JP2007074304A - Relay station selecting device and relay station selecting method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a relay station selecting device which improves a communication capacity of a communication system for transmitting a radio signal from a source node to a target node via a plurality of relay stations, more than conventional communication capacities. <P>SOLUTION: A relay station selecting device includes; a means which receives signals meaning pass loss coefficients of backward channels between the source node and the relay stations and those of forward channels between the relay stations and the target node, from a plurality of relay stations; a means which generates first and second composite matrixes (H and G) including a plurality of channel matrixes; a means which calculates matrix expressions of matrixes including square matrixes (GG<SP>H</SP>, (GH)(GH)<SP>H</SP>, etc) derived from the first and second composite matrixes and conjugate transposed matrixes of the composite matrixes; a means which compares communication capacities calculated on the basis of matrix expressions for respective combinations including one or more relay station candidates and selects one combination (I<SB>1</SB>, ..., I<SB>opt</SB>); and a means which transmits a signal showing relay stations included in the selected combination and the number of the relay stations, to one or more relay stations. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention generally relates to the technical field of wireless communication, and more particularly to a relay station selection apparatus and relay station selection method for transmitting a radio signal from a source node to a target node through a plurality of relay stations.

  In a communication system including a wireless link such as a cellular system or a wireless local area network (LAN), the communication capacity of the system is drastically reduced if the channel state between the transmitting station and the receiving station is not good. The multi-hop method is one method for dealing with this problem, and this method improves the received signal power-to-noise power ratio (SNR) at the receiving station by providing a relay station between the transmitting station and the receiving station. (See Patent Documents 1 and 2 for conventional multi-hop systems.)

In the multi-hop method, it seems that the SNR at the receiving station can be improved as the number of hops is increased. However, if the number of hops is increased, a transmission delay is generated accordingly. One conventional method for dealing with this is to determine one optimum signal transmission route by selecting an optimum relay station from various relay stations. The optimum relay station is selected based on specific criteria such as the number of hops and the SNR of the communication path. However, since the signal quality at the receiving station depends on one specific path, there is a concern that sufficient signal quality cannot be ensured depending on the communication situation. In this regard, it is also conceivable that a plurality of relay stations are routed in parallel instead of a specific one (see, for example, Patent Document 3 for this type of technology).
JP 2001-292093 A JP-A-2005-65267 JP 2004-56787 A

  However, in the related art including the invention described in Non-Patent Document 3, when selecting a plurality of relay stations, each route is merely optimized based on the SNR, the number of hops, and the like of each route. Therefore, in the conventional method, the gain and communication capacity of the entire system when a plurality of relay stations are used at the same time may not be optimized. This is because the sum of the communication capacities of the individually optimized paths does not necessarily become the optimal communication capacity of the entire system.

  The present invention has been made in order to address at least one of the above-described problems, and its problem is to increase the communication capacity of a communication system that transmits a radio signal from a source node to a target node through a plurality of relay stations. It is another object of the present invention to provide a relay station selection device and a relay station selection method that improve the performance of the relay station.

  In the present invention, a relay station selection device used in a communication system that transmits a radio signal from a source node to a target node through a plurality of relay stations is used. The apparatus includes means for receiving a signal indicating a path loss coefficient of a backward channel between a source node and a relay station and a path loss coefficient of a forward channel between the relay station and a target node from a plurality of relay stations, and a plurality of channel matrices, respectively. Means for generating first and second composite matrices; means for calculating a determinant of a matrix including a square matrix derived from a conjugate transpose of the first and second composite matrices and the composite matrix; Means for comparing communication capacities calculated based on determinants for each combination including relay station candidates and selecting one combination, and a signal indicating the number of relay stations and relay stations included in the selected combination For transmitting to one or more relay stations.

  According to the present invention, it is possible to improve the communication capacity of a communication system that transmits a radio signal from a source node to a target node through a plurality of relay stations.

According to one aspect of the present invention, first and second composite matrices ( H , G ) each including a plurality of channel matrices are generated and derived from the first and second composite matrices and the conjugate transpose of the composite matrix. A determinant of a matrix including a square matrix ( GG ^H , ( GH ) ( GH ) ^H, etc.) to be calculated, and communication calculated based on the determinant for each combination including one or more relay station candidates By comparing the capacities C _{(I1,..., IL)} , one combination of relay nodes (I ₁ ,..., I _opt ) is selected.

  By appropriately grasping the upper limit of the communication capacity of the entire system using the composite matrix, the communication capacity of the entire system can be surely increased as compared with the conventional system.

  The first one of the composite matrix may include a plurality of products of the channel matrix and the path loss coefficient of the rear channel, and the second one of the composite matrix may include a plurality of products of the channel matrix and the path loss coefficient of the front channel.

  The relay station selection device may or may not receive signals indicating channel matrices of the backward and forward channels from the relay station candidates. When receiving, the communication capacity can be more accurately evaluated based on an accurate channel matrix. If not received, the traffic volume and processing load can be reduced accordingly. The channel matrix may be appropriately derived by using prior knowledge about the channel matrix.

  The relay station selection apparatus may further include means for calculating an integrated path loss coefficient expressed as a function of the product of the path loss coefficients of the front and rear channels for each of the relay station candidates. Selection means may further be provided for selecting relay station candidates corresponding to the integrated path loss coefficients belonging to a predetermined numerical range. Storage means for storing a table indicating a correspondence relationship between the relay station, the number of relay stations, and the communication capacity may be further provided. By limiting the relay station candidates included in the combination to the relay station candidates selected by the selection unit, it is possible to greatly improve the calculation efficiency.

  A table indicating the correspondence relationship between the relay station, the number of relay stations, and the communication capacity is referred to, and one relay station is selected by comparing the communication capacity corresponding to a certain number of relay stations, so that more relay stations are supported. Another relay station may be selected by comparing the communication capacities. If the communication capacity corresponding to the selected one relay station is larger than the communication capacity corresponding to the selected another relay station, the larger number of relay stations and the other relay station It may be determined that the signal is not used for transmission. Thereby, it is possible to efficiently and reliably derive the optimum combination of relay stations among given relay candidates.

FIG. 1 shows a communication system according to an embodiment of the present invention. The communication system includes a source node 11, a plurality (K) of relay stations 12-1 to 12 -K, and a target node 13. A source node may be called a source, and a destination node may be called a target node or a destination node. The distinction between the communication nodes is convenient, and all the nodes may have the same function. The communication system forms a wireless ad hoc network, and typically all communication nodes are mobile stations, but the invention is not limited to such aspects. The source node 11 is a communication node that transmits information, and is also called a transmission source. The relay stations 12-1 to 12 -K are communication nodes that relay signals transmitted from the source node and transmit them to the target node. The destination node 13 is a communication node that is a destination of information transmission by the source node 11, and receives a relay signal transmitted from one or more relay stations. In a general ad hoc network, information is transmitted from a source to a target node with one or more hops via one or more relay stations. In this embodiment, the number of hops is limited to 1, and any relay station can directly communicate with the source node and the destination node. Since the number of hops is kept small, transmission delay due to hopping can be suppressed. Since a plurality of relay stations relay signals at the same time, it is possible to cope with signal quality degradation caused by a channel state of a specific path. For convenience of explanation, a propagation path (channel) between the source node 11 and the relay stations 12-1 to 12 -K is called a “backward channel” or “backward channel” (BC). The transfer characteristic of the backward channel between the source node 11 and a specific relay station 12-i (1 ≦ i ≦ K) is expressed by a channel matrix H _i . Similarly, the channel between the relay stations 12-1 to 12 -K and the target node is called “forward channel” or “forward channel” (FC). Transfer characteristics of the forward channel during a particular relay station 12-i (1 ≦ i ≦ K) and a destination node are represented by the channel matrix G _i.

  FIG. 2 shows a functional block diagram of the relay station selection device. The relay station selection device selects all or a part of the K relay stations, and the selected relay station actually relays the signal transmitted from the source node to the target node. An unselected relay station may be used to relay another signal. The relay station selection device 20 is typically provided in the source node as in the present embodiment, but may be provided in another device. The relay station selection device 20 includes a pattern generation unit 22, a communication capacity calculation unit 23, and a pattern selection unit 26.

The pattern generation unit 22 designates possible relay station combinations based on the path loss and channel matrix received from each relay station (12-1 to K in FIG. 1), and details of the combination (simultaneous transmission of relay signals in parallel). The number of relay stations to be transmitted (L), relay station identification information (Ii), etc.) are output. The output combinations (I ₁ , I ₂ ,..., I _L ) are not necessarily optimal and are only possible choices among the assumed relay stations. The path loss is an average propagation loss of the channel, and is mainly determined by distance fluctuation, shadowing, etc., and does not depend on the influence of instantaneous fading. For example, the path loss may be calculated based on the difference between the transmission power and the reception power of the pilot channel over a certain period. As described above, there are two types of propagation paths associated with each relay station: a backward channel and a forward channel. Therefore, the relay station selection apparatus receives, from each relay station, a path loss coefficient λ ^FC _k and a channel matrix G _k for the forward channel in addition to the path loss coefficient λ ^BC _k and the channel matrix H _k for the backward channel (k = 1 to K). .

The communication capacity calculation unit 23 includes a composite matrix generation unit 24 and a capacity calculation unit 25. Based on the combination of relay stations (I ₁ ,..., I _L ) and the path loss notified from the pattern generation unit 22, the composite matrix generation unit 24 uses two composite matrices H expressed by the following equation (1). , G is created.

即ち、組み合わせに含まれる全ての中継局の各々（Ｉ_ｉ）について、後方チャネルに関するパスロス係数λ^ＢＣ _Ｉｉとチャネル行列Ｈ_Ｉｉとの積を算出し、Ｌ個の積を列方向に並べることで後方チャネルの複合行列Ｈが導出される。同様に、組み合わせに含まれる全ての中継局の各々（Ｉ_ｉ）について、前方チャネルに関するパスロス係数λ^ＦＣ _Ｉｉとチャネル行列Ｇ_Ｉｉとの積を算出し、Ｌ個の積を列方向に並べることで前方チャネルの複合行列Ｇが導出される。説明の便宜上、複合行列中のチャネル行列は列方向に並べられているが、このことは本発明に必須ではない。

That is, for each of the relay stations (I _i ) included in the combination, the product of the path loss coefficient λ ^BC _Ii for the rear channel and the channel matrix H _Ii is calculated, and the L products are rearranged in the column direction. A composite matrix H for the channel is derived. Similarly, for each relay station (I _i ) included in the combination, the product of the path loss coefficient λ ^FC _Ii for the forward channel and the channel matrix G _Ii is calculated, and L products are arranged in the column direction. A composite matrix G for the forward channel is derived. For convenience of explanation, the channel matrices in the composite matrix are arranged in the column direction, but this is not essential to the present invention.

The capacity calculator 25 estimates the communication capacity C _{(I1,..., IL)} of the entire system according to the following equation (2) for the specified combination of relay stations (I ₁ ,..., I _L ).

ここで、上付き文字の「Ｈ」は共役転置をとる操作を表し、σ^２は中継局や目的ノードでのノイズ電力を表し、Ｉは単位行列を表し、Ｅ（・）はカッコ内の変数のアンサンブル平均値を表し、ｄｅｔ（・）はカッコ内の行列の行列式を表す。ノイズ電力は厳密には装置毎に異なるが、本実施例では簡明化のため全ての通信ノードでノイズ電力は等しいことが仮定されている。（２）式は、（３）式で表現される量について、（１）式で表現される複合行列を適用することで導出することができる。

Here, the superscript “H” represents a conjugate transposition operation, σ ² represents noise power at a relay station or a target node, I represents a unit matrix, and E (•) represents a variable in parentheses. Represents the ensemble average value, and det (·) represents the determinant of the matrix in parentheses. Strictly speaking, the noise power differs from device to device, but in this embodiment, it is assumed that the noise power is equal in all communication nodes for the sake of simplicity. The expression (2) can be derived by applying the composite matrix expressed by the expression (1) to the quantity expressed by the expression (3).

ここで、Ａｓ＝Ｅ（ｓｓ^Ｈ）であり、Ａｒ＝Ｅ（ｒｒ^Ｈ）であり、Ａｕ＝Ｅ（ｕｕ^Ｈ）であり。ｓはソースノードからの送信信号を表わすベクトルであり、ｒは目的ノードでの受信信号を表すベクトルである。ｕはｓとｒを合成したベクトルの複素共役を表し、ｕ＝（ｓ，ｒ）^＊で定義される量である。例えば、ｓ，ｒが共に１行４列のベクトルであったとすると、ｕは１行８列のベクトルになる。

Here, As = E (ss ^H ), Ar = E (rr ^H ), and Au = E (uu ^H ). s is a vector representing a transmission signal from the source node, and r is a vector representing a reception signal at the target node. u represents a complex conjugate of a vector obtained by combining s and r, and is an amount defined by u = (s, r) ^* . For example, if both s and r are vectors of 1 row and 4 columns, u is a vector of 1 row and 8 columns.

The pattern selection unit 26 calculates the communication capacity {C _{(I1,..., IL)} } (I ₁ , I ₂ ,..., I _L ∈ (1, 2,..., K)) calculated for each combination of relay stations. The combination of relay stations giving the largest capacity (I ₁ , I ₂ ,..., I _opt ) is determined and the determined content is output (see equation (4)). The output contents include the optimum number of relay stations and the identification information of the optimum relay station.

図３は図1に示されるような通信システムで行われる通信方法を示すフローチャートである。ステップＳ３１ではソースノードが目的ノードに通信要求信号及びパイロット信号を直接的に送信する。即ち目的ノードは中継局からではなくソースノードから送信された信号を直接的に受信する。

FIG. 3 is a flowchart showing a communication method performed in the communication system as shown in FIG. In step S31, the source node directly transmits a communication request signal and a pilot signal to the target node. That is, the target node directly receives a signal transmitted from the source node, not from the relay station.

  In step S32, the target node returns the communication request response signal and the pilot signal directly to the source node.

  In step S33, the reception quality of the received pilot signal is measured. For example, the reception quality may be measured by a signal power to noise power ratio (SNR) as in the present embodiment, or may be measured by an amount capable of evaluating SIR or other channel conditions. If the SNR is better than a predetermined threshold, the channel condition between the source node and the destination node is good. In this case, the target node directly receives the signal transmitted from the source node. On the other hand, if the SNR is not better than a predetermined threshold, the channel state is not good enough to allow direct communication. In this case, communication is performed via the relay station. In this case, the flow proceeds to step S34.

  In step S34, the source node broadcasts a relay request signal and a pilot signal to one or more relay stations (this broadcast may be called broadcast).

  In step S35, each relay station receives the relay request signal and the pilot signal from the source node, and performs channel estimation, path loss coefficient measurement, and the like for the backward channel.

  In step S36, the source node requests the target node to transmit a pilot signal, and in response thereto, the target node broadcasts the pilot signal to each relay station.

  In step S37, each relay station receives a pilot signal from the target node, and performs channel estimation, measurement of a path loss coefficient, and the like for the forward channel.

In step S38, each relay station notifies the source node of path loss coefficients and the like of the backward and forward channels. In this embodiment, each relay station also notifies the source node of information about channel matrices H _i and G _i of the rear channel and the front channel. For convenience of explanation, it has been described that the processes of steps S34, S35, S36, and S37 are performed in order, but this is not essential to the present invention. After the information about the front channel is obtained, the information about the front channel may be obtained, or the information about all of them or the arrangement part may be obtained at the same time.

In step S39, appropriate relay station combinations (I ₁ , I ₂ ,..., I _opt ) are determined by the relay station selection device (FIG. 2) provided in the source node. First, a combination of relay stations (I ₁ , I ₂ ,..., I _L ) including the number of L relay stations is generated. L is a parameter that takes a value in the range of 1 to K. Based on the path loss coefficient and the channel matrix reported from the relay station, a composite matrix H for the rear channel and a composite matrix G for the forward channel as shown in Equation (1) are calculated. The communication capacity is calculated according to equation (2) for each combination of relay stations. For example, assume that relay stations 12-1, 12-2, and 12-3 are notified to the source node as relay station candidates. In this case, the possible combinations are (12-1), (12-2), (12-3) when the number of relay stations is 1, and (12-) when the number of relay stations is 2. 1, 12-2), (12-2, 12-3), (12-3, 12-1), and if the number of relay stations is 3, (12-1, 12-2, 12 -3). The communication capacity of equation (2) is calculated for all these possible combinations. In general, when K relay stations are relay candidates, there are (2 ^K −1) combinations. Among the calculated communication capacities, the combination that gives the maximum capacity is determined, and it is determined as the optimum combination (I ₁ ,..., I _opt ).

  Depending on the value of the maximum communication capacity, it may be determined that it is better to perform direct communication. In that case, the target node directly receives the signal from the source node.

In step S40, information (I _i ) specifying the relay stations included in the selected optimal combination and the number of relay stations are notified to each relay station. Thereafter, as shown in step S41, a signal is transmitted from the source node, received by one or more selected relay stations, a relay signal is transmitted from each relay station, and is received by the target node.

  Equation (2) appropriately represents the capacity of the entire system, which is not simply the sum of the individual communication capacities for individual relay stations. Accordingly, the communication capacity of the entire system can be more appropriately evaluated, and the communication capacity can be improved.

  In the first embodiment, the channel matrix is calculated at each relay station and notified to the source node. However, in the second embodiment of the present invention, the relay station does not notify the channel matrix. In this embodiment, the channel matrices of the rear channel and the front channel are estimated by the source node. For example, the relay node can estimate the forward and backward channels by sending a pilot signal from the source node and the destination node to the relay node. The estimated channel matrix is substituted into equation (2), the average value for all samples is calculated, and the optimum communication capacity is determined.

  Further, the channel matrix of the rear channel and the front channel may be expressed by a covariance matrix. This is desirable from the viewpoint of appropriately estimating the channel matrix based on the statistical properties based on the average correlation value.

As described above, when the number of candidate relay stations is K, the total number of possible combinations is 2 ^K −1, which may be a very large number. Therefore, when the number of candidates is large, it may be difficult to derive an appropriate combination of relay stations. The third embodiment of the present invention can address such concerns.

  FIG. 4 shows a relay station selection apparatus according to an embodiment of the present invention. The relay station selection apparatus 40 includes a relay station classification unit 41, a cluster selection unit 42, a lookup table 44, and a relay station selection unit 45.

The relay station classification unit 41 classifies candidate relay stations into one or more clusters based on the path loss coefficient notified from each relay station. Specifically, an integrated path loss coefficient (integrated path loss coefficient) λ _k as shown in the following equation (5) is calculated.

ここで、Ｐは送信電力を表し、ｋは中継ノードを指定するパラメータを表し（ｋ＝１，．．．，Ｋ）、σ^２はノイズ電力を表す。（５）式から明らかなように統合パスロス係数は、前方及び後方チャネル双方のパスロス係数の積の関数であるので双方のチャネル状態に依存する。中継局の各々について統合パスロス係数λ_ｋが算出される。中継局分類部４１は、算出された統合パスロス係数λ_ｋを１以上のクラスターに分類する。例えば統合パスロス係数λ_ｋが閾値ν_１より小さい場合にはそれは第1クラスターに分類され、閾値ν_１以上ν_２未満ならばそれは第２クラスターに分類され、閾値ν_２以上ν_３未満ならばそれは第３クラスターに分類され、以下同様にして適切なクラスターに統合パスロス係数が分類される。図５はこのようにして分類された統合パスロス係数の一例を示す。

Here, P represents transmission power, k represents a parameter for designating a relay node (k = 1,..., K), and σ ² represents noise power. As apparent from the equation (5), the integrated path loss coefficient is a function of the product of the path loss coefficients of both the front and rear channels, and therefore depends on both channel states. An integrated path loss coefficient λ _k is calculated for each relay station. The relay station classification unit 41 classifies the calculated integrated path loss coefficient λ _k into one or more clusters. For example, if the integrated path loss coefficient λ _k is smaller than the threshold ν ₁ , it is classified as the first cluster, if it is greater than or equal to the threshold ν _{1 and} less than ν _2, it is classified as the second cluster, and if it is greater than or equal to the threshold ν _{2 and} less than ν _3, it is In the same manner, the integrated path loss coefficient is classified into an appropriate cluster. FIG. 5 shows an example of integrated path loss coefficients classified in this way.

  The cluster selection unit 42 in FIG. 4 selects one or more relay stations that belong to the smallest number of numerical ranges (first cluster), and excludes other relay stations from relay candidates.

The look-up table 44 represents the correspondence relationship between the number L of relay stations, the relay station ρ (λ _k ), and the communication capacity C _k ^ρ . Various values prepared in the lookup table are limited to those related to the relay station selected by the cluster selection unit 42. Accordingly, calculation of communication capacity or the like is unnecessary for combinations including relay stations excluded from relay candidates, and calculation efficiency can be improved. In the look-up table in the present embodiment, the kth relay station is designated by the ratio ρ of the path loss coefficients of the front and rear channels.

ρ = ρ (k) = (λ ^BC _k ) / (λ ^FC _k )
A specific example of the lookup table is shown in FIG.

  The relay station selection unit 45 in FIG. 4 determines the optimum number of relay stations and the combination of relay stations while referring to the lookup table, and outputs the contents of the combination.

  FIG. 7 is a flowchart of a relay node selection method according to an embodiment of the present invention. Although the flow starts from step S71, it is assumed that calculation of the integrated path loss coefficient, cluster classification, and selection of the optimal cluster have already been completed at this point. Therefore, the procedure described below is mainly performed by the relay station selection unit 45 (FIG. 4). For convenience of explanation, it is assumed that the first cluster as shown in FIG. 5 has been selected and the lookup table as shown in FIG. 6 has been created. In this case, the path loss coefficient is assumed as follows.

k = 2: (λ ^BC ₂ , λ ^FC ₂ ) = (0.1, 1.0), ρ (2) = 0.1
k = 7: (λ ^BC ₇ , λ ^FC ₇ ) = (0.2,1.0), ρ (7) = 0.2
k = 5: (λ ^BC ₅ , λ ^FC ₅ ) = (0.3, 1.0), ρ (5) = 0.3
k = 4: (λ ^BC ₄ , λ ^FC ₄ ) = (1.0, 0.1), ρ (4) = 10
Under these conditions, the communication capacity C _k ^{ρ of the} entire system as shown in the equation (2) is calculated for all possible combinations of relay candidates, and prepared in the lookup table.

  In step S72, the number of relay stations L and the relay station parameter i are each initially set to 1.

In step S73, the communication capacity C _k ^ρ corresponding to the number of relay stations L = 1 is compared, and the relay station associated with the maximum capacity MAX {C _k ^ρ } is selected. In the example of FIG. 6, 6.3 is the largest among the capacity values corresponding to L = 1, and therefore the relay station ρ = 0.3 (k = 5) corresponding to 6.3 is selected.

Step S74 In selected relay station numbered I ₁ associated, the relay station number is determined in the relay station actually relayed, are excluded from the relay candidates for further selecting relay station. In the example of FIG. 6, I ₁ = 5, and the relay station of k = 5 is determined as the relay station that actually relays the signal, and is excluded from the relay candidates.

In step S75, the communication capacity C _{k + 1} ^ρ for the number of relay stations L + 1 is compared, and the relay station associated with the maximum capacity MAX {C _{k + 1} ^ρ } is found. In the example of FIG. 6, paying attention to the communication capacity of a relay station other than the selected ρ = 0.3 among the communication capacities corresponding to the number of relay stations L + 1 = 2, the maximum capacity 6.8 is among them. Found.

In step S76, the previous maximum capacity C _k ^ρ is compared with the current maximum capacity MAX {C _{k + 1} ^ρ }. If the current maximum capacity is larger, the flow proceeds to step S77. In the example of FIG. 6, the maximum capacity 6.3 when L = 1 is compared with the maximum capacity 6.8 when L = 2.

In step S77, a relay station corresponding to MAX {C _{k + 1} ^ρ } is selected, and a number I _{i + 1} is associated therewith, and the relay station is determined as a relay station that actually relays a signal. And it is excluded from relay candidates. In the example of FIG. 6, the number I ₂ is associated with the relay station ρ = 0.2 (k = 7) corresponding to the maximum capacity 6.8 corresponding to L = 2, and is determined as an actual relay station.

  In step S78, the number L of relay stations and the parameter i are incremented by one, and the process returns to step S75, and the operation already described is performed again. In the example of FIG. 6 at present, attention is paid to the communication capacities of relay stations other than the selected ρ = 0.3 and 0.2 among the communication capacities corresponding to the number of relay stations L + 1 = 3. A capacity of 7.4 is found.

  In the next step S76, the maximum capacity 6.8 when L = 2 is compared with the maximum capacity 7.4 when L = 3.

Numbered _{I 3} associated with the relay station [rho = 10 corresponding to the maximum capacity 7.4 corresponding to the next step S77 the L = 3 (k = 4) , is determined on the actual relay station.

  In the next step S78, the number L of relay stations and the parameter i are incremented by one, and the process returns to step S75 to perform the operation already described. This time, paying attention to the communication capacities of relay stations other than the selected ρ = 0.3, 0.2 and 10 among the communication capacities corresponding to the number of relay stations L + 1 = 4, the maximum capacity 6 .5 is found.

In the next step S76, the previous maximum capacity C _k ^ρ is compared with the current maximum capacity MAX {C _{k + 1} ^ρ }. If the current maximum capacity is larger, the flow proceeds to step S77, but if not, the flow proceeds to step S79 and ends. In the example of FIG. 6, the maximum capacity 7.4 when L = 3 is compared with the maximum capacity 6.5 when L = 4. Since the former is larger than the latter, the flow ends.

In the example of FIG. 6, the communication capacity of the entire system increases as the number of relay stations is increased to 1, 2, 3, and increases accordingly, but when the number of relay stations is increased to 4, it decreases from the case of 3. End up. Therefore, the optimum combination of relay stations is (I ₁ , I ₂ , I ₃ ) = (12−5, 12−7, 12−4), and the number L of relay stations is 3. As described above, according to the present embodiment, the number of relay candidates can be reduced by dividing into clusters, and an optimal combination can be derived reliably and efficiently.

1 is a diagram illustrating a communication system according to an embodiment of the present invention. It is a figure which shows the functional block diagram of the relay station selection apparatus by one Example of this invention. It is a figure which shows the flowchart of the communication method by one Example of this invention. It is a figure which shows the functional block diagram of the relay station selection apparatus by one Example of this invention. It is a figure which shows the cluster classification example of the integrated path loss coefficient by one Example of this invention. It is a figure which shows an example of the look-up table used by one Example of this invention. It is a figure which shows the flowchart of the relay station selection method by one Example of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Source node 12-1-K Relay station 13 Target node 20 Relay station selection apparatus 22 Pattern generation part 23 Communication capacity calculation part 24 Complex matrix generation part 25 Capacity calculation part 26 Pattern selection part 40 Relay station selection apparatus 41 Relay station classification part 42 Cluster selection unit 44 Look-up table 45 Relay station selection unit H _k Channel matrix of rear channel G _k Channel matrix of front channel λ ^BC _k Path loss coefficient of rear channel λ ^FC _k Path loss coefficient of front channel
H , G composite matrix λ _k integrated path loss coefficient

Claims (12)

A relay station selection device used in a communication system for transmitting a radio signal from a source node to a target node through a plurality of relay stations,
Means for receiving, from a plurality of relay stations, a signal indicating a path loss coefficient of a backward channel between the source node and the relay station and a path loss coefficient of a forward channel between the relay station and the target node;
Means for generating first and second composite matrices each including a plurality of channel matrices;
Means for calculating a determinant of a matrix including a square matrix derived from a first and second composite matrix and a conjugate transpose of the composite matrix;
Means for comparing the communication capacities calculated based on determinants for each of the combinations including one or more relay station candidates and selecting one combination;
Means for transmitting a signal indicating the number of relay stations and relay stations included in the selected combination to one or more relay stations;
A relay station selection device comprising:   The relay station selection device according to claim 1, wherein the relay station selection device is provided in a source node. The means for generating the composite matrix includes: a first composite matrix including a plurality of products of a channel matrix and a path loss coefficient of a rear channel; and a second composite matrix including a plurality of products of a channel matrix and a path loss coefficient of a front channel; The relay station selection device according to claim 1, wherein: The relay station selection apparatus according to claim 1, wherein the receiving means also receives signals indicating channel matrices of backward and forward channels from relay station candidates. The relay station selection apparatus according to claim 1, further comprising means for calculating an integrated path loss coefficient expressed by a function of a product of the path loss coefficients of the front and rear channels for each candidate relay station. The relay station selection device according to claim 5, further comprising selection means for selecting a relay station candidate corresponding to an integrated path loss coefficient belonging to a predetermined numerical range. The relay station selection apparatus according to claim 6, wherein relay station candidates included in the combination are limited to relay station candidates selected by the selection unit. The relay station selection device according to claim 1, further comprising storage means for storing a table indicating a correspondence relationship between the relay station, the number of relay stations, and the communication capacity. A relay station selection method used in a communication system for transmitting a radio signal from a source node to a target node through a plurality of relay stations,
Receiving from a plurality of relay stations signals indicating a path loss coefficient of a backward channel between the source node and the relay station and a path loss coefficient of a forward channel between the relay station and the target node;
Generating first and second composite matrices each including a plurality of channel matrices;
Calculating a determinant of a matrix including a square matrix derived from a first and second composite matrix and a conjugate transpose of the composite matrix;
Comparing communication capacities calculated based on determinants for each combination including one or more relay station candidates and selecting one combination;
Transmitting a signal indicating the number of relay stations and relay stations included in the selected combination to one or more relay stations;
A relay station selection method comprising: Selecting the one combination comprises:
Refer to the table showing the correspondence relationship between the relay station, the number of relay stations, and the communication capacity, compare the communication capacity corresponding to a certain number of relay stations, select one relay station, and correspond to a larger number of relay stations Comparing the communication capacity and selecting another relay station;
If the communication capacity corresponding to the selected one relay station is larger than the communication capacity corresponding to the selected another relay station, the larger number of relay stations and the other relay station Determining not to adopt for transmission;
The relay station selection method according to claim 9, further comprising: A relay station selection method used in a communication system for transmitting a radio signal from a source node to a target node through a plurality of relay stations,
Each relay station estimating a channel matrix of a backward channel between the source node and the relay station and a channel matrix of a forward channel between the relay station and the target node;
Generating first and second composite matrices each including the plurality of channel matrices;
Calculating a determinant of a matrix including a square matrix derived from a first and second composite matrix and a conjugate transpose of the composite matrix;
Comparing communication capacities calculated based on channel determinant estimates for each of the combinations including one or more relay station candidates and selecting one combination;
Transmitting a signal indicating the number of relay stations and relay stations included in the selected combination to one or more relay stations;
A relay station selection method comprising: A relay station selection method used in a communication system for transmitting a radio signal from a source node to a target node through a plurality of relay stations,
A step in which a source node transmits a covariance matrix of a backward channel to each relay station to each relay station, and a step in which a target node transmits a covariance matrix of a forward channel to each relay station to each relay station; ,
Generating first and second composite matrices each including the plurality of channel matrices;
Calculating a determinant of a matrix including a square matrix derived from a first and second composite matrix and a conjugate transpose of the composite matrix;
Comparing communication capacities calculated based on channel determinant estimates for each of the combinations including one or more relay station candidates and selecting one combination;
Transmitting a signal indicating the number of relay stations and relay stations included in the selected combination to one or more relay stations;
A relay station selection method comprising:

Images