KR100913101B1 - Method for signal transmitting applying for MIMO - Google Patents

Abstract

In the present invention, a signal transmission control method applied to a MIMO (Multi-Input and Multi-Output) system includes a modulation and coding set by a transmitter on a stream corresponding to a plurality of antennas. And a step of determining the actual stream to be transmitted according to the modulation and coding combination, and transmitting the determined stream to a receiving side. In the input and multi output system, there is an excellent effect of reducing the control signal to perform more efficient communication.

Multiple Input Multiple Output (MIMO), Modulation and Code Combination (MCS), Antenna Reference Rate Control (PARC), Stream Reference Rate Control (PSRC)

Description

Signal transmission method applied to multiple input multiple output system {Method for signal transmitting applying for MIMO}

The present invention relates to a signal processing method applied to a multi-input and multi-output (MIMO) system, and to a signal processing method for transmitting a stream using only some of a plurality of channels.

Hereinafter, a channel coding method and a modulation method of a general multiple input multiple output system will be briefly described.

In the V-BLAST (Vertical Bell Laboratories Layered Space Time) system, one of the MIMO (Multi-Input Multi-Output) systems, the symbol to be transmitted to each antenna is Methods of differentiating the channel coding method and the modulation method include an antenna reference rate control (PARC) and a stream reference rate control (PSRC) method.

Reference documents in the prior art describing a method of receiving and processing symbols independently transmitted by each antenna in the prior art are as follows. P.W. Wolniansky, G. J. Foschini, G.D. Golden and R. A. Valenzuela, “V-BLAST: An Architecture for Realizing Very High Data Rates Over the Rich-Scattering Wireless Channel”, IEE Electronics Letters, vol. 35, no. 1, pp. 14-16, January, 1999.

1 is a block diagram illustrating an embodiment of a transmission side of a Per Antenna Rate Control (PARC) system. As shown in FIG. 1, the data stream undergoes a demultiplexing process before being transmitted through a plurality of transmit antennas. In this case, the number of bits allocated to each transmit antenna may vary according to a data rate transmitted through each antenna.

The data stream undergoes demultiplexing, becomes a sub-stream corresponding to each transmit antenna, and is mapped to symbols through coding and interleaving processes. These symbols are demultiplexed again prior to spreading again using respective spreading codes. Coding is performed only in the temporal dimension, and is not as robust coding as space-time coding used in single-rate systems. However, coding in the time domain has a function of removing post-decoding interference cancellation, thereby improving the performance of the receiver.

2 is a block diagram illustrating an embodiment of a receiving side of a Per Antenna Rate Control (PARC) system. If data is coded through demultiplexing as shown in FIG. 1, signals transmitted through each transmit antenna can be independently decoded. That is, as shown in FIG. 2, after despreading and demultiplexing the received signal, a signal for one antenna is detected, reallocated, deinterleaved, and decoded. Based on these decoded bits, the signal to this antenna is reconstructed and removed from the received signal stored in the buffer.

FIG. 3 is a block diagram illustrating a transmission side of a Per Stream Rate Control (PSRC) system according to an embodiment. As shown in FIG. 3, until the data stream weight is multiplied, it is the same as the transmitting side of the Per Antenna Rate Control (PARC). However, in Per Antenna Rate Control (PARC), each stream is mapped one-to-one to each antenna, whereas in Per Stream Rate Control (PSRC), the size of the number of transmit antennas in each stream is orthogonal. Each of the orthogonal weight vectors is multiplied, and the weighted streams are distributed to each antenna and transmitted. When looking at only one stream, it is transmitted in the same manner as Tx diversity, and multiple streams are superimposed on the same transmission antenna combination.

4 is a block diagram of an embodiment of a receiving side of a Per Stream Rate Control (PSRC) system. As shown in Fig. 4, since the transmitting end multiplies each stream by an orthogonal weight vector and transmits it through each antenna, when the receiving side multiplies the Hermitian of the weight vector, each stream can be received without interference. You will be able to. Accordingly, in the Per Stream Rate Control (PSRC) system, an interference cancellation process performed at the receiving side of the Per Antenna Rate Control (PARC) is not required.

According to the prior art, as many streams are transmitted according to the number of transmit antennas. Therefore, to support this, a control signal is required for each stream. In other words, as a control signal sent from the receiving side to the transmitting side, channel state information (CQI; channel quality information), HARQ-ACK/NACK, etc. for each stream for selection of modulation and code set (MCS) are provided by the receiving side. It is transmitted to the sending side.

Meanwhile, in the case of the Per Stream Rate Control (PSRC) method, transmission of a weight vector for each stream is additionally required, and modulation information and transmission for each stream from the transmitting side to the receiving side It is necessary to transmit control signals such as transport block size and HARQ processing information. As the number of streams increases, the control signal increases, and the receiving side requires a receiving antenna equal to or greater than the number of transmission streams.

In particular, in the case of the Per Stream Rate Control (PSRC) method, looking at the distribution of eigen values in a system with the number of transmit/receive antennas (4,4), only the eigen values of the two revisions are transmitted. For the other two, the modulation and code set (MCS) of almost the lowest data rate is selected or data transmission is not possible frequently.

In the prior art, the control signal is increased by performing transmission for all streams without taking this point into account, and thus there is an inefficient problem in data transmission.

The present invention has been proposed to solve the above problems, and in a MIMO (Multi Input and Multi Output) system, efficient communication is achieved by reducing a control signal by transmitting a stream using only some channels. It is an object of the present invention to provide a method for controlling a transmission signal to perform.

The present invention for achieving the above object is a signal transmission control method applied to a multi-input multiple-output (MIMO; Multi-Input and Multi-Output) system, a transmitting side modulates and modulates a stream corresponding to a plurality of antennas. And determining a coding combination (Modulation and Code Set), determining an actual stream to be transmitted according to the modulation and coding combination, and transmitting the determined stream to a receiving side.

In addition, in the present invention, in a signal transmission control method applied to a multi-input and multi-output (MIMO) system, a transmitting side calculates a weight vector for streams corresponding to a plurality of antennas. And determining a stream to be transmitted according to the weight vector, and transmitting the determined stream to a receiver.
In addition, as an aspect of the present invention, a signal transmission method applied to a MIMO (Multi-Input and Multi-Output) system includes the steps of transmitting a signal for measuring a beamforming parameter and a signal It may include receiving a beamforming parameter measured based on and transmitting a signal using at least two antennas using the beamforming parameter.
In addition, as another aspect of the present invention, a signal transmission method applied to a multi-input and multi-output (MIMO) system includes transmitting a signal for determining a beamforming parameter and an MCS to be used at a transmitting side. It may include receiving the measured beamforming parameter and MCS based on the step and the signal, and transmitting a signal using the beamforming parameter and the MCS.

As described above, the present invention has an excellent effect of enabling more efficient communication by reducing a control signal in a MIMO (Multi Input and Multi Output) system.

The above objects, features, and advantages will become more apparent through the following detailed description in connection with the accompanying drawings. Hereinafter, a preferred embodiment according to the present invention will be described in detail with reference to the accompanying drawings.

As described in the configuration of the prior art, on the transmitting side of the V-BLAST (Vertical Bell Laboratories Layered Space Time) system, each transmitting antenna does not use a separate signal processing or space-time code. Data is transmitted in different signals through different antennas. That is, the transmitting side performs only the function of transmitting data as different signals through different antennas. In addition, the receiving side needs to detect data through an appropriate signal processing procedure for the signal transmitted from each transmission antenna. The signal processing method at the receiving side can be summarized as follows.

In order to describe the signal processing method at the receiving side in detail, first, a multi-input multi-output (MIMO) system having M transmit antennas and N receive antennas is assumed. At this time, a signal vector transmitted differently through M transmit antennas is a When H is the channel matrix passed through before the transmission signal vector is received by the receiving side, it is received by the receiving side of the MIMO (Multi-Input Multi-Output) system with N receiving antennas. The signal vector r ₁ can be defined as in Equation 1.

r ₁ = Ha + v

,

,...

, 즉 M*1 벡터)은 각각 다른 채널

를 거치고, 수신측에서는 N 개의 안테나를 통해 신호가 수신된다. 수학식 1 에서 v 는 가우시안 잡음을 나타낸 것으로 수신측 각 안테나에 유기되므로 N+1 벡터가 된다.Since the transmitting side uses M antennas and the receiving side uses N antennas, signals transmitted from the M transmitting antennas are received by the N receiving antennas through different paths. Therefore, H , which is a channel matrix, becomes an N*M matrix. That is, signals transmitted differently through M antennas (

,

,...

, I.e. M*1 vector) is a different channel

After passing through, the signal is received through N antennas at the receiving side. In Equation 1, v denotes Gaussian noise and is induced by each antenna on the receiving side, so it becomes an N+1 vector.

,

,...

)를 수신측에서 검출한 신호 벡터를 s (

,

,...,

) 라고 나타낸다. 이렇게 송신측의 각각 다른 안테나에서 송신된 신호를 수신측에서 검출하기 위해 수신측의 N 개의 안테나에 가중치 벡터(weight vector)를 곱하게 되는데 이 가중치 벡터를 w 로 나타낸다. 이 때, 송신측의 안테나에서 각각 다른 신호가 송신되므로 수신측에서 이를 검출하기 위해서는 M 개의 가중치 벡터가 필요하게 된다. 수신측의 각 안테나에 곱해 줄 가중치 벡터는 수학식 2 와 같은 성질을 만족하여야 한다.The signal received through the N antennas in this way passes through a signal processing unit on the receiving side having the following signal search algorithm. First, the transmitted signal a vector (

,

,...

), the signal vector detected by the receiver is s (

,

,...,

). So there is to multiply the weight vector (weight vector) in the N antennas at the receiving end for a signal transmitted from different antennas on the transmission side to detect the receiving side represents the weight vector by w. At this time, since different signals are transmitted from the antennas of the transmitting side, M weight vectors are required to detect them at the receiving side. The weight vector to be multiplied by each antenna of the receiver must satisfy the same property as in Equation 2.

w _i ^H H _j = 0 (j≥i)

w _i ^H H _j = 1 (j=i)

In Equation 2, H _j represents the j-th column vector of H. That is, Equation 2 shows that the weight vector w _i to be multiplied by the receiving antenna in order to detect i transmission data has a value of 1 only for the product of H with the j-th column vector, and the product of H with the remaining column vector It indicates that it becomes 0 for. That is, the weight vector w _i for receiving data transmitted from the i-th transmit antenna is calculated to remove the influence of the signal transmitted from the other transmit antenna. Since the transmitted signals are sequentially detected, the influence of signals that have already been detected before obtaining a weight vector to be used for current detection is excluded.

A weight vector that satisfies the property of Equation 2 can be obtained as follows. First, the signal vector received at the receiving side shown in Equation 1 may be expressed as Equation 3.

r ₁ = a ₁ H ₁ + a ₂ H ₂ + ㆍㆍㆍ+ a _M H _M

Signals transmitted from each of the transmitting antennas are received by the receiving side through different channels, respectively, and Equation 3 expresses this in the form of a linear sum. As can be seen from Equation 3, when detecting the first transmission signal, it is preferable to remove the influence of the second to M-th signals and multiply the receiving antenna side by a weight vector that can be received. In addition, this principle can be applied to each of the other transmission signals. To satisfy this requirement, the weight vector can be updated as follows.

First, when the weight vector update starts, the Moore-Penrose pseudoinverse matrix of the first H matrix is obtained, and this matrix is expressed as H ⁺ or G ₁ . That is, it can be expressed as in Equation 4.

G ₁ = H ⁺

를 검출한다. 이렇게 K 번째 안테나에서 송신된 신호를 검출하면, 상기 수학식 3 에서 K번째 신호의 영향을 가감한다. 즉 수학식 5 와 같은 연산을 수행한다. Find out which row vector the vector norm of the row vector is the smallest among each row vector of the following G _{1 matrix.} If the row vector represented by the smallest value among the vector norms of the row vector is K , the weight vector w _i for detecting the K-th transmission signal is selected as the K-th row of the G _{1 matrix.} Then, the received vector r ₁ is multiplied by the weight vector w _i to detect the signal transmitted from the K-th antenna. At this time, since the receiving side already knows the modulation method used by the transmitting side (e.g., digital modulation method such as QPSK, QAM, etc.), it determines which constellation it belongs to and finally transmits it from the K-th transmitting antenna. signal

Is detected. When the signal transmitted from the K-th antenna is detected in this way, the influence of the K-th signal is added or subtracted in Equation 3 above. That is, the same operation as in Equation 5 is performed.

r ₂ = r ₁ - H _k

r ₂ = r ₁ - H _k

R ₂ in Equation 5 Represents the received vector to be used in the second update. After that, G ₂ , that is, the G matrix to be used to obtain the second weight vector becomes a Moore-Penrose pseudoinverse matrix of a matrix in which the K-th column of the H ^{+ matrix is all zeros.} That is, it can be written as in Equation 6.

=

=

는 H ⁺ 행렬 중 K 번째 열을 모두 0 으로 만든 행렬의 Moore-Penrose pseudoinverse 행렬을 의미한다. In Equation 6

Denotes a Moore-Penrose pseudoinverse matrix in which all K-th columns of the H ^{+ matrix are made 0.}

행렬에 대해서

행렬의 행벡터들의 벡터 놈(vector norm) 중에서 가장 작은 값으로 나타나는 행벡터가 어느 것인지를 알아내고, 이것을 V 라고 하면 V번째 송신 신호를 검출하기 위한 가중치 벡터 w _v 는

행렬의 V 번째 행으로 선택 한다. 그리고, 수신벡터 r ₂ 와 가중치 벡터 w _v 를 곱해서 V 번째 안테나에서 송신된 신호를 검출한다. 그러면, 송신측에서 사용하는 변조 방법 (예를 들면 QPSK, QAM 등의 디지털 변조 방식)을 수신측에서도 알고 있으므로 어느 위치(constellation)에 속한 것인지를 판단하여 마지막으로 V 번째 송신 안테나에서 송신된 신호

를 검출한다. 이와 같이 V 번째 안테나에서 송신된 신호를 검출하면 상기 수학식 5 에서 V 번째 신호를 가감한다. 즉 수학식 7 과 같은 연산을 수행한다.

About the matrix

Find out which row vector appears as the smallest value among the vector norm of the matrix row vectors, and if this is V , the weight vector w _v for detecting the V-th transmission signal is

Select as the V-th row of the matrix. Then, a signal transmitted from the V-th antenna is detected by multiplying the reception vector r ₂ and the weight vector w _v. Then, since the receiving side knows the modulation method used by the transmitting side (e.g., digital modulation method such as QPSK, QAM, etc.), it is determined which constellation belongs to and finally the signal transmitted from the V-th transmitting antenna.

Is detected. When the signal transmitted from the V-th antenna is detected as described above, the V-th signal is added or subtracted in Equation 5 above. That is, the same operation as in Equation 7 is performed.

r ₃ = r ₂ - H _ㅍ

r ₃ = r ₂ - H _ㅍ

In this case, r ₃ represents a received vector to be used in the third update. The above process is performed until all the signals transmitted from each of the M antennas are obtained.

Unlike this method of applying the same channel coding and modulation method to the symbols transmitted from each antenna, the channel coding and modulation methods of the symbols to be transmitted from each antenna are transmitted differently by reflecting the channel conditions of each antenna. A (PARC; Per Antenna Rate Control) method was proposed.

The antenna reference rate control (PARC) transmitting side may transmit independently encoded signals for each transmission antenna. In other words, the antenna reference rate control (PARC; Per Antenna Rate Control) is the previous single rate (Single Rate) Multiple Input Multiple Output (MIMO) technology in that data transmission rates (Modulation and Coding) may be different for each antenna. It is different from BLAST. The Per Antenna Rate Control (PARC) enables more precise control in controlling the data transmission rate independently for each antenna, thereby increasing the throughput of the entire system.

In this case, the number of bits required to inform the state of each antenna channel is required more than the techniques proposed for single rate multiple input multiple output (MIMO), but a reference set can be determined. That is, in the antenna reference rate control (PARC; Per Antenna Rate Control), the signal-to-interference noise ratio (SINR) of each transmitting antenna received from the receiving antenna is determined to determine an effective modulation and coding combination (MCS) in each antenna. ; Calculate Signal to Interference Noise Ratio). At this time, to select the channel coding and modulation method used by each antenna, measure the signal to interference noise ratio (SINR) of the signal received from each antenna, and based on the value, the channel coding to be used by each antenna. Choose a combination of and modulation methods.

Table 1 shows an example of a combination of a transmission rate and a modulation and coding set (MCS) corresponding to the (4,4) system.

bps/Hz Data rate (Mbps) Modulation method Coding rate 3 7.2 16 QAM 3/4 2 4.8 16 QAM 1/2 1.5 3.6 QPSK 3/4 One 2.4 QPSK 1/2 0.5 1.2 QPSK 3/4

(4,4) In the case of per antenna rate control (PARC), in the highest geometries (indexes 1 to 38), the code re-use order is selected as 4, For lower geometries (indexes 39 to 54), selection and code re-use order 2 are used at the same time. And, for the lowest geometries, selective diversity and a single antenna transmission scheme are used.

Table 2 shows an embodiment of a transmission rate combination in a system using a transmit antenna and four receive antennas.

index Data rate:Mbps 1st antenna 2nd antenna 3rd antenna 4th antenna One 28.8 3 3 3 3 2 26.4 3 3 2 3 3 26.4 3 2 3 3 4 26.4 2 3 3 3 5 24.0 2 3 3 2 6 24.0 2 3 2 3 7 24.0 2 2 3 3 8 21.6 2 2 3 2 9 21.6 2 2 2 3 10 19.2 2 2 2 2 11 22.8 2 1.5 3 3 12 20.4 2 1.5 2 3 13 18.0 2 1.5 2 2 14 19.2 2 1.5 1.5 2 15 16.8 2 One 2 2 16 25.2 1.5 3 3 3 17 22.8 1.5 3 2 3 18 22.8 1.5 2 3 3 19 20.4 1.5 2 2 3 20 18.0 1.5 2 2 2 21 19.2 1.5 2 1.5 2 22 21.6 1.5 1.5 3 3

23 21.6 1.5 1.5 3 3 24 16.8 1.5 1.5 2 2 25 14.4 1.5 1.5 2 One 26 15.6 1.5 1.5 1.5 2 27 15.6 1.5 One 2 2 28 24.0 One 3 3 3 29 21.6 One 3 2 3 30 21.6 One 2 3 3 31 19.2 One 2 2 3 32 16.8 One 2 2 2 33 15.6 One 2 2 1.5 34 15.6 One 2 1.5 2 35 18.0 One 1.5 2 3 36 15.6 One 1.5 2 2 37 20.4 0.5 2 3 3 38 15.6 0.5 2 2 2 39 14.4 3 3 40 14.4 3 3 41 12.0 2 2 42 12.0 3 2 43 12.0 2 3 44 12.0 2 3 45 9.6 2 2 46 9.6 2 2 47 8.4 2 1.5 48 8.4 2 1.5 49 10.8 1.5 3 50 10.8 1.5 3 51 8.4 1.5 2 52 8.4 1.5 2 53 9.6 One 3 54 9.6 One 3 55 7.2 3 56 7.2 3 57 4.8 2 58 4.8 2 59 3.6 1.5 60 3.6 1.5 61 2.4 One 62 2.4 One 63 1.2 0.5 64 1.2 0.5

Hereinafter, a description will be given of per stream rate control (PSRC). When using M transmit antennas, the transmitting side constructs a signal vector s composed of M symbols as follows, and transmits each symbol through different transmit antennas.

s = [, ,ㆍㆍㆍ ]^T

s = [, ,ㆍㆍㆍ] ^T

The Per Stream Rate Control (PSRC) is a method for transmitting each symbol through an independent channel in a situation in which correlation between antennas and correlation between symbols exist in an actual communication situation. That is, taking into account that the transmitting side is composed of an antenna array, each symbol is multiplied by an eigenvector of the channel matrix to be transmitted.

On the other hand, after examining the size of each eigenvalue value through eigenvalue decomposition of the channel matrix estimated by the receiving side of the multiple input/output communication system, use a higher order modulation method such as 64QAM or 16QAM for an antenna with a large eigenvalue value. And, for an antenna having a relatively small eigenvalue, a low order modulation method such as binary phase shift keying (BPSK) or quadrature phase shift keying (QPSK) is used.

That is, instead of using the same modulation scheme for all of the symbols transmitted from each transmit antenna, the channel condition of each antenna is fed back from the terminal to determine the modulation scheme to be used by each antenna according to the channel condition. At this time, the receiving side must also have the same algorithm to know the modulation method used by each antenna when receiving.

In addition, a symbol is not transmitted for an antenna whose eigenvalue values obtained through the eigenvalue decomposition of the channel matrix are smaller than a specific value, thereby eliminating the possibility that the symbol transmitted from the transmitting antenna may cause an error at the receiving side. However, even if this method is used, since symbols will be allocated to use a higher order modulation scheme to an antenna having a good channel condition, the number of bits to be transmitted in all transmit antennas increases, which rather helps to increase system throughput.

In addition, in an antenna that is useless to transmit a symbol, the overall communication quality can be improved by processing the symbol in advance so that the symbol is not transmitted. That is, after estimating the channel matrix at the receiving side, the receiving side feeds back each eigenvalue separated by eigenvalue decomposition to the transmitting side, or by comparing the eigenvalue values, the transmitting and receiving side has a predetermined modulation scheme allocation table, An index corresponding to the modulation method to be used may be fed back.

A method of forming a beam according to each symbol in Per Stream Rate Control (PSRC) may be expressed as Equation 9.

S = w ₁ s ₁ + w ₂ s ₂ + ㆍㆍㆍ+ w _M s _M

Here, w _i represents a weight vector for beamforming each symbol, s ₁ to s _{M are} data symbols, and S is a signal vector after beam formation for each symbol.

The symbol transmitted from each transmit antenna receives feedback from the terminal of the channel condition of each antenna and determines a modulation scheme to be used by each antenna according to the channel condition. Therefore, s ₁ to s _M in Equation 9 use different modulation schemes by comparing the eigenvalues of the channel matrix. That is, through the eigenvalue decomposition of the channel matrix, a higher order modulation method is allocated to an antenna having a large eigenvalue value, and a lower order modulation method is allocated to an antenna having a relatively small eigenvalue value.

In this case, a method of obtaining a weight vector to be multiplied by each symbol is as follows. First, a multiple-input multiple-output (MIMO) system with M transmit antennas and N receive antennas is assumed. At this time, when H is the mobile communication channel matrix through which signal vectors transmitted differently through M transmit antennas are received before being received by the receiving side, when the receiving side has M receiving antennas, the channel matrix H is N*M. It becomes a matrix. When each antenna of the transmitting side transmits a pilot symbol or a separate pilot channel known in advance by the transmitting/receiving side, the receiving side can estimate each component of the channel matrix H.

Eigen-Decomposition is performed on the channel matrix H at the receiving side. In the present invention, since a system in which the number of antennas in the antenna array on the transmitting side is larger than the number of antennas in the antenna array on the receiving side is assumed, the channel matrix H cannot be a square matrix, and the channel matrix itself You cannot perform eigenvalue decomposition. Therefore, the eigenvalue decomposition of H ^H H is performed as shown in Equation 10. Here, H ^H represents the Hermitian operation on the vector H.

H ^H H = λ ₁ e ₁ + λ ₂ e ₂ + ㆍㆍㆍλ _M e _m

In Equation 10, λ _i represents eigen-values of the matrix H ^H H , and e _i represents eigen-vectors. Since each eigen-vector is generally orthogonal to each other, when a symbol is to be transmitted at once according to the number of antennas on the transmitting side, a signal can be transmitted by multiplying each symbol by an independent weight vector.

As shown in Equation 9, when the transmitting side performs beamforming for each symbol and transmits the signal, the receiving side performs the following signal processing process. In other words, since the signal is transmitted by multiplying each symbol by an independent weight vector, the signal processing of the conventional BLAST receiver cannot be used. Therefore, a zero-forcing or minimum mean square error (MMSE) method is used. First , after estimating S , each symbol is detected by multiplying the conjugate value of the weight vector multiplied by each symbol at the transmitting side. Zero-forcing, MMSE can be summarized in the following equation.

First, each symbol is multiplied by a weight vector and a signal is transmitted by the transmitting side, and the receiving side receives the signal, and the received signal is as shown in Equation 11.

R = HS + n

In Equation 11, n denotes additive white Gaussian noise (AWGN).

라고 하면

는 수학식 12 와 같이 나타낼 수 있다. A signal vector obtained by estimating the transmitted signal vector S by multiplying each symbol by a weight vector using zero-forcing is obtained.

If you say

Can be expressed as in Equation 12.

= [ H ^H H ]^-1 H ^H R

= [ H ^H H ] ^-1 H ^H R

는 수학식 13 과 같이 나타낼 수 있다.Meanwhile, a signal vector transmitted by multiplying each symbol by a weight vector using a minimum mean square error (MMSE) method

Can be expressed as in Equation 13.

= [α I + H ^H H ]^-1 H ^H R

= [α I + H ^H H ] ^-1 H ^H R

In Equation 13, α denotes a signal-to-interference noise ratio, and I denotes an identity matrix.

를 추정한 후, 송신측에서 각 심볼에 곱하여 전송한 가중치 벡터의 공액(conjugate) 값을 다시

에 곱하여 전송단에서 전송한 심볼들인 s₁ 부터 s_M까지의 추정치

부터

를 수학식 14 에 따라 추정할 수 있다. In this way, a signal vector transmitted by multiplying each symbol by a weight vector using a zero-forcing or minimum mean square error (MMSE) method.

After estimating, the conjugate value of the weight vector transmitted by multiplying each symbol by the transmitting side is again

The estimated value from s _{1 to s M} , which are symbols transmitted by the transmitter by multiplying by

from

Can be estimated according to Equation 14.

= w ₁ ^H ㆍㆍㆍ _M = w _M ^H

= w ₁ ^H ㆍㆍㆍ _M = w _M ^H

부터

을 각각 사용된 변조(modulation) 방법에 맞게 복조(demodulation)하여 각 심볼로 할당하기 전의 비트들을 알아내고, 한편 각 안테나에서 전송된 심볼의 비트들을 알아낸 다음 멀티플렉싱을 통해 송신측에서 전송한 비트 스트림을 알아낸다.Estimated after this

from

The bit streams transmitted from the transmitting side through multiplexing after finding the bits of the symbols transmitted from each symbol by demodulating them according to the respective modulation methods used. Find out.

In addition, when each antenna transmits different symbols, the eigen-vectors of the channel matrix estimated by the receiving side are multiplied by each symbol and transmitted, and the eigenvalue decomposition at the receiving side to consider the channel conditions of each antenna together. By comparing the eigenvalues of the channel matrix obtained through Relatively, each antenna selects a channel coding and modulation method of a symbol to be transmitted.

In order to obtain the Joule weight vector by multiplying the symbol to be transmitted independently from each transmit antenna, the eigenvalue decomposition of the channel matrix estimated by the receiver must be performed. ) Can be obtained together, so no additional calculation is required. That is, instead of using the same modulation method for all symbols transmitted from each transmit antenna, the channel condition of each antenna is fed back from the terminal to determine the modulation method to be used by each antenna according to the channel condition. do.

Therefore, s ₁ to s _M in Equation 2 use different modulation schemes by comparing the eigenvalues of the channel matrix. That is, through the decomposition of the eigenvalues of the channel matrix, a higher order modulation method is assigned to an antenna with a large eigenvalue value, and a lower order modulation method is assigned to an antenna with a relatively small eigenvalue value. do.

In this case, a method of determining a modulation and code set (MCS) to be allocated to each antenna can be described as follows. As in Equation 10, in order to determine a modulation scheme to be allocated to each antenna with eigenvalue values obtained through eigenvalue decomposition, a relative ratio of each eigenvalue value is first determined.

Through this relative ratio, if the smallest eigen-value is less than or equal to a specific threshold, the antenna prevents the symbol from being transmitted, and the minimum eigen-value excluding the excluded eigen-value. value), the lowest modulation scheme and coding combination is used. Meanwhile, for the antenna having the largest eigen-value, the highest order modulation scheme and coding combination is used.

For intermediate values, an intermediate modulation and coding combination is used through comparison of relative values. In addition, it is also possible to use a combination of modulation and coding according to the channel situation after having the necessary criteria as shown in Table 2 in advance after confirming various channel conditions through simulation in advance through an experiment.

Otherwise, the types of modulation and coding that can be actually used will be limited, so if the combination is determined as shown in Table 1, the combination of modulation and coding to be used in each antenna is indexed because both the transmitting and receiving sides know the same criterion. You might be able to tell. Alternatively, the entire eigenvalue values may be slowly fed back in consideration that the eigenvalue values of the channel matrix change relatively slowly. In this case, modulation and coding to be used in each antenna must be determined with the eigenvalue returned from the transmitting side.

A multiple-input multiple-output (MIMO) system with the number of transmit/receive antennas (M, N) uses a method based on transmission of M streams. That is, if the number of transmission antennas is M, the number of channels that can be transmitted is M, but it does not often occur when all channels are good at every moment. Accordingly, the present invention is based on transmission of P (P<M) streams out of M.

That is, P or less streams are transmitted using only P channels, and a signal required to control this is P independent control signals and index information of a channel selected from M channels. Accordingly, while M independent control signals are required in the conventional method, the method according to the present invention requires only P control signals and an index of a channel selected from M number of control signals.

Equation 15 shows an embodiment in which the present invention is applied to a Per Stream Rate Control (PSRC) method.

S = w ₁ s ₁ + w ₂ s ₂ + ㆍㆍㆍ+ w _p s _p

In Equation 15, w _p corresponds to the P-th in the order of the size of the eigen-value corresponding to the eigen-vector of the channel matrix. That is, eigen-vectors whose corresponding eigen-values are after the Pth in order of magnitude are excluded from use.

The present invention can significantly reduce complex control signals by limiting the number of transport streams to less than the effective number of streams in a multiple-input multiple-output (MIMO) system using a method of varying modulation and code combination (MCS) for each stream.

In particular, in a system such as PSRC, the use of channels with low eigen-values is excluded in advance, so that throughput loss is very low, and the amount of transmission of the weight vector transmitted to the transmitter can be reduced. Therefore, it is possible to reduce a feedback delay or quantization error, which may be a problem in actual implementation. In addition, since the number of transport streams is reduced from M to P, the number of reception antennas of the terminal for receiving them may be reduced to P. Assuming that the number of transmit/receive antennas is (4, 4), if only two streams are transmitted, the amount of the control signal can be reduced to 1/2, and the number of receiving antennas can be reduced to two.

The method of the present invention as described above may be implemented as a program and stored in a recording medium (CD-ROM, RAM, floppy disk, hard disk, magneto-optical disk, etc.) in a form that can be read by a computer. This process can be easily performed by a person of ordinary skill in the art to which the present invention pertains, and thus will not be described in detail.

The present invention described above is capable of various substitutions, modifications, and changes without departing from the technical spirit of the present invention to those of ordinary skill in the art. It is not limited by the drawings.

1 is a block diagram illustrating an embodiment of a transmission side of an antenna reference rate control (PARC) system.

FIG. 2 is a block diagram illustrating an embodiment of a receiving side of a Per Antenna Rate Control (PARC) system.

3 is a block diagram of an embodiment of a transmission side of a Per Stream Rate Control (PSRC) system.

Figure 4 is a configuration diagram showing an embodiment of a receiving side of a stream reference rate control (PSRC; Per Stream Rate Control) system.

Claims (7)

As a transmission control method applied to a multi-input multi-output (MIMO) system, Transmitting a signal used for channel matrix measurement for the M transmit antennas through the M transmit antennas, in a transmitter having M transmit antennas; Receiving, at the transmitter, information about the channel matrix estimated by the receiver receiving the transmitted signal based on the transmitted signal; And At the transmitter, selecting P transmit antennas less than M of the M transmit antennas based on eigenvalues of the channel matrix; And In the transmitter, a second transmission step of multiplying P streams by a weight vector for the selected P transmit antennas and transmitting through the selected P transmit antennas Including, Modulation and coding schemes are independently applied to the selected P transmit antennas. Transmission control method. The method of claim 1, In the second transmission step, weight vectors for the selected P transmit antennas are further transmitted, The eigen value is calculated by performing eigen decomposition on the channel matrix, and the weight vector is calculated from the eigen vector calculated by performing the eigen decomposition. Transmission control method. As a transmission control method applied to a multi-input multi-output (MIMO) system, Receiving at a receiver a signal transmitted from a transmitter having M transmit antennas, the signal being used for channel matrix measurement for the M transmit antennas; Sending, at the receiver, information about the channel matrix estimated based on the received signal to the transmitter; And Receiving, at the receiver, P streams from P transmit antennas smaller than M of the M transmit antennas of the transmitter, wherein the P streams are multiplied by weight vectors for the P transmit antennas, Second receiving step Including, The P transmit antennas are selected at the transmitter based on the eigenvalues of the channel matrix, Modulation and coding schemes are independently applied to the selected P transmit antennas. Transmission control method. The method of claim 3, In the second receiving step, a weight vector for the P transmit antennas is further received. The eigen value is calculated by performing eigen decomposition on the channel matrix, and the weight vector is calculated from the eigen vector calculated by performing the eigen decomposition. Transmission control method. delete delete delete

Images