KR101400855B1 - Apparatus and method for calculating channel quality information per stream in multiple input multiple output wireless communication system - Google Patents

Abstract

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to generation of channel quality information (CQI) for each stream in a multiple input multiple output (MIMO) wireless communication system and includes a receiver for receiving a signal from a transmitter through a plurality of antennas, An estimator for estimating a channel for each antenna with the transmitter and constructing a channel matrix, and a generator for generating channel-specific channel quality information by deriving an effective noise for each stream using a lattice reduction technique Feedback information for a closed loop (CL) multi-input / output system can be generated by generating channel-specific channel quality information using effective noise calculated using the lattice reduction technique.

(MIMO), channel quality information (CQI), signal to interference and noise ratio (SINR), lattice reduction, effective noise noise)

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to an apparatus and a method for generating channel quality information per stream in a MIMO wireless communication system,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a MIMO (Multiple Input Multiple Output) wireless communication system, and more particularly, to an apparatus and method for generating stream-specific channel quality information in a MIMO wireless communication system.

2. Description of the Related Art [0002] Recently, as a demand for high-speed and high-quality data transmission has increased, a MIMO (Multiple Input Multiple Output) wireless communication system using a plurality of transmitting and receiving antennas has been attracting attention. The MIMO technique performs communication using a plurality of streams through a plurality of antennas, thereby greatly improving channel capacity as compared with the case of using a single antenna. For example, if both the transmitting and receiving ends use M transmit and receive antennas, the channel between each antenna is independent, and the bandwidth and total transmit power are fixed, the average channel capacity increases by M times as compared to a single antenna.

Recently, the use of a CL (Close Loop) MIMO system has been considered. In the closed loop MIMO system, the transmitter acquires the channel state of the receiver and determines a modulation and coding scheme (MCS) level for each stream based on the channel state of the receiver. To this end, the receiver feeds back channel quality information (CQI) for each stream to the transmitter. Therefore, the receiver must generate channel quality information for each stream using its channel information.

When the receiving end uses the Minimum Mean Square Error (MMSE) detection technique or the MMSE-OSIC (MMSE Ordered Successive Interference Cancellation) detection technique, the per-stream channel quality information (e.g., Signal to Interference and Noise Ratio (SINR) Ratio) generation is performed easily. On the other hand, when the receiving end uses the ML (Maximum Likelihood) detection technique or the lattice-reduction-aided detection technique, since the signals for each stream are bundled and detected as one unit, Generation is very difficult. Therefore, in order to apply the ML scheme or the grating reduction utilization scheme to the closed loop MIMO system, an alternative is needed to generate feedback information suitable for the ML scheme or the grating reduction utilization scheme.

It is therefore an object of the present invention to provide an apparatus and method for generating stream quality channel quality information (CQI) for a closed loop (CL) scheme in a multiple input multiple output (MIMO) Method.

It is another object of the present invention to provide an apparatus and method for generating per-stream channel quality information for a lattice-reduction-aided detection technique in a multi-input / output wireless communication system.

It is still another object of the present invention to provide an apparatus and method for generating per-stream channel quality information for a maximum likelihood (ML) detection technique in a MIMO wireless communication system.

According to an aspect of the present invention, there is provided a receiving apparatus in a multiple input multiple output (MIMO) wireless communication system including a receiver for receiving a signal from a transmitting terminal through a plurality of antennas, An estimator for estimating a channel for each antenna with the transmitter and constructing a channel matrix and an effective noise for each stream when a lattice reduction technique is applied, And a generator for generating channel quality information (CQI) for each stream by using the channel quality information.

According to a second aspect of the present invention, there is provided a method of generating channel quality information for each stream in a MIMO wireless communication system, comprising: receiving a signal from a transmitter through a plurality of antennas; Generating a channel matrix by estimating a channel for each antenna with the transmitting end; deriving an effective noise for each stream when a lattice reduction technique is applied; generating channel quality information for each stream using the error component for each stream; The method comprising the steps of:

delete

delete

Channel quality information (CQI) for each stream is generated by using effective noise calculated using a lattice reduction technique in a multiple input multiple output (MIMO) wireless communication system , And closed loop (CL) multi-input / output systems.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Reference will now be made in detail to the preferred embodiments of the present invention, examples of which are illustrated in the accompanying drawings. In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

The present invention relates to a technique for generating channel-specific channel quality information for a lattice-reduction-aided detection technique and ML (Maximum Likelihood) detection technique in a multiple input multiple output (MIMO) .

The process of generating channel quality information for each stream according to the present invention proceeds in three stages. The present invention is characterized in that the first step includes scaling a modulation constellation, a step of calculating a symbol error rate (SER) of each stream, and a step of calculating a symbol error rate (SR) And proceeds to the step of determining the final channel quality information. In the second stage, the per-stream symbol error rate is calculated for each possible modulation scheme combination. At this time, a symbol error rate for each stream is calculated by a different method depending on the type of lattice reduction matrix corresponding to the combination of modulation schemes. Hereinafter, the present invention will be described with respect to a process of generating channel quality information for each stream according to the present invention, using Equation (1).

로 스케일링되면, 채널행렬의 k번째 열은

로 스케일링된다. 즉, 채 널행렬

의 k번째 열인

는

로 변경된다. 스케일링 값(factor)

는 변조방식에 따라 다르게 결정된다. 그리고, 스케일링된 변조 성상도는 각 점의 값이 정수가 되도록 쉬프트(shift)된다. 예를 들어, 도 1의 (a)와 같이 4개의 점

,

,

,

을 포함하는 성상도를 가정할 때, 상기 도 1의 (a)와 같은 성상도는 상기 도 1의 (b)와 같이 스케일링 및 쉬프팅된다.First, the constellation is scaled such that the distance between neighboring points in the constellation is '1'. Accordingly, the channel matrix is also scaled. For example, if the modulation constellation of the k < th >

, The kth column of the channel matrix < RTI ID = 0.0 >

Lt; / RTI > That is,

The kth column of

The

. Scaling factor

Is determined differently depending on the modulation method. Then, the scaled modulation constellation is shifted so that the value of each point becomes an integer. For example, as shown in Fig. 1 (a), four points

,

,

,

The constellation shown in FIG. 1 (a) is scaled and shifted as shown in FIG. 1 (b).

개의 스트림들이 존재하고,

개의 변조방식이 사용가능할 때, 총

가지의 변조방식 조합(combination)들이 발생한다. 이하 설명에서, 상기

가지의 변조방식 조합들을 하나의 집합으로 구성하고, i번째 변조방식 조합을

라 정의한다. 상기

는

이며,

는 i번째 변조방식 조합에서 k번째 스트림에 대응되는 변조방식을 의미한다.

There are two streams,

When modulation schemes are available,

Combinations of modulation schemes of branches occur. In the following description,

The combination of the modulation methods of the branch is formed into one set, and the combination of the i-th modulation method

. remind

The

Lt;

Denotes a modulation scheme corresponding to the k-th stream in the i-th modulation scheme combination.

을 가정할 때, 스케일링된 채널행렬 및 스케일링된 송신신호 각각은 하기 <수학식 1>과 같이 표현된다.Any combination of modulation schemes

, The scaled channel matrix and the scaled transmission signal are expressed by Equation (1).

는 스케일링된 채널행렬, 상기

는 i번째 변조방식 조합에 대한 채널행렬의 m번째 열에 대한 스케일링 값, 상기

은 채널행렬의 m번째 열, 상기

는 스케일링된 송신신호, 상기

은 m번째 스트림을 통한 수신신호, 상기

는 쉬프팅 값을 의미한다. 여기서, 상기

는 복소수(complex number)이다.In Equation (1) above,

Is a scaled channel matrix,

Is the scaling value for the m < th > column of the channel matrix for the i-th modulation scheme combination,

Is the m-th column of the channel matrix,

A scaled transmit signal,

Is the reception signal through the m &lt; th &gt; stream,

Means the shifting value. Here,

Is a complex number.

The received signal can be expressed by Equation (1).

는 스케일링된 수신신호, 상기

은 수신신호, 상기

는 쉬프팅 값, 상기

은 스트림 개수, 상기

는 총 송신 전력, 상기

는 스케일링된 채널행렬, 상기

는 스케일링된 송신신호, 상기

은 잡음을 의미한다.In Equation (2) above,

A scaled received signal,

A received signal,

The shifting value,

The number of streams,

Total transmission power,

Is a scaled channel matrix,

A scaled transmit signal,

Means noise.

When the lattice reduction technique is applied, Equation (2) can be expressed as Equation (3) below.

는 스케일링된 수신신호, 상기

은 스트림 개수, 상기

는 총 송신 전력, 상기

는 스케일링된 채널행렬, 상기

는 격자 감소 행렬, 상기

는 가판정(tentative decision) 대상 신호, 즉, 격자 감소 기법에 의한 검출 대상 신호, 상기

은 잡음을 의미한다. 여기서, 상기

는

이며, 상기

및 상기

내 모든 원소들의 실수부 및 허수부는 정수이다.In Equation (3) above,

A scaled received signal,

The number of streams,

Total transmission power,

Is a scaled channel matrix,

A grid reduction matrix,

A tentative decision target signal, that is, a detection target signal by a lattice reduction technique,

Means noise. Here,

The

, And

And

The real and imaginary parts of all of the elements are integers.

를 검출한 경우의 오류 확률이 필요하다. 본 발명은 상기

를 검출하기 위한 기법으로서 ZF(Zero Forcing) 검출 기법 또는 SQRD(Sorted QR Decomposition) 검출 기법을 고려하며, 이하 본 발명은 검출 기법에 따라 구분되어 설명된다.At this time,

Error probability is required. The present invention relates to

ZF (Zero Forcing) detection technique or SQRD (Sorted QR Decomposition) detection technique is considered as a technique for detecting the QPSK signal. Hereinafter, the present invention will be described according to the detection technique.

The case where the ZF detection technique is used will be described as follows.

As a result of performing the lattice reduction technique, a lattice reduction matrix is obtained. At this time, different effective channel matrices are generated depending on the modulation scheme combination, and different lattice reduction matrices are obtained depending on what is the effective channel matrix. At this time, the present invention distinguishes a lattice reduction matrix having the form of a unit matrix and a lattice reduction matrix having no unit matrix form, and applies different schemes according to the type.

First, a case where the lattice reduction matrix is not an identity matrix will be described as follows.

를 가판정하면 하기 <수학식 4>와 같다.If the lattice reduction matrix is not a unitary matrix, we use the ZF detection scheme

Is expressed as Equation (4) below.

는 격자 감소 기법에 의한 검출 대상 신호의 검출 값, 상기

는 반올림 연산자, 상기

은 스트림 개수, 상기

는 총 송신 전력, 상기

는 격자 감소 행렬, 상기

는 스케일링된 채널행렬, 상기

는 격자 감소 기법에 의한 검출 대상 신호, 상기

은 잡음, 상기

은 유효 잡음을 의미한다. 여기서, 상기

은 가우시안(Guassian) 분포를 따르며, 상기

의 평균은 0, 상기

의 분산은

이다.In Equation (4) above,

The detection value of the detection target signal by the lattice reduction technique,

Is a rounding operator,

The number of streams,

Total transmission power,

A grid reduction matrix,

Is a scaled channel matrix,

A signal to be detected by the lattice reduction technique,

Noise,

Means effective noise. Here,

Is in accordance with the Guassian distribution,

Is 0,

The dispersion of

to be.

In the detection result shown in Equation (4), the only element that affects the detection error is effective noise, and the effective noise is expressed by Equation (5).

은 유효 잡음, 상기

은 스트림 개수, 상기

는 총 송신 전력, 상기

는 격자 감소 행렬, 상기

는 스케일링된 채널행렬, 상기

은 잡음을 의미한다. 여기서, 상기

은 가우시안 분포를 따르며, 상기

의 평균은 0, 상기

의 분산은

이다.In Equation (5) above,

Is an effective noise,

The number of streams,

Total transmission power,

A grid reduction matrix,

Is a scaled channel matrix,

Means noise. Here,

Is in accordance with the Gaussian distribution,

Is 0,

The dispersion of

to be.

The transmission signal is determined according to Equation (6) based on Equation (4).

는 검출된 송신신호, 상기

는 격자 감소 행렬, 상기

는 격자 감소 기법에 의한 검출 대상 신호의 검출 값, 상기

는 격자 감소 기법에 의한 검출 대상 신호, 상기

는 반올림 연산자, 상기

는 스케일링된 송신신호, 상기

은 유효 잡음을 의미한다.In Equation (6) above,

The transmitted signal,

A grid reduction matrix,

The detection value of the detection target signal by the lattice reduction technique,

A signal to be detected by the lattice reduction technique,

Is a rounding operator,

A scaled transmit signal,

Means effective noise.

In calculating the error probability for the transmission signal detection, the case where the transmission signal is at the interior constellation point and the case where the transmission constellation point is at the exterior constellation point are processed in different ways. Here, the external constellation point refers to points located at the outermost among the constellation points of the modulation method, and the constellation point refers to points except for the external constellation point. For example, in the case of a modulation scheme having four constellation points such as Quadrature Phase Shift Keying (QPSK), the scaling and shifted constellation points are as shown in FIG. (A) 201, B (203), C (205), and C (205) which are located at the outermost positions are referred to as A (201), B D (207) are external constellation points. At this time, since there are no constellation points except for A (201), B (203), C (205) and D (207), in the case of the QPSK modulation scheme, Do not.

First, if the transmission signal is one of the constellation constellation points, the detection error probability of the k-th stream is given by Equation (7).

는 k번째 스트림에 대한 송신신호의 검출 값, 상기

는 상기

의 오류 확률, 상기

는 격자 감소 행렬의 k번째 행, 상기

는 반올림 연산자, 상기

은 유효 잡음, 상기

는 사건 A가 일어날 확률을 의미한다.In Equation (7) above,

Is a detection value of a transmission signal for the k < th > stream,

Quot;

The error probability of

Is the kth row of the lattice reduction matrix,

Is a rounding operator,

Is an effective noise,

Is the probability that event A will occur.

If the transmitted signal is one of the outer constellation points, the detection error probability of the kth stream is given by Equation (8).

는 k번째 스트림에 대한 송신신호의 검출 값, 상기

는 상기

의 검출 오류 확률, 상기

는 k번째 스트림 에 대한 송신신호, 상기

는 격자 감소 행렬의 k번째 행, 상기

은 유효 잡음, 상기

는 사건 A가 일어날 확률을 의미한다.In Equation (8) above,

Is a detection value of a transmission signal for the k < th > stream,

Quot;

The detection error probability of

Is a transmission signal for the k &lt; th &gt; stream,

Is the kth row of the lattice reduction matrix,

Is an effective noise,

Is the probability that event A will occur.

이 가능한 영역의 정수 점(any integer point in feasible region)이 되는 확률이 외부 성상도 점에 대한 k번째 스트림의 검출 성공 확률이다. 여기서, 가능한 영역은 해당 외부 성상도 점을 가장 가까운 외부 성상도 점으로 갖는 외부 성상도 점 외의 정수 점들을 모두 포함하는 영역을 의미한다. 예를 들어, 상기 도 2에서, 점 C(205)의 가능한 영역은 빗금친 영역이다.According to Equation (8) above,

The probability that a certain integer point in feasible region is the probability of detection of the kth stream for the external constellation point. Here, the possible region means an area including all the integer constellation points having the outer constellation point closest to the outer constellation point. For example, in FIG. 2, a possible region of the point C 205 is a hatched region.

As shown in Equation (7) and Equation (8), the error probability is calculated according to the position of the transmission signal. Accordingly, the detection error probability of each of the possible transmission signals in the k-th stream is averaged to thereby calculate the detection error probability of the k-th stream. This can be expressed by Equation (9).

는 k번째 스트림에 대한 송신신호의 검출 값, 상기

는 상기

의 오류 확률, 상기

는 조건 A에서 상기

의 검출 오류 확률, 상기

는 k번째 스트림에 대한 송신신호, 상기

는 i번째 변조방식 조합 중 k번째 스트림의 변조방식의 성상도에서 내부 성상도 점들의 집합, 상기

는 i번째 변조방식 조합 중 k번째 스트림의 변조방식의 성상도에서 외부 성상도 점들의 집합, 상기

는 i번째 변조방식 조합 중 k번째 스트림의 변조방식의 성상도 내 점의 개수를 의미한다.In Equation (9) above,

Is a detection value of a transmission signal for the k < th > stream,

Quot;

The error probability of

Lt; RTI ID = 0.0 &gt;

The detection error probability of

Is a transmission signal for the k &lt; th &gt; stream,

Denotes a set of inner constellation points in the constellation of the modulation scheme of the k-th stream among the i-th modulation scheme combination,

Is a set of external constellation points in the constellation of the modulation method of the k-th stream among the i-th modulation scheme combination,

Denotes the number of constellation points in the modulation scheme of the k-th stream among the i-th modulation scheme combination.

It is very difficult to directly calculate the detection error probability of the kth stream. Accordingly, the present invention calculates a lower boundary of the detection success probability of the k-th stream and calculates a detection probability of the k-th stream using the minimum limit of the detection success probability of the k-th stream as follows: Obtain the upper boundary.

First, the detection error probability of the k-th stream corresponding to the inner constellation point is expressed by Equation (10).

는 k번째 스트림에 대한 송신신호의 검출 값, 상기

는 조건 A에서 상기

의 검출 오류 확률, 상기

는 조건 A에서 상기

의 검출 성공 확률, 상기

는 k번째 스트림에 대한 송신신호, 상기

는 i번째 변조방식 조합 중 k번째 스트림의 변조방식의 성상도에서 내부 성상도 점들의 집합, 상기

는 격자 감소 행렬의 k번째 행, 상기

는 반올림 연산자, 상기

은 유효 잡음, 상기

는 사건 A가 일어날 확률, 상기

는 j번째 스트림에 대한 유효 잡음을 의미한다.In Equation (10) above,

Is a detection value of a transmission signal for the k < th > stream,

Lt; RTI ID = 0.0 &gt;

The detection error probability of

Lt; RTI ID = 0.0 &gt;

The probability of detection success of

Is a transmission signal for the k &lt; th &gt; stream,

Denotes a set of inner constellation points in the constellation of the modulation scheme of the k-th stream among the i-th modulation scheme combination,

Is the kth row of the lattice reduction matrix,

Is a rounding operator,

Is an effective noise,

Is the probability of occurrence of event A,

Denotes effective noise for the jth stream.

The detection success probability of the kth stream corresponding to the external constellation point is expressed by Equation (11) below.

는 k번째 스트림에 대한 송신신호의 검출 값, 상기

는 조건 A에서 상기

의 검출 오류 확률, 상기

는 조건 A에서 상기

의 검출 성공 확률, 상기

는 k번째 스트림에 대한 송신신호, 상기

는 i번째 변조방식 조합 중 k번째 스트림의 변조방식의 성상도에서 외부 성상도 점들의 집합, 상기

는 격자 감소 행렬의 k번째 행, 상기

는 반올림 연산자, 상기

은 유효 잡음, 상기

은 가능한 l번째 성상도 점, 상기

는 사건 A가 일어날 확률을 의미한다.In Equation (11) above,

Is a detection value of a transmission signal for the k < th > stream,

Lt; RTI ID = 0.0 &gt;

The detection error probability of

Lt; RTI ID = 0.0 &gt;

The probability of detection success of

Is a transmission signal for the k &lt; th &gt; stream,

Is a set of external constellation points in the constellation of the modulation method of the k-th stream among the i-th modulation scheme combination,

Is the kth row of the lattice reduction matrix,

Is a rounding operator,

Is an effective noise,

Is the possible first constellation point,

Is the probability that event A will occur.

은 0, ±1, ±j, ±1±j 등 9가지이다. 이와 같이, 유효 잡음의 반올림 값은 9^M개의 다양한 값들로 나타나며, 각 값은 가우시안 분포에 따른다. 상기 유효 잡음의 반올림 값들 각각에 대하여, 성공적인 검출 값

이 산출된다. 그리고, 상기 성공적인 검출 값이 가능한 송신신호 영역에 포함되면, 송신신호의 올바른 검출이 수행된다. 즉, 모든 외부 성상도 점에 대응되는 송신신호들의 검출 성공 확률을 평균화함으로써, 외부 성상도 점에 대응되는 k번째 스트림의 검출 성공 확률이 얻어진다.In Equation (11), each effective noise component included in each term is independent. When the appropriate signal-to-noise ratio region is considered, the variance of the effective noise is relatively large, and the curve of the effective noise probability density function is relatively sharp. At this time, as to the frequency of occurrence according to the value of the effective noise, the frequency of occurrence becomes lower as the value of the effective noise is larger. Therefore, the rounding value of effective noise having a high occurrence frequency

There are 9 types such as 0, ± 1, ± j, ± 1 ± j. Thus, the rounded value of the effective noise appears to 9 ^M of different values, each value to be in accordance with the Gaussian distribution. For each rounded value of the effective noise, a successful detection value

. Then, if the successful detection value is included in the possible transmission signal region, correct detection of the transmission signal is performed. That is, a detection success probability of a kth stream corresponding to an external constellation point is obtained by averaging the detection success probabilities of transmission signals corresponding to all external constellation points.

Using Equation (10) and Equation (11), the maximum limit of the detection error probability for the k-th stream is expressed as Equation (12).

는 i번째 변조방식 조합 중 k번째 스트림에 대한 심벌 에러율의 최대 한계, 상기

는 i번째 변조방식 조합 중 k번째 스트림의 변조방식의 성상도 내 점의 개수, 상기

는 k번째 스트림에 대한 송신신호, 상기

는 i번째 변조방식 조합 중 k번째 스트림의 변조방식의 성상도에서 내부 성상도 점들의 집합, 상기

는 j번째 스트림에 대한 유효 잡음, 상기

는 i번째 변조방식 조합 중 k번째 스트림의 변조방식의 성상도에서 외부 성상도 점들의 집합, 상기

는 격자 감소 행렬의 k번째 행, 상기

은 가능한 l번째 성상도 점,상기

은 유효 잡음, 상기

는 사건 A가 일어날 확률을 의미한다.In Equation (12) above,

Is the maximum limit of the symbol error rate for the kth stream among the combinations of the i < th > modulation scheme,

Is the number of the constellation inner points of the modulation scheme of the kth stream among the i th modulation scheme combination,

Is a transmission signal for the k &lt; th &gt; stream,

Denotes a set of inner constellation points in the constellation of the modulation scheme of the k-th stream among the i-th modulation scheme combination,

Is the effective noise for the jth stream,

Is a set of external constellation points in the constellation of the modulation method of the k-th stream among the i-th modulation scheme combination,

Is the kth row of the lattice reduction matrix,

Is the possible first constellation point,

Is an effective noise,

Is the probability that event A will occur.

Next, a case where the lattice reduction matrix is a unit matrix will be described.

를 가판정하면 하기 <수학식 13>와 같다.If the channel state is good and the lattice reduction matrix is a unitary matrix,

The following equation (13) is obtained.

는 격자 감소 기법에 의한 검출 대상 신호의 검출 값, 상기

는 총 송신 전력, 상기

은 스트림 개수, 상기

는 스케일링된 채널행렬, 상기

는 격자 감소 기법에 의한 검출 대상 신호, 상기

은 유효 잡음, 상기

는 반올림 연산자를 의미한다.In Equation (13) above,

The detection value of the detection target signal by the lattice reduction technique,

Total transmission power,

The number of streams,

Is a scaled channel matrix,

A signal to be detected by the lattice reduction technique,

Is an effective noise,

Means the rounding operator.

In the detection result shown in Equation (13), the only element that affects the detection error is effective noise, and the effective noise is expressed by Equation (14).

은 유효 잡음, 상기

는 총 송신 전력, 상기

은 스트림 개수, 상기

는 스케일링된 채널행렬, 상기

은 잡음을 의미한다.In Equation (14) above,

Is an effective noise,

Total transmission power,

The number of streams,

Is a scaled channel matrix,

Means noise.

Based on Equation (13), the transmission signal is detected as Equation (15).

는 검출된 송신신호, 상기

는 격자 감소 기법에 의한 검출 대상 신호의 검출 값, 상기

는 격자 감소 기법에 의한 검출 대상 신호, 상기

는 반올림 연산자, 상기

은 유효 잡음, 상기

는 스케 일링된 송신신호를 의미한다.In Equation (15) above,

The transmitted signal,

The detection value of the detection target signal by the lattice reduction technique,

A signal to be detected by the lattice reduction technique,

Is a rounding operator,

Is an effective noise,

Denotes a scaled transmission signal.

Accordingly, the detection value of the transmission signal for the k-th stream corresponding to the internal constellation point is expressed by Equation (16).

는 k번째 스트림에 대한 송신신호의 검출 값, 상기

는 조건 A에서 상기

의 검출 오류 확률, 상기

는 k번째 스트림에 대한 송신신호, 상기

는 i번째 변조방식 조합 중 k번째 스트림의 변조방식의 성상도에서 내부 성상도 점들의 집합, 상기

는 k번째 스트림에 대한 유효 잡음, 상기

는 사건 A가 일어날 확률을 의미한다.In Equation (16) above,

Is a detection value of a transmission signal for the k < th > stream,

Lt; RTI ID = 0.0 &gt;

The detection error probability of

Is a transmission signal for the k &lt; th &gt; stream,

Denotes a set of inner constellation points in the constellation of the modulation scheme of the k-th stream among the i-th modulation scheme combination,

Is the effective noise for the k &lt; th &gt; stream,

Is the probability that event A will occur.

The detected value of the transmission signal for the k-th stream corresponding to the external constellation point is represented by Equation (17).

는 k번째 스트림에 대한 송신신호의 검출 값, 상기

는 조건 A에서 상기

의 검출 오류 확률, 상기

는 k번째 스트림에 대한 송신신호, 상기

는 i번째 변조방식 조합 중 k번째 스트림의 변조방식의 성상도에서 외부 성상도 점들의 집합, 상기

는 k번째 스트림에 대한 유효 잡음, 상기

는 반올림 연산자, 상기

는 사건 A가 일어날 확률을 의미한다.In Equation (17) above,

Is a detection value of a transmission signal for the k < th > stream,

Lt; RTI ID = 0.0 &gt;

The detection error probability of

Is a transmission signal for the k &lt; th &gt; stream,

Is a set of external constellation points in the constellation of the modulation method of the k-th stream among the i-th modulation scheme combination,

Is the effective noise for the k &lt; th &gt; stream,

Is a rounding operator,

Is the probability that event A will occur.

In this case, the probability factor included in Equation (16) can be defined as Equation (18).

는 가우시안 Q 함수, 상기

는 k번째 스트림에 대한 유효 잡음, 상기

는 사건 A가 일어날 확률을 의미한다.In Equation 18,

Is a Gaussian Q function,

Is the effective noise for the k < th > stream,

Is the probability that event A will occur.

Using the definition as in Equation (18), when the 4-QAM scheme is used, the detection value of the transmission signal for the k-th stream corresponding to the point of external constellation is expressed by Equation (19).

는 k번째 스트림에 대한 송신신호의 검출 값, 상기

는 조건 A에서 상기

의 검출 오류 확률, 상기

는 k번째 스트림에 대한 송신신호, 상기

는 i번째 변조방식 조합 중 k번째 스트림의 변조방식의 성상도에서 외부 성상도 점들의 집합, 상기

는 가우시안 Q 함수, 상기

는 사건 A가 일어날 확률을 의미한다.In Equation (19) above,

Is a detection value of a transmission signal for the k < th > stream,

Lt; RTI ID = 0.0 &gt;

The detection error probability of

Is a transmission signal for the k &lt; th &gt; stream,

Is a set of external constellation points in the constellation of the modulation method of the k-th stream among the i-th modulation scheme combination,

Is a Gaussian Q function,

Is the probability that event A will occur.

Using the definition as in Equation (18), in the case of using the 16-QAM scheme, the detection error probability of the transmission signal for the kth stream corresponding to the point of external constellation is expressed by Equation (20) .

는 k번째 스트림에 대한 송신신호의 검출 값, 상기

는 조건 A에서 상기

의 검출 오류 확률, 상기

는 k번째 스트림에 대한 송신신호, 상기

는 i번째 변조방식 조합 중 k번째 스트림의 변조방식의 성상도에서 외부 성상도 점들의 집합, 상기

는 가우시안 Q 함수를 의미한다.In Equation (20) above,

Is a detection value of a transmission signal for the k < th > stream,

Lt; RTI ID = 0.0 &gt;

The detection error probability of

Is a transmission signal for the k &lt; th &gt; stream,

Is a set of external constellation points in the constellation of the modulation method of the k-th stream among the i-th modulation scheme combination,

Denotes a Gaussian Q function.

The detection error probability corresponding to the inner constellation point calculated using Equation (16) and the detection constancy corresponding to the outer constellation point calculated using Equation (20) , A maximum limit of the detection error probability when the lattice reduction matrix is the unit matrix is obtained.

The case where the SQRD detection scheme is used will be described below.

First, the effective channel matrix is QR decomposed as shown in Equation (21).

은 스트림 개수, 상기

는 총 송신 전력, 상기

는 스케일링된 채널행렬, 상기

는 격자 감소 행렬, 상기

는 유효 채널 행렬을 QR분해한 결과 얻어지는 Q행렬, 상기

는 유효 채널 행렬을 QR분해한 결과 얻어지는 R행렬을 의미한다.In Equation 21,

The number of streams,

Total transmission power,

Is a scaled channel matrix,

A grid reduction matrix,

A Q matrix obtained by QR decomposition of an effective channel matrix,

Denotes an R matrix obtained as a result of QR decomposition of an effective channel matrix.

Accordingly, the received signal is transformed as shown in Equation (22).

는 유효 채널 행렬을 QR분해한 결과 얻어지는 Q행렬의 허미션 행렬과 수신신호를 곱한 행렬, 상기

는 유효 채널 행렬을 QR분해한 결과 얻어지는 Q행렬의 허미션(Hermitian) 행렬, 상기

는 수신 신호, 상기

는 유효 채널 행렬을 QR분해한 결과 얻어지는 R행렬, 상기

는 격자 감소 기법에 의한 검출 대상 신호, 상기

은 잡음, 상기

은 유효 잡음을 의미한다.In Equation 22,

A matrix obtained by multiplying a received signal by a hermetion matrix of a Q matrix obtained as a result of QR decomposition of an effective channel matrix,

A Hermitian matrix of a Q matrix obtained as a result of QR decomposition of an effective channel matrix,

A reception signal,

An R matrix obtained by QR decomposition of an effective channel matrix,

A signal to be detected by the lattice reduction technique,

Noise,

Means effective noise.

가 상삼각행렬(upper triangular matrix)와 곱해진 형태가 되어 간단히 검출될 수 있다.By multiplying the received signal by the hermetian matrix of the Q matrix as in Equation (22), the signal to be detected

Can be simply detected by being multiplied with an upper triangular matrix.

In this case, the noise component is expressed by Equation (23).

은 유효 잡음, 상기

는 유효 채널 행렬을 QR분해한 결과 얻어지는 Q행렬의 허미션 행렬, 상기

은 잡음을 의미한다.In Equation 23,

Is an effective noise,

A hermetion matrix of a Q matrix obtained as a result of QR decomposition of an effective channel matrix,

Means noise.

가 얻어된다.From the modified received signal as in Equation (22), Equation (24)

Is obtained.

는 격자 감소 기법에 의한 k번째 스트림에 대한 검출 대상 신호의 검출 값, 상기

는 유효 채널 행렬을 QR분해한 결과 얻어지는 Q행렬의 허미션 행렬과 수신신호를 곱한 행렬의 k번째 스트림에 대응되는 신호, 상기

는 유효 채널 행렬을 QR 분해한 결과 얻어지는 R행렬의 j행k열 원소, 상기

는 격자 감소 기법에 의한 k번째 스트림에 대한 검출 대상 신호, 상기

는 k번째 스트림에 대한 유효 잡음을 의미한다.In Equation 24,

The detection value of the detection target signal for the k < th > stream by the lattice reduction technique,

A signal corresponding to a k-th stream of a matrix obtained by multiplying a received signal by a hermetion matrix of a Q matrix obtained as a result of QR decomposition of an effective channel matrix,

K column element of the R matrix obtained as a result of QR decomposition of the effective channel matrix,

Is a detection target signal for the k &lt; th &gt; stream by the lattice reduction technique,

Denotes an effective noise for the k-th stream.

Assuming that perfect interference cancellation is performed, Equation (24) is simplified as Equation (25).

는 격자 감소 기법에 의한 k번째 스트림에 대한 검출 대상 신호의 검출 값, 상기

는 유효 채널 행렬을 QR 분해한 결과 얻어지는 R행렬의 k행k열 원소, 상기

는 격자 감소 기법에 의한 k번째 스트림에 대한 검출 대상 신호, 상기

는 k번째 스트림에 대한 유효 잡음, 상기

는 k번째 스트림에 대한 2차 유효 잡음을 의미한다.In Equation 25,

The detection value of the detection target signal for the k < th > stream by the lattice reduction technique,

K column k column element of the R matrix obtained as a result of QR decomposition of the effective channel matrix,

Is a detection target signal for the k &lt; th &gt; stream by the lattice reduction technique,

Is the effective noise for the k &lt; th &gt; stream,

Denotes a second-order effective noise for the k-th stream.

At this time, the noise component affecting the detection error is expressed by Equation (26).

는 k번째 스트림에 대한 2차 유효 잡음, 상기

는 평균이 a이고 분산이 b인 가우시안 분포, 상기

는 잡음 전력, 상 기

는 유효 채널 행렬을 QR 분해한 결과 얻어지는 R행렬의 k행k열 원소를 의미한다.In Equation 26,

Is the second-order effective noise for the k-th stream,

A Gaussian distribution with mean a and variance b,

Is the noise power,

Denotes the k-th row k-th column of the R matrix obtained as a result of QR decomposition of the effective channel matrix.

Then, the transmission signal is detected as shown in Equation (27) below.

는 송신신호의 검출 값, 상기

는 격자 감소 행렬, 상기

는 격자 감소 기법에 따른 검출 대상 신호의 검출 값, 상기

는 격자 감소 기법에 따른 검출 대상 신호, 상기

은 2차 유효 잡음, 상기

는 송신신호를 의미한다.In Equation 27,

A detection value of a transmission signal,

A grid reduction matrix,

The detection value of the detection target signal according to the lattice reduction technique,

A detection target signal according to the lattice reduction technique,

Is the second effective noise,

Denotes a transmission signal.

If the lattice reduction matrix is not an identity matrix, the minimum limit of detection probability of the kth stream is given by Equation (28).

는 k번째 스트림에 대한 송신신호의 검출값, 상기

는 상기

의 검출 성공 확률, 상기

는 격자 감소 행렬의 k번째 행, 상기

는 간섭 제거된 j번째 스트림에 대한 유효 잡음, 상기

는 사건 A가 일어날 확률을 의미한다.In Equation 28,

Is a detection value of a transmission signal for the k < th > stream,

Quot;

The probability of detection success of

Is the kth row of the lattice reduction matrix,

&Lt; / RTI &gt; is the effective noise for the j &lt;

Is the probability that event A will occur.

Using Equation (28), the maximum limit of the detection value of the transmission signal for the k-th stream is expressed by Equation (29).

는 k번째 스트림에 대한 송신신호, 상기

는

의 검출 오류 확률, 상기

는 k번째 스트림에 대한 송신신호의 검출 값, 상기

는 상기

의 오류 확률, 상기

는 i번째 변조방식 조합 중 k번째 변조방식을 적용한 k번째 스트림의 신호대 잡음비 최대 한계, 상기

는 간섭 제거된 j번째 스트림에 대한 유효 잡음, 상기

는 사건 A가 일어날 확률을 의미한다.In Equation 29,

Is a transmission signal for the k < th > stream,

The

The detection error probability of

Is a detection value of a transmission signal for the k &lt; th &gt; stream,

Quot;

The error probability of

The maximum signal-to-noise ratio limit of the k-th stream to which the k-th modulation scheme is applied in the i-th modulation scheme combination,

&Lt; / RTI &gt; is the effective noise for the j &lt;

Is the probability that event A will occur.

If the lattice reduction matrix is a unitary matrix, the minimum limit of the probability of detection of the kth stream is given by Equation (30).

는 k번째 스트림에 대한 송신신호의 검출 값, 상기

는 상기

의 검출 성공 확률, 상기

는 간섭 제거된 j번째 스트림에 대한 유효 잡음, 상기

는 사건 A가 일어날 확률을 의미한다.In Equation (30) above,

Is a detection value of a transmission signal for the k < th > stream,

Quot;

The probability of detection success of

&Lt; / RTI &gt; is the effective noise for the j &lt;

Is the probability that event A will occur.

Using Equation (30), the maximum limit of the detection error probability of the kth stream is expressed by Equation (31).

는 k번째 스트림에 대한 송신신호, 상기

는

의 검출 오류 확률, 상기

는 k번째 스트림에 대한 송신신호의 검출 값, 상기

는 상기

의 오류 확률, 상기

는 i번째 변조방식 조합 중 k번째 변조방식을 적용한 k번째 스트림의 신호대 잡음 비 최대 한계, 상기

는 간섭 제거된 j번째 스트림에 대한 유효 잡음, 상기

는 사건 A가 일어날 확률을 의미한다.In Equation 31,

Is a transmission signal for the k < th > stream,

The

The detection error probability of

Is a detection value of a transmission signal for the k &lt; th &gt; stream,

Quot;

The error probability of

The maximum signal-to-noise ratio limit of the k-th stream applying the k-th modulation scheme among the combinations of the i-th modulation scheme,

&Lt; / RTI &gt; is the effective noise for the j &lt;

Is the probability that event A will occur.

As described above, if the maximum limit of the detection error probability of the kth stream is calculated according to the ZF detection scheme or the SQRD detection scheme, the effective signal band interference and noise ratio of the kth stream is calculated using the maximum limit of the detection error probability of the kth stream (32) < EMI ID = 32.0 >

는 i번째 변조방식 조합 중 k번째 스트림의 유효 신호대 간섭 및 잡음비, 상기

는 심벌 에러율을 신호대 간섭 및 잡음비로 변환하는 함수, 상기

는 i번째 변조방식 조합 중 k번째 스트림에 대한 심벌 에러율의 최대 한계, 상기

는 i번째 변조방식 조합 중 k번째 스트림의 변조방식을 의미한다.In Equation 32,

The effective signal band interference and noise ratio of the kth stream among the i th modulation scheme combination,

A function for converting the symbol error rate to the signal-to-interference and noise ratio,

Is the maximum limit of the symbol error rate for the kth stream among the combinations of the i < th > modulation scheme,

Denotes the modulation scheme of the k-th stream among the i-th modulation scheme combination.

The inverse function represented by Equation (32) differs depending on the modulation method, and is expressed by Equation (33), for example.

는 변조방식이 mod인 k번째 스트림의 유효 신호대 간섭 및 잡음비, 상기

는 가우시안 Q 함수의 역함수, 상기

는 변조방식이 mod인 k번째 스트림에 대한 심벌 에러율의 최대 한계를 의미한다.In Equation 33,

The effective signal-to-noise and noise ratio of the k-th stream whose modulation scheme is mod,

Is the inverse of the Gaussian Q function,

Denotes the maximum limit of the symbol error rate for the k < th > stream whose modulation scheme is mod.

For each possible combination of modulation schemes, the signal-to-interference and noise ratio for each stream is calculated as described above, and then the signal-to-interference and noise ratio for each modulation scheme combination is determined as the final channel quality information. First, the combination of modulation schemes that do not satisfy the minimum required signal-to-interference and noise ratio of the modulation scheme is excluded from the candidate. For example, if the modulation scheme of the mth stream among the nth modulation scheme combination is 4-QAM (Quadreture Amplitude Modulation) and the minimum required signal-to-noise and interference ratio of the 4-QAM is 14.6 dB, the calculated signal- Is less than 14.6 dB, the nth modulation scheme combination is excluded from the candidate. Then, the signal-to-interference and noise ratio of the modulation scheme combination having the highest sum rate among the modulation scheme combinations satisfying the minimum required signal-to-interference and noise ratio is determined as the final channel quality information.

Hereinafter, the present invention will be described in detail with reference to the drawings as to the structure and operation procedure of a receiving terminal for generating channel quality information as described above.

3 is a block diagram of a receiving end in a MIMO wireless communication system according to an embodiment of the present invention.

3, the receiver includes a plurality of RF (Radio Frequency) receivers 302-1 to 302-N, a MIMO detector 304, a channel estimator 306, and a CQI generator 308 .

Each of the plurality of RF receivers 302-1 to 302-N converts an RF band signal received through a corresponding antenna into a baseband signal. The MIMO detector 304 detects stream-by-stream transmission signals from the reception signals provided from the plurality of RF receivers 302-1 to 302-N. For example, the maximum likelihood (ML) detection method, the lattice reduction utilization detection method, the Minimum Mean Square Error-Ordered (MMSE-OSIC) method, and the MIMO- Successive Interference Cancellation) detection technique is used. The channel estimator 306 estimates a channel for each antenna with a transmitter using a predetermined signal such as a pilot signal.

The CQI generator 308 generates channel quality information for each stream using the channel information provided from the channel estimator 306. In other words, the CQI generator 308 derives an error component for each stream using a lattice reduction technique, thereby generating channel quality information for each stream. Here, the channel quality information indicates a signal-to-interference and noise ratio. 4, the CQI generator 308 includes a channel transformer 402, an LR operator 404, a classifier 406, a first SER calculator 408, a second SER calculator 410, a SINR transformer A rate calculator 412, a rate adder 414, and a CQI determiner 416.

The channel transformer 402 generates a list of possible modulation scheme combinations and scales the channel matrix provided from the channel estimator 306 according to each of the modulation scheme combinations. In other words, the channel transformer 402 identifies a scaling value that makes the distance between adjacent points in the constellation of the modulation scheme for each stream corresponding to each column of the channel matrix to be '1' The channel matrix is scaled as shown in Equation (1). This produces scaled channel matrices corresponding to each of the possible modulation scheme combinations.

)들을 생성한다. 상기 격자 감소 행렬은 자신 및 자신의 역행렬 내 모든 원소들의 실수부 및 허수부가 정수인 행렬이며, 채널행렬에 따라 다르게 산출된다. 상기 격자 감소 행렬을 생성하는 상세한 과정은 다수의 선행된 연구들을 통해 공개되어 있으며, 다양한 방식들 중 어떠한 방식을 사용하더라도 본 발명의 요지에 어긋나지 않으므로, 본 발명은 이에 대한 설명은 생략한다. 예를 들어, 상기 격자 감소 행렬을 생성하는 과정은 논문 「H.Yao and G.W.Wornell,"lattice-reduction-aided detectors for MIMO communication systems," IEEE Proc.GLOBECOM 2002, vol.1, pp.424-428, Nov.2002」 및 논문 「 Y.H.Gan and W.H.Mow,"Complex lattice reduction algorithms for low-complexity MIMO detection," IEEE Proc.GLOBECOM 2005, vol.5, pp.2953-2957, Nov.2005」에 나타나 있다.The LR operator 404 multiplies a grid reduction matrix (corresponding to each scaled channel matrix)

). The lattice reduction matrix is a matrix whose real part and imaginary part are integers of all the elements in itself and its inverse matrix, and is calculated differently according to the channel matrix. The detailed process of generating the lattice reduction matrix is disclosed through a number of previous studies, and any method of any of the various schemes is not contradicted by the gist of the present invention, so that the description of the present invention will be omitted. For example, the process of generating the lattice reduction matrix is described in H. Yao and GW Wornell, " lattice-reduction-aided detectors for MIMO communication systems, "IEEE Proc. GLOBECOM 2002, vol. 1, pp. 424-428, Nov.2002 "and the paper" YHGan and WHMow, "Complex lattice reduction algorithms for low-complexity MIMO detection, IEEE Proc. GLOBECOM 2005, vol.5, pp.2953-2957, Nov.2005.

The classifier 406 confirms whether the matrix is the unit matrix for the lattice reduction matrices corresponding to the respective modulation scheme combinations. If the lattice reduction matrix is a unitary matrix, the classifier 406 outputs the information of the combination of the modulation schemes to the first SER calculator 408. On the other hand, if the lattice reduction matrix is not a unit matrix, the classifier 406 outputs information on the combination of the modulation schemes to the second SER calculator 410.

The first SER calculator 408 calculates a maximum limit of a symbol error rate per stream according to a first scheme for the unit matrix case. At this time, a method of calculating the maximum limit of the symbol error rate per stream is changed according to a detection technique used for signal detection.

First, when the ZF detection scheme is used, the operation of the first SER calculator 408 based on one stream will be described. The first SER calculator 408 calculates a detection error probability corresponding to an internal constellation diagram and a detection error probability corresponding to an external constellation diagram for a target stream for which a symbol error rate is to be determined, The 1SER calculator 408 calculates the maximum limit of the symbol error rate of the target stream by summing the detection error probabilities and dividing by the number of constellation points of the modulation scheme corresponding to the target stream. At this time, when there is no internal constellation point in the target stream, the first SER calculator 408 does not calculate the detection error probability corresponding to the internal constellation diagram for the target stream. Here, the detection error probability corresponding to the inner constellation point is calculated by subtracting the probability that the rounding value of the effective noise for the target stream is 0 from 1, for example, as shown in Equation (16) . The detection error probability for each stream corresponding to the external constellation point is calculated by subtracting the probability that the sum of the transmission signal and the effective noise is within an area in which the sum of the transmission signal and the effective noise is 1 and for example, . The maximum limit of the symbol error rate of the target stream is calculated by Equation (12).

Next, when the SQRD detection scheme is used, the operation of the first SER calculator 408 based on one stream will be described as follows. The first SER calculator 408 calculates a second effective noise using the R matrix obtained by QR decomposition of the effective channel matrix and then calculates a second effective noise using a target stream to be searched for at 1 and a target stream to be interfered The maximum limit of the symbol error rate of the target stream is calculated by subtracting the probability that the secondary effective noise is 0 for all of the streams. For example, the effective noise for the k-th stream is obtained by multiplying the effective noise for the k-th stream by the k-th row k-th row element of the R matrix . For example, the maximum limit of the symbol error rate per stream is calculated according to Equation (30) and Equation (31).

The second SER calculator 410 calculates a maximum limit of a symbol error rate per stream according to a second scheme for a case where the non-unit matrix is not used. At this time, a method of calculating the maximum limit of the symbol error rate per stream is changed according to a detection technique used for signal detection.

First, when the ZF detection scheme is used, the operation of the second SER calculator 410 based on one stream will be described as follows. The second SER calculator 410 calculates a detection error probability corresponding to an internal constellation point and a detection error probability corresponding to an external constellation point for a target stream for which a symbol error rate is to be calculated. The second SER calculator 410 calculates the maximum limit of the symbol error rate of the target stream by summing the detection error probabilities and dividing by the number of constellation points of the modulation scheme corresponding to the target stream. At this time, if there is no internal constellation point in the target stream, the second SET calculator 410 does not calculate the detection error probability corresponding to the internal constellation diagram for the target stream. Here, the detection error probability for each stream corresponding to the inner constellation point is calculated by subtracting the probability that the rounding values of the effective noise for all the streams are 0 at 1, and for example, as shown in Equation (10) . The detection error probability for each stream corresponding to the external constellation point is 1, one of the integer points in the area where the sum of the rounding of the effective noise and the product of the row corresponding to the object stream and the transmission signal in the lattice reduction matrix is Is calculated by subtracting the probability of occurrence, for example, from Equation (11). The maximum limit of the symbol error rate of the target stream is calculated by Equation (12).

Next, when the SQRD detection scheme is used, the operation of the second SER calculator 410 based on one stream will be described as follows. The second SER calculator 410 calculates the secondary effective noise using the R matrix obtained by QR decomposition of the effective channel matrix, and then subtracts the probability that the secondary effective noise for all the streams is 0 at 1 Thereby calculating the maximum limit of the symbol error rate of the target stream. For example, the effective noise for the k-th stream is obtained by multiplying the effective noise for the k-th stream by the k-th row k-th row element of the R matrix . For example, the maximum limit of the symbol error rate of the target stream is calculated as Equation (29).

The SINR converter 412 converts a symbol error rate calculated by the first SER calculator 408 and the second SER calculator 410 into a signal-to-interference and noise ratio. The conversion to the signal-to-interference and noise ratio is performed differently depending on the modulation method. For example, when the modulation scheme is 4-QAM or 16-QAM, the conversion to the signal-to-interference and noise ratio is performed through one of the equations shown in Equation (11). Since the equation for converting the symbol error rate to the signal-to-interference and noise ratio is a well-known equation, the description of the equation for the other modulation schemes is omitted.

The rate adder 414 calculates a stream rate sum of each modulation scheme combination. At this time, the rate adder 414 sets the rate of a stream having a signal-to-interference and noise ratio lower than a threshold value to zero. For example, if the modulation scheme of the k-th stream is m, the signal-to-interference and noise ratio of the k-th stream generated by the SINR converter 412 is A [dB], and the modulation scheme m is used If the required signal-to-interference and noise ratio threshold is B [dB], if the A [dB] is less than the B [dB], the rate adder 414 multiplies the kth stream among the i- Set the transmission rate to zero.

The CQI determiner 416 determines the final channel quality information using a sum rate for each stream of the modulation scheme combinations calculated by the rate adder 414. That is, the CQI determiner 416 determines the signal-to-interference and noise-ratio set of the modulation scheme combination having the highest sum of the transmission rates as the final channel quality information.

The channel quality information for each stream generated through the above-described configuration is fed back to the transmitting end to be used for scheduling of the transmitting end and determination of the transmission signal modulation scheme. However, since the modulation scheme is simultaneously determined in the process of generating the channel quality information for each stream through the above-described configuration, the modulation scheme information may be directly fed back. Therefore, although not shown, the receiving end further includes a feedback transmitter for transmitting feedback information to the transmitting end, and the feedback transmitter transmits the stream-based channel quality information or modulation scheme information to the transmitting end.

FIG. 5 illustrates a process of generating channel quality information for each stream of a receiver in a MIMO wireless communication system according to an embodiment of the present invention.

Referring to FIG. 5, the receiver generates a list of possible modulation scheme combinations in step 501. Through subsequent steps, a signal-to-noise and signal-to-noise ratio corresponding to each of the possible modulation scheme combinations is generated.

After generating the list of possible modulation scheme combinations, the receiver proceeds to step 503 and scales the channel matrix of the i-th modulation scheme combination. At the start of this procedure, the variable i is initialized to one.

)을 생성한다. 상기 격자 감소 행렬은 자신 및 자신의 역행렬 내 모든 원소들의 실수부 및 허수부가 정수인 행렬이며, 채널행렬에 따라 다르게 산출된다. 상기 격자 감소 행렬을 생성하는 상세한 과정은 다수의 선행된 연구들을 통해 공개되어 있으며, 다양한 방식들 중 어떠한 방식을 사용하더라도 본 발명의 요지에 어긋나지 않으므로, 본 발명은 이에 대한 설명은 생략한다. After the channel matrix is scaled, the receiver proceeds to step 505 where a grid reduction matrix corresponding to the channel matrix scaled by the grid reduction technique

). The lattice reduction matrix is a matrix whose real part and imaginary part are integers of all the elements in itself and its inverse matrix, and is calculated differently according to the channel matrix. The detailed process of generating the lattice reduction matrix is disclosed through a number of previous studies, and any method of any of the various schemes is not contradicted by the gist of the present invention, so that the description of the present invention will be omitted.

After generating the lattice reduction matrix, the receiving end proceeds to step 507 and checks whether the lattice reduction matrix generated is a unit matrix.

If the lattice reduction matrix is a unit matrix, the receiver proceeds to step 509 and calculates a maximum limit of a symbol error rate per stream according to a first scheme for a case where the matrix is not a unit matrix. At this time, a method of calculating the maximum limit of the symbol error rate per stream is changed according to a detection technique used for signal detection.

First, when the ZF detection scheme is used, the receiver calculates a detection error probability for each stream corresponding to a stream detection error probability and an external constellation point corresponding to an internal constellation point, And the maximum number of symbol error rates per stream is calculated by dividing the number of constellation constellations corresponding to the streams by the number of constellation points. At this time, if there is no internal constellation point in the target stream for which the symbol error rate is to be determined, the receiving end does not calculate the detection error probability corresponding to the internal constellation point for the target stream. Here, the detection error probability for each stream corresponding to the inner constellation point is calculated by subtracting the probability that the rounding value of the effective noise for the stream for which the symbol error rate is to be obtained is 0, and for example, Lt; 16 &gt;. The detection error probability for each stream corresponding to the external constellation point is calculated by subtracting the probability that the sum of the transmission signal and the effective noise is within an area in which the sum of the transmission signal and the effective noise is 1 and for example, (18). &Quot; (18) &quot; The maximum limit of the symbol error rate per stream is calculated by Equation (12).

Next, when the SQRD detection scheme is used, the receiver calculates the second-order effective noise using the R matrix obtained by QR decomposition of the effective channel matrix, By calculating a maximum limit of the symbol error rate per stream by subtracting the probability that the secondary effective noise is 0 for all of the streams interfered by the stream error rate. For example, the effective noise for the k-th stream is obtained by multiplying the effective noise for the k-th stream by the k-th row k-th row element of the R matrix . For example, the maximum limit of the symbol error rate per stream is calculated according to Equation (30) and Equation (31).

If the lattice reduction matrix is not a unit matrix, the receiver proceeds to step 511 and calculates a maximum limit of a symbol error rate per stream according to a second scheme for a unit matrix. At this time, a method of calculating the maximum limit of the symbol error rate per stream is changed according to a detection technique used for signal detection.

First, when the ZF detection scheme is used, the receiver calculates a detection error probability for each stream corresponding to a stream detection error probability and an external constellation point corresponding to an internal constellation point, And the maximum number of symbol error rates per stream is calculated by dividing the number of constellation constellations corresponding to the streams by the number of constellation points. At this time, if there is no internal constellation point in the target stream for which the symbol error rate is to be determined, the receiving end does not calculate the detection error probability corresponding to the internal constellation point for the target stream. Here, the detection error probability for each stream corresponding to the inner constellation point is calculated by subtracting the probability that the rounding values of the effective noise for all the streams are 0 at 1, and for example, (11). &Quot; (11) &quot; The detection error probability for each stream corresponding to the external constellation point is 1, and the product of the rounded value of the effective noise and the symbol error rate in the lattice reduction matrix and the corresponding row, Is calculated by subtracting the probability of one of the integer points, for example, as shown in Equation (11) and Equation (13). The maximum limit of the symbol error rate per stream is calculated by Equation (12).

Next, when the SQRD detection scheme is used, the receiver calculates a second effective noise using the R matrix obtained by QR decomposition of the effective channel matrix, and then calculates the second effective noise for all the streams at 1 And calculates the maximum limit of the symbol error rate per stream by subtracting the probability of zero. For example, the effective noise for the k-th stream is obtained by multiplying the effective noise for the k-th stream by the k-th row k-th row element of the R matrix . For example, the maximum limit of the symbol error rate per stream is calculated according to Equation (28) and Equation (29).

In step 509 or step 511, the receiver calculates a maximum error rate for each stream, and then proceeds to step 513 and converts the symbol error rate per stream into the signal-to-interference and noise ratio for each stream. The conversion to the signal-to-interference and noise ratio is performed differently depending on the modulation method. For example, when the modulation scheme is 4-QAM or 16-QAM, the conversion to the signal-to-interference and noise ratio is performed through one of the equations shown in Equation (33). Since the above equation for converting the symbol error rate into the signal-to-interference and noise ratio is widely known, description of the formula for the other modulation schemes is omitted.

After converting the symbol error rate for each stream into the signal-to-interference and noise ratio for each stream, the receiver proceeds to step 515 and sets the transmission rate of a stream having a signal-to-interference and noise ratio lower than a threshold value to zero. For example, if the modulation scheme of the k-th stream among the i-th modulation scheme combination is m, the signal-to-interference and noise ratio of the k-th stream generated in step 513 is A [dB], the signal- And the threshold value of the noise ratio is B [dB], if the A [dB] is less than the B [dB], the receiver sets the transmission rate of the kth stream among the i th modulation scheme combination to zero.

Then, the receiver proceeds to step 517 to calculate a stream rate sum of the i-th modulation scheme combination. That is, the receiver calculates a transmission rate for each of the modulation schemes included in the i-th modulation scheme, and sums the calculated transmission rates. However, the transmission rate of the stream set to 0 in step 515 is not calculated.

After calculating the sum of the stream rates of the combination of the i-th modulation scheme, the receiving end proceeds to step 519 and checks whether the sum of the transmission rates is calculated through steps 503 to 517 for all the modulation scheme combinations.

If the sum of the transmission rates is not calculated for all the modulation scheme combinations, the receiver proceeds to step 521 and increments i by 1, and then returns to step 503.

On the other hand, if the sum of the transmission rates is calculated for all the modulation scheme combinations, the receiver proceeds to step 523 and determines the final signal quality information as the signal-to-interference and noise-ratio set for each stream of the modulation scheme combination having the maximum transmission rate sum.

After determining the final channel quality information, the receiving end proceeds to step 525 and feeds back the final channel quality information. At this time, since the modulation scheme for each stream is also determined simultaneously in the process of generating the channel quality information, the modulation scheme information may be fed back.

4 and 5, the receiving end calculates the symbol error rate in a different manner according to the shape of the lattice reduction matrix, calculates the symbol error rate by dividing the external constellation diagram and internal constellation diagram Respectively. However, according to another embodiment of the present invention, the receiving end calculates the symbol error rate without distinguishing the type of the lattice reduction matrix and the type of the constellation point. In this case, the detailed configuration of the CQI generator 308 included in the receiver and the operation procedure of the receiver are as follows.

6 is a block diagram of a channel quality information generator in a MIMO wireless communication system according to another embodiment of the present invention.

6, the CQI generator 308 includes a channel transformer 602, an LR calculator 604, a SER calculator 606, a SINR converter 608, and a CQI determiner 610, do.

The channel modifier 602 scales the channel matrix provided from the channel estimator 106 to generate a scaled channel matrix corresponding to each of the modulation scheme combinations. In other words, the channel transformer 602 identifies a scaling value that makes the distance between adjacent points in the constellation of the modulation scheme for each stream corresponding to each column of the channel matrix to be '1' The channel matrix is scaled as shown in Equation (1). This produces scaled channel matrices corresponding to each possible combination of modulation schemes.

)들을 생성한다. 상기 격자감소 행렬은 자신 및 자신의 역행 렬 내 모든 원소들의 실수부 및 허수부가 정수인 행렬이며, 채널행렬에 따라 다르게 산출된다. 상기 격자감소 행렬을 생성하는 상세한 과정은 다수의 선행된 연구들을 통해 공개되어 있으며, 다양한 방식들 중 어떠한 방식을 사용하더라도 본 발명의 요지에 어긋나지 않으므로, 본 발명은 이에 대한 설명은 생략한다. 예를 들어, 상기 격자감소 행렬을 생성하는 과정은 논문 「H.Yao and G.W.Wornell,"lattice-reduction-aided detectors for MIMO communication systems," IEEE Proc.GLOBECOM 2002, vol.1, pp.424-428, Nov.2002」 및 논문 「Y.H.Gan and W.H.Mow,"Complex lattice reduction algorithms for low-complexity MIMO detection," IEEE Proc.GLOBECOM 2005, vol.5, pp.2953-2957, Nov.2005」에 나타나 있다.The LR operator 604 multiplies a grid reduction matrix (corresponding to each scaled channel matrix)

). The lattice reduction matrix is a matrix whose real part and imaginary part are integers of all the elements in itself and its own inverse matrix, and is calculated differently according to the channel matrix. The detailed process of generating the lattice reduction matrix is disclosed through a number of previous studies, and any method of any of the various schemes is not contradicted by the gist of the present invention, so that the description of the present invention will be omitted. For example, the process of generating the lattice reduction matrix is described in H. Yao and GW Wornell, " lattice-reduction-aided detectors for MIMO communication systems, "IEEE Proc. GLOBECOM 2002, vol. 1, pp. 424-428, Nov.2002 "and the paper" YHGan and WHMow, "Complex lattice reduction algorithms for low-complexity MIMO detection, IEEE Proc. GLOBECOM 2005, vol.5, pp.2953-2957, Nov.2005.

The SER calculator 606 calculates the effective noise using the lattice reduction matrix calculated by the LR calculator 604 and calculates the maximum limit of the symbol error rate for each stream using the effective noise. Here, the effective noise is calculated according to Equation (5), and the maximum limit of the symbol error rate per stream is calculated as Equation (34).

는 수신신호 중 k번째 스트림에 대한 신 호, 상기

는 상기

의 검출 오류 확률, 상기

는 k번째 스트림에 대한 송신신호의 검출 값, 상기

는 상기

의 검출 오류 확률, 상기

는 i번째 변조방식 조합 중 k번째 스트림에 대한 심벌 에러율의 최대 한계, 상기

는 사건 A가 일어날 확률, 상기

는 k번째 스트림에 대한 유효 잡음을 의미한다.In Equation 34,

The signal for the k < th > stream of the received signal,

Quot;

The detection error probability of

Is a detection value of a transmission signal for the k &lt; th &gt; stream,

Quot;

The detection error probability of

Is the maximum limit of the symbol error rate for the kth stream among the combinations of the i &lt; th &gt; modulation scheme,

Is the probability of occurrence of event A,

Denotes an effective noise for the k-th stream.

The SINR converter 608 converts the symbol error rate calculated by the SER calculator 606 into the signal-to-interference and noise ratio. The conversion to the signal-to-interference and noise ratio is performed differently depending on the modulation method. For example, when the modulation scheme is 4-QAM or 16-QAM, the conversion to the signal-to-interference and noise ratio is performed through one of the equations shown in Equation (33). Since the equation for converting the symbol error rate to the signal-to-interference and noise ratio is a well-known equation, the description of the equation for the other modulation schemes is omitted.

The CQI determiner 610 determines one modulation scheme combination having a signal-to-interference and noise-ratio set to be used as the final channel quality information among a plurality of modulation scheme combinations. To this end, the CQI determiner 610 excludes a signal-to-interference and noise-ratio set that does not satisfy the minimum required signal-to-interference and noise ratio of the modulation scheme from the candidates. The CQI determiner 610 determines the set of the SINRs of the modulation scheme combination having the highest total rate among the modulation scheme combinations satisfying the minimum required signal-to-interference and noise ratio as the final channel quality information.

The channel quality information for each stream generated through the above-described configuration is fed back to the transmitting end to be used for scheduling of the transmitting end and determination of the transmission signal modulation scheme. However, since the modulation scheme is simultaneously determined in the process of generating the channel quality information for each stream through the above-described configuration, the modulation scheme information may be directly fed back. Therefore, although not shown, the receiving end further includes a feedback transmitter for transmitting feedback information to the transmitting end, and the feedback transmitter transmits the stream-based channel quality information or modulation scheme information to the transmitting end.

FIG. 7 illustrates a process of generating channel quality information for each stream of a receiver in a MIMO wireless communication system according to another embodiment of the present invention.

Referring to FIG. 7, in step 701, the receiver generates a list of possible modulation scheme combinations. In the following steps, the signal-to-noise and signal-to-noise ratios for each of the possible combinations of modulation schemes are generated.

After generating the list of possible modulation scheme combinations, the receiver proceeds to step 703 and scales the channel matrix. In other words, the receiver checks a scaling value that makes the inter-point distance within the constellation of the modulation scheme for each stream to '1', and scales the channel matrix according to Equation (1) according to the scaling value. This produces scaled channel matrices corresponding to each possible combination of modulation schemes.

)들을 생성한 다. 상기 격자감소 행렬은 자신 및 자신의 역행렬 내 모든 원소들의 실수부 및 허수부가 정수인 행렬이며, 채널행렬에 따라 다르게 산출된다. 상기 격자감소 행렬을 생성하는 상세한 과정은 다수의 선행된 연구들을 통해 공개되어 있으며, 다양한 방식들 중 어떠한 방식을 사용하더라도 본 발명의 요지에 어긋나지 않으므로, 본 발명은 이에 대한 설명은 생략한다. After scaling the channel matrix, the receiving end proceeds to step 705 and uses the lattice reduction technique to calculate a lattice reduction matrix (corresponding to each scaled channel matrix)

). The lattice reduction matrix is a matrix whose real part and imaginary part are integers of all the elements in itself and its inverse matrix, and is calculated differently according to the channel matrix. The detailed process of generating the lattice reduction matrix is disclosed through a number of previous studies, and any method of any of the various schemes is not contradicted by the gist of the present invention, so that the description of the present invention will be omitted.

After generating the lattice reduction matrix, the receiving end derives an error component for each stream of the modulation scheme combinations using the lattice reduction matrix in step 707, And the error rate is calculated. That is, the receiver calculates the effective noise using the lattice reduction matrix, and calculates the maximum limit of the symbol error rate for each stream using the effective noise. Here, the effective noise is calculated according to Equation (5), and the maximum limit of the symbol error rate per stream is calculated as Equation (34).

After calculating the maximum limit of the symbol error rate per stream, the receiving terminal proceeds to step 709 and converts the symbol error rate calculated in step 707 into a signal-to-interference and noise ratio. The conversion to the signal-to-interference and noise ratio is performed differently depending on the modulation method. For example, when the modulation scheme is 4-QAM or 16-QAM, the conversion to the signal-to-interference and noise ratio is performed through one of the equations shown in Equation (33). Since the equation for converting the symbol error rate to the signal-to-interference and noise ratio is a well-known equation, the description of the equation for the other modulation schemes is omitted.

After converting the symbol error rate to a signal-to-interference and noise ratio, the receiver proceeds to step 711 and excludes a signal-to-interference and noise-ratio set that does not satisfy the minimum required signal-to-interference and noise ratio of the modulation scheme from the candidate.

In step 713, the receiver calculates a sum rate for each modulation scheme combination.
After calculating the transmission rate for each combination of modulation schemes, the receiver proceeds to step 715. In step 715, the receiver computes a signal-to-interference and noise ratio set of the modulation scheme combination having the highest sum rate among the modulation scheme combinations satisfying the minimum required signal- As the final channel quality information.

The channel quality information for each stream generated through the above procedure is fed back to the transmitting end and used for scheduling of the transmitting end and determination of the transmission signal modulation scheme. However, since the modulation scheme is simultaneously determined in the procedure for generating the channel quality information for each stream, the modulation scheme information may be directly fed back. Accordingly, after the above-described procedure, the receiving end sends the effective signal-to-noise ratio or the modulation scheme information for each stream to the transmitting end.

While the present invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not limited to the disclosed embodiments, but is capable of various modifications within the scope of the invention. Therefore, the scope of the present invention should not be limited by the illustrated embodiments, but should be determined by the scope of the appended claims and equivalents thereof.

1 is a diagram illustrating constellation scaling and shifting examples in a multiple input multiple output (MIMO) wireless communication system according to the present invention;

2 is a diagram illustrating an example of a feasible region of an external constellation point in a MIMO wireless communication system according to the present invention;

3 is a block diagram of a receiving end in a MIMO wireless communication system according to an embodiment of the present invention;

4 is a block diagram of a channel quality information generator in a MIMO wireless communication system according to an embodiment of the present invention.

5 is a diagram illustrating a procedure of generating channel quality information for each stream of a receiver in a MIMO wireless communication system according to an embodiment of the present invention.

6 is a block diagram of a channel quality information generator in a MIMO wireless communication system according to another embodiment of the present invention.

7 is a flowchart illustrating a channel quality information generation process for each stream of a receiver in a MIMO wireless communication system according to another embodiment of the present invention.

Claims (45)

In a receiving end apparatus in a multiple input multiple output (MIMO) wireless communication system, A receiver for receiving a signal from a transmitting end via a plurality of antennas, An estimator for estimating a channel for each antenna with the transmitter using the received signal to construct a channel matrix; Effective noise for each stream is derived according to a result of applying a lattice reduction technique to the channel matrix and channel quality information (CQI) for each stream is generated using effective noise for each stream , &Lt; / RTI &gt; Wherein the lattice reduction technique is a technique for scaling the channel matrix and the received signal by converting a distance between neighboring points in the constellation to an integer. The method according to claim 1, The generator comprising: A calculator for calculating a lattice reduction matrix corresponding to each of the channel matrices according to possible combinations of modulation schemes, Calculating effective noise when using the lattice reduction technique for each of the modulation scheme combinations and calculating at least one symbol error rate (SER) for each of the combinations of modulation schemes using the effective noise; The output device, A converter for converting the symbol error rate into a signal-to-interference and noise ratio (SINR) And a determiner for determining a signal-to-interference and noise-ratio set for each stream having a maximum transmission rate among sets of signal-to-interference and noise-ratio sets for each of the modulation scheme combinations as channel quality information for each of the final streams. 3. The method of claim 2, Wherein the at least one calculator uses the effective noise to calculate a maximum limit of the per-stream symbol error rate. The method of claim 3, Wherein the determiner determines a maximum sum rate (sum) of at least one combination of modulation schemes satisfying the minimum required signal-to-interference and noise ratio, except for at least one combination of modulation schemes that does not satisfy the minimum required signal- rate of the modulation scheme combination is determined as the channel quality information for each of the final streams. 5. The method of claim 4, Wherein the at least one calculator comprises: A first calculator for calculating a maximum limit of a symbol error rate per stream of a modulation scheme combination having a lattice reduction matrix of a unit matrix form, And a second calculator for calculating a maximum limit of a symbol error rate per stream of a modulation scheme combination having a lattice reduction matrix of a type other than a unit matrix. 6. The method of claim 5, Wherein the first calculator and the second calculator calculate detection error probabilities corresponding to an external contellation point and, if an interior contellation point exists, And averages the detection error probabilities on a stream-by-stream basis, thereby calculating a maximum limit of a symbol error rate per stream. The method according to claim 6, The first calculator calculates a probability that the sum of the transmission signal and the effective noise in the target stream for which the symbol error rate is to be calculated is an integer point in a region where the sum of the transmission signal and the effective noise can be obtained and derives a detection error probability corresponding to the external constellation point from the probability and, If there is an inner constellation point, calculating a probability that the rounding value of the effective noise for the target stream is 0, and deriving a detection error probability corresponding to the inner constellation point from the probability. The method according to claim 6, Wherein the second calculator calculates a probability that one of the integer points in the area where the rounding of the effective noise and the sum of the transmission signals are available and the multiplication of the corresponding stream and the target stream for which the symbol error rate is to be calculated, Deriving a detection error probability corresponding to the external constellation point from a probability, If there is an inner constellation point, calculates a probability that rounding values of effective noise for all streams are 0, and derives a detection error probability corresponding to the inner constellation point from the probability. 6. The method of claim 5, Wherein the first calculator and the second calculator calculate a second effective noise by dividing the effective noise matrix by a diagonal element of an R matrix obtained as a result of QR decomposition of the effective channel matrix, To calculate a maximum limit of a symbol error rate per stream. 10. The method of claim 9, Wherein the first calculator calculates a probability that the second effective noise is zero for both the target stream for which the symbol error rate is to be sought and the streams that are interfered by the target stream and determines from the probability the maximum limit Lt; / RTI &gt; 10. The method of claim 9, Wherein the second calculator calculates a probability that the secondary effective noises for all streams are zero and derives a maximum limit of the per-stream symbol error rate from the probability. The method according to claim 1, Further comprising a transmitter for feeding back modulation scheme combination information corresponding to the stream-based channel quality information or the stream-based channel quality information to the transmitter. A method of generating channel quality information (CQI) for each stream in a multiple input multiple output (MIMO) wireless communication system, Receiving signals from a transmitter through a plurality of antennas, Constructing a channel matrix by estimating a channel for each antenna with the transmitter using a received signal; Deriving an effective noise per stream according to a result of applying a lattice reduction technique to the channel matrix; And generating channel quality information for each stream using the stream-specific effective noise, Wherein the lattice reduction technique is a technique of scaling the channel matrix and the received signal by converting a distance between neighboring points in the constellation into an integer. 14. The method of claim 13, Wherein the generating of the per-stream channel quality information comprises: Calculating a lattice reduction matrix corresponding to each of the channel matrices according to possible combinations of modulation schemes; Calculating effective noise when a lattice reduction technique is used for each of the streams of the modulation scheme combinations; Calculating a symbol error rate (SER) for each stream using the effective noise; Converting the symbol error rate to a signal-to-interference and noise ratio (SINR) And determining a set of signal-to-interference and noise ratios for each stream having a maximum transmission rate among sets of signal-to-interference and noise-ratio sets for each of the modulation scheme combinations as channel quality information for each of the final streams. 15. The method of claim 14, The process of calculating the symbol error rate per stream may include: And using the effective noise to calculate a maximum limit of the symbol error rate per stream. 16. The method of claim 15, Determining a set of signal-to-interference and noise ratios for each stream having the maximum data rate as channel quality information for each final stream, Excluding at least one modulation scheme combination that does not satisfy the minimum required signal-to-interference and noise ratio of the modulation scheme; And determining a set of signal-to-interference and noise-ratio for each stream of the modulation scheme combination having the highest sum rate among at least one modulation scheme combination that is not excluded as channel quality information for each of the final streams . 17. The method of claim 16, The process of calculating the symbol error rate per stream may include: Calculating a maximum limit of a symbol error rate per stream according to a first scheme when the lattice reduction matrix is an identity matrix, And calculating a maximum limit of a symbol error rate per stream according to a second scheme when the lattice reduction matrix is not a unit matrix. 18. The method of claim 17, The process of calculating the symbol error rate per stream may include: Calculating detection error probabilities corresponding to an external contellation point, Calculating detection error probabilities corresponding to the internal constellation point when an interior contellation point exists; And averaging the detection error probabilities for each stream to calculate a maximum limit of a symbol error rate per stream. 19. The method of claim 18, Wherein the step of calculating a maximum limit of a symbol error rate per stream according to the first scheme comprises: Calculating a probability of an integer point in a region where a sum of a transmission signal and an effective noise is available for a target stream for which a symbol error rate is to be calculated and deriving a detection error probability corresponding to the external constellation point from the probability; Calculating a probability that the rounding value of the effective noise of the target stream is 0 when there is an internal constellation point and deriving a detection error probability corresponding to the internal constellation point from the probability; Way. 19. The method of claim 18, Wherein the step of calculating a maximum limit of a symbol error rate per stream according to the second scheme comprises: Calculating a rounding value of effective noise and a probability of one of integer points in a region where a sum of a transmission signal and a product of a target stream and a corresponding signal to be searched for a symbol error rate in a lattice reduction matrix is feasible, Detecting a probability of a detection error corresponding to a point; Calculating a probability that rounding values of effective noise for all the streams are 0 if there is an internal constellation point and deriving a detection error probability corresponding to the internal constellation point from the probability; How to. 18. The method of claim 17, The process of calculating the symbol error rate per stream may include: Calculating a second effective noise by dividing the effective channel matrix by the diagonal elements of the R matrix obtained as a result of QR decomposition, And calculating a maximum limit of a symbol error rate per stream using the secondary effective noise. 22. The method of claim 21, Wherein the step of calculating a maximum limit of a symbol error rate per stream according to the first scheme comprises: Calculating a probability that the second effective noise is 0 for both the target stream for which the symbol error rate is to be determined and the streams that are interfered by the target stream; And deriving a maximum limit of a symbol error rate for the target stream from the probability. 22. The method of claim 21, Wherein the step of calculating a maximum limit of a symbol error rate per stream according to the second scheme comprises: Calculating a probability that the secondary effective noises for all the streams are 0, And deriving a maximum limit of the per-stream symbol error rate from the probability. 14. The method of claim 13, And feeding back the modulation scheme combination information corresponding to the stream-based channel quality information or the stream-based channel quality information to the transmitting end. delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete

Images