KR20080039714A - Apparatus and method for scheduling in mimo system - Google Patents

Abstract

A scheduling apparatus for scheduling in a MIMO(Multi Input Multi Output) system and a method thereof are provided to prevent performance from being deteriorated due to a channel estimation error by carrying out scheduling in consideration of channel uncertainty. A scheduling apparatus in a MIMO system comprises a user channel combination part(201), an MSE(Mean Squared Error) calculation part(202), a user selection part(203), and an OFDM modulator(230). The user channel combination part combines quantized channel information, which is fed back from terminals, and forms channel matrixes for all possible user combinations. The MSE calculation part receives the channel matrix and calculates the MSE values of Rx signals for all users or combinations of partial users. The user selection part selects a user group which shows the minimum MSE value. The OFDM modulator transmits the data of the selected user group.

Description

Scheduling apparatus and method in multi-antenna system {APPARATUS AND METHOD FOR SCHEDULING IN MIMO SYSTEM}

1 is a simplified MIMO system configuration according to the present invention,

2 is an apparatus block diagram for performing scheduling in a multi-antenna system according to an embodiment of the present invention;

3A is a flowchart of a method of performing scheduling in a base station according to an embodiment of the present invention;

3B is a flowchart of a method of performing scheduling in a terminal according to an embodiment of the present invention.

4 is a configuration diagram of a MIMO mobile communication system for performing a scheduling operation according to an embodiment of the present invention;

5 is a scheduling performance graph according to the prior art and scheduling in consideration of uncertainty in accordance with an embodiment of the present invention.

The present invention relates to an apparatus and method for user scheduling in a multi-antenna system, and more particularly, to a scheduling apparatus and method considering an error of channel information in a multi-antenna system.

Multi-input multi-output system (MIMO) is a technique for increasing frequency efficiency by sharing the spatial resources secured by multi-antennas simultaneously by multiple users. In the case of downlink transmitting information from the base station to the mobile station, the base station transmits information of several users at the same time through a preprocessing process, and each mobile station receives its own information from the received signal arriving through its channel. At this time, in order to secure a high channel capacity, a scheduling scheme for allocating resources adaptively to the channel conditions of users who change every moment may be essential. In order to allocate resources to users at the base station, it is necessary to know downlink channel information from the base station to the mobile station, and assuming that the channel information is accurate, the resource is allocated to a group of users that maximizes the channel capacity or signal noise ratio (SNR). Assign it.

If the base station accurately knows the downlink channel information between mobile stations and the channel does not change over time, the conventional scheduling technique may provide good performance. However, the base station cannot know the downlink channel accurately and cannot accurately predict the channel changing over time.

In a frequency division duplexing (FDD) environment, since downlink channel and uplink channel frequencies are different, downlink channel information cannot be predicted using uplink channel information. Therefore, in order for the base station to know the downlink channel information, the mobile station should acquire the downlink channel information, quantize it, and feed it back through the uplink channel. However, information loss is inevitable in this quantization process. In addition, since the downlink channel information cannot be fed back at every moment, it is not possible to cope with a change in the downlink channel occurring therebetween. That is, the downlink channel information of the base station is different from the actual downlink channel information.

Conventional scheduling techniques do not consider the possibility of error of channel information at all in using downlink channel information of a base station which may be different from actual channel information.

Therefore, there is a need for an optimal scheduling apparatus and method that utilizes erroneous channel information.

Accordingly, an object of the present invention is to provide an optimal scheduling apparatus and method considering an error in downlink channel information of a base station.

Another object of the present invention is to provide a scheduling apparatus and method for considering a channel information error by using variance information of the channel information error.

Another object of the present invention is to provide an apparatus and method for reducing performance degradation due to channel error by scheduling in consideration of channel uncertainty.

An apparatus according to the present invention for achieving the above object, in the base station scheduling apparatus of a multi-antenna system, a user channel combination unit for configuring a channel matrix by combining the quantized channel information fed back from the terminal, and the user channel combination A MSE calculation unit for calculating a mean squared error (MSE) value of a received signal for all user combinations or partial user combinations to reduce complexity, and the calculated mean squared error values And a user selector for selecting a minimum user group, and an OFDM modulator for transmitting data of the selected user group.

In accordance with an aspect of the present invention, a method of scheduling a base station of a multi-antenna system includes: configuring a channel matrix by combining quantized channel information fed back from terminals, and all of the based on the channel information. Calculating a mean squared error (MSE) value of the received signal for the user combination, selecting a user group that minimizes the calculated mean squared error value; And transmitting data of the selected user group.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In describing the present invention, when it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted.

Hereinafter, the present invention will be described a scheduling apparatus and method for obtaining the mean square error (MSE) of the user group in all cases based on the quantized channel information to select the user group that minimizes the MSE value and send the information. Shall be.

1 shows a simplified MIMO system according to the present invention.

는

번째 사용자에게 전송하고자 하는 데이터 심볼이고,

는

번째 사용자의 심볼에 곱해지는 가중치 벡터(weight vector)이다. 그리고,

는

번째 사용자의 채널 계수이다. 이때,

개의 송신안테나를 통해 전송되는 송신신호는 하기 <수학식 1>과 같이 채널 인벌션(channel inversion)에 기초하는 것으로 한다.Referring to FIG. 1, a system model of one antenna of a mobile communication terminal in a multi-user multi-antenna downlink system is shown.

Is

Data symbol to send to the first user,

Is

The weight vector multiplied by the symbol of the first user. And,

Is

Channel coefficient of the first user. At this time,

The transmission signal transmitted through the two transmission antennas is based on channel inversion as shown in Equation 1 below.

단, 이다.

However, it is.

는 사용자 수이고,

는 전체 송신전력이고,

은 추정된 채널 행렬,

은

의 허미시안 행렬(Hermitian Matrix),

는 사용자에게 전송하고자 하는 데이터 심벌이다.here,

Is the number of users,

Is the total transmit power,

Is the estimated channel matrix,

silver

Hermitian Matrix of,

Is the data symbol to be sent to the user.

If the mean square error (MSE) of the transmission signal of Equation 1 is minimized, Equation 2 may be expressed.

,

,

는 사용자 수이고,

는 전체 송신전력이고,

는 노이즈 전력이고,

은 추정된 채널 행렬,

은

의 허미시안 행렬(Hermitian Matrix),

는 단위행렬이고,

는 사용자에게 전송하고자 하는 데이터 심벌이다.here,

Is the number of users,

Is the total transmit power,

Is noise power,

Is the estimated channel matrix,

silver

Hermitian Matrix of,

Is the unit matrix,

Is the data symbol to be sent to the user.

If the mean square error (MSE) is minimized in consideration of the uncertainty in the transmission signal of Equation 2, it may be represented by Equation 3 below.

이다.only,

to be.

는 사용자 수이고,

는 전체 송신전력이고,

는 노이즈 전력이고,

는 채널 에러의 분산(variance)이고,

은 추정된 채널 행렬,

은

의 허미시안 행렬(Hermitian Matrix),

는 단위행렬이고,

는 사용자에게 전송하고자 하는 데이터 심벌이다.here,

Is the number of users,

Is the total transmit power,

Is noise power,

Is the variance of the channel error,

Is the estimated channel matrix,

silver

Hermitian Matrix of,

Is the unit matrix,

Is the data symbol to be sent to the user.

For reference, all errors in channel information known by the base station may be regarded as occurring in the quantization process, and it is well known to estimate the quantization error of the quantizer. For example, a channel coefficient is generated in a Gaussian and then subjected to a quantizer to estimate the variance by averaging the square of the difference between the quantized value and the original value. Therefore, it can be seen that when the quantizer is determined, the dispersion is also determined.

2 is a block diagram of an apparatus for performing scheduling in a multi-antenna system according to an exemplary embodiment of the present invention. The following description will be given by taking an example of scheduling from the base station to each terminal.

As shown in FIG. 5, the base station includes a scheduler 200, an encoder 210, a subchannel allocator 220, and an OFDM modulator 230.

First, the scheduler 200 includes a user channel combiner 201, an MSE calculator 202, and a user selector 203 to check the size and channel information of data to be transmitted from data queues. . Thereafter, the scheduler 200 selects a user according to the size of the data and channel information.

The user channel combiner 201 constructs a channel matrix for all possible user combinations. The MSE calculator 202 calculates a mean square error (MSE) of the received signal using the channel matrix provided by the user channel combiner 201. The mean squared error is derived from the MSE of the received signal and is represented by Equation 4 below using the channel matrix as a variable. The user selector 203 receives the result of the MSE calculator 202 and selects a user group having the minimum MSE.

The encoder 210 receives data of a user group provided from the scheduler 200 and encodes the data at a corresponding code rate.

The subchannel allocator 220 maps and outputs a radio resource (subchannel) according to the user group information provided from the scheduler 200.

The OFDM modulator 230 converts the parallel signal into a frequency component by performing an IFFT that converts a serially output modulated signal into a parallel signal.

3A is a flowchart of a method of performing scheduling in a base station according to an embodiment of the present invention.

) 수신하여 가능한 모든 조합의 그룹에 대한 수신신호의 MSE 값을 하기 <수학식 4>을 이용하여 계산한다.Referring to FIG. 3, the base station proceeds to step 301 in which channel information (

) The MSE value of the received signal for all possible combinations of groups is calculated using Equation 4 below.

으로 구한다.

Obtain as

는 사용자 수이고,

는 전체 송신전력이고,

는 노이즈 전력이고,

는 채널 에러의 분산(variance)이고,

은 추정된 채널 행렬,

은

의 허미시안 행렬(Hermitian Matrix),

는 단위행렬이다.here,

Is the number of users,

Is the total transmit power,

Is noise power,

Is the variance of the channel error,

Is the estimated channel matrix,

silver

Hermitian Matrix of,

Is the unit matrix.

는 그룹에 속한 사용자들의 채널만 모은 행렬이다.In Equation 4 above

Is a matrix of channels of users belonging to the group.

In step 303, the base station selects a user group having a minimum MSE value.

In step 308, the base station transmits the information to the selected group of users. The transmission signal is represented by Equation (3).

The base station then terminates the algorithm of the present invention.

3B is a flowchart of a method for performing scheduling in a terminal according to an exemplary embodiment of the present invention.

를 알게된다.Referring to FIG. 3B, in step 300, the terminal quantizes an estimated value by channel estimation of a downlink to start scheduling. Where each channel estimate is

Get to know.

)를 피드백한다. 여기서, 상기 기지국은 전체 채널 행렬(

)을 알게 된다.Subsequently, the terminal proceeds to step 302 where the quantized channel information (

Feedback. Here, the base station is a full channel matrix (

You get to know

.In other words,

.

In operation 304, data is received by scheduling of the base station.

Thereafter, the terminal terminates the scheduling procedure of the present invention.

4 is a block diagram illustrating a MIMO mobile communication system for performing a scheduling operation according to an embodiment of the present invention.

명의 사용자들이 기지국을 통해 데이터 서비스 중이며 기지국은 하나의 시간 슬롯 동안에

개의 안테나를 통해

명의 사용자 단말들(430, 431,432)에게 해당하는 데이터를 전송한다.

즉, 스케줄러(400)는 매 시간구간마다

명의 사용자 버퍼들(420,421,422)에서 최적의 채널 환경을 가지는

명의 사용자 단말들(430,431,432)을 위한 사용자 데이터를 선택하여 해당하는 안테나를 통해 전달한다.Referring to FIG. 4 above,

Users are serving data through a base station, and the base station

Through antennas

Data corresponding to the two user terminals 430, 431, and 432 is transmitted.

That is, the scheduler 400 every hour

Optimal Channel Environment in Two User Buffers 420,421,422

User data for the two user terminals 430, 431, and 432 are selected and transmitted through the corresponding antenna.

5 is a graph illustrating scheduling performance according to the present invention and scheduling performance according to the prior art according to an exemplary embodiment of the present invention.

Referring to FIG. 5, in the present embodiment, information is transmitted to two mobile communication terminals at the same time by using two antennas at a base station. Each of the graphs of FIG. 4 shows BER performance when the total number of users is 4 and 8. Conv. Refers to a conventional scheduling technique, and this embodiment considers scheduling to maximize channel capacity. Prop. Is a proposed scheduling technique.

Meanwhile, in the detailed description of the present invention, specific embodiments have been described, but various modifications are possible without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined not only by the scope of the following claims, but also by the equivalents of the claims.

As described above, the present invention can reduce performance degradation due to channel estimation error by scheduling in consideration of channel uncertainty in the transmitter. Therefore, lower BER performance can be obtained when using the same transmission power.

Claims (16)

In the base station scheduling apparatus of a multi-antenna system, A user channel combination unit constituting a channel matrix by combining the quantized channel information fed back from the terminals; An MSE calculation unit receiving a channel matrix from the user channel combination unit and calculating a mean squared error (MSE) value of the received signal for all user or partial user combinations; A user selection unit for selecting a user group that minimizes the calculated mean squared error value; And an OFDM modulator for transmitting data of the selected user group. The method of claim 1, The average squared error value is calculated by Equation 5 below.

only,

to be. here,

Is the number of users,

Is the total transmit power,

Is noise power,

Is the variance of the channel error,

Is the estimated channel matrix,

silver

Hermitian Matrix of,

Is the unit matrix. The method of claim 2, Distribution of the channel error (

) Generates channel coefficients as Gaussian and averages the square of the difference between the quantized value and the original value through a quantizer. The method of claim 1, The transmission signal of the OFDM modulator is a scheduling apparatus characterized in that represented by the following equation (6).

only,

to be. here,

Is the number of users,

Is the total transmit power,

Is noise power,

Is the variance of the channel error,

Is the estimated channel matrix,

silver

Hermitian Matrix of,

Is the unit matrix,

Is the data symbol to be sent to the user. In the base station scheduling method of a multiple antenna system, Configuring a channel matrix by combining the quantized channel information fed back from the terminals; Calculating a mean squared error (MSE) value of the received signal for all user combinations based on the channel information; Selecting a user group that minimizes the calculated mean squared error value; And transmitting the data of the selected user group. The method of claim 5, wherein The average squared error value is calculated by Equation (7).

only,

to be. here,

Is the number of users,

Is the total transmit power,

Is noise power,

Is the variance of the channel error,

Is the estimated channel matrix,

silver

Hermitian Matrix of,

Is the unit matrix. The method of claim 6, Distribution of the channel error (

) Is a method of generating channel coefficients as Gaussians and then averaging the square of the difference between the quantized value and the original value through a quantizer. The method of claim 5, wherein In the process of transmitting data of the selected user group, a transmission signal is represented by Equation (8).

only,

to be. here,

Is the number of users,

Is the total transmit power,

Is noise power,

Is the variance of the channel error,

Is the estimated channel matrix,

silver

Hermitian Matrix of,

Is the unit matrix,

Is the data symbol to be sent to the user. In the scheduling apparatus of a multiple antenna system, A terminal for performing downlink estimation and quantization and feeding back the quantized channel information to a base station; A channel matrix is constructed by combining the quantized channel information fed back from the terminals, and the mean squared error (MSE) value of the received signal is calculated for all user combinations by receiving the channel matrix from the user channel combiner. And a base station for selecting a user group that minimizes the calculated mean squared error value and transmitting data of the selected user group to the terminal. The method of claim 9, The average squared error value is calculated by the following Equation (9).

only,

to be. here,

Is the number of users,

Is the total transmit power,

Is noise power,

Is the variance of the channel error,

Is the estimated channel matrix,

silver

Hermitian Matrix of,

Is the unit matrix. The method of claim 10, Distribution of the channel error (

) Generates channel coefficients as Gaussian and averages the square of the difference between the quantized value and the original value through a quantizer. The method of claim 9, The transmission signal of the base station is a scheduling device, characterized in that represented by the following equation (10).

only,

to be. here,

Is the number of users,

Is the total transmit power,

Is noise power,

Is the variance of the channel error,

Is the estimated channel matrix,

silver

Hermitian Matrix of,

Is the unit matrix,

Is the data symbol to be sent to the user. In the scheduling method of a multiple antenna system, Channel estimation and quantization of the downlink by each of the terminals; Feeding back the quantized channel information by the terminal; Calculating, by a base station, a mean squared error (MSE) value for all user combinations based on the feedback channel information; Selecting a user group that minimizes the calculated mean squared error value; And scheduling, by the base station, transmitting data of the selected user group. The method of claim 13, The average squared error value is calculated by the following Equation (11).

only,

to be. here,

Is the number of users,

Is the total transmit power,

Is noise power,

Is the variance of the channel error,

Is the estimated channel matrix,

silver

Hermitian Matrix of,

Is the unit matrix. The method of claim 14, Distribution of the channel error (

) Is a method of generating channel coefficients as Gaussians and then averaging the square of the difference between the quantized value and the original value through a quantizer. The method of claim 13, In the process of transmitting data of the selected user group by the base station, the transmission signal is represented by Equation 12 below.

only,

to be. here,

Is the number of users,

Is the total transmit power,

Is noise power,

Is the variance of the channel error,

Is the estimated channel matrix,

silver

Hermitian Matrix of,

Is the unit matrix,

Is the data symbol to be sent to the user.

Images