KR101571103B1 - Apparatus and method for transmitting linearly in distributed mimo system - Google Patents

Abstract

The present invention relates to an optimal linear transmission apparatus and method in a multiple input multiple output (MIMO) system. A method for data transmission of a base station in a distributed MIMO system in which geographically dispersed base stations are connected by wired or leased lines according to the present invention is characterized in that MMSE (Minumum Mean Square Error) transmission is performed in consideration of channel information and interference power per user. Calculating a preprocessing matrix, calculating a power control variable that maximizes a total transmission capacity while satisfying a power constraint condition per base station, allocating power to data to be transmitted using the calculated power control variable, Performing a pre-processing on the data to which the power is allocated using the calculated MMSE transmission pre-processing matrix, and transmitting the pre-processed transmission data.

Distributed Multiple Input Multiple Output (MIMO), Minumum Mean Square Error (MMSE), Maximum Power Constraint, Power Control, Simulated Annealing

Description

[0001] APPARATUS AND METHOD FOR TRANSMITTING LINEARLY IN DISTRIBUTED MIMO SYSTEM [0002]

The present invention relates to a transmission apparatus and method for increasing the total transmission capacity for multiple users in a downlink transmission in a distributed MIMO system, and more particularly, (DPC), known as the upper limit of the theoretical performance, by transmitting a linear multi-user signal optimized according to a minimum mean square error (MMSE) And more particularly, to a transmission apparatus and method that are close to each other.

2. Description of the Related Art [0002] Recently, as a demand for high-speed and high-quality data transmission has been increased, a technique of satisfying such a demand has attracted a great deal of attention as a multiple input / output (MIMO) technique using a plurality of transmitting / receiving antennas . The MIMO technique performs communication using a plurality of channels due to a plurality of antennas, thereby greatly improving the maximum transmission capacity compared with the case of using a single antenna.

Among the systems to which the MIMO technique is applied, a distributed MIMO system utilizing distributed units is a system in which a plurality of units distributed in a geographically distributed manner are connected by a wired or a dedicated line to share information of terminals, And transmits the data by using it. Here, the unit may be one of a base station, a relay station, and a separate amplifier. In the following description, the base station will be described as an example.

In a distributed transmission scheme of the distributed MIMO system, a single-antenna transmission (hereinafter referred to as 'SAT') scheme in which one base station transmits a signal to one UE during a given transmission interval, A plurality of base stations can simultaneously transmit signals to a plurality of terminals. In this case, the calculation of the maximum transmission capacity for the latter scheme is complicated, and a series of optimization processes including power control of the transmission signals for each terminal is required.

The main feature distinguishing the distributed MIMO system from the existing centralized MIMO system is that, in the case of the centralized MIMO system, a plurality of transmission antennas exist in one base station and a plurality of terminals are spaced the same distance on average from the base station The average received power and the average interference power of each terminal are equal to each other. On the other hand, in the case of the distributed MIMO system, the average received power and the average interference power for terminals allocated in a cell from a geographically dispersed base station are all It is different. This implies that the theoretical transmission capacity derived under the structure of a conventional centralized MIMO system can not be directly applied to the structure of a distributed MIMO system and must be derived by a partial deformation or another approach.

를 넘지 않아야한다는 전체 전력 제약(Total Power Constraint : 이하 'TPC'라 칭함) 조건이 있었으나, 분산 MIMO 시스템을 포함하여 셀룰러 시스템의 송신 안테나마다 존재하는 고유한 전력증폭기 특성을 감안할 때, 송신 안테나당 전력 제약(Per-Antenna Power Constraint : 이하 'PAPC'라 칭함) 조건이 있음을 가정하는 것이 보다 현실적이고 중요한 요소라는 견해가 지배적이다. 상기 분산 MIMO 시스템의 경우, 지리적으로 분산된 기지국마다 독립적인 PAPC 조건이 적용되는 것이 타당하며, 이러한 PAPC 조건 하에서의 다중 사용자 전송 시 최대 전송 용량에 관한 연구 결과는 미미한 실정이다.Another great feature of the distributed MIMO system is that each base station has its own maximum transmittable power constraint. In the conventional centralized MIMO system, the sum of the average signal power transmitted for each transmission antenna is the total power

The total power constraint (hereinafter referred to as " TPC ") is required. However, considering the characteristics of a power amplifier unique to each transmission antenna of a cellular system including a distributed MIMO system, It is more and more important to assume that there is a constraint (Per-Antenna Power Constraint (PAPC)) condition. In the case of the distributed MIMO system, it is reasonable that an independent PAPC condition is applied to each geographically dispersed base station, and the result of research on the maximum transmission capacity in the case of multiuser transmission under the PAPC condition is insufficient.

The maximum transmission capacity of the distributed MIMO system has been analyzed and presented based on the above two main characteristics. Therefore, there is a need for a transmission scheme that can increase the total transmission capacity for multiple users in a PAPC condition applied to each base station and in a cellular system in which different interference power exists for each terminal.

It is an object of the present invention to provide an optimal linear transmission apparatus and method in a distributed MIMO system.

It is another object of the present invention to provide a transmission apparatus and method for increasing the total transmission capacity for multiple users in downlink transmission in a distributed MIMO system.

Another object of the present invention is to provide a method and apparatus for transmitting a multi-user signal that is optimized in accordance with a minimum mean square error (MMSE) condition, taking into account independent maximum power constraints for each geographically dispersed base station , Which is close to the performance of an optimal nonlinear dirty paper coding (DPC) known as the upper limit of the theoretical performance.

It is a further object of the present invention to provide a method and a system for deriving an MMSE conditional formula in a distributed MIMO system in an environment having PAPC and different interference powers for each user and using the MMSE conditional formula as a transmission pre- And a second step of calculating the maximum transmission capacity by performing optimization on the second transmission method based on the MMSE.

It is another object of the present invention to provide a method of using a simulated annealing (SA) algorithm as a power control method for an optimization problem in a downlink transmission in a distributed MIMO system.

According to an aspect of the present invention, there is provided a method for transmitting data of a base station in a distributed MIMO system in which geographically distributed base stations are connected by a wired or leased line, Calculating a MMSE transmission pre-processing matrix by calculating a power control variable that maximizes a total transmission capacity while satisfying a power constraint condition per base station; Allocating power to data to be transmitted, performing pre-processing on the data to which the power is allocated using the calculated MMSE transmission pre-processing matrix, and transmitting the pre-processed transmission data .

According to one aspect of the present invention, there is provided an apparatus for data transmission in a base station in a multiple input multiple output (MIMO) system in which geographically distributed base stations are connected by a wire or a dedicated line, An MMSE transmission pre-processing matrix calculator for calculating a MMSE transmission pre-processing matrix in consideration of channel information and interference power per user, a power control variable for maximizing a total transmission capacity while satisfying power constraints per base station A power allocation unit for allocating power to data to be transmitted using the calculated power control variable; and a pre-processing unit for performing pre-processing on the data to which the power is allocated using the calculated MMSE transmission pre-processing matrix And a preprocessing unit for transmitting the preprocessed transmission data. It is gong.

As described above, the present invention is advantageous in that a high transmission capacity can be achieved with low complexity under a power constraint condition per base station by applying the MMSE-based two-stage transmission scheme for downlink multi-user transmission in the distributed MIMO system. That is, based on the MMSE transmission pre-processing matrix, which is channel adaptive and uniquely determined by the ZF filter type in the high SINR and the transmission matched filter type in the low SINR, a trade-off relationship between transmission complexity and performance There is an advantage in that the transmission capacity can be greatly increased through an appropriate level of additional power control techniques. Also, when nonlinear dirty paper coding is applicable, it can be combined with the proposed scheme to further increase the transmission capacity.

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.

Hereinafter, a transmission apparatus and method for increasing a total transmission capacity for multiple users in a downlink transmission in a distributed MIMO system will be described. In particular, the present invention relates to a method and apparatus for optimizing a linear multi-user (MMSE) optimized in accordance with a minimum mean square error (MMSE) condition, taking into account independent maximum power constraints for each geographically dispersed base station Transmission apparatus and method that approximates the performance of an optimal non-linear paper-based coding (hereinafter referred to as " DPC "), known as the upper limit of the theoretical performance, by transmitting a signal.

Hereinafter, the distributed MIMO system according to the present invention will be described as an example of a distributed MIMO system divided into sectors, but is also applicable to a distributed MIMO system not divided into sectors.

1 is a diagram illustrating a distributed MIMO system divided into sectors according to the present invention.

1, the distributed MIMO system includes at least one base station 111, 112, and 113 that are geographically distributed and can communicate with at least one terminal 101, 102, and 103, , 112, 113) for communicating with a plurality of terminals (101, 102, 103). Here, the hexagons centering on each of the base stations 111, 112, and 113 exemplify coverage that can be transmitted when the base stations 111, 112, and 113 independently transmit signals.

The base stations 111, 112 and 113 are connected to each other via a wire or a dedicated line to share information of the terminals 101, 102 and 103 and to transmit the same frequency band (i.e., a frequency reuse factor of 1) use. Each of the base stations 111, 112, and 113 has three sector antennas, and each sector antenna independently performs signal transmission in the corresponding sector region. Here, a plurality of antennas may be provided for each sector in each of the base stations 111, 112, and 113, and the terminals 101, 102, and 103 may also have a plurality of antennas. It is assumed that there is one antenna per antenna.

The neighboring three base stations 111, 112 and 113 perform multiuser cooperation transmission on the terminals 101, 102 and 103 in the three sector areas adjacent to each other. At this time, for each of the base stations 111, 112 and 113, one sector antenna corresponding to the three sector areas adjacent to each other participates in multi-user cooperative transmission for the terminals 101, 102 and 103 do. Similarly, for each of the base stations 111, 112, and 113, the remaining two sector antennas other than the one sector antenna also cooperatively transmit with neighboring base stations in different directions. Each of the terminals 101, 102, and 103 randomly and uniformly distributes in the corresponding sector region, and the neighboring base stations 111, 112, and 113 Quot;) < / RTI >

Meanwhile, a DPC known as an optimal scheme for downlink multi-user transmission under the existing TPC will be described as follows. The DPC, which is known as the upper limit of the theoretical performance, is a nonlinear signal processing process, and has a complexity that is not suitable for practical application. The linear multi-user transmission scheme considering such an implementation aspect includes a Zero Forcing (ZF) beam forming technique and an MMSE based transmission technique (or Wiener filtering). Among them, the MMSE based transmission scheme is known as an optimal linear transmission scheme under an environment having the same interference power for each TPC and user.

Therefore, in the present invention, the first step of deriving the MMSE conditional expression in an environment having PAPC and different interference power for each user and using the MMSE conditional expression as a transmission pre-processing matrix, and further optimizing the power control variable for each user, The proposed method is based on the MMSE based two - stage transmission method.

In order to calculate the optimal DPC performance, a received signal model for performing downlink multi-user transmission to K terminals having M receive antennas with a single receive antenna is expressed as Equation (1) .

는 수신 신호 벡터이고, 상기

는 송수신 안테나 사이에 형성되는 네트워크 채널 행렬로서, 원소

는 k번째 단말과 m번째 기지국 사이의 채널 이득을 의미한다. 상기

는 송신 선처리 행렬로서 최적화의 대상이 된다. 상기

는 k번째 단말로 전송하고자 하는 복소 심볼이며, 상기

는 k번째 단말이 수신하게 되는, 열잡음을 포함하는 외부의 복소 간섭 신호이다. Here, all the matrices and vectors have a complex symbol as an element. remind

Is a received signal vector,

Is a network channel matrix formed between the transmitting and receiving antennas,

Denotes the channel gain between the k-th UE and the m-th BS. remind

Is a target of optimization as a transmission preprocessing matrix. remind

Is a complex symbol to be transmitted to the k < th > terminal,

Is an external complex interference signal including a thermal noise to be received by the k < th >

라 하면, 기지국 당 최대 전송 가능한 전력 제약(per-BS power constraint) 조건식, 즉 PAPC에 의한 제약 조건식은 하기 <수학식 2>와 같이 나타낼 수 있다. Here, the maximum transmittable power per each base station is

The per-BS power constraint condition equation, i.e., the constraint condition formula by the PAPC, can be expressed as Equation (2).

에 대한 전송 용량 최대화를 목적으로 하는 최적화 문제는 하기 <수학식 3>과 같이 형성된다.Here, an arbitrary transmission preprocessing matrix satisfying Equation (2)

The optimization problem for maximizing the transmission capacity for the mobile station is formed as shown in Equation (3) below.

은

의 분산으로서 k번째 단말에서의 간섭 신호 전력값이다. 상기 <수학식 3>을 이용하여 전체 탐색을 통해, 단말마다의 쉐넌(Shannon) 전송 용량을 합산한 전체 전송 용량을 최대화하는

를 결정하면, 상기 <수학식 1>의 수신 신호 모델 하에서의 최적의 DPC 성능을 얻을 수 있다. 여기서, 로그 항의 분모의 합산 부분에서

인 부분만 간섭 전력 성분으로 적용되는 것이 DPC에 의해 얻을 수 있는 이득이다. 이와 같이, DPC는 주어진 채널 행렬

의 행 간의 정렬 순서에 따라 그 성능이 영향을 받으므로,

의 행들간의 위치를 서로 바꾸는 총 K!(= K(K-1)…1) 가지 수의 사용자 정렬에 따른 상기 <수학식 1>을 이용하여, 사실상 동일한 K! 가지 수의 다른 최적화 문제를 풀고, 그 중 가장 최대의 목적 함수 값을 선택하면, 주어진 채널 정보 하에서의 최적의 DPC 성능이 계산된다. 예를 들어,

의 행들 중 행 1과 2가 교환되면, 그에 따라

의 열 1과 2가, 그리고

과

가 서로 교환되어 상기 <수학식 3>에 적용된다.Here,

silver

Is the interference signal power value at the k < th > The total transmission capacity obtained by adding the Shannon transmission capacity for each terminal is maximized through the entire search using Equation (3)

The optimal DPC performance under the received signal model of Equation (1) can be obtained. Here, in the summation part of the denominator of the log term

Is the gain that can be obtained by the DPC. Thus, the DPC is a channel matrix

Since the performance is affected according to the sorting order of the rows of < RTI ID = 0.0 >

K = K (K-1) ... 1) number of columns in which the positions of the rows of K! By solving other optimization problems of the number of branches and selecting the largest objective function value among them, the optimal DPC performance under given channel information is calculated. E.g,

&Lt; / RTI > of the rows of < RTI ID = 0.0 >

Columns 1 and 2, and

and

Are exchanged with each other and applied to Equation (3).

Instead of the above nonlinear transmission scheme, considering an actual implementation complexity, an optimal linear transmission scheme maximizing a transmission capacity of the entire user may be applied, and Equation (3) may be expressed as Equation (4) have.

로만 바뀌었음을 확인할 수 있으며, 이는 DPC에 의한 효과만을 제거하고 마찬가지의 방법으로 선처리 행렬

에 대한 전체 탐색을 수행함을 의미한다. 최적화 변수의 개수는, 상기 <수학식 3>과 <수학식 4> 모두 행렬

의 원소 개수에 각 원소의 실수부 및 허수부 각각이 최적화 변수가 되므로, 2MK이고, 상기 <수학식 3>의 경우 사용자 정렬 가지수 K!만큼 최적화 과정을 반복한다는 차이가 있다. 상기 <수학식 3>과 <수학식 4>에서 선처리 행렬

에 대한 효과적인 탐색 방안으로 시뮬레이티드 어넬링(Simulated Annealing : SA) 알고리듬이 적용 가능하다. That is, when Equation (3) is compared with Equation (3), the Equation (4)

This is because only the effect of the DPC is removed and the pre-processing matrix

Quot; < / RTI > The number of optimization variables is expressed by Equation (3) and Equation (4)

Is 2MK since the real number part and the imaginary part of each element are the optimization variable in the number of elements in the equation (3). In the case of Equation (3), the optimization process is repeated by K! In the equations (3) and (4), the preprocessing matrix

Simulated Annealing (SA) algorithm can be applied as an effective search algorithm for

에 대한 최적화 문제를 풀고, 그 중 가장 최대의 목적 함수 값을 선택하면, 상기 <수학식 1>의 수신 신호 모델 하에서의 최적의 성능이 계산될 수 있다. 하지만, 실제적으로 상기 <수학식 1>의 수신 신호 모델 하에서 상기 송신 선처리 행렬

를 최적화하기란 어려운 문제점이 있다. Here, as shown in Equation (4), by applying an optimal linear transmission scheme that maximizes the transmission capacity of the entire user, a transmission preprocessing matrix

The optimum performance under the received signal model of Equation (1) can be calculated by solving the optimization problem for Equation (1) and selecting the maximum objective function value among them. However, in practice, under the received signal model of Equation (1), the transmit pre-

There is a problem in that it is difficult to optimize.

를, MMSE 송신 선처리 행렬

와 전력 제어를 위한 대각 행렬

로 분할하여 최적화를 수행하기로 한다. Therefore, in the present invention, the transmission pre-processing matrix in the received signal model of Equation (1)

, An MMSE transmit pre-processing matrix

And a diagonal matrix for power control

And the optimization is performed.

로 변형된 수신 신호 모델을 고려하며, 이는 하기 <수학식 5>와 같이 표현할 수 있다. That is, in the present invention,

, Which can be expressed as Equation (5). &Quot; (5) &quot;

는 MMSE 송신 선처리 행렬로서, PAPC 및 사용자별 간섭 전력을 고려하여 계산된다. 상기 MMSE 송신 선처리 행렬은 채널 정보를 송수신단에서 모두 알고 있는 경우 송신부에서 선처리 하는 것으로서, 특히 하향링크 다중 사용 자 전송과 같이 수신 단말 간의 협력적 신호 처리가 불가능한 경우에 효과적으로 사용될 수 있다. 상기

는 전력 제어를 위한 최적화 변수로서, PAPC 및 전체 전송 용량을 최대화를 고려하여 결정된다. Here,

Is an MMSE transmission pre-processing matrix, and is calculated in consideration of PAPC and per-user interference power. The MMSE transmission pre-processing matrix pre-processes the channel information when the channel information is known at the transmitting and receiving end, and can be effectively used when cooperative signal processing between receiving terminals is impossible, such as downlink multiple user transmission. remind

Is an optimization variable for power control, and is determined in consideration of maximizing PAPC and total transmission capacity.

2 is a flowchart illustrating a MMSE-based linear transmission method for a multi-user in a base station of a distributed MIMO system according to an embodiment of the present invention.

Referring to FIG. 2, in step 201, the BS periodically receives feedback information from the UEs. Here, the feedback information indicates channel information.

In step 203, the BS determines whether there is data to be transmitted. If there is no data to be transmitted, the base station returns to step 201 and repeats the following steps. On the other hand, if there is data to be transmitted, the BS calculates an MMSE transmission pre-processing matrix using the channel information in step 205, and calculates the MMSE transmission pre-processing matrix considering PAPC and per-user interference power.

First, the MMSE transmission pre-processing matrix may be derived as Equation (6) using an MMSE conditional expression of Gauss-Markoff theorem.

는 송신 신호를 나타내고, 상기

는 수신 신호를 나타낸다. 또한, 상기

는 송신 신호

와 수신 신호

간의 상호 상관 행렬을 나타내고, 상기

는 수신 신호

의 자기 상관 행렬을 나타내며, 각각 하기 <수학식 7>과 같이 유도할 수 있다.Here,

Indicates a transmission signal,

Represents a received signal. In addition,

Lt; RTI ID =

And the received signal

Correlation matrix, and

Lt; RTI ID =

And can be derived as Equation (7), respectively.

Substituting Equation (7) into Equation (6), an optimal MMSE transmission pre-processing matrix G as shown in Equation (8) below can be calculated.

는 지금까지 잘 알려진 최적 MMSE 송신 선처리 행렬의 다중 단위행렬 합산 형태가 아님을 알 수 있다. 즉, 기존에 사용자마다 동일한 간섭량을 가정하는 MIMO 시스템에서는 이 부분을

형태로, 스칼라

만큼 곱해진 단위행렬을 더해줌으로써, 채널 행렬

의 의사역행렬

대신 정규화된 역행렬(regularized inverse matrix)을 사용하여 왔다. 그러나 분산 MIMO 시스템과 같이 사용자마다 상이한 간섭 전력이 존재하는 경우, 상기 <수학식 8>과 같은 MMSE 송신 선처리 행렬을 사용하는 것이 바람직함을 알 수 있고, 그 효과는 기존의 정규화된 역행렬과 같이, 해당 MMSE 송신 선처리 행렬을 사용하는 필터가 높은 신호 대 간섭 및 잡음 비(Signal to Interference plus Noise Ratio : 이하 'SINR'이라 칭함)에서는

형태의 ZF 필터로 동작하며, 낮은 SINR에서는

형태의 송신 정합 필터로 동작하게 된다. 상기 <수학식 8>의 최적 MMSE 송신 선처리 행렬을 사용하는 필터는, 분산 MIMO 구조와 같이 각 단말에 수신되는 외부 간섭 신호들이 독립-비동일 분포(idependent but non-identically distributed : i.n.d.)를 따르는 환경을 정확히 반영하는 송신 필터로서, 집중형 MIMO 구조의 독립-동일 분포(independent and identically distributed : i.i.d.) 환경에서의 송신 필터와 차별화된다.Here, the inverse matrix part of the MMSE transmission pre-

Is not a multi-unit matrix summation form of the well-known optimal MMSE transmit pre-processing matrix. That is, in a conventional MIMO system assuming the same amount of interference for each user,

In the form of scalar

By adding the multiplied unit matrix,

Pseudo-inverse of

Instead, a regularized inverse matrix has been used. However, it is preferable to use an MMSE transmission pre-processing matrix such as Equation (8) when there is different interference power for each user as in the distributed MIMO system. The effect is similar to that of the conventional normalized inverse matrix, In a high signal-to-interference plus noise ratio (SINR) filter using a corresponding MMSE transmit pre-processing matrix,

Type ZF filter, and at low SINR

Type transmission matched filter. The filter using the optimal MMSE transmission pre-processing matrix of Equation (8) may be applied to an environment in which external interference signals received at each terminal are ideally but non-identically distributed (ind) And differentiates it from a transmit filter in an independent and identically distributed (iid) environment of a centralized MIMO structure.

및 사용자별 간섭 전력

에 대해 유일하게 결정되는 MMSE 송신 선처리 행렬을 사용할 때, PAPC에 의한 제약 조 건식을 만족시키기 위한 전력 제어 과정이 필요하다. 여기서, 상기 PAPC에 의한 제약 조건식은 하기 <수학식 9>와 같이 나타낼 수 있다. Then, the base station transmits the channel matrix

And user-specific interference power

There is a need for a power control procedure to satisfy the constraint condition by the PAPC when using an MMSE transmit pre-determined matrix that is uniquely determined for the PAPC. Here, the constraint condition formula by the PAPC can be expressed as Equation (9).

를 조절함에 따라 상기 <수학식 9>의 PAPC에 의한 제약 조건식을 만족시키면서 전체 전송 용량을 최대화하는

들을 결정하는 과정이다. 따라서, 상기 기지국은 207단계에서, 상기 PAPC에 의한 제약 조건식을 만족시키면서 전체 전송 용량을 최대화하는 최적화 변수

들을 결정하며, 이를 위한 최적화 문제는 하기 <수학식 10>과 같이 형성된다.Here, the power control process may include:

To maximize the total transmission capacity while satisfying the constraint condition by PAPC in Equation (9)

. Accordingly, in step 207, the BS determines whether or not an optimization variable < RTI ID = 0.0 >

And the optimization problem for this is defined as Equation (10).

를

와

의 두 가지 행렬로 분리하여 각각을 최적화하는 문제로 분할함으로써, 상기 <수학식 4>에서의 총 2MK 개의 최적화 변수에 대한 최적화를 상기 <수학식 10>에서의 오직 K 개의 최적화 변수

에 대한 최적화로 간략화하였다. 즉, 탐색 복잡도를 반영하는 최적화 변수의 개수가 2M 배 만큼 감소됨을 알 수 있다. 여기서, 상기 <수학식 10>의 최적화 문제에 대한 최적화 변수

결정 방안은 이후 도 3의 전력 제어 알고리듬을 통해 자세히 설명하기로 한다. In Equation (10), as compared with Equation (4), a linear preprocessing matrix

To

Wow

The optimization of the 2MK total optimization variables in Equation (4) is performed by solving only the K optimization variables in Equation (10)

As shown in Fig. That is, it can be seen that the number of optimization variables reflecting the search complexity is reduced by 2M times. Here, the optimization variable for the optimization problem of Equation (10)

The decision scheme will be described later in detail with reference to the power control algorithm of FIG.

In step 209, the BS allocates power to the data to be transmitted using the determined optimization variable.

Then, the BS pre-processes the power-allocated data using the calculated MMSE transmission pre-processing matrix in step 211, and transmits the pre-processed data in step 213.

Thereafter, the base station terminates the algorithm according to the present invention.

결정 방법은, 하기 <수학식 11>과 같이, 모든 전력 제어 변수들, 즉

들을 같은 값

로 할당하되, M 개의 기지국 인덱스에 대 한 각각의 PAPC 중 가장 큰

이 최대 전력

에 도달하도록

를 맞춰주는 방식이다. On the other hand, as a non-optimal power control scheme for the optimization problem of Equation (10), a very simple optimization variable

The determination method is to calculate all power control variables, i.e., < RTI ID = 0.0 >

The same value

, The largest of the PAPCs for the M base station indices

This maximum power

To reach

.

This scheme is the most basic MMSE transmission scheme and can be used as an initial value of the optimal power control algorithm proposed in the present invention.

The power control method for the optimization problem of Equation (10) may include various power control methods. However, in the present invention, the simulated annealing (SA) ) Algorithm is modified and presented in accordance with the optimization problem of Equation (10). The simulated annealing algorithm is a probabilistic search algorithm that can be searched in all regions of a solution, probabil- ically determines whether to accept a solution by probabil- ically generating a new solution in the current solution Repeat the process to perform global optimization.

3 is a flowchart illustrating an optimal power control method using a simulated annealing algorithm in a base station of a distributed MIMO system according to an embodiment of the present invention.

, 가중치

, 온도 함수(temperature function)

, 가중치

를 포함하며, 각각 1, 1, 10, 0.99의 값으로 초기 설정한다. 상기

는 다음 상태값을 계산함에 있어 적용되는 가중치이고, 상기

는 복잡도 및 전송 용량 간의 트레이드 오프 관계를 고려하여 적용되는 가중치로서, 상기

를 높게하여 오랜 시간이 소요되더라도 좀 더 정확한 전력 제어를 수행할 수 있고, 상기

를 낮게하여 다소 부정확하더라도 짧은 시간 내에 전력 제어를 수행할 수 있다. Referring to FIG. 3, in step 301, the base station sets an initial parameter for optimal power control. Here, the parameter is a repeated index

, weight

Temperature function,

, weight

And are initially set to values of 1, 1, 10, and 0.99, respectively. remind

Is a weight applied in calculating the next state value,

Is a weight applied in consideration of the trade-off relationship between the complexity and the transmission capacity,

It is possible to perform more accurate power control even when a long time is required,

The power control can be performed within a short time even if it is somewhat inaccurate.

 In step 303, the BS sets the current state value of the transmission capacity according to the optimization variable and the optimization variable. Here, the current state value of the transmission capacity according to the optimization variable and the optimization variable is set as shown in Equation (12).

 In step 305, the BS sets an initial optimal solution for the transmission capacity according to the optimization variable and the optimization variable. Here, the initial optimal solution is set as shown in Equation (13).

In step 307, the BS sets the next state value of the transmission capacity according to the optimization variable and the optimization variable. Here, the next state value of the transmission capacity according to the optimization variable and the optimization variable is set as Equation (14).

는 구간 [a, b]에서 균등 분포를 따르는 임의의 값을 랜덤하게 발생함을 의미한다. here,

Means that random values occur along the uniform distribution in the interval [a, b].

이 상기 설정된 최적해

보다 큰지 여부를 검사한다. 상기 설정된 전송 용량의 다음 상태 값

이 상기 설정된 최적해

보다 크지 않을 시, 상기 기지 국은 313단계로 바로 진행한다. 반면, 상기 설정된 전송 용량의 다음 상태 값

이 상기 설정된 최적해

보다 클 시, 상기 기지국은 311단계에서 상기 설정된 최적화 변수의 다음 상태 값

과 상기 최적화 변수에 따른 전송 용량의 다음 상태 값

으로, 상기 최적해

와

로 각각 갱신한 후, 상기 313단계로 진행한다. Then, in step 309, the BS calculates a next state value

And the set optimal solution

Is greater than &lt; / RTI &gt; The next state value of the set transmission capacity

And the set optimal solution

The base station proceeds directly to step 313. On the other hand, the next state value of the set transmission capacity

And the set optimal solution

The BS determines the next state value of the set optimization variable in step 311,

And the next state value of the transmission capacity according to the optimization variable

, The optimal solution

Wow

Respectively, and then proceeds to step 313.

이 상기 설정된 전송 용량의 현재 상태 값

보다 큰지 여부를 검사한다. Then, in step 313, the BS calculates a next state value

The current state value of the set transmission capacity

Is greater than &lt; / RTI &gt;

이 상기 설정된 전송 용량의 현재 상태 값

보다 클 시, 상기 기지국은 317단계에서 상기 설정된 최적화 변수의 다음 상태 값

과 상기 최적화 변수에 따른 전송 용량의 다음 상태 값

으로, 상기 설정된 최적화 변수의 현재 상태 값

과 최적화 변수에 따른 전송 용량의 현재 상태 값

을 각각 갱신한 후, 319단계로 진행한다. In step 313, the next state value of the set transmission capacity

The current state value of the set transmission capacity

The BS determines the next state value of the set optimization variable in step 317,

And the next state value of the transmission capacity according to the optimization variable

, The current state value of the set optimization variable

And the current state value of the transmission capacity according to the optimization variable

Respectively, and then proceeds to step 319.

이 상 기 설정된 전송 용량의 현재 상태 값

보다 크지 않을 시, 상기 기지국은 315단계에서 비록 다음 상태의 전송 용량이 현재 상태의 전송 용량보다 낮은 값을 가지더라도 상기 두 전송 용량 값의 차이에 따라 확률적으로 다음 상태의 최적화 변수 및 전송 용량 값을 현재 상태의 최적화 변수 및 전송 용량 값으로 갱신한다. 즉, 상기 기지국은 상기 설정된 전송 용량의 다음 상태 값

이 하기 <수학식 15>를 만족하는지 여부를 검사하고, 상기 설정된 전송 용량의 다음 상태 값

이 하기 <수학식 15>를 만족할 시, 상기 317단계로 진행하여 이하 단계를 반복 수행한다. 반면, 상기 설정된 전송 용량의 다음 상태 값

이 하기 <수학식 15>를 만족하지 않을 시, 상기 기지국은 상기 319단계로 바로 진행한다. On the other hand, if it is determined in step 313 that the next state value

The current status value of the above transmission capacity

If the transmission capacity in the next state is lower than the transmission capacity in the current state, the base station probabilistically determines the next optimization variable and the transmission capacity value in step 315 according to the difference between the two transmission capacity values To the current optimization variable and the transmission capacity value. That is, the BS determines the next state value of the set transmission capacity

(15) is satisfied, and if the next state value of the set transmission capacity

If Equation (15) is satisfied, the process proceeds to step 317 and the following steps are repeated. On the other hand, the next state value of the set transmission capacity

(15) is not satisfied, the BS proceeds directly to step 319.

가 반복 인덱스

의 증가에 따라 서서히 감소하기 때문에, 반복 과정 초기에는 현재 상태 값의 갱신이 빈번히 발생하게 되고, 반복 인덱스

가 증가함에 따라 현재 상태 값을 강제로 갱신하는 확률이 계속 낮아지게 된다. 따라서 결국에는 하나의 지역적 최적점으로 수렴하게 되며, 점근적 분석(asymptotic analysis)에 의하면

가 1에 점근할수록 그 최적값이 전역적 최적점과 일치하게 된다.That is, when the portion calculated as an exponential function in Equation (15) is larger than a random value between 0 and 1, the BS calculates an optimization variable and a transmission capacity value of the next state as an optimization variable and a transmission capacity value . This makes it possible to generalize the search to converge to a global optimum, which is the same effect as the whole search, while avoiding convergence to a local optimum that is found first in the optimization process. At this time,

Repeat Index

, The current state value is updated frequently at the beginning of the iterative process,

The probability of forcibly updating the current state value is continuously lowered. Therefore, it eventually converges to one local optimal point, and asymptotic analysis

Becomes closer to 1, the optimal value becomes equal to the global optimal point.

이 수렴되었는지 여부를 검사한다. 여기서, 상기 전송 용량의 최적해

이 수렴되었는지 여부는 반복 인덱스

가 정해진 횟수까지 증가하였음에도 불구하고, 상기 전송 용량의 최적해

이 갱신되지 않았는지 여부를 확인함으로써 검사할 수 있다. Thereafter, in step 319, the BS calculates an optimal transmission capacity

Is converged. Here, the optimum value of the transmission capacity

Whether convergence has converged or not is determined by the iteration index

Is increased to a predetermined number, the optimal value of the transmission capacity

Can be inspected by checking whether or not it has been updated.

이 수렴되지 않았을 시, 상기 기지국은 321단계에서 상기 파라미터를 갱신하고, 상기 307단계로 돌아가 이하 단계를 반복 수행한다. 즉, 상기 반복 인덱스

를

로 갱신하고, 이에 따라 상기

와

를 각각

,

로 갱신한다. In step 319, the optimal size of the transmission capacity

The BS updates the parameter in step 321 and returns to step 307 to repeat the following steps. That is,

To

, And accordingly,

Wow

Respectively

,

.

이 수렴되었을 시, 상기 기지국은 본 발명에 따른 알고리즘을 종료한다. On the other hand, in step 319,

Is converged, the base station ends the algorithm according to the present invention.

값을 적절한 수준으로 낮추면서 저복잡도 준최적의 전력 제어를 수행하는 것이 가능하다. On the other hand, in the simulated annealing algorithm, considering the trade-off relationship between the complexity and the transmission capacity

It is possible to perform low-complexity sub-optimal power control while lowering the value to an appropriate level.

의 열 별 원소 간 전력 비에 따라 미리 정해 놓은 탐색표(Look-Up Table : LUT)를 참조하여, PAPC 제약 조건식 하에서 초기치

로부터의 증감

를 일대일 대응 형태로 결정하는 방식도 가능하다.In order to reduce the complexity to the maximum, an optimal MMSE transmission pre-processing matrix, which is the first step of the proposed method, is obtained,

The look-up table (LUT) determined according to the row-to-row power ratios of the columns of the PAPC constraint expression,

Increase or decrease from

In a one-to-one correspondence manner.

4 is a block diagram illustrating a transmission apparatus of a base station in a distributed MIMO system according to an embodiment of the present invention.

The base station includes a power allocation unit 402, a pre-processing unit 404, an optimization variable determining unit 406, an MMSE transmission pre-processing matrix calculating unit 408, and a feedback receiving unit 410.

Referring to FIG. 4, the power allocating unit 402 receives an optimization variable from the optimizing variable determiner 406, allocates power to data to be transmitted using the optimizing variable, To the pre-processing unit (404).

The pre-processing unit 404 receives an MMSE transmission pre-processing matrix from the MMSE transmission pre-processing matrix calculation unit 408, preprocesses the power-allocated data using the MMSE transmission pre-processing matrix, and transmits the pre- To the terminal.

The optimization parameter determiner 406 receives the channel information of the UEs from the feedback receiver 410 and receives the MMSE transmission pre-processing matrix from the MMSE transmission pre-processor matrix calculator 408, satisfies the constraint condition by the PAPC Determines an optimization variable that maximizes the total transmission capacity, and provides the determined optimization variable to the power allocation unit 402. [

The MMSE transmission pre-processing matrix calculator 408 receives the channel information of the UEs from the feedback receiver 410, calculates an MMSE transmission pre-processing matrix using the received channel preamble, and transmits the calculated MMSE transmission pre- And an optimization variable determining unit 406. [ At this time, the MMSE transmission pre-processing matrix is calculated in consideration of PAPC and user-specific interference power.

The feedback receiver 410 periodically receives feedback information from the UEs and provides the feedback information to the optimization parameter determiner 406 and the MMSE transmission pre-processing matrix calculator 408. Here, the feedback information indicates channel information.

In order to evaluate the performance of the MMSE-based two-stage transmission method proposed in the present invention, the following simulation environment is defined as shown in Table 1 below.

Key parameters value Total number of base stations 61 Number of simultaneous transmission base stations 3 Number of interference base stations 58 Number of transmit antennas per base station sector One Number of concurrent transmission terminals 3 Number of receiving antennas per terminal One Transmission link Downlink Channel model Flat Rayleigh fading Channel coding No coding Channel estimation Perfect channel information Modulation method N / A (capacity formula used)

Base station marginal power

43 [dBm] Minimum distance between base stations 500 [m] Path loss index 3.76 (w / o shadowing) System bandwidth 10 [MHz] Thermal noise -104 [dBm]

FIG. 5 is a graph showing the transmission capacity of the repeated index in the optimal power control method using the simulated annealing algorithm proposed in the present invention.

Referring to FIG. 5, it can be seen that the transmission capacity of the current state changes frequently at the beginning of the iteration process, but the transmission capacity converges to the optimal value after about 200 iterations.

6 is a graph illustrating a comparison of the performance of the transmission capacity between the linear transmission method based on the MMSE for the multi-user and the linear transmission method according to the present invention as a cumulative distribution function of the transmission capacity per base station.

의 의사역행렬로 사용하고 전력 제어 행렬

를 상기 <수학식 11>과 같이 동일한 전력

로 할당한 경우이며, 'MMSE'로 표기된 방식은 전력 제어 행렬

를 상기 <수학식 11>과 같이 동일한 전력으로 할당하되 송신 선처리 행렬을 상기 <수학식 8>에서 제안한 최적 MMSE 송신 선처리 행렬로 사용한 경우이다. 'ZF + PC'로 표기된 방식은 상기 'ZF'로 표기된 방식에서 추가로 본 발명에서 제안하는 최적의 전력 제어를 수행한 경우이며, 이때는 ZF의 특성상 전력 제어를 위한 최적화 문제가 표준적인 컨벡스(convex) 문제로 형성되어 서브그래디언트(subgradient) 방식 등 최적성을 보장하는 다양한 컨벡스 최적화 기법이 사용 가능하다. 'SAT'로 표기된 방식은 각각의 기지국이 최대 전력

로 해당 섹터에 존재하는 단말에게 전력 제어 과정 없이 독립적인 신호 전송을 하는 경우로서, 서빙 기지국을 제외한 다른 모든 기지국으로부터 단말로 수신되는 신호가 간섭신호가 되는 경우이다. 최적의 전력 제어를 수행하지 않고 사용자마다 동일한 전력으로 할당하는 'ZF' 및 'MMSE' 방식의 경우 SAT 보다 성능이 열화되는 경우가 발생하는데, 이는 SAT의 경우 항상 최대의 송신 전력

로 송신하지만, 'ZF' 및 'MMSE' 방식에서는 최대 전력

로 송신 가능한 하나의 기지국을 제외하고는

이하의 송신 전력으로 전송이 이루어지게 되고 이러한 송신 전력의 손실이 다중 사용자 전송으로부터 얻는 효과보다 더 큰 영향을 주기 때문이다.Referring to FIG. 6, the method indicated by 'MMSE + PC' is an MMSE-based two-step transmission method proposed in the present invention. In the MMSE-based two-step transmission method, 'optimal linear precoder' It can be confirmed that the performance of the transmission method is close to that of the transmission method. Also, the method marked with 'ZF'

And the power control matrix

Is equal to the power &lt; RTI ID = 0.0 &gt;

, And the method indicated by 'MMSE' is a case where the power control matrix

As the Equation (11), and the transmission pre-processing matrix is used as the optimal MMSE transmission pre-processing matrix proposed by Equation (8). The method indicated by 'ZF + PC' is a case where the optimal power control proposed in the present invention is further performed in the manner indicated by 'ZF'. In this case, the optimization problem for power control is a convex ), And various convex optimization techniques can be used to ensure optimality such as a sub-gradient scheme. The scheme labeled 'SAT' indicates that each base station has a maximum power

Is a case where an independent signal is transmitted to a terminal existing in a corresponding sector without a power control process and a signal received from all the other base stations except the serving base station is an interference signal. In the case of 'ZF' and 'MMSE' schemes, which allocate the same power to each user without performing optimal power control, performance sometimes deteriorates compared to SAT. In SAT,

However, in the 'ZF' and 'MMSE' schemes, the maximum power

Except for one base station that can transmit

And the loss of such transmission power has a greater effect than the effect obtained from the multi-user transmission.

FIG. 7 is a graph illustrating a comparison of the performance of the transmission capacity between the linear transmission method based on MMSE for the multi-user and the DPC transmission method according to the present invention as a cumulative distribution function of transmission capacity per base station.

Referring to FIG. 7, the method indicated by 'MMSE + PC' is a MMSE-based two-step transmission method proposed by the present invention, and shows a transmission capacity of 3.3 bps / Hz / BS in the median of the cumulative distribution function . It can be seen that the proposed method exhibits a performance difference of less than 0.1 bps / Hz / BS compared to the sub-optimal scheme using 'ZF-DPC', ie, user alignment and QR decomposition. In addition, it can be seen that the proposed method according to the present invention shows a performance difference of less than 0.3 bps / Hz / BS compared with the method indicated by 'Optimal DPC', that is, the optimal nonlinear DPC method shown in Equation (3) have. The method indicated by 'MMSE-DPC' is a method in which non-linear DPC is applied in addition to the method proposed in the present invention. The performance difference is within 0.05 bps / Hz / BS as compared with the optimal DPC scheme by selecting the maximum transmission capacity among them.

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 illustrates a distributed MIMO system divided into sectors according to the present invention,

FIG. 2 is a flowchart illustrating a MMSE-based linear transmission method for a multi-user in a base station of a distributed MIMO system according to an embodiment of the present invention.

3 is a flowchart illustrating an optimal power control method using a simulated adjacent algorithm in a base station of a distributed MIMO system according to an embodiment of the present invention.

4 is a block diagram illustrating a transmission apparatus of a base station in a distributed MIMO system according to an embodiment of the present invention.

FIG. 5 is a graph showing a transmission capacity of a repeated index in an optimal power control method using a simulated annealing algorithm proposed in the present invention,

FIG. 6 is a graph illustrating a comparison of the performance of the transmission capacity between the linear transmission method based on the MMSE for the multi-user and the linear transmission method according to the present invention as a cumulative distribution function of the transmission capacity per base station,

FIG. 7 is a graph illustrating a comparison of the performance of the transmission capacity between the linear transmission method based on the MMSE for the multi-user and the DPC transmission method according to the present invention as a cumulative distribution function of the transmission capacity per base station.

Claims (18)

A method for data transmission of a base station in a multiple input multiple output (MIMO) system, Determining a minimum mean square error (MMSE) transmission pre-processing matrix considering channel information and interference power per user, Determining power control variables that maximize the total transmission capacity by repeatedly updating state values of the power control variable and the transmission capacity; Adjusting power of signals to be transmitted to each user using power control variables that maximize the total transmission capacity; Pre-processing the signals using the MMSE transmission pre-processing matrix; And transmitting the pre-processed signals, Wherein the power control parameter and the state value of the transmission capacity are updated when the next state value of the transmission capacity is larger than the current state value of the transmission capacity, and when the next state value is not larger than the current state value, Updated, Wherein the power control variables maximizing the total transmission capacity are determined such that a maximum of the transmission power values of the plurality of base stations reaches a maximum power value that can be transmitted per base station. The method according to claim 1, Wherein the MMSE transmission pre-processing matrix is determined according to Equation (16).

Here,

Indicates a transmission signal,

Represents a received signal. In addition,

Lt; RTI ID =

And the received signal

Correlation matrix, and

Lt; RTI ID =

&Lt; / RTI &gt; remind

Is a network channel matrix formed between the transmitting and receiving antennas,

Is the interference power of the k &lt; th &gt; user,

Represents the transmission power of the m-th base station. The method according to claim 1, Wherein the power control variables are determined by searching for a solution to the optimization problem that maximizes the total transmission capacity using Equation (17): < EMI ID = 17.0 >

Here, M represents the number of base stations having a single transmit antenna, and K represents the number of terminals having a single receive antenna. remind

Is the interference power of the k-th user,

Is the channel gain between the k-th UE and the m-th base station,

Is an element of the MMSE transmission pre-processing matrix, and is an element between the k-th terminal and the m-th base station. remind

Is a power control variable for the k &lt; th &gt; terminal,

Is the maximum power of the base station. The method according to claim 1, Wherein the step of determining the power control variables maximizing the total transmission capacity comprises: And a step of repeatedly updating the power control variables and determining whether or not convergence occurs, Wherein the power control variables are updated based on a weighted value multiplied by a random value that follows an initial power control variable value and an even distribution. The method of claim 3, Wherein the step of determining the power control variables maximizing the total transmission capacity comprises: Initializing a current state value of the transmission power according to the power control variable and the power control variable; Initializing an optimal solution of the power control variable and the transmission capacity according to the power control variable as the current state value; Setting the next state value of the transmission power according to the power control variable and the power control variable; Updating an optimal solution of the power control variable and the transmission capacity to a next state value of the power control variable and the transmission capacity when the next state value of the transmission capacity is greater than the optimal solution of the transmission capacity; Updating the current state value of the power control variable and the transmission capacity to the next state value of the power control variable and the transmission capacity when the next state value of the transmission capacity is greater than the current state value of the transmission capacity; And determining the optimal solution of the transmission capacity as a solution to the optimization problem of maximizing the total transmission capacity when the optimal solution of the transmission capacity is not updated more than a predetermined number of repetitions. 6. The method of claim 5, Wherein the current state value of the power control variable and the transmission capacity according to the power control variable is initially set using Equation (18).

6. The method of claim 5, Wherein the power control variable and the next state value of the transmission capacity according to the power control variable are set using Equation (19).

here,

Means to randomly generate an arbitrary value following the even distribution in the interval [a, b], and

Is a repeated index,

Is a weight. 6. The method of claim 5, And when the next state value of the transmission capacity is not larger than the current state value of the transmission capacity, the power control variable and the transmission capacity are determined as the next state value of the transmission capacity in accordance with the difference between the next state value and the current state value. And updating the current state value of the control variable and the transmission capacity. The method according to claim 1, Further comprising the step of periodically receiving feedback of the channel information from the terminal. 1. An apparatus for data transmission in a base station in a multiple input multiple output (MIMO) system, An MMSE transmission pre-processing matrix calculator for determining a minimum mean square error (MMSE) transmission pre-processing matrix considering channel information and interference power per user, An optimization variable determining unit that determines power control variables that maximize the total transmission capacity by repeatedly updating the state values of the power control variable and the transmission capacity; A power allocator for adjusting power of signals to be transmitted to each user by using power control variables for maximizing the total transmission capacity; And a pre-processing unit for pre-processing the signals using the MMSE transmission pre-processing matrix and transmitting the pre-processed signals, Wherein the power control parameter and the state value of the transmission capacity are updated when the next state value of the transmission capacity is larger than the current state value of the transmission capacity, and when the next state value is not larger than the current state value, Updated, Wherein the power control variables maximizing the total transmission capacity are determined such that a maximum of the transmission power values of the plurality of base stations reaches a maximum power value that can be transmitted per base station. 11. The method of claim 10, Wherein the MMSE transmission pre-processing matrix is determined according to Equation (20).

Here,

Indicates a transmission signal,

Represents a received signal. In addition,

Lt; RTI ID =

And the received signal

Correlation matrix, and

Lt; RTI ID =

&Lt; / RTI &gt; remind

Is a network channel matrix formed between the transmitting and receiving antennas,

Is the interference power of the k &lt; th &gt; user,

Represents the transmission power of the m-th base station. 11. The method of claim 10, Wherein the power control variables maximizing the total transmission capacity are determined by searching a solution of an optimization problem maximizing the total transmission capacity using Equation (21).

Here, M represents the number of base stations having a single transmit antenna, and K represents the number of terminals having a single receive antenna. remind

Is the interference power of the k-th user,

Is the channel gain between the k-th UE and the m-th base station,

Is an element of the MMSE transmission pre-processing matrix, and is an element between the k-th terminal and the m-th base station. remind

Is a power control variable for the k &lt; th &gt; terminal,

Is the maximum power of the base station. 11. The method of claim 10, Wherein the optimization variable determiner repeatedly updates power control variables for maximizing the total transmission capacity and determines whether to converge, Wherein the power control variables maximizing the total transmission capacity are updated based on a weighted value multiplied by an initial power control variable value and a random value along an even distribution. 13. The method of claim 12, The power allocating unit allocates, Means for initializing the power control variable and a current state value of a transmission capacity according to the power control variable; Means for initializing, as the current state value, an optimal solution of the power control variable and a transmission capacity according to the power control variable; Means for setting the power control variable and a next state value of a transmission capacity according to the power control variable; Means for updating an optimal solution of the power control variable and the transmission capacity to the next state value of the power control variable and the transmission capacity when the next state value of the transmission capacity is greater than the optimal solution of the transmission capacity; Means for updating the current state value of the power control variable and the transmission capacity to the next state value of the power control variable and the transmission capacity when the next state value of the transmission capacity is greater than the current state value of the transmission capacity; And means for determining the optimal solution of the transmission capacity as a solution to the optimization problem that maximizes the total transmission capacity when the optimal solution of the transmission capacity is not updated more than a predetermined number of repetitions. 15. The method of claim 14, Wherein the current state value of the power control variable and the transmission capacity according to the power control variable is initially set using Equation (22).

15. The method of claim 14, Wherein the power control parameter and the next state value of the transmission capacity according to the power control variable are set using Equation (23).

here,

Means to randomly generate an arbitrary value following the even distribution in the interval [a, b], and

Is a repeated index,

Is a weight. 15. The method of claim 14, And when the next state value of the transmission capacity is not larger than the current state value of the transmission capacity, the power control variable and the transmission capacity are determined as the next state value of the transmission capacity in accordance with the difference between the next state value and the current state value. And means for updating the current state value of the control variable and the transmission capacity. 11. The method of claim 10, Further comprising: a feedback receiver for periodically receiving feedback of the channel information from the terminal.

Images