KR101482925B1

KR101482925B1 - Zero-forcing Beamformer Design Device and Method in MISO Broadcast Channel Based on k-regularity

Info

Publication number: KR101482925B1
Application number: KR1020130079153A
Authority: KR
Inventors: 성영철; 서준영; 박주호; 이길원
Original assignee: 한국과학기술원
Priority date: 2013-07-05
Filing date: 2013-07-05
Publication date: 2015-01-14

Abstract

On the MISO Broadcast Channel

A zero-forcing beamforming apparatus and method based on normality are disclosed. A storage unit for storing channel information and a number of antennas to be selected by a data stream (RF chain), and a control unit for controlling the number of rows in the matrix of the channel information to be equal to the number of antennas based on the stored channel information and the number of stored antennas The number of rows in the matrix of the channel information is equal to the number of the antennas, and the number of rows in the matrix of the channel information is minimized To optimize zero forcing beam formation.

Description

[0001] The present invention relates to a zero-forcing beamforming apparatus and method based on k-normality in a MISO broadcast channel,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of designing a beamformer in a multiple input single output (MISO) environment, and more particularly to a method of designing a beamformer in a MISO broadcast channel

To a zero-forcing beamforming apparatus and method based on normality.

Aiming at designing a transmitter beamformer that supports multiple users in a massive multiple-input multiple-output (MIMO) system with a very large number of antennas directly at the base station while lowering the hardware complexity of the base station do. In Massive MIMO, when the number of data streams (RF chain) equal to the number of antennas is used for digital beamforming, the power consumption and the hardware size of the transmitter are greatly increased.

An antenna selection technique was used in a structure in which the number of existing data streams is smaller than that of an antenna. The antenna selection technique is a technique for selecting and transmitting only a part of antennas available in a single user MIMO system. For example,

There are

less

If there are two data streams, each data stream does not overlap and only one antenna is selected

Of the antennas, the best channel condition

This means that only one antenna is selected and actually used. This antenna selection technique does not use all the antennas but it is a technology that can reduce the hardware complexity while obtaining the same diversity order as the case of using all antennas. However,

There were a number of cases of branch selection, which resulted in a significant amount of computation. For this reason, a fast antenna selection algorithm has been proposed which can significantly reduce the computational complexity even with some performance degradation. However, antenna selection techniques ultimately lead to significant performance degradation compared to using all existing antennas. In order to compensate for the performance degradation of the antenna technique, there is no significant difference between the antenna selection technique and the hardware complexity, but with the performance comparable to the optimal eigen-beamforming technique

- Normal beamforming technique is proposed.

The technical problem to be solved by the present invention is

- It is aimed to have normal beamforming in MISO environment where there are many users.

Another technical problem to be solved by the present invention is to reduce the amount of output required to obtain an optimal solution, while the performance is not different from the existing one.

A zero forcing beamforming apparatus includes a storage unit for storing channel information and a number of antennas to be selected by a data stream (RF chain), and a number of rows in the matrix of channel information based on the stored channel information and the number of stored antennas. When the number of rows in the matrix of channel information is removed until the number of rows is equal to the number of antennas, a minimum performance And a control unit for optimizing the zero forcing beam formation by removing the rows one by one so as to have degradation.

Wherein the controller removes a row having the highest performance when the removed matrix is removed or a row having a minimum performance degradation when the removed matrix is removed so as to adjust the signal to noise ratio The signal transmission rate can be optimized.

The optimization maximizes the desired signal, the interference is zero, and the power constraint can be satisfied.

The controller may calculate the following equation to optimize the zero forcing beam formation:

here,

Is the size

From the base station

Channel users, and channel vector

Is an index set

Quot; means the number of elements of "

Denotes a channel matrix,

Is an index set,

The channel matrix

in

And the remaining matrix and channel matrix

in

Quot; is a matrix that is left after removing a row corresponding to an index element included in "

The channel vector

of

And the remaining information after removing the row corresponding to the index element included in the index element.

The control unit may express the row having the highest performance by the following equation.

here,

The

And the remaining channel matrix

.

The controller may express the performance as: < EMI ID =

here,

The

A column vector which is a transpose of the ith row,

The

Of the identity matrix of

Column.

The control unit can express the row having the lowest performance by the following equation:

here,

The

For the second user

Lt; RTI ID = 0.0 > a < / RTI > row.

The control unit can express the performance degradation by the following equation: < EMI ID =

here,

The

Lt; / RTI >

The

&Quot;

The

Lt; / RTI >

The

&Quot;

The

Lt; / RTI >

The

&Quot;

The

Lt; / RTI >

The

.

The controller may design a beamforming vector to optimize the zero forcing beamforming as follows:

here,

The

Th beamforming vector,

The channel matrix

in

And the remaining matrix after removing the ith column.

The zero forcing beamforming method includes the steps of storing channel information and a number of antennas to be selected by a data stream (RF chain), determining a number of rows of the matrix of channel information based on the stored channel information and the number of stored antennas, Removing the rows one by one so that the remaining matrix has the highest performance when the number of removed channel identifiers is equal to the number of the removed channel information, and designing the beamforming vectors based on the matrixes of the remaining channel information.

The zero forcing beamforming method includes the steps of storing channel information and a number of antennas to be selected by a data stream (RF chain), determining a number of rows of the matrix of channel information based on the stored channel information and the number of stored antennas, Removing a row having a minimum performance degradation when it is removed until it is equal to the number of channel information to be removed, and designing a beamforming vector based on the matrix of channel information remaining and removed.

In the MISO broadcast channel according to the present invention

According to a zero-forcing beamforming apparatus and method based on normality

- Normal beamforming can be in a MISO environment where multiple users are present.

In addition, the performance may be different from the existing one, while reducing the amount of output to obtain the optimal solution.

1 is a configuration diagram illustrating linear beamforming according to an embodiment of the present invention.
Figure 2 is a block diagram of an embodiment of the present invention

- is a configuration diagram showing a normal beamformer structure;
3 is a block diagram illustrating a zero forcing beamforming apparatus according to an embodiment of the present invention.
4 is a flowchart illustrating a first algorithm of a zero-forcing beamforming apparatus according to an embodiment of the present invention.
5 is a flowchart illustrating a second algorithm of a zero-forcing beamforming apparatus according to an embodiment of the present invention.
6 is an exemplary diagram illustrating an average sum transmission amount for SNR according to an embodiment of the present invention.
7 is an exemplary diagram illustrating an amount of output for increasing the number of antennas according to an embodiment of the present invention.

1 is a configuration diagram illustrating linear beamforming according to an embodiment of the present invention.

Referring to FIG. 1, linear beamforming is a technique of allowing a beam of an antenna to be limited to a corresponding terminal in a manner of a smart antenna.

The present invention

A base station with one transmit antenna has one antenna

(MISO) system that supports a number of users, and the number of antennas of the base station is much larger than the number of users

) Can be considered. In addition,

To support users

Lt; RTI ID = 0.0 > beamforming, < / RTI > In this situation

The received signal received by the ith user is expressed by Equation (1).

here,

Is the size

From the base station

The channel vector between the first and second users,

Means that all terminals are collected. The channel vector may be included in the channel information.

Means a complex Gaussian noise with an average of 0 and a variance of 1. Also

Is a transmission vector transmitted from the base station,

Data symbol for the < RTI ID = 0.0 >

&Lt; / RTI >

here,

ego

to be. Using Equation (2), the transmission signal vector (

) Can be represented by a determinant such as Equation (3).

here,

Is a channel matrix and may be a matrix of channel information.

Equation (3) assumes that data symbols sent by each user are independent of each other (i.e.,

), And assuming that the power used for each user's data is the same (i.e.,

)can do. In Equation (3), it can be assumed that the transmitting end and the receiving end know perfectly information about the channel.

Equation (4) is one of the performance values mainly used in this situation

And the signal-to-interference plus noise ratio (SINR) of the second user.

If the base station has to support multiple users, it is important that the base station considers the interference between each user. Also, one of the most useful techniques for base station to control interference may be zero forcing beamforming.

Figure 2 is a block diagram of an embodiment of the present invention

- is a configuration diagram showing a normal beamformer structure;

Referring to Figure 2,

- The normal beamformer structure consists of each data stream

this

After the number of complex gains are multiplied,

Of the antennas

And the information can be transmitted. Also

The normal beamformer structure can transmit information to the antenna by summing signals assigned to the same antenna.

- Normal beamforming matrix and method

- Can be defined in normal beamforming structure.

The normal beamforming matrix is a beamforming matrix

Each row of

If the nonzero components and the remaining components are all zero,

To

- It can be defined as regular beamforming matrix.

- The regular beamforming method

- It can be defined as a method of transmitting information by linear combination with a data stream with a regular beamforming matrix.

3 is a block diagram illustrating a zero forcing beamforming apparatus according to an embodiment of the present invention.

Referring to FIG. 3, the zero-forcing beamforming apparatus 1 performs zero-

- It can be applied to normal beamforming structure, and it can be a beamforming device that guarantees transmission amount without interference to each user. The zero-forcing beamforming apparatus 1 may be a personal computer system such as a desktop, laptop, tablet or handheld computer mounted on a transmitting end. The zero-forcing beamforming apparatus 1 may be a high-end computer such as a supercomputer or a large-sized computer mounted on a transmitting end. The zero forcing beamforming apparatus 1 may include an input unit 310, a control unit 320, an output unit 330, and a storage unit 340.

The input unit 310 may receive an input value necessary for forming a zero forcing beamforming. The input value may be channel information and the number of antennas. The number of antennas

Number of antennas

Lt; / RTI > That is,

May be the number of antennas selected for each data stream. The input value may also be a setting value when the zero-forcing beamforming apparatus 1 is operated for the first time. Accordingly, if the set value is not changed, the input value is stored in the storage unit 340, and the input value can be called up when the input value is needed. If the input value is changed, the input unit 310 can receive a new input value and simultaneously store the input value in the storage unit 320.

The output unit 330 may output the result of forming the zero-forcing beamforming. The output unit 330 outputs the resultant beamforming vector value and

Can be output. The output unit 330 outputs a beamforming vector value, which is a result value,

Lt; / RTI > Therefore, the transmitting end is based on the optimized beamforming vector value output from the zero-forcing beamforming apparatus 1

It is possible to transmit information to the antenna of FIG.

The optimization may be to maximize the desired signal, the interference is zero, and satisfy the power constraint.

The storage unit 340 may store at least one of the input value, the algorithm, and the variable value for zero-forcing beamforming. The storage unit 340 may store an input value received from the input unit 310. [ Also, when the input value is changed, the storage unit 340 can erase the existing input value and update the new input value.

The control unit 320 receives the channel information received from the input unit 310,

Can be used to form a zero forcing beamforming. Also, a received signal including channel information input from the input unit 310 can be expressed as Equation (1). The controller 320 may be a MISO broadcast channel model in which a plurality of users exist. MISO has a transmitter

There may be a case where there are two antennas and there is a single antenna at the receiver. The controller 320 may reduce the amount of output in obtaining an optimal solution, and the performance may not differ. In addition, the controller 320 may determine the final beamforming vector value

Can be calculated.

Zero forcing beamforming satisfies the condition of Equation (5) since it serves to make interference between users zero.

The result of Equation (5) is that the

The signal-to-interference and noise beams of the first user can be expressed as: " (6) "

In addition, the instantaneous rate of Equation (6) can be calculated by a log function such as Equation (7).

Zero Forcing Beam Forming Technique

The existing zero forcing beamforming technique prior to applying the normal beamforming situation can be calculated as Equation (8) using a pseudo inverse matrix of the channel matrix.

here,

The matrix is a power normalization matrix to satisfy power constraints. (In other words,

)

The control unit 320 may not use the pseudo inverse matrix for the zero forcing beam forming problem as in Equation (8). Further, the control unit 320 can express the optimization problem as: " (9) " The optimization problem may be a zero forcing beamforming problem that maximizes the desired signal, interference is zero, and satisfies the power constraint.

The control unit 320 determines the optimization problem of Equation (9)

If the condition is satisfied, the solution exists uniquely, and the solution can be calculated using Equation (10) using a projection operator.

here,

The channel matrix

in

And the remaining matrix after removing the ith column. Also,

The channel matrix

in

And the remaining matrix is removed. In each case

Instead, when the index set is entered, it means the matrix that has been removed by removing the column or rows corresponding to the elements of the index set. The other notations used are as follows.

The control unit 320

- Constraint condition caused by normal beamforming can be set as a zero norm condition as in Equation (11).

Accordingly, the control unit 320 determines

The optimization problem of the zero forcing beamforming considered in the normal beamforming situation can be further constrained as shown in Equation (12).

As a result of examining whether there exists a solution to the optimization problem of the expression (12)

The solution of the optimization problem may exist. Also

end

The position of the non-zero component of the solution is determined, the solution can be determined uniquely.

Accordingly, the control unit 320

- In the case of normal beamforming,

Calculating the nonzero element position may be the biggest goal. That is, each data stream selects

It may be a goal to calculate the number of antennas. This can be expressed as Equation (13).

In Equation (13), when the constraint conditions are substituted into the objective function, the optimization problem can be solved as shown in Equation (14).

here,

Is an index set

The number of elements of the element.

The channel matrix

in

And the remaining matrix and channel matrix

Denotes a matrix left after removing a row corresponding to an index element included in the index set S,

The channel vector

Index set of

&Quot;< / RTI >

In the case of Equation (14), an optimal solution can be obtained by using a brute-force search algorithm considering all cases. However, the compulsory decryption search compulsion decryption search algorithm

Since there are a number of possible cases of branches, when the number of antennas of a base station increases,

As the number increases, the output for solving the problem may increase sharply. Accordingly, the control unit 320 can use two algorithms based on a continuous removal method that can greatly reduce the amount of output.

Equation (14) shows that maximizing the cost function

Index < / RTI >

, The channel matrix < RTI ID = 0.0 >

To maximize the cost function

Quot; < / RTI >< RTI ID = 0.0 >

When the control unit 320

Selecting one row at a time has a very high throughput

You can remove each row one by one.

In the first algorithm, the control unit 320 may select and remove a row maximizing the performance that can be obtained by the remaining matrix when a specific row is removed for each step. In addition, the control unit 320 may perform the removal process

Repeated times can be made.

The control unit 320

Channel matrix

Which can be obtained by

(Cost function in Equation (14)),

Can be expressed by Equation (15).

The controller 320 can find and remove a row having a maximum performance that can be obtained with the remaining matrix by removing a specific row using the first algorithm. When the control unit 320

If the second row is removed,

Performance Obtained by

Is expressed by Equation (16).

The control unit 320 determines whether or not

To maximize the value

Can be found.

The control unit 320

Of rows

Rows can be obtained as shown in Equation (17)

It can be defined that removing the ith row is the optimal solution. The control unit 320 determines in the next step that it is determined in Equation (17)

Lt; / RTI >

To

And the same process can be repeated. The control unit 320 performs the above-

In order for a row to be selected,

Times. In the case of the existing coercive search algorithm,

However, since the controller 320 can consider a much smaller number of cases, the amount of output can be reduced and the performance of the conventional algorithm can be maintained.

In order to more efficiently implement the first algorithm,

The value can be calculated and updated effectively. The control unit 320

Diet of

And

Since there is no need to express in terms of the expression of the expression, the same result as the expression (18) can be obtained by using the algebraic calculation.

here

The

A column vector which is a transpose of the ith row,

The

Of the identity matrix of

Column.

In the second algorithm, when a specific row is removed, the control unit 320 can select and remove one row in which the performance degradation is minimized. In addition, the control unit 320 may perform the removal process

Repeat times. The control unit 320

To

For the second user

The second row may be removed or the performance may be degraded,

(19). &Quot; (19) "

Equation (19) is similar to the first algorithm

Of rows

Choosing the best row of

In < / RTI >

Remove the second row.

Equation (20) shows that the above-described removal method is the same as the first algorithm

To get a row of

Times.

Therefore, the controller 320 may have exactly the same performance using the first and second algorithms. However, the second algorithm may be advantageous in that the amount of output can be reduced more than the first algorithm by efficiently developing mathematical expressions.

The control unit 320, through logarithmic expansion

Is calculated as shown in Equation (21)

The second row can be expressed by an expression of the second row.

Here, each term is as follows.

The control unit 320 determines whether or not the existing point-

- Normal beamforming is extended to a MISO broadcast channel model with multiple users,

- We can propose a regular beamforming method. In addition, the control unit 320 requires a very high output in order to obtain an optimal solution. Two algorithms can be proposed, in which the amount of output is reduced and the performance is hardly differentiated.

4 is a flowchart illustrating a first algorithm of a zero-forcing beamforming apparatus according to an embodiment of the present invention.

Referring to FIG. 4, the zero-forcing beamforming apparatus 1 can support a large number of users using the first algorithm and reduce the complexity. The first algorithm can find and remove rows in which the specific row is removed and the performance obtained by the remaining matrix becomes the maximum.

The control unit 320 initializes the values of the variables (S100). The control unit 320 determines the position of the antenna,

Can be initialized (

). The control unit 320 may initialize all beamforming vectors to zero. Since the control unit 320 can independently design the beamforming vectors of each user with the characteristics of the zero forcing beam forming,

Order beamforming vectors of the second user can be designed in order.

The control unit 320 initializes the user index (S102). The control unit 320 receives the user index

The value of

. &Lt; / RTI >

The controller 320 stores the channel matrix (S104). The control unit 320 determines

And the channel matrix < RTI ID = 0.0 >

The channel information may be stored in the storage unit 340.

The control unit 320 determines

Quot; 1 " (S106). The control unit 320 determines

Can be performed in order, and the variable indicating this

You can specify a value of 1.

The control unit 320 determines that the remaining matrix is the highest performance

(S108). The control unit 320

Values are from 1

And then,

To maximize the value

Can be calculated as shown in [Equation 22].

The control unit 320 determines whether the remaining matrix is the highest performance

(S110). The control unit 320 may optimize the signal transmission rate according to the signal-to-noise ratio among the matrixes of the channel information by removing the row in which the remaining matrix has the highest performance.

The control unit 320

A set of values and indices

(S112). Index set

And

Updating the value can be done as follows.

The control unit 320 determines

And

(S114). The control unit 320 determines whether or not

To remove the variable

And

Can be updated.

The control unit 320 determines whether the number of antennas is greater than a value obtained by subtracting the number of antennas selected by the data stream from the number of antennas,

(Step S116). The control unit 320 determines

this

If it is greater than or equal to the threshold value, the process proceeds to the next step. Otherwise, the process can proceed to the step S108.

The control unit 320

Th beamforming vector

(S118). The control unit 320 may design the beamforming vector value corresponding to the finally remaining index set as Equation (23).

The control unit 320 determines

The value of the number of users

(S120). The control unit 320 determines

The value of the number of users

And if they are not the same

The value may be updated as follows and the process may proceed to step S102.

5 is a flowchart illustrating a second algorithm of a zero-forcing beamforming apparatus according to an embodiment of the present invention.

Referring to FIG. 5, the zero-forcing beamforming apparatus 1 can support a large number of users using the first algorithm and reduce the complexity.

The control unit 320 initializes the values of the variables (S200). The control unit 320 determines the position of the antenna,

Can be initialized (

Order beamforming vectors of the second user can be designed in order.

The control unit 320 initializes the user index (S202). The control unit 320 receives the user's index

The value of

. &Lt; / RTI >

The controller 320 stores the channel matrix (S204). The control unit 320 determines

And assigns the channel information assumed to be known to the channel matrix

Lt; / RTI >

The control unit 320 determines

Quot; 1 " (S206). The control unit 320 determines

Will remove the rows in order, and the variables representing them

You can specify a value of 1.

The control unit 320 controls the performance

(S208). The control unit 320

Values are from 1

, And is calculated as in Equation (21)

Can be calculated.

The control unit 320 may be configured to have a minimum performance degradation

(S210). The control unit 320

Lt; RTI ID = 0.0 > row

Can be calculated as shown in Equation (20).

(S212). The control unit 320 may optimize the signal transmission rate according to the signal-to-noise ratio among the matrixes of the channel information by removing the row having the minimum performance degradation.

The control unit 320

A set of values and indices

Is updated as follows (S214).

The control unit 320 determines

And

(S216). The control unit 320 determines whether or not

To remove the variable

And

Can be updated as follows.

(S218). &Lt; / RTI > The control unit 320 determines

this

If it is greater than or equal to the threshold value, the flow advances to the next step. Otherwise, the flow advances to step S208.

The control unit 320

Th beamforming vector

(S220). The control unit 320 may design the beamforming vector value corresponding to the finally remaining index set as Equation (23).

The control unit 320 determines

The value of the number of users

(S222). The control unit 320 determines

The value of

If not, it can be terminated. If not

The value may be updated as follows and the process may proceed to step S204.

6 is an exemplary diagram illustrating an average sum transmission amount for SNR according to an embodiment of the present invention.

Referring to FIG. 6, the first algorithm and the second algorithm proposed in the zero forcing beamforming apparatus 1

, A forced decryption search algorithm, a random selection algorithm,

&Lt; / RTI >

FIG. 6 may show that the performance of the first algorithm and the second algorithm are exactly the same. The first algorithm and the second algorithm can perform much better than the random selection algorithm that arbitrarily selects the antenna. The first algorithm and the second algorithm are used when all the antennas are connected,

It is possible to show that the performance is not greatly deteriorated as compared with the case of FIG. Also, the first and second algorithms can show that there is almost no difference in performance from the forced decryption search algorithm which considers all cases.

7 is an exemplary diagram illustrating an amount of output for increasing the number of antennas according to an embodiment of the present invention.

Referring to FIG. 7, the first algorithm and the second algorithm proposed in the zero-forcing beamforming apparatus 1

, The number of antennas

Can be compared with the output of the forced decryption search algorithm.

FIG. 7 may show that the first algorithm and the second algorithm have much lower throughput than the forced decryption search algorithm. The second algorithm can show that it has a lower throughput in efficient mathematical expansion than the first algorithm. Therefore, the first and second algorithms can show that there is almost no difference in performance between the best compulsory retrieval algorithm and the best comprehensible retrieval algorithm that takes all cases into consideration while having a much lower throughput.

The present invention can also be embodied as computer-readable codes on a computer-readable recording medium. A computer-readable recording medium includes all kinds of recording apparatuses in which data that can be read by a computer apparatus is stored. Examples of the computer-readable recording medium include a ROM, a RAM, a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like, and may be implemented in the form of a carrier wave (for example, transmission via the Internet) . The computer-readable recording medium may also be distributed to networked computer devices so that computer readable code can be stored and executed in a distributed manner.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is clearly understood that the same is by way of illustration and example only and is not to be taken by way of limitation in the embodiment in which said invention is directed. It will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the scope of the appended claims.

1: Zero forcing beam forming apparatus 310: Input unit
320: control unit 330: output unit
340:

Claims

A storage unit for storing channel information and a number of antennas to be selected by the data stream (RF chain); And
Removing one row of the matrix having the highest performance when the number of rows of the channel information is removed until the number of rows of the matrix is equal to the number of the antennas based on the stored channel information and the number of the stored antennas, And a controller for optimizing zero-forcing beam formation by removing rows one by one to minimize the performance degradation when the number of rows of the channel information matrix is equal to the number of antennas,
Wherein,
The signal rate of the matrix of the channel information is reduced according to a signal to noise ratio by removing a row that has the highest performance when the matrix is removed or a row that has a minimum performance degradation when the matrix is removed In a MISO broadcast channel characterized by < RTI ID = 0.0 >

- A zero-forcing beamforming device based on normality.

delete

The method according to claim 1,
Characterized in that the optimization maximizes the desired signal, the interference is zero and satisfies the power constraint. &Lt; RTI ID = 0.0 >

- A zero-forcing beamforming device based on normality.

The method according to claim 1,
Wherein the controller calculates the following equation to optimize the zero forcing beam formation: < RTI ID = 0.0 >

- Zero forcing beamforming device based on regularity:

here,

Is the size

From the base station

Channel users, and channel vector

Is an index set

Quot; means the number of elements of "

Denotes a channel matrix,

Is an index set,

The channel matrix

in

And the remaining matrix and channel matrix

in

The channel vector

of

The method according to claim 1,
Wherein the controller indicates the highest performance row by the following equation: < RTI ID = 0.0 >

- Zero forcing beamforming device based on regularity:

here,

The

And the remaining channel matrix

.

6. The method of claim 5,
Wherein the controller is configured to perform the MISO broadcast on the MISO broadcast channel,

- Zero forcing beamforming device based on regularity:

here,

The

A column vector which is a transpose of the ith row,

The

Of the identity matrix of

Column.

The method according to claim 1,
Wherein the controller indicates the lowest performance row by the following equation: < RTI ID = 0.0 > MISO < / RTI >

- Zero forcing beamforming device based on regularity:

here,

The

For the second user

Lt; RTI ID = 0.0 > a < / RTI > row.

8. The method of claim 7,
Wherein the controller decides the performance degradation by the following equation: < RTI ID = 0.0 > MISO < / RTI >

- Zero forcing beamforming device based on regularity:

here,

The

Lt; / RTI >

The

&Quot;

The

Lt; / RTI >

The

&Quot;

The

Lt; / RTI >

The

&Quot;

The

Lt; / RTI >

The

.

The method according to claim 1,
Wherein the controller is configured to design a beamforming vector to optimize the zero forcing beamforming as follows: < RTI ID = 0.0 >

- Zero forcing beamforming device based on regularity:

here,

The

Th beamforming vector,

The channel matrix

in

And the remaining matrix after removing the ith column.

Storing channel information and the number of antennas to be selected by the data stream (RF chain);
Removing one row at a time when the remaining matrix is maximized when the number of rows of the matrix of channel information is equal to the number of antennas based on the stored channel information and the number of stored antennas ; And
And designing a beamforming vector based on the removed channel information matrix,
The step of removing the rows one by one,
And removing a row in which the remaining matrix has the highest performance when the channel matrix is removed, thereby optimizing a signal transmission rate according to a signal-to-noise ratio of the matrix of the channel information.

A zero - forcing beamforming method based on normality.

Storing channel information and the number of antennas to be selected by the data stream (RF chain);
Removing a row having a minimum performance degradation when the number of rows of the channel information is removed until the number of rows is equal to the number of the antennas based on the stored channel information and the number of stored antennas; And
And designing a beamforming vector based on the removed channel information matrix,
The step of removing the rows one by one,
And removing a row having a minimum performance degradation when the signal is removed, thereby optimizing a signal transmission rate according to a signal-to-noise ratio among matrices of the channel information.

A zero - forcing beamforming method based on normality.