KR101765127B1 - Apparatus and method for clustering for coordinated multi point ommunication in cellular communication system - Google Patents

Abstract

The present invention relates to a clustering method and apparatus for cooperative communication in a cellular communication system, and a method of clustering a base station for cooperative communication in a central processor includes collecting channel state information for a plurality of base stations, Comprising the steps of: constructing a cluster by considering a difference between a capacity for cooperative communication between base stations and a capacity for a single cell communication of the predetermined number of base stations; So that the cluster can be configured.

Description

[0001] APPARATUS AND METHOD FOR CLUSTERING FOR COORDINATED MULTI POINT OMMUNICATION IN CELLULAR COMMUNICATION SYSTEM [0002]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a cellular communication system capable of cooperation between base stations, and more particularly, to a method and apparatus for configuring a cluster for cooperative communication between base stations.

If neighboring cells use the same frequency in a cellular communication environment, interference in the cell boundary region can be seriously generated. Accordingly, in the cellular communication system, various methods for improving cell boundary capacity while reducing inter-cell interference are being provided. For example, Coordinated Multi-Point (CoMP) scheme is provided in which neighboring cells cooperate in consideration of an instantaneous channel value and a traffic situation of a terminal located at a cell boundary.

For the cooperative communication scheme, it is necessary to share information between adjacent base stations to perform cooperative communication. However, the process of sharing information between the base stations may cause problems such as overload, delay, and difficulty in network construction. Therefore, in recent years, a method of forming a cluster between adjacent base stations so that cooperative communication can be performed between a limited number of base stations has been studied.

In particular, the cluster configuration method can be roughly classified into three types. That is, the cluster configuration method can be classified into a static clustering method, a full-search based dynamic clustering method, and a greedy-search based dynamic clustering method.

The static clustering scheme is a scheme of initially configuring adjacent base stations as clusters and keeping the cluster initially configured. The static clustering scheme can easily and quickly construct clusters, but it is difficult to completely eliminate interference between clusters, and there is a high probability that a user located at a cluster boundary is exposed to interference and the performance is degraded. In addition, since the static clustering method keeps the initial cluster, it is not efficient for time-varying channels, and diversity gain can not be obtained.

The full-search-based dynamic clustering method (hereinafter, referred to as "all-region search method") is a method of configuring clusters in consideration of a user environment that varies with time in a central processing unit, And selects a cluster that maximizes a data transmission capacity of a user among clusters that can be obtained through a combination of all the base stations every predetermined period. However, since the transmission capacity needs to be calculated for all possible combinations of base stations, the entire area search scheme is very complex and difficult to implement realistically.

The greedy-search-based dynamic clustering method (hereinafter referred to as a greedy search method) is a method of configuring clusters in consideration of a user environment that varies with time in a central processing unit as in the full area search method , More specifically, selects one base station and then selects a base station that maximizes the data transmission capacity of the user when cooperating with the selected base station. Unlike the full area search method, the greedy search method has the advantage of reducing the complexity by not considering all possible combinations of base stations, but it has a problem of not being able to construct an appropriate cluster. That is, the user transmission capacity may increase by reducing the interference through cooperation, but may also increase when each user is located at the cell center. Therefore, when selecting a base station that maximizes the transmission capacity of a user, there is a problem in that the user is likely to select a base station located at the center of the cell, thereby forming an inefficient cluster.

Therefore, it is necessary to provide a method of efficiently clustering base stations while considering a user environment changing with time.

It is an object of the present invention to provide a method and apparatus for configuring a cluster for cooperative communication between base stations in a cellular communication system.

It is another object of the present invention to provide a method and apparatus for clustering base stations that maximize capacity gain in cooperation in a cellular communication system.

It is still another object of the present invention to provide a method and an apparatus for clustering base stations with high interference between each other in a cellular communication system.

According to a first aspect of the present invention, there is provided a method for clustered base stations for cooperative communication in a central processor of a cellular communication system, the method comprising: collecting channel state information for a plurality of base stations; And configuring a cluster by taking into consideration the difference between the capacity for the cooperative communication of the base stations and the capacity for the single cell communication of the predetermined number of base stations.

According to a second aspect of the present invention, there is provided an apparatus for clustering base stations for cooperative communication in a central processor of a cellular communication system, comprising: a channel information collecting unit for collecting channel state information for a plurality of base stations; And a base station clustering unit for forming a cluster considering a difference between a capacity for a cooperative communication of a predetermined number of base stations and a capacity for a single cell communication of the predetermined number of base stations.

The present invention can achieve high performance while clustering base stations that maximize the capacity gain due to cooperation in a cellular communication system or by clustering base stations with large interference between them, There is an effect that the cluster can be constituted by efficiently reflecting the environment of the user changing according to the flow.

1 is a diagram showing a system configuration for configuring a cluster in a cellular communication system according to the present invention;
Figure 2 shows a simplified procedure for establishing a cluster and cooperating communication in a cellular communication system according to the present invention;
3 is a block diagram of a base station and a central processor in a cellular communication system according to the present invention;
4 is a detailed block diagram of a base station clustering unit in a central processor according to an embodiment of the present invention;
5 is a detailed block diagram of a base station clustering unit in a central processor according to another embodiment of the present invention;
6 is a diagram illustrating a clustering procedure of a central processor according to an embodiment of the present invention;
FIG. 7 illustrates a clustering procedure of a central processor according to another embodiment of the present invention; FIG.
FIG. 8 is a diagram illustrating a correlation between interference between base stations performing cooperative communication and non-cooperative communication according to another embodiment of the present invention; FIG.
9 is a diagram illustrating a correlation between a capacity gain for a case where the cooperative communication between the base stations is performed and a case where the cooperative communication is not performed according to another embodiment of the present invention;
FIGS. 10A and 10B are diagrams showing data rates for the case of clustering according to the conventional technique and the case of clustering according to the present invention,
FIGS. 11A and 11B are diagrams showing data rates of edge users in the case of clustering according to the conventional technique and the case of clustering according to the present invention, and FIGS.
Figure 12 shows the complexity and performance for a case of clustering according to the prior art and a case of clustering according to the present invention.

Hereinafter, preferred embodiments of the present invention will be described with reference to 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. In addition, the terms described below are defined in consideration of the functions of the present invention, which may vary depending on the intention of the user, the operator, or the custom. Therefore, the definition should be based on the contents throughout this specification.

Hereinafter, a method and an apparatus for configuring a base station cluster for performing cooperative communication by reflecting a channel environment of a user that changes with time in a cellular communication system will be described.

In the present invention, the cooperative communication is performed in consideration of a case where a plurality of base stations perform cooperative communication and a case where the plurality of base stations do not perform cooperative communication and use single cell communication, And selects the base stations whose gain is maximized to construct a cluster.

In particular, the present invention proposes two methods of constituting clusters.

One is a method of constructing a cluster by selecting base stations whose capacity is greatest when a plurality of base stations perform cooperative communication and when the plurality of base stations do not perform cooperative communication, (Max-Capacity Difference) method.

The other is to consider the interference between two base stations having a proportional relation to the capacity difference when two base stations perform cooperative communication and when the two base stations do not perform cooperative communication, (Hereinafter referred to as an " Interference Weight ") scheme.

Before describing the method of configuring the cluster, a cellular communication system according to the present invention will be described.

1 shows a system configuration for configuring a cluster in a cellular communication system according to the present invention. Here, although not shown, each base station transmits channel information about all the nearby users to which the signal is received to the central processor. That is, it is assumed that the central processor knows all the channel information. It is also assumed that all base stations in the cluster share channel information and user data with each other.

As shown in FIG. 1, the present invention describes a clustering method for an uplink when each base station supports K users in an environment where there are N cells. In the present invention, each cluster is composed of B base stations, so that each cluster includes a total of BK users. At this time, the cluster is reconstructed every predetermined time. In the present invention, it is assumed that each base station includes N _b antennas, and each user includes one antenna. It is also assumed that each cluster receives a user's signal through cooperation between the base stations using a linear receiver. Here, the linear receiver includes both a ZF (Zero-Forcing) receiver and a linear receiver such as a Minimum Mean-Square Error (MMSE) receiver.

Then, in order to cluster each base station in the central processor, a received signal of the cluster will be described.

According to the above assumption, the reception signal of the cluster G can be expressed by the following equation (1).

는 상기 클러스터 G에 대한 다른 클러스터의 간섭을 나타낸다.Here, y _G denotes a received signal of the cluster G, and H _G denotes a channel between all base stations and users belonging to the cluster G. Here, H _G has a size of BN _b x BK. Also, s _G is a vector representing a signal transmitted by all users belonging to the cluster G, and n _G is a noise vector of BN _b × 1 for the cluster G. Where the noise vector n _G follows a normal distribution where each component is independent, the mean is zero, and the variance is σ ² _n . In addition,

Represents the interference of the other cluster with respect to the cluster G.

When the reception signal of the cluster G is detected using a linear receiver, the following equation (2) can be obtained.

는 상기 클러스터 G에 대한 다른 클러스터의 간섭을 나타낸다.Here, W _G denotes a linear receiver, H _G denotes a channel between all base stations and users belonging to the cluster G, and s _G denotes a signal transmitted by all users belonging to the cluster G. N _G denotes a noise vector of BN _b × 1 for the cluster G,

Represents the interference of the other cluster with respect to the cluster G.

Signal-to-interference-noise ratio of the user b (i) b, located in the base station belonging to the cell, assuming that the linear receiver W _G The equation (3) and ZF (Zero-Forcing) receiver, such as down, cluster G (SINR: Signal-to- Interference-plus-Noise-Ratio) can be expressed by Equation (4) below.

Here, with use of the linear receiver W _G, a signal of user i belonging to the cluster is expressed by G 1, the signals of the other users who belong to the cluster G are both knurled (nulling).

Therefore, the signal-to-interference noise ratio of the user b (i) located in the cell of the base station b belonging to the cluster G can be simply expressed by Equation (5) below.

는 클러스터에 속한 기지국 b의 셀에 위치한 사용자 b(i)의 신호대 간섭 잡음비를 나타낸다.
Here,

Represents the signal-to-interference noise ratio of the user b (i) located in the cell of the base station b belonging to the cluster.

Hereinafter, the two clustering methods proposed by the present invention will be described in detail based on the above description.

First, a Max-Capacity Difference (Max-CD) method will be described.

As shown in Equation (6), the Max-CD scheme is a scheme in which the capacity when a plurality of base stations perform cooperative communication and the capacity when a plurality of base stations perform single cell communication without cooperative communication The cluster is configured to maximize the difference.

^(C) denotes a capacity sum of users belonging to the cluster G when the B base stations perform cooperating communication, and C ^(N) denotes a capacity sum of users belonging to the cluster G, Represents a capacity sum of users belonging to the cluster G when the B base stations perform single cell communication. In the following description, for convenience of description, the C ^(C) is referred to as a capacity when the corresponding base stations perform cooperative communication, and the C ^(N) is referred to as a capacity when the corresponding base stations perform single cell communication .

Here, the difference between C ^(C) and C ^(N) can be expressed as Equation (7) using Equation (5 ⁾ and Equation (6) can be summarized as Equation (8) .

는 클러스터에 속한 기지국들이 협력 통신하는 경우에 기지국 b의 셀에 위치한 사용자 b(i)의 신호대 간섭 잡음비를 나타내고, 상기

는 클러스터에 속한 기지국들이 단일 셀 통신하는 경우에 기지국 b 셀에 위치한 사용자 b(i)의 신호대 간섭 잡음비를 나타낸다. 또한, 상기 F_{b ,i}는 단일 셀 통신 시 사용하는 선형 수신필터 F_b에서 i번째 사용자에게 해당하는 i번째 행 벡터를 나타내며, 상기 W_{G ,b(i)}는 협력 통신 시 사용하는 선형 수신필터 W_G에서 기지국 b에 속하는 i번째 사용자에 해당하는 행 벡터를 나타낸다. 또한, 상기 h_β(k),G는 기지국 β에 속하는 k번째 사용자와 클러스터 G에 속하는 모든 기지국 사이의 채널을 나타내는 BN_b×1 벡터이며, h_β(k),b는 기지국 β에 속하는 k번째 사용자와 기지국 b사이의 채널을 나타내는 N_b×1 벡터이다.In the above Equations 7 and 8,

Indicates a signal-to-interference noise ratio of a user b (i) located in a cell of the base station b when the base stations belonging to the cluster cooperatively communicate with each other,

Denotes the signal-to-noise interference ratio of the user b (i) located in the cell b of the base station when the base stations belonging to the cluster perform single cell communication. In addition, F _{b , i} denotes an i-th row vector corresponding to an i-th user in a linear reception filter F _b used for single cell communication, and W _{G , b (i)} denotes a linear reception filter W _G denotes a row vector corresponding to the i-th user belonging to the base station b. In addition, the h _{β (k), G} is a BN _b × 1 vector representing the k-th user and the channels between all base stations that belong to the cluster G belonging to the base β, h _{β (k), b} is a k belonging to the base station β Is an N _b x 1 vector representing the channel between the ith user and the base station b.

That is, since the Max-CD clustering scheme is required to form a cluster for all base stations, all the candidate clusters capable of being combined with all of the base stations are generated, and then the candidate clusters generated based on Equation (8) The candidate cluster having the largest capacity difference is selected, and the selected candidate cluster is determined as the final cluster. The Max-CD clustering scheme regenerates all possible candidate clusters with respect to the remaining base stations not included in the final cluster, and performs cooperation among the generated candidate clusters based on Equation (8) The candidate cluster having the largest capacity difference is selected, and the selected candidate cluster is determined as the final cluster, so that all the base stations are clustered. Here, such a clustering method can be called a serial-parallel hybrid search.

The Max-CD scheme is designed in consideration of the fact that the capacity of a cell in a cell having a good channel environment is large when a cooperative communication is performed, but the capacity gain is small. In the Max-CD scheme, So that the communication performance of users with poor channel environment can be improved.

Next, the IW (Interference Weight) method will be described.

As shown in Equation (7), the IW scheme is a scheme in which the capacity between two base stations performing cooperative communication and the capacity between two base stations having a proportional relationship with a capacity difference when the two base stations do not perform cooperative communication Considering the interference, the base stations having large interference are selected to constitute a cluster. That is, although the IW scheme has an advantage that the Max-CD scheme can improve the performance compared to the clustering schemes conventionally provided, there is a problem that complexity increases due to consideration of all combinations of base stations. It is designed to maintain high performance while reducing the complexity of the method. Therefore, the IW scheme also aims to maximize the difference in capacity between a case where a plurality of base stations perform cooperative communication and a case where cooperative communication is not performed.

The IW scheme considers two base stations without considering the combination of all base stations. That is, the IW scheme forms a cluster based on the interference between the two base stations having a proportional relationship with the capacity difference, taking into consideration the capacity difference in case of cooperative communication between two base stations and in case of no cooperative communication.

Equation (9) represents the difference between the capacity when two base stations cooperate with each other and the capacity when there is no cooperative communication.

Equation (9) is a simplification on the assumption that the SINR of the user is larger than 1 in the actual environment, and thus the SINR of the user has a value larger than 1. Here, b1 and b2 represent two base stations, respectively.

The numerator part shown in Equation (9) can be expressed by dividing the cell belonging to the cluster and the cell belonging to the cluster into Equation (10).

는 기지국 b1과 기지국 b2가 협력 통신을 하지 않는 경우에 상기 기지국 b1과 기지국 b2가 아닌 다른 기지국에 속한 사용자들로부터의 간섭을 나타내며, 상기

는 상기 기지국 b1과 기지국 b2가 협력 통신을 하는 경우에 상기 기지국 b1과 기지국 b2가 아닌 다른 기지국에 속한 사용자들로부터의 간섭을 나타낸다. 여기서, 도 8을 참조하면, 상기 두 가지의 간섭은 서로 비례하여 증가하는 알 수 있다. 여기서, 상기 도 8은 상기 두 기지국이 협력통신을 수행하는 경우에 다른 기지국들로부터의 간섭(Other Cell Interference with cooperation)과 상기 두 기지국이 협력통신을 수행하지 않는 경우에 다른 기지국들로부터의 간섭(Other Cell Interference without cooperation)을 나타낸다.Here,

Indicates interference from users belonging to base stations other than the base station b1 and the base station b2 when the base station b1 and the base station b2 do not cooperate with each other,

Indicates interference from users belonging to base stations other than the base station b1 and the base station b2 when the base station b1 and the base station b2 perform cooperative communication. Referring to FIG. 8, it can be seen that the two types of interference increase in proportion to each other. FIG. 8 is a diagram illustrating an example of a case where the two base stations perform cooperative communication with each other, and in the case where the two base stations do not perform cooperative communication, interference from other base stations Other Cell Interference without cooperation.

은 상기 두 기지국이 협력 통신을 수행한 경우의 캐패시티와 상기 두 기지국이 협력 통신을 수행하지 않는 경우의 캐패시티 간의 차이를 나타내는 캐패시티 이득에 따라 증가하는 것을 알 수 있다. 즉, 두 기지국 간의 간섭이 커질수록 두 기지국에 대한 캐패시티 이득이 커지는 특징이 있다. Here, referring to FIG. 9, when the two base stations do not perform cooperative communication, when the interference between the base station b1 and the base station b2

Is increased according to the capacity gain indicating the difference between the capacity when the two base stations perform cooperative communication and the capacity when the two base stations do not perform cooperative communication. That is, as the interference between the two base stations increases, the capacity gain for the two base stations increases.

Accordingly, in the present invention, a cluster is configured by considering the interference between two base stations using this feature. To this end, in the present invention, the interference weights are set according to the following Equation (11) in consideration of the interference between the two base stations and the noise passed through the reception filter, and the base stations having the large interference weights are clustered.

Here, the F _{b, i} are in the linear receive filter F _b used by the base station b denotes the i-th row vector corresponding to the i th user, W _b1 is i belonging to the base station b1 in the linear receive filter W _bi used when cooperative communication Represents the row vector corresponding to the ith user. Also, h _{b2 (k), b1} represents a channel between the k-th user and the base station b2 belonging to the base station b1.

That is, the IW scheme clusters each base station using an interference weight that increases according to the capacity gains of the two base stations, as shown in Equation (11). More specifically, the IW scheme selects a base station as a reference among all the base stations, and then configures a cluster by selecting a predetermined number of base stations whose interference weights with respect to the reference base station are the largest calculated. Thereafter, the IW scheme selects a base station again from the remaining base stations not included in the cluster, and selects a predetermined number of base stations whose interference weights with the selected reference base station are the largest calculated, To cluster all base stations. Here, such a clustering method can be referred to as a direct search. Herein, the base station selection method as a reference of the cluster may use a method of selecting an arbitrary base station. In a method of preferentially selecting a base station having a poor channel environment, that is, a sum of channel eigenvalues between a base station and a user is compared A method of selecting a base station having the smallest value may be used. The method of selecting an arbitrary base station can guarantee fairness among the base stations so that all users can use it fairly. In a method of preferentially selecting a base station having a poor environment, It is advantageous to improve the performance of cell boundary users.

As described above, in the present invention, the central processor performs clustering using either the Max-CD scheme or the IW scheme.

Hereinafter, the system configuration for performing clustering in the central processor and the configuration and operation of the central processor will be described in detail.

2 shows a simplified procedure for establishing a cluster and cooperating communication in a cellular communication system according to the present invention. Here, although two base stations and two terminals belonging to each base station are illustrated for convenience of explanation, it is natural that the cellular communication system includes two or more base stations and a plurality of terminals are included in the base station. In addition, the procedure of FIG. 2 to be described below will be repeatedly performed at predetermined intervals in order to consider the channel environment of the user which changes with time.

Referring to FIG. 2, in step 210, the BS 1 202 and the BS 206 transmit a reference signal, which is received from MSs existing in a cell region and a neighboring cell region, And proceeds to step 212 to transmit the channel state information including the estimated channel information to the central processor 204. [

In step 214, the central processor 204 determines whether a predetermined number of base stations are cooperating with each other based on the channel state information received from the first base station 202, the second base station 206 and a plurality of other base stations (not shown) (Not shown) in consideration of a difference in capacity when performing single cell communication without performing cooperative communication with the base station 1 202, the base station 2 206, and other multiple base stations . Here, the central processor can perform clustering on the BSs using a Max-CD scheme or an IW scheme.

In step 216, the central processor 204 transmits the clustering information and the reception filter information to the first base station 202, the second base station 206, and a plurality of other base stations (not shown). Here, the clustering information may include information of other base stations belonging to the same cluster as the cluster to which each base station belongs.

When the first UE 200 belonging to the first BS 202 and the second BS 208 belonging to the second BS 206 transmit the uplink signals in step 218, 202 and 206 detect signals of the first terminal 200 and the second terminal 208 in cooperation with other base stations belonging to the cluster.

3 shows a block configuration of a base station and a central processor in a cellular communication system according to the present invention.

3, each of the plurality of base stations 300-1 to 300-N includes a channel estimation unit 302, a cooperative signal detection unit 304, and a transmission / reception unit 306, The central processor 350 includes a transmitting / receiving unit 352, a channel information collecting unit 354, and a base station clustering unit 356.

First, the configuration of the plurality of base stations 300-1 to 300-N will be described. The channel estimator 302 estimates a channel quality of a corresponding terminal by using a reference signal of the terminals received through the transceiver 306 And provides the channel state information indicating the channel estimation result to the transceiving unit 306 and the cooperative-based signal detecting unit 304. [

The cooperative base signal detector 304 receives the clustering information and the reception filter information from the transceiver 306 and performs a function for detecting a signal of the terminal in cooperation with other base stations belonging to the clustering. Here, the cooperative-based signal detector 304 may detect a signal of the UE using a linear receiver based on the channel state information.

The transceiver 306 transmits and receives signals to and from the terminal, the neighboring base station, and the central processor 350. In particular, a reference signal received from terminals belonging to a cell region of the base station or terminals belonging to an adjacent cell region is received and provided to the channel estimation unit 302. The channel estimation unit 302 receives a reference signal received from the channel estimation unit 302, Information to the central processor 350. In addition, the transceiver 306 provides clustering information and reception filter information received from the central processor 350 and a signal received from a terminal belonging to its cell area to the cooperative base signal detector 304.

Next, referring to the configuration of the central processor, the transceiver unit 352 performs a function of transmitting and receiving signals with the base station. In particular, the transceiver 352 receives channel state information from a plurality of base stations, provides the channel state information to the channel information collector 354, receives base station clustering information from the base station clustering unit 356, Lt; / RTI >

The channel information collection unit 354 collects channel state information from each base station through the transmission / reception unit 352, identifies the base stations to be clustered, requests the base station clustering unit 356 to confirm that the base stations are clustered do.

The base station clustering unit 356 receives channel state information on base stations to be clustered through the channel information collecting unit 354 and performs a cooperative communication with a predetermined number of base stations based on the received channel state information And performs the clustering in consideration of the difference in the capacities when single cell communication is performed without cooperative communication. In particular, the base station clustering unit 356 may perform clustering on the BSs using a Max-CD scheme or an IW scheme. The base station clustering unit 356 may be configured as shown in FIG. 4 when the Max-CD scheme is used, and may be configured as shown in FIG. 5 when the IW scheme is used. Therefore, the operation of the base station clustering unit 356 in the case of using the Max-CD scheme and the case of using the IW scheme will be described in detail with reference to FIG. 4 and FIG. When clustering is completed for all base stations, the base station clustering unit 356 transmits clustering information and reception filter information used for cooperative communication to each of the base stations, and each base station performs cooperative communication with other base stations .

FIG. 4 shows a detailed block configuration of a base station clustering unit in a central processor according to an embodiment of the present invention. FIG. 4 shows a block configuration when the base station clustering unit 356 uses the Max-CD scheme.

4, the base station clustering unit 356 using the Max-CD scheme includes a clustering combination generating unit 400, a capacity gain calculating unit 410, and a cluster determining unit 420.

The clustering combination generator 400 generates candidate clusters based on all the base stations to be clusters by selecting the base stations corresponding to the number of the clusters to be included in one cluster. At this time, the candidate clusters include all clusters configurable through a combination. For example, if the total number of base stations is 7 and the number of base stations to be included in one cluster is 3, the number of candidate clusters will be ₇ C ₃ . The clustering combination generator 400 provides the generated candidate clusters to the capacity gain calculator 410.

The capacity gain calculation unit 410 calculates a capacity gain for each of the candidate clusters input from the clustering combination generation unit 400. Here, as shown in Equations (6) and (8), the capacity gain for the candidate cluster G is determined by the capacity when the base stations belonging to the candidate cluster G perform cooperative communication and the base stations belonging to the candidate cluster G And the difference in capacity when single cell communication is performed. The capacity gain calculator 410 provides the calculated capacity gains to the cluster determiner 420 for each of the candidate clusters.

When the capacity gain for each candidate cluster is provided from the capacity gain calculation unit 410, the cluster determination unit 420 selects a candidate cluster having the largest capacity gain among the candidate clusters, . Thereafter, the cluster determination unit 420 transmits a signal indicating that the remaining base stations not included in the determined cluster are to be re-clustered to the clustering combination generation unit 400. That is, the cluster determination unit 420 provides information on the non-clustered base stations to the clustering combination generator 400 until the clustering for all the base stations is performed, and the capacity gain calculator 410 And the final cluster is determined by comparing the capacity gains of the candidate clusters output from the candidate clusters.

FIG. 5 shows a detailed block configuration of a base station clustering unit in a central processor according to another embodiment of the present invention. 5 illustrates a block configuration when the base station clustering unit 356 uses the IW scheme.

Referring to FIG. 5, the base station clustering unit 356 includes a base station determining unit 500, an interference weight calculating unit 510, and a cluster determining unit 520.

The base station determination unit 500 selects one base station to be a cluster reference among all base stations to be clustered. Here, the BS determining unit 500 may select an arbitrary BS among the BSs to be clusters as a reference BS, and may select a BS that is the worst channel environment among the BSs to be clustering. The BS determining unit 500 provides information on the selected one reference BS and information on other BSs to the interference weight calculating unit 510.

The interference weight calculation unit 510 calculates interference weights for the base station and the other base stations determined by the base station decision unit 500. [ That is, the interference weight calculator 510 calculates the interference weights between the reference base station and the selected other base station as shown in Equation (11) after selecting one of the other base stations sequentially or arbitrarily, It repeatedly performs for the base stations. Then, the interference weight calculation unit 510 provides the calculated interference weights to the cluster determination unit 520 for each of the base stations and the other base stations.

The cluster determining unit 520 compares the interference weights inputted from the interference weight calculating unit 510, selects a predetermined number of base stations having a large interference weight, selects one of the selected base stations As shown in FIG. Here, when the number of base stations to be included in the cluster is B, the cluster determining unit 520 can select B-1 base stations in descending order of the interference weight. After the clusters are formed, the cluster determining unit 520 transmits a signal indicating that clustering for the base stations not included in the cluster is to be performed again when the base stations that are not included in the cluster exist, ). That is, the cluster determination unit 520 provides information on the non-clustered base stations to the base station determination unit 500 until the clustering for all the base stations is performed, and the interference weight calculation unit 510 And the operation of configuring the cluster by comparing the output interference weights is repeatedly performed.

FIG. 6 illustrates a clustering procedure of a central processor according to an embodiment of the present invention. FIG. 6 illustrates an operation procedure when the central processor performs clustering using the Max-CD scheme. The process of FIG. 6, which will be described below, may be repeated at predetermined intervals according to the design method of the designer, or may be performed at a preset time.

Referring to FIG. 6, the central processor receives channel state information from a plurality of base stations in step 601, and proceeds to step 603 where base station information on the received base stations, that is, , And all possible candidate clusters are generated. For example, if the total number of base stations is 7 and the number of base stations to be included in one cluster is 3, the central processor generates ₇ C ₃ candidate clusters.

Then, in step 605, the central processing unit performs a capacity gain for each of the candidate clusters, that is, the capacity when the base stations included in the candidate cluster perform cooperative communication and the base stations included in the candidate cluster perform cooperative communication And calculates the difference between the capacities when performing single cell communication without performing the single cell communication. Here, the capacity gains for the candidate clusters can be calculated as shown in Equations (6) and (8).

In step 607, the central processor selects a candidate cluster having the largest capacity gain among the candidate clusters and determines the final cluster as a final cluster. In step 609, the central processor determines whether clustering has been performed for all the base stations. That is, the central processor determines whether there is no more candidate cluster that is not determined as the final cluster.

If the clustering is not performed for all of the base stations, the central processor proceeds to step 613 to regenerate all candidate clusters that can be combined with the unclustered base stations, and returns to step 605, I do.

On the other hand, if clustering is performed on all the base stations, the central processor proceeds to step 611, and transmits information on the determined cluster to each base station, and terminates the algorithm according to the present invention.

FIG. 7 illustrates a clustering procedure of a central processor according to another embodiment of the present invention. FIG. 7 shows a procedure for the case where the central processor performs clustering using the IW scheme. The process of FIG. 7 described below may be repeatedly performed at predetermined intervals according to the design method of the designer, or may be performed at a preset time.

Referring to FIG. 7, the central processor receives channel state information from a plurality of base stations in step 701, and proceeds to step 703 where base stations, that is, all of the base stations to be clustered, And selects one base station to be a reference. Here, the central processor may select a base station as a reference base station among the base stations to be clustered, and may select a base station that is the worst channel environment among the base stations as the clustering target. Here, for convenience of explanation, the base stations other than the reference base station among the base stations to be clusters are referred to as candidate base stations.

In step 705, the central processor calculates interference weights between the reference BS and the candidate BSs. That is, the central processor sequentially or randomly selects one candidate base station among the candidate base stations, calculates the interference weights between the reference base station and the selected candidate base station as shown in Equation (11) . ≪ / RTI >

In step 707, the central processor compares the interference weights calculated in step 705, selects a predetermined number of candidate base stations having a large interference weight, and proceeds to step 709. In step 709, It is determined that the candidate base stations are configured as one cluster. Here, if the number of base stations to be included in the cluster is B, the central processor can select B-1 candidate base stations in descending order of the interference weight.

In step 711, the central processor determines whether clustering has been performed on all base stations. That is, the central processor checks whether or not a candidate cluster that is not determined to be composed of the cluster is no longer present.

If clustering is not performed for all the base stations, the central processor proceeds to step 415 to determine a reference base station among the non-clustered base stations, and then returns to step 705 to perform the following steps again.

On the other hand, if clustering is performed on all the base stations, the central processor proceeds to step 713 and transmits information on the determined cluster to each base station, and ends the algorithm according to the present invention.

As described above, in the present invention, base stations are clustered using a Max-CD scheme and an IW scheme in a central processor.

Hereinafter, the performance of the clustering using the Max-CD scheme and the IW scheme according to the present invention will be described.

FIGS. 10A and 10B show data rates for the case of clustering according to the conventional technique and the case of clustering according to the present invention. 10A shows a data rate for the case where the total number of base stations is 7 and the three base stations are grouped and clustered. FIG. 10B shows a case where the total number of base stations is 19 and three base stations are grouped and clustered And the data rate for the data. In FIGS. 10A and 10B, IW-random denotes a case of randomly selecting a reference base station in the IW scheme, and IW-weak denotes a case in which the reference base station is selected as a base station having a poor channel environment in the IW scheme.

Referring to FIGS. 10A and 10B, it can be seen that the Max-CD method proposed by the present invention has higher performance than the conventional greedy bath search method. Of course, the performance of the entire area search scheme is similar to that of the Max-CD scheme of the present invention, but the entire area search scheme has a much higher complexity than the Max-CD.

FIGS. 11A and 11B show data rates of edge users for clustering according to the prior art and clustering according to the present invention.

FIGS. 11A and 11B show data rates of cell boundary users when clustering according to the related art and clustering according to the present invention. 11A shows a data rate of a cell boundary user for the case where the total number of base stations is seven and the three base stations are clustered together. In FIG. 11B, the total number of base stations is 19, and three base stations are grouped And the data rate of the cell boundary user for the case of clustering. In FIGS. 11A and 11B, IW-random denotes a case of randomly selecting a reference base station in the IW scheme, and IW-weak denotes a case in which the reference base station is selected as a base station having a poor channel environment in the IW scheme.

11A and 11B, it can be seen that the Max-CD, IW-random, and IW-weak methods proposed by the present invention have higher performance than the conventional greedy search method. That is, the conventional greedy search method aims to maximize the sum of capacities, so that the base stations of the users with good channel environment are clustered preferentially so that the performance of the users located in the cell boundary and having poor channel environment . On the other hand, with the aim of maximizing the capacity gain, the schemes of the present invention can improve the performance of users at the cell boundary by clustering in consideration of users with poor channel environments located at cell boundaries.

FIG. 12 shows the complexity and performance of the case of clustering according to the prior art and the case of clustering according to the present invention.

Referring to FIG. 12, the Max-CD scheme proposed in the present invention has higher complexity than the conventional greedy search scheme, but achieves high performance. The IW scheme proposed by the present invention is superior to the greedy search scheme It has a very low complexity and has the advantage of achieving high performance. In particular, the IW-weak scheme achieves very high performance in terms of data rate for cell edge users.

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. 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.

356: base station clustering unit 400: clustering combination generating unit
410: Capacity gain calculation unit 420: Cluster determination unit
500: base station determination unit 510: interference weight calculation unit
520: Cluster determination unit

Claims (18)

A method for clustered base stations for cooperative communications in a central processor of a cellular communication system,
Collecting channel state information for a plurality of base stations,
Comprising the steps of: configuring a cluster in consideration of a difference between a capacity of a predetermined number of base stations performing cooperative communication and a capacity of a predetermined number of base stations communicating in a single cell;
Wherein the difference between the capacities is a difference of the largest capacity sum or a difference of the capacity sum proportional to the interference.
The method according to claim 1,
The process of configuring the cluster in consideration of the difference between the capacities,
Generating all combinable candidate clusters for the plurality of base stations;
Selecting a candidate cluster having a largest difference between a capacity sum of users corresponding to a case where the base stations belonging to the candidate cluster cooperate with each other and a capacity sum corresponding to a case where the base stations belonging to the candidate cluster belong to a single cell communication;
And constructing the selected candidate cluster as one cluster.
3. The method of claim 2,
All the candidate clusters capable of being combined with the remaining base stations not included in the cluster are regenerated until the clustering for the plurality of base stations is completed so that the candidate clusters having the largest difference between the two capacities are constituted as one cluster And repeating the process.
The method according to claim 1,
The process of configuring the cluster in consideration of the difference between the capacities,
And constructing a cluster by considering interference between two base stations,
Wherein interference between the two base stations is proportional to a difference between a capacity sum of corresponding users when the two base stations cooperate and a corresponding capacity sum when the two base stations perform single cell communication.
5. The method of claim 4,
The process of configuring a cluster in consideration of interference between the two base stations,
Selecting a reference base station,
Calculating interference weights indicating interference with the reference base station for the plurality of base stations;
Selecting a predetermined number of base stations based on the calculated interference weight value;
And configuring the reference base station and the selected predetermined number of base stations as one cluster.
6. The method of claim 5,
Wherein the reference base station arbitrarily selects among the plurality of base stations.
6. The method of claim 5,
Wherein the reference base station preferentially selects a base station having a smallest sum of channel eigenvalues between the base station and the user among the plurality of base stations.
5. The method of claim 4,
Selecting one base station among the remaining base stations not included in the cluster until the clustering for the plurality of base stations is completed, selecting a predetermined number of base stations based on the interference weight with the selected base station, And a cluster of a plurality of clusters are repeatedly performed.
6. The method of claim 5,
Wherein the interference weight is calculated in consideration of at least one of interference and reception noise passed through one of the plurality of base stations and the reference base station.
An apparatus for clustering base stations for cooperative communication in a central processor of a cellular communication system,
A channel information collecting unit for collecting channel state information for a plurality of base stations,
And a base station clustering unit configured to form a cluster in consideration of a capacity of a predetermined number of base stations performing cooperative communication and a difference between capacities of the predetermined number of base stations in case of single cell communication,
Wherein the difference between the capacities is the difference of the largest capacity sum or the difference of the capacity sum proportional to the interference.
11. The method of claim 10,
Wherein the base station clustering unit generates all combinable candidate clusters for the plurality of base stations and then calculates a capacity sum of users corresponding to the cooperative communication between the base stations belonging to the candidate cluster, Selects a candidate cluster having the largest difference in capacity sum corresponding to the communication, and determines to configure the selected candidate cluster as one cluster.
12. The method of claim 11,
The base station clustering unit recreates all possible candidate clusters that can be combined with the remaining base stations not included in the cluster until the clustering for the plurality of base stations is completed to obtain a candidate cluster having the largest difference between the two capacities And a step of performing a process of configuring a plurality of clusters of clusters.
11. The method of claim 10,
The base station clustering unit configures a cluster in consideration of interference between two base stations,
Wherein the interference between the two base stations is proportional to a difference between a capacity sum of corresponding users when the two base stations cooperate and a corresponding capacity sum when the two base stations perform single cell communication.
14. The method of claim 13,
The base station clustering unit calculates an interference weight indicating interference with the reference base station to the plurality of base stations after selecting the reference base station, selects a predetermined number of base stations based on the calculated interference weight value, Wherein the base station and the selected number of base stations are configured as one cluster.
14. The method of claim 13,
Wherein the base station clustering unit selects an arbitrary base station among the plurality of base stations and determines the base station as a reference base station.
14. The method of claim 13,
Wherein the base station clustering unit preferentially selects a base station having the smallest sum of channel eigenvalues between the base station and the user among the plurality of base stations and determines the base station as a reference base station.
14. The method of claim 13,
Wherein the base station clustering unit selects one reference base station among remaining base stations not included in the cluster until the clustering for the plurality of base stations is completed, And selecting one of the base stations and configuring the base stations as one cluster.
15. The method of claim 14,
Wherein the interference weight calculation unit considers at least one of interference and reception noise transmitted through one of the plurality of base stations and the reference base station.

Images