1,2,3, with W2 (i), W3 (i), W4 ( In i), W2 (0), W3 (0), and W4 (0) are generated from the vectors [1 -1 -1 -1] ^T , and W2 (1), W3 (1), and W4 (1) Is generated from the vectors [1 1 -1 1] ^T and W2 (2), W3 (2), and W4 (2 ) Is generated from the vector [1 -1 1 1] ^T , and W2 (3), W3 (3), and W4 (3) are generated from the vector [1 1 1 -1] ^T.

The present invention described above can be used in an open-loop spatial division multiplexing method for rank adaptive 4Tx systems.

According to the present invention, since the number of phase shift based precoding matrices is limited according to rank, the complexity of the wireless mobile communication system is reduced.

1 is a block diagram of an orthogonal frequency division multiplexing system having multiple transmit / receive antennas.

2 is a block diagram of a transmitter of a multiple antenna system using a conventional cyclic delay diversity scheme.

3 is a block diagram of a transmitter of a multi-antenna system using a conventional phase shift diversity scheme.

4 is a block diagram of a transmitting and receiving end of a multiple antenna system using a conventional precoding technique.

FIG. 5 is a block diagram illustrating a main configuration of a transceiver for performing phase shift based precoding.

6 graphically illustrates two applications of phase shift based precoding or phase shift diversity.

7 illustrates a series of steps for forming a baseband signal representing a downlink physical channel.

FIG. 8 is a block diagram of an SCW OFDM transmitter to which a phase shift based precoding technique is applied, and FIG. 9 is a block diagram of an MCW OFDM transmitter.

Some of the additional advantages, objects, and features of the invention are set forth in the detailed description set forth below, while others may be learned from the practice of the invention, or may be learned by those skilled in the art by the following description. It is self-evident to them. The objects and other advantages of the invention can be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

Reference will now be made in detail to preferred embodiments of the invention, examples of which are illustrated in the accompanying drawings. Wherever possible, the same reference numbers will be used throughout the drawings to refer to the same or similar parts.

Embodiments according to the present invention described below cyclically change the precoding matrices in a codebook according to the data index with a large-delay large-delay CCD (CDD) by limiting the number of precoding matrices according to rank. In order to minimize the coding and decoding complexity of the open-loop spatial multiplexing scheme in the 4Tx system.

Phase Shift Based Precoding Matrix

FIG. 5 is a block diagram illustrating a main configuration of a transceiver for performing phase shift based precoding.

Phase shift based precoding transmits all streams to be transmitted through the entire antenna but multiplies the sequences of different phases. In general, when a phase sequence is generated using a small cyclic delay value, the channel selectivity becomes larger or smaller depending on the frequency domain while generating frequency selectivity in the channel when viewed from the receiver.

As shown in FIG. 5, the transmitter secures scheduling gain by allocating a user terminal to a portion in which a frequency is increased and a channel state is improved in a frequency band fluctuating according to a relatively small cyclic delay value. In this case, a phase shift based precoding matrix is used to apply a cyclic delay value that increases or decreases uniformly for each antenna.

The phase shift based precoding matrix P can be expressed as follows.

Here, k denotes an index of a subcarrier or an index of a specific frequency band, and w ^k _{i, j} (i = 1, ..., N _t , j = 1, ..., R) represents the complex weight determined by k. In addition, N _t represents the number of transmit antennas, and R represents the spatial multiplexing rate. Here, the complex weight may have a different value depending on the OFDM symbol multiplied by the antenna and the index of the corresponding subcarrier. The complex weight may be determined according to at least one of channel conditions and presence or absence of feedback information.

Meanwhile, the precoding matrix P of Equation 1 is preferably designed as a unitary matrix to reduce the loss of channel capacity in the multi-antenna system. In this case, the channel capacity of the multi-antenna open-loop system is expressed as an equation in order to determine the configuration conditions of the unitary matrix.

Here, H is a multi-antenna channel matrix having a size of N _r x N _t and N _r represents the number of receiving antennas. The phase shift based precoding matrix P is applied to Equation 2 as follows.

As shown in Equation 3, in order to avoid loss in channel capacity, since PP ^H must be a single matrix, the phase shift based precoding matrix P must satisfy the following conditions.

In order for the phase shift based precoding matrix P to become a unit matrix, two conditions, that is, a power constraint and an orthogonal constraint, must be satisfied at the same time. The power constraint is to make each column of the matrix to be 1, and the orthogonal constraint is to make orthogonality between each column of the matrix. Each of these expressions is expressed as follows.

Next, an example of a generalized equation of a 2 × 2 phase shift based precoding matrix will be presented, and a relation for satisfying the two conditions will be described. Equation 7 shows a general equation of a phase shift based precoding matrix having two transmit antennas and a spatial multiplexing rate of two.

Here, α _i , β _i ( i = 1, 2) have a real value, θ _i (i = 1, 2, 3, 4) represents a phase value, and k represents a subcarrier index of an OFDM signal. In order to implement such a precoding matrix as a unitary matrix, power constraint conditions of Equation 8 and orthogonal constraints of Equation 9 must be satisfied.

Where the * symbol indicates a conjugate complex number. An example of a 2 × 2 phase shift based precoding matrix that satisfies Equations 7 to 9 is as follows.

Here, θ ₂ and θ ₃ have the same relationship as in Equation 11 under orthogonal constraints.

The precoding matrix may be stored in the form of a codebook in the memory of the transmitter and the receiver. The codebook may include various precoding matrices generated through finite different θ ₂ values. Here, the value of θ ₂ may be appropriately set according to the channel condition and the presence or absence of feedback information. If θ ₂ is used when the feedback information is used, θ ₂ is set high when the feedback information is not used. Frequency diversity gain can be obtained.

Meanwhile, frequency diversity gain or frequency scheduling gain may be obtained according to the magnitude of delay samples applied to the phase shift based precoding. 6 graphically illustrates two applications of phase shift based precoding according to the size of a delay sample.

As shown in FIG. 6, when a large value of delay sample (or cyclic delay) is used, the frequency selectivity period is shortened, thereby increasing frequency selectivity and thus, the channel code may obtain frequency diversity gain. This is preferably used in open loop systems where the temporal change of the channel is severe and the reliability of feedback information is poor.

In addition, in the case of using a small value of the delay sample, there is a portion where the size of the channel increases and decreases in the frequency selective channel changed in the flat fading channel. Accordingly, the channel size of a certain subcarrier region of the OFDM signal becomes large, and the channel size of the other subcarrier region becomes small.

In this case, in an orthogonal frequency division multiple access (OFDMA) system that accommodates multiple users, signal to noise ratio (SNR) may be increased by transmitting a signal through a predetermined frequency band having a larger channel size for each user. In addition, since frequency bands with increased channel sizes are often different for each user, a multi-user diversity scheduling gain is gained from the system's point of view. On the other hand, since the receiver only needs to transmit CQI (Channel Quality Indicator) information of the subcarrier region that can be allocated to each resource simply as feedback information, the feedback information is relatively reduced.

The delay sample (or cyclic delay) for the phase shift based precoding may be a predetermined value to the transceiver, or may be a value delivered by the receiver to the transmitter through feedback. In addition, the spatial multiplexing rate (R) may also be a predetermined value for the transceiver, but the receiver may periodically determine the channel state to calculate the spatial multiplexing rate and feed it back to the transmitter. Multiplexing rates can also be calculated and changed.

Generalized Phase Shift Diversity Matrix

The above-described phase shift based precoding matrix is expressed in the form of Equation 12 for a system having an antenna number N _t ( N _t is a natural number of 2 or more) and a spatial multiplexing rate R ( R is a natural number of 1 or more). Can be. Since this can be seen as a generalized representation of a conventional phase shift diversity scheme, the multi-antenna technique according to Equation 12 will be referred to as generalized phase shift diversity (GPSD).

here, Denotes a GPSD matrix for a kth subcarrier of a MIMO-OFDM signal having N _t transmit antennas and a spatial multiplexing rate of R, Is It is a unitary matrix (second matrix) that satisfies Δ is used to minimize interference between subcarrier symbols corresponding to each antenna. In particular, to maintain the unit matrix characteristics of the diagonal matrix (primary matrix) for phase shifting It is desirable for itself to satisfy the condition of the unit matrix. In Equation 12, the phase angles θ _i , i = 1, ..., N _t in the frequency domain have the following relationship with the delay time τ _i , i = 1, ..., N _t in the time domain.

Here, N _fft represents the number of subcarriers of the OFDM signal.

As a modified example of Equation 12, the GPSD matrix may be obtained as follows.

When the GPSD matrix is constructed in the equation (14), since the symbols of each data stream (or OFDM subcarrier) are shifted by the same phase, the matrix may be easily constructed. That is, the GPSD matrix of Equation 12 has rows of the same phase, whereas the GPSD matrix of Equation 14 has columns of the same phase, so that each subcarrier symbol is shifted by the same phase. Expanding Equation 14, the GPSD matrix can be obtained in the following manner.

According to Equation 15, since rows and columns of the GPSD matrix have independent phases, more various frequency diversity gains can be obtained.

As an example of Equations 12, 14, and 15, the GPSD determinant of a system having two transmit antennas and using a 1-bit codebook is expressed as follows.

When the value of α is determined in Equation 16, the value of β is easily determined. Therefore, information on the value of α may be set to two appropriate values and the information may be fed back to the codebook index. For example, when the feedback index is 0, α may be 0.2, and when the feedback index is 1, α may be 0.8.

In Equation 12, 14, and 15, the unitary matrix ( For example, a predetermined precoding matrix for obtaining a signal-to-noise ratio (SNR) gain may be used, and a Walsh Hadamard matrix or a DFT matrix may be used as the precoding matrix. Among them, an example of the GPSD matrix according to Equation 12 when the Walsh Hadamard matrix is used is as follows.

Equation (17) assumes a system having four transmit antennas and a spatial multiplexing rate of 4, wherein a specific transmission antenna is selected by appropriately reconstructing the second matrix, or a rank adaptation is performed. )can do.

On the other hand, the unitary matrix of the formula (12), (14), (15) ) May be provided in the form of a codebook at the transmitting end and the receiving end. In this case, the transmitter receives feedback information of the codebook from the receiver, selects a second matrix of the corresponding index from the codebook provided by the receiver, and then uses a phase shift-based precoding matrix using one of Equations 12, 14, and 15. Configure

Unitary matrices of Equations 12, 14, and 15 ( An example of the GPSD matrix in the case where 2 x 2 and 4 x 4 Walsh codes are used as follows is as follows.

Time-variable Generalized Phase Shift Diversity

In the GPSD matrices of Equations 12, 14, and 15, the phase angle θ _i and / or the unitary matrix U of the diagonal matrix may change with time. For example, the time-variable GPSD for Equation 12 may be expressed as follows.

here, Denotes the GPSD matrix for the kth subcarrier of the MIMO-OFDM signal having N _t transmit antennas and R spatial multiplexing rate at a specific time t, Is It is a unitary matrix (fourth matrix) that satisfies the value and is used to minimize the interference between subcarrier symbols corresponding to each antenna. In particular, to maintain the unit matrix characteristics of the diagonal matrix (third matrix) for phase shifting It is desirable for itself to satisfy the condition of the unit matrix. In Equation 18, the following relationship holds for the phase angles θ _i (t), i = 1, ..., N _t and the delay times τ _i (t), i = 1, ..., N _t .

Here, N _fft represents the number of subcarriers of the OFDM signal.

As shown in Equations 18 and 19, the time delay sample value and the unitary matrix may change over time, and the unit of time may be an OFDM symbol unit or a certain unit of time.

An example of a GPSD matrix using 2 × 2 and 4 × 4 Walsh codes as a unitary matrix for obtaining a time-variable GPSD is shown in Tables 3 and 4 below.

In Embodiment 3, a time-variable GPSD matrix for Equation 12 is introduced, but the same applies to the diagonal matrix and the unit matrix in Equations 14 and 15, respectively. Therefore, in the following embodiment, Equation 12 will be described as an example, but it is obvious to those skilled in the art that the same may be extended to Equations 14 and 15.

Expansion of generalized phase shift diversity

In the second embodiment, an extended GPSD matrix may be configured by adding a third matrix corresponding to a precoding matrix to a GPSD matrix including a diagonal matrix and a unit matrix. This is expressed as an equation.

The extended GPSD matrix is characterized in that a precoding matrix P having a size of N _t x R is added before the diagonal matrix, and thus the size of the diagonal matrix is changed to R x R as compared with Equation 12. The added precoding matrix ( ) May be set differently for a specific frequency band or a specific subcarrier symbol, and in an open loop system, it is preferable to set a fixed matrix. Such a precoding matrix ( In addition, a more optimized signal-to-noise ratio (SNR) gain can be obtained.

Alternatively, a codebook including a plurality of precoding matrices P may be provided at the transmitting end and the receiving end.

Meanwhile, in the extended GPSD matrix, at least one of the precoding matrix P, the phase angle θ of the diagonal matrix, and the unit matrix U may change with time. To this end, when the index of the next precoding matrix P is fed back in a predetermined time unit or a predetermined subcarrier unit, a specific precoding matrix P corresponding to the index may be selected from a predetermined codebook.

The extended GPSD determinant according to the present embodiment can be expressed as follows.

As an example of the extended GPSD matrix, the determinant of a multi-antenna system having two and four transmit antennas is as follows. Here, although the DFT matrix is used as the unit matrix U, the matrix is not necessarily limited thereto. Any matrix that satisfies the unit condition such as a Walsh Hadamard code may be used.

In addition, as another example of the extended GPSD matrix, a determinant of a multi-antenna system having four transmit antennas is as follows.

In equation (24), the extended GPSD matrix is added before the N _t x N _t diagonal matrix (D ₁₎ and a precoding matrix of the N _t x R size of the size (P) is a diagonal matrix (D ₂₎ than in equation (12) Therefore, the size of the diagonal matrix D ₂ is characterized in that it is changed to R x R. The added precoding matrix ( ) May be set differently for a specific frequency band or a specific subcarrier symbol, and in an open loop system, it is preferable to set a fixed matrix.

In this case, two kinds of phase angles can be simultaneously shifted in one system through the diagonal matrix D ₁ and the diagonal matrix D ₂ . In one example, the diagonal matrix (D ₁₎ through application of a phase shift of a value, and a diagonal matrix (D ₂₎ when applying a phase shift of a value through, to obtain a multi-user diversity scheduling gain by the electronic And the latter can obtain frequency diversity gain. In this case, the diagonal matrix D ₁ is used to improve the performance of the system, and the diagonal matrix D ₂ may be used for averaging channels between streams. In addition, by applying a large phase shift through the diagonal matrix D ₁ , the frequency diversity gain is increased, and a large phase shift is applied through the diagonal matrix D ₂ to average the channels between streams. Can be. This gain can be obtained from the structure of equation (21).

In this case, the matrix P of Equation 21 may be transformed into a subcarrier unit or a frequency resource unit without feedback information from the receiver. If this form is expressed as an equation, it can be expressed as follows.

In Equation 25 Is used to increase the frequency diversity gain by using different precoding matrix (P) for each resource index k , and to average the channel between each stream through diagonal matrix and single matrix (U).

<Example 5>

Use Codebook Subset Restriction Technique

For example, a codebook containing N _c of the precoding matrix when used to apply the codebook subset restriction techniques that use only a portion of the codebook according to a base station or a terminal N _c of the precoding matrix is the number of N _restrict It should be shortened to precoding matrix. Here, the codebook subset limiting technique can be used to reduce multi-cell interference or to reduce complexity. Here, the condition of N _restrict ≤ N _c must always be satisfied. For example, suppose the total number of precoding matrices in the codebook is N _c = 6. And, for example, a codebook determined to use only four precoding matrices out of six precoding matrices Can be expressed as Equation 26 below.

In Equation 26 above Is Equivalent codebook that rearranges the index of the codebook. When using the codebook subset limiting method of Equation 26, in order to reduce the reception complexity, only the precoding matrix of the codebook consisting of {1, -1, j, -j} among the codebooks may be used as a subset. The size of the element may have a different value depending on the normalization factor.

Use recursive iteration of precoding matrices in codebook

For example, if a precoding matrix set defined in a transmission / reception period at a specific time is defined in advance, it may be expressed as Equation 27.

In Equation 27, the set of precoding matrices includes N _c precoding matrices. Equation 27 above may be simplified to a form like Equation 28 below.

That is, Equations 27 and 28 represent codebooks. The precoding matrices within the loop are repeatedly used according to subcarriers or resource indexes.

And, in Equation 28 above Is a mix of data streams, May be referred to as a data stream substitution matrix and may be selected according to the spatial multiplexing rate R as shown in Equation 27. Can be expressed in a simple form as shown in Equation 29 below.

Spatial multiplexing rate 2:

Spatial multiplexing ratio 3:

Spatial Multiplexing Ratio 4:

The method of cyclically repeating the precoding matrices in the above-described codebook may be used in the codebook to which the codebook restriction technique is applied. For example, in Equation 26 Equation 28 may be expressed as Equation 30 below.

K in Equation 30 above represents a subcarrier or frequency resource index and in the above case, N _restrict = 4. That is, Equation 30 represents a codebook in which the precoding matrix is limited. It shows how to use the precoding matrices in the loop repeatedly according to subcarriers or resource indexes.

Iteratively use the precoding matrix in the codebook in a predetermined unit

Equation 28 may be expressed as Equation 31 below according to the frequency resource setting.

In Equation 31, k may be a subcarrier index or a virtual resource index. If k is a subcarrier index, Equation 31 shows that the precoding matrix is changed for each v subcarriers. In addition, when k is a virtual resource index, Equation 31 shows that the precoding matrix is changed for each virtual resource.

Equation 31 shows a case in which the precoding matrix can be changed within N _c precoding matrices. The value of ν can be determined using the same as the spatial multiplexing ratio of the precoding matrix. For example, it can be used in the form of ν = R.

In addition, even when the codebook subset limitation scheme described through Equation 26 is applied, it will be obvious that the precoding matrix can be changed to a predetermined number of subcarriers or virtual resource units as described above. This is shown in Equation 32 below.

Even in the case of Equation 32, one the same as the pre-coding matrices that can show a form of transforming the Equation 31 into ν ν unit according to the value, however, pre-coding matrix is N _restrict (≤N _c) of the precoding matrix There will be differences in that

On the other hand, when the frequency diversity scheme is applied by applying the cyclic repetition of the precoding matrix for each specific frequency resource using the codebook subset restriction scheme of the fifth embodiment, the frequency diversity gain is increased according to the number of cyclically repeated precoding matrices. Can vary. Various embodiments of the codebook subset restriction scheme are described below.

Codebook Subset Limitation Scheme Based on Spatial Multiplexing Rate

Subsets can be defined differently according to spatial multiplexing. For example, if the spatial multiplexing rate is low, the frequency diversity gain can be maximized by using a large number of subsets, and when the spatial multiplexing rate is high, the number of subsets can be used to reduce complexity while maintaining performance. .

Equation 33 shows an example of a method of defining a codebook subset having a different size according to each spatial multiplexing rate.

In Equation 33 above Denotes the number of precoding matrices of the subset of codebooks according to the spatial multiplexing rate R. As a result, when the precoding matrices are cyclically used for the codebook to which the codebook subset restriction scheme of the fifth embodiment is applied, the complexity of the receiver can be reduced and the performance can be improved.

Codebook Subset Limitation Scheme with Channel Coding Rate

The subset may be defined differently according to the channel coding rate. For example, the frequency diversity gain can usually obtain high performance when the channel coding rate is low, and may deteriorate when the channel coding rate is high. Therefore, in the same spatial multiplexing environment, codebook subsets of different sizes can be optimized according to channel coding rates.

Codebook Subset Limitation Scheme with Retransmission

The subset can be defined differently in consideration of retransmission. For example, by using a subset other than the codebook subset used for the first transmission, the probability of success of the retransmission can be increased. Accordingly, the performance of the system can be improved by using a cyclic iteration method of the precoding matrix using the same subset but different subsets depending on whether the retransmission is performed or the number of retransmissions.

Expansion of Generalized Phase Shift Diversity Using Power Antenna-Specific Power Control

For the precoding schemes, a power value having a different size according to frequency or time for each transmitting antenna may be used to improve performance or to efficiently use power.

For example, a power control scheme for each transmit antenna may be applied by using Equation 28, Equation 30, Equation 31, and Equation 32. In particular, an application example to the embodiments of Equations 31 and 32 may be expressed as Equation 34 and Equation 35 below.

In Equation 34 above Is a function of mixing the data streams as described above, and may also be expressed in the form of Equation 29. And, Denotes a power control diagonal matrix that allows power of different sizes to be transmitted for each transmit antenna according to the m-th frequency region or t-time in the diagonal matrix. Also, Denotes a power control factor used at t-time in the m-th frequency domain of the i-th transmission antenna.

Equation 34 represents a method of applying power control for each transmitting antenna to a scheme using cyclic repetition using a codebook having N _c precoding matrices, and Equation 35 below is a part of the codebook in Equation 32. It represents the method of applying power control for each transmitting antenna to the method using cyclic repetition using the set limiting technique.

In Equation 35 , And Each bar represents the same as in Equation 34 above. However, there will be a difference in that the precoding matrix is cyclically repeated in N _restrict (≦ N _c ) precoding matrices.

Hereinafter, a structure that the downlink physical channel can have will be described.

7 illustrates a series of steps for forming a baseband signal representing a downlink physical channel.

(1) scrambling the encrypted bits in each code word to be transmitted on the physical channel

(2) modulating the scrambled bits to generate modulation symbols having complex values

(3) mapping modulation symbols having complex values onto one or more transport layers

(4) precoding modulation symbols with complex values on each layer for transmission on the antenna port

(5) mapping modulation symbols having complex values for each antenna port to resource elements

(6) generating a time domain OFDM signal having a complex value for each antenna port

In scrambling block 201, M ^{( q )} _bit bits ( b ^{( q )} (0), ..., b ^{( q )} ( M ^{( q )} _bits ⁾ transmitted on a physical channel in one subframe. Bits in each codeword q with &lt; RTI ID = 0.0 &gt; 1)) &lt; / RTI &gt; Can be converted into

Here, the scrambling sequence c ^{( q )} ( i ) may be defined as a Gold sequence of length 31. This scrambling sequence generation stage may be initialized at the start of each subframe.

In modulation mapping block 202, for each codeword q, one block of scrambled bits ( ) Are modulated by QPSK, 16QAM, and 64QAM for PDSCH and PMCH, so that modulation symbols d ^{( q )} (0), ..., d ^{( q )} ( M ^{( q )} with complex values of one block _symb -1)).

In layer mapping block 203, modulation symbols with complex values for each codeword to be transmitted may be mapped onto one or more layers. The modulation symbols d ^{( q )} (0), ..., d ^{( q )} ( M ^{( q )} _symb -1) with complex values for codeword q are given by layer x ( i ) = [ x ^{(0 )} ( i ) ... x ^{( ν −1)} ( i )] may be mapped onto ^T. Here, i = 0, 1, ..., M ^layer _symb -1, ν is the number of layers, M ^layer _symb may be the number of modulation symbols per layer.

In precoding block 204, the vector x ( i ) = [ x ⁽⁰⁾ ( i ) ... x ^{( v -1)} ( i )] of one block ^T , i = 0, 1, ..., With M ^layer _symb -1 as input, the vector of a block mapped to the resource of each antenna port y ( i ) = [... y ^{( p )} ( i ) ...] ^T , i = 0,1 , ..., M ^ap _symb -1 Where y ^{( p )} ( i ) represents the signal for antenna port p.

In the resource element mapping block 205, for each antenna port used for transmission of the physical channel, symbols having a complex value of one block y ^{( p )} (0), ..., y ^{( p )} ( M ^ap _symb -1)), it can be mapped to a (k, l) th resource element of a physical resource block corresponding to the virtual resource block allocated for transmission. At this time, starting from y ^{( p )} (0) may be sequentially mapped. In this case, the virtual resource block may not be used for PCFICH, PHICH, PDCCH, PBCH, synchronization signal or reference signals.

The precoding block 204 is described in more detail below.

For transmission on a single antenna port, precoding can be defined by y ^{( p )} ( i ) = x ⁽⁰⁾ ( i ). Here, i = 0, 1, ..., M ^ap _symb -1, M ^ap _symb = M ^layer _symb .

Precoding for spatial multiplexing can be used with layer mapping for spatial multiplexing. In this case, spatial multiplexing supports two or four antenna ports, and the antenna ports used are p ∈ {0,1} or p ∈ {0,1,2,3}, respectively.

Without cyclic delay diversity (CDD), the precoding for spatial multiplexing can be defined by equation (37).

Here, the size of the precoding matrix W (i) is an P x ν, 0,1 = a i, ..., M _symb ^ap -1, M _symb ^ap = M _symb ^layer.

The value of the precoding matrix W (i) (or the index of the precoding matrix W (i) ) may be selected from codebooks configured in the eNodeB and the UE. The eNodeB may restrict the UE to select a precoder only within one subset of the elements in the codebook (hereafter codebook subset restriction).

In this case, the codebook may be selected from Table 5 or Table 6.

Table 5 is for transmission on two antenna ports, and the precoding matrix W ( i ) may be selected from Table 5 or a subset of Table 5. For the closed loop spatial multiplexed transmission mode, codebook index zero may not be used when the number of layers is two.

Table 6 is for transmission on four antenna ports, where the precoding matrix W ( i ) can be selected from Table 2 or a subset of Table 2. In Table 6, W _n ^{s} represents a matrix defined by columns given by the set {s} from the matrix W _n = I −2 u _n u _n ^H / u _n ^H u _n . Where I is a 4x4 unitary matrix and the vector u _n is given by Table 2. For example, when codebook index = 0 and the number of layers is ν = 2, u ₀ = [1 -1 -1 -1] ^T and W ₀ = I -2 u ₀ u ₀ ^H / u ₀ ^H u ₀ is given by a 4x4 matrix. In this case, the matrix W _n ^{14} is a 4x2 matrix composed of two columns indicated by {s} = {1, 4} among the matrix W ₀ , that is, the first column and the fourth column. .

Hereinafter, precoding for large delay CDD will be described.

In a 3GPP LTE system, the large-delay CDD as an open loop space division multiplexed transmission scheme when the rank is greater than 1 can be defined as follows. That is, for large-delay CDDs, the precoding for spatial division multiplexing may be defined as in Equation 38.

Here, the size of the precoding matrix W (i) is an P x ν (only, i = 0,1, ..., M _symb ^ap -1, M _symb ^ap = M _symb ^layer).

For different layer numbers ν , a diagonal matrix D (i) of size ν x ν that supports cyclic delay diversity and a matrix U of size ν x ν can be given by Table 7.

The value of the precoding matrix W (i) (or the index of the precoding matrix) may be selected from codebooks configured in the eNodeB and the UE. The eNodeB may restrict the UE to select a precoder only within one subset of the elements in the codebook (hereafter codebook subset restriction. The codebook configured for 2Tx may be selected from Table 5). And a codebook configured for 4Tx may be selected from Table 6.

Hereinafter, 4Tx of precoding for large delay CDD will be described in detail.

For 4Tx, the precoding matrix W (i) must be one of the precoding elements in the codebook defined in Table 6. The open loop SM operation for the 4Tx antenna system is defined as follows.

(1) PMI feedback is not used. The large-delay CCD operation is specified only as an open loop spatial multiplexing defined as follows. Node-B recursively assigns different precoders selected from a fixed codebook subset to different data subcarriers in the scheduled subbands. Different precoders are used once for each γ data resource element (RE) subcarriers. Where γ is the transmission rank. In particular, the precoder W ( i ) for data index i is selected by W ( i ) = C _k . Where k is the precoder index Is given by Where k = 1, 2, ..., N, and N is the size of the codebook subset represented by {C ₁ , C ₂ , ..., C _N }.

(2) The codebook subset represented by {C ₁ , C ₂ ,..., C _N } may be defined as in Embodiments 1, 2, and 3 below.

Example 1

N may have the same value regardless of rank. For example, if N = 4 regardless of rank, the codebook subset may be selected from the codebooks defined in Table 8 defined below. Here, the selected precoding matrices in the codebook have a maximum codeword distance in order to maximize frequency diversity gain.

Example 2

In order to reduce the complexity of encoding and decoding, N may have different values according to rank. In general, as the rank value increases, the frequency diversity gain resulting from precoding matrix cycling decreases. Therefore, it is desirable to use smaller N values as the rank value becomes larger. For example, N values may be defined such that N has 4, 2, and 1 values for ranks 2, 3, and 4, respectively. In this case, the codebook subset may be selected from codebooks defined in Table 9 below.

Alternatively, N values may be defined for ranks 2, 3, and 4 so that N has values of 4, 3, and 2, respectively. In this case, the codebook subset may be selected from codebooks defined in Table 10 below.

Example 3

Since the frequency diversity gain resulting from precoding matrix cycling is marginal when the rank has a maximum value, a fixed matrix (i.e., N = 1) is used for full rank transmission, and a plurality of The precoding matrix may use a scheme used when the rank number is smaller than the number of transmit antennas. For example, if a fixed matrix is used for full rank and is defined as N = 4 for rank 2, rank 3, the codebook subset may be selected from the codebooks defined in Table 11 below.

<Example 4>

Transceiver with Phase-Transition based Precoding

In general, a communication system includes a transmitter and a receiver. Here, the transmitter and the receiver may be referred to as a transceiver that performs both a transmission function and a reception function. However, in order to clarify the description of the feedback, one of the general data transmission is called a transmitter, and the other transmitting the feedback data to the transmitter is called a receiver.

In downlink, a transmitter may be part of a base station and a receiver may be part of a terminal. In uplink, a transmitter may be part of a terminal and a receiver may be part of a base station. The base station may include a plurality of receivers and a plurality of transmitters, and the terminal may also include a plurality of receivers and a plurality of transmitters. In general, since each component of the receiver performs a reverse function of each component of the transmitter corresponding thereto, only the transmitter will be described in detail below.

FIG. 8 is a block diagram of an embodiment of an SCW OFDM transmitter to which a phase shift based precoding technique is applied, and FIG. 9 is a block diagram of an embodiment of an MCW OFDM transmitter.

Other configurations, including channel encoders 510 and 610, interleavers 520 and 620, fast inverse Fourier transformers (IFFTs) 550 and 650 and analog converters 560 and 660, are the same as those in FIG. The description will be omitted, and only the precoders 540 and 640 will be described in detail here.

Precoders 540 and 640 include precoding matrix determination modules 541 and 641 and precoding modules 542 and 642.

The precoding matrix determination module 541, 641 determines a phase shift based precoding matrix in the form of Equations 12, 14, 15 and 20, 21. Since a detailed method of determining a precoding matrix has been described in detail, a description thereof is omitted here. The phase shift based precoding matrix determined in the form of equations (12), (14), (15), and (20), (21) can be used to determine the precoding matrix and / or You can change the phase angle and / or unit matrix.

In addition, the precoding matrix determination module 541, 641 may select at least one of the precoding matrix and the unit matrix based on information fed back from the receiving end, wherein the feedback information includes a matrix index for a predetermined codebook. It is desirable to.

The precoding modules 542 and 642 perform precoding by multiplying the determined phase shift based precoding matrix by a corresponding subcarrier of an OFDM symbol.

The embodiments described above are the components and features of the present invention are combined in a predetermined form. Each component or feature is to be considered optional unless stated otherwise. Each component or feature may be embodied in a form that is not combined with other components or features. It is also possible to combine some of the components and / or features to form an embodiment of the invention. The order of the operations described in the embodiments of the present invention may be changed. Some components or features of one embodiment may be included in another embodiment or may be replaced with corresponding components or features of another embodiment. It is obvious that the claims may be combined to form an embodiment by combining claims that do not have an explicit citation relationship in the claims or as new claims by post-application correction.

Embodiments according to the present invention may be implemented by various means, for example, hardware, firmware, software, or a combination thereof. In the case of a hardware implementation, an embodiment of the present invention may include one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), FPGAs ( field programmable gate arrays), processors, controllers, microcontrollers, microprocessors, and the like.

In the case of implementation by firmware or software, an embodiment of the present invention may be implemented in the form of a module, procedure, function, etc. that performs the functions or operations described above. The software code may be stored in a memory unit and driven by a processor. The memory unit may be located inside or outside the processor, and may exchange data with the processor by various known means.

It is apparent to those skilled in the art that the present invention can be embodied in other specific forms without departing from the spirit and essential features of the present invention. Accordingly, the above detailed description should not be interpreted as limiting in all aspects and should be considered as illustrative. The scope of the invention should be determined by reasonable interpretation of the appended claims, and all changes within the equivalent scope of the invention are included in the scope of the invention.

The present invention can be used in a multi-antenna system using a plurality of subcarriers.

Claims (8)

A method of performing precoding in a multi-antenna system using a plurality of subcarriers and using a phase shift based matrix, Performing precoding on a subcarrier or a virtual resource using the phase shift based matrix, The phase shift based matrix is composed of a product of a precoding matrix, a diagonal matrix for phase shift, and a unitary matrix, The codebook composed of the precoding matrix has four precoding matrices W2 (i) for rank 2, i = 0,1,2,3, and four precoding matrices W3 for rank 3. i) (where i = 0,1,2,3) and one precoding matrix W4 (i) for rank 4 (where i = 0), How to perform precoding. The method of claim 1, Among the W2 (i), W3 (i), and W4 (i), W2 (0), W3 (0), and W4 (0) are generated from a vector [1 -1 -1 1] ^T , and W2 (1 ) And W3 (1) are generated from the vector [1 -1 1 -1] ^T , W2 (2) and W3 (2) are generated from the vector [1 1 -1 -1] ^T , and W2 (3) and W3 (3) is generated from a vector [1 1 1 1] ^T. A method of performing precoding in a multi-antenna system using a plurality of subcarriers and using a phase shift based matrix, Performing precoding on a subcarrier or a virtual resource using the phase shift based matrix, The phase shift based matrix is composed of a product of a precoding matrix, a diagonal matrix for phase shift, and a unitary matrix, The codebook composed of the precoding matrix has a configuration in which the number of precoding matrices defined for the rank decreases as the rank value increases. How to perform precoding. The method of claim 3, The codebook has four precoding matrices W2 (i) for rank 2 (where i = 0,1,2,3) and three precoding matrices W3 (i) for rank 3 (where i With = 0,1,2, and with two precoding matrices W4 (i) for rank 4, i = 0,1, How to perform precoding. The method of claim 4, wherein Among the W2 (i), W3 (i), and W4 (i), W2 (0), W3 (0), and W4 (0) are generated from a vector [1 -1 -1 1] ^T , and W2 (1 ), W3 (1), W4 (1) are generated from the vector [1 -1 1 -1] ^T , W2 (2) and W3 (2) are generated from the vector [1 1 -1 -1] ^T , W2 (3) is generated from a vector [1 1 1 1] ^T. The method of claim 3, The codebook has four precoding matrices W2 (i) for rank 2 (where i = 0,1,2,3) and two precoding matrices W3 (i) for rank 3 (where i With = 0,1) and one precoding matrix W4 (i) for rank 4, with i = 0 How to perform precoding. The method of claim 6, Among the W2 (i), W3 (i), and W4 (i), W2 (0), W3 (0), and W4 (0) are generated from a vector [1 -1 -1 1] ^T , and W2 (1 ) And W3 (1) are generated from the vector [1 -1 1 -1] ^T , W2 (2) is generated from the vector [1 1 -1 -1] ^T , and W2 (3) is generated from the vector [1 1 1 1] A method of performing precoding, generated from ^T. A method of performing precoding in a multi-antenna system using a plurality of subcarriers and using a phase shift based matrix, Performing precoding on a subcarrier or a virtual resource using the phase shift based matrix, The phase shift based matrix is composed of a product of a precoding matrix, a diagonal matrix for phase shift, and a unitary matrix, The codebook composed of the precoding matrix has four precoding matrices W2 (i) for rank 2, i = 0,1,2,3, and four precoding matrices W3 for rank 3. i) (where i = 0,1,2,3) and four precoding matrices W4 (i) for rank 4 (where i = 0,1,2,3), Of the W2 (i), W3 (i), and W4 (i), W2 (0), W3 (0), and W4 (0) are generated from a vector [1 -1 -1 -1] ^T , and W2 ( 1), W3 (1), and W4 (1) are generated from the vector [1 1 -1 1] ^T , and W2 (2), W3 (2), and W4 (2) are generated from the vector [1 -1 1 1 ] Is generated from ^T , and W2 (3), W3 (3), and W4 (3) are generated from the vector [1 1 1 -1] ^T , How to perform precoding.