JP6265485B2 - Communication device, control method, and program - Google Patents

Description

  The present invention relates to multi-user MIMO technology.

  A multi-input multi-output (MIMO) technique is known in which a transmitting / receiving station simultaneously transmits / receives a plurality of data streams using a plurality of antennas to increase the speed of data transmission / reception. By using MIMO, ideally, the wireless communication capacity can be increased in proportion to the number of antennas. For this reason, it is considered that a large number of antennas such as 112 (see Non-Patent Document 1) and 128 (see Non-Patent Document 2) are arranged in the base station to greatly increase the radio communication capacity. By arranging a large number of antennas in the base station in this way, high-speed communication can be performed simultaneously with one or more user terminals having one or more antennas.

  In MIMO, the radio communication capacity can be increased by increasing the number of antennas as described above. However, in order to obtain the effect, a channel between each of a plurality of transmitting antennas and each of a plurality of receiving antennas is used. Need to be estimated. Channel estimation is performed, for example, by observing a pilot signal transmitted from the base station with one or more user terminals and feeding back the observation result to the base station. However, in such a system, since the amount of information that needs to be fed back becomes very large, there is a problem that a large amount of communication resources must be allocated for this feedback.

  On the other hand, in a time division duplex system, it is possible to estimate a downlink channel by measuring, for example, an uplink signal using the symmetry of the channel between the uplink and the downlink. In this case, the base station can estimate the downlink channel, for example, by observing pilot signals transmitted by a plurality of user terminals in the uplink.

J. et al. Hoydis, C.I. Hoek, T .; Wild and S.W. ten Brink, “Channel Measurements for Large Antenna Arrays”, IEEE International Symposium on Wireless Communications Systems (ISWCS) August 2012. S. Payami and F.M. Tufvesson, “Channel Measurements and Analysis for Very Large Array Systems At 2.6 GHz”, 6th European Conference on AntennaChapter 12 T.A. L. Marzetta, “Noncooperative Cellular Wireless with Unlimited Numbers of Base Station Antennas”, IEEE Trans. Wireless Communications, vol. 9, no. 11, pp. 3590-3600, November 2010

  In a MIMO environment, user terminals may use different numbers of antennas for transmission and reception. For example, the user terminal may use only one antenna for transmission while using two antennas for reception. In this case, considering the case where N antennas are used on the base station side, the size of the channel matrix is 2 rows and N columns in the downlink, and N rows and 1 column in the uplink. Therefore, in such a case, it is difficult to estimate the downlink channel matrix from the uplink channel matrix. Therefore, in this case, it may be difficult to obtain a sufficient MIMO effect.

  The present invention has been made in view of the above problems, and an object thereof is to provide an effective MIMO transmission technique in an environment in which a communication apparatus uses a different number of antennas for transmission and reception.

  In order to achieve the above object, a communication apparatus according to the present invention uses a first plurality of antennas to communicate with a counterpart apparatus that has a second plurality of antennas and uses only a part of them for transmission. A communication apparatus, wherein signals transmitted from the part of the second plurality of antennas are received by each of the first plurality of antennas, whereby each of the first plurality of antennas and the first antenna Estimation means for estimating a first channel matrix having a value of a channel between each of the two antennas as a component, and a signal transmitted from each of the first plurality of antennas, A first profile estimated in the counterpart device based on a second channel matrix having elements of channel values when received by at least one of the remaining portions of the second plurality of antennas. Acquisition means for acquiring information identifying a coding matrix from the counterpart device; calculation means for calculating a second precoding matrix by connecting the first channel matrix and the first precoding matrix; Transmitting means for transmitting each element of the transmission signal vector obtained by multiplying the second precoding matrix by a transmission data vector including transmission data as an element from each of the first plurality of antennas. .

  Another communication apparatus according to the present invention includes a first plurality of antennas, and uses only a part of the first plurality of antennas for signal transmission and uses the first plurality of antennas for signal reception. A communication apparatus that performs communication with a counterpart apparatus having a second plurality of antennas using at least one of the part of the plurality of antennas and the remaining part of the plurality of antennas, the second plurality of antennas Based on a first channel matrix whose elements are channel values when signals transmitted from each of the antennas are received at at least one of the remaining portions of the first plurality of antennas, An estimation unit that estimates one precoding matrix, and a transmission unit that transmits information specifying the first precoding matrix to the counterpart device, wherein the first plurality of antennas in the counterpart device Before Based on the second channel matrix and the first precoding matrix as elements of channel values when signals transmitted from each of the parts are received by the second plurality of antennas, respectively. 2 precoding matrices are calculated, and each transmission signal vector element obtained by multiplying the transmission data vector including transmission data as an element by the second precoding matrix is transmitted from each of the second plurality of antennas. It is characterized by that.

  According to the present invention, it is possible to provide an effective MIMO transmission technique in an environment where a communication apparatus uses different numbers of antennas for transmission and reception.

The figure which shows the structural example of a radio | wireless communications system. The block diagram which shows the example of a function structure of a base station and a user terminal. The block diagram which shows the example of the hardware constitutions of a base station and a user terminal.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

(Wireless communication system)
FIG. 1 shows a configuration example of a wireless communication system according to this embodiment. The wireless communication system is configured by a plurality of wireless communication devices, and includes, for example, one base station and one or more user terminals that are communication partner devices of the base station. Although not shown, it is assumed that there are also base stations in adjacent cells and user terminals connected to the base stations.

  The base station has multiple antennas and can simultaneously transmit signals to one or more user terminals and similarly receive signals from one or more user terminals simultaneously. The user terminal has one or a plurality of antennas. It is assumed that the base station and one or more user terminals communicate by time division duplex. That is, the time when the base station transmits a signal is a time when the user terminal receives a signal from the base station without transmitting a signal, and the time when the user terminal transmits a signal is when the user terminal does not transmit a signal. Assume that a signal from a terminal is received.

  In the present embodiment, at least some of the user terminals use different numbers of antennas for transmission and reception. For example, a user terminal has two antennas, uses both to receive signals, and uses only one of them to transmit signals. For example, a certain user terminal has four antennas, two of which are used for signal transmission, and one of the remaining two is used for signal reception in addition to the two. Not all antennas have to be used at all times, such as using three. In the following, it is assumed that the antenna used for transmission in the user terminal is also used for reception. This is because the symmetry of the channel matrix is used. However, at least a part of the antenna used for transmission may be used for reception, and the remaining part may be dedicated for transmission.

  Below, the process of the base station and user terminal in this embodiment is demonstrated. FIG. 1 shows an example in which a plurality of user terminals are included in the wireless communication system. However, if there is one or more user terminals that use different numbers of antennas for transmission and reception, The following discussion can be applied. Therefore, in the following, the flow of processing when there is one user terminal will be described first, and then processing when there are a plurality of user terminals will be described.

(Outline of processing)
In the following, it is assumed that the base station has N antennas, all of which are used for transmission and reception. The user terminal also has N antennas, all of which are used for reception, while some of them (M, M <N) are used for transmission.

  In this case, a signal transmitted from each of the N antennas of the base station is received by each of the N antennas of the user terminal, and a channel matrix H whose elements are the channel values at this time has N rows and N columns. It becomes the matrix of. On the other hand, a signal transmitted from each of the M antennas of the user terminal is received by each of the N antennas of the base station, and a channel matrix G whose elements are the channel values at this time has N rows and M columns. It becomes. Therefore, the base station can obtain the uplink channel matrix G from the signal transmitted from the user terminal even in a time division duplex (TDD) system. The channel matrix H cannot be obtained. Therefore, in this state, the base station cannot calculate a precoding matrix for efficiently performing MIMO transmission based on the channel matrix H.

  Therefore, in this embodiment, in order to deal with such a problem, information feedback is received from the user terminal regarding the precoding matrix related to the remaining components of the channel matrix H. In the following, it will be described in detail how this information to be fed back is defined and how it is transmitted.

・・・（１）
となることが分かっている（非特許文献３参照）。なお、上付きのＨは行列の共役転置を表し、Ｉ_NはＮ行Ｎ列の単位行列を示す。このため、非特許文献３では、十分に大きいＮに対して、最適なプリコーディング行列はＨ^Hであるとしている。すなわち、Ｎ個の送信データからなる送信データベクトルをｓとすると、ｘ＝Ｈ^Hｓとなるように、Ｎ本の送信アンテナのそれぞれから送信される信号を要素とする送信信号ベクトルを形成するのが最適となる。実際、このようにすることにより、雑音ベクトルをｎとし、Ｎ本の受信アンテナで受信される信号を要素とする受信信号ベクトルをｙとすると、
ｙ＝Ｈｘ＋ｎ＝ＨＨ^Hｓ＋ｎ＝ｓ＋ｎ
となり、アンテナ間の干渉の影響がなくなる。 (Precoding matrix)
According to the law of large numbers, when the number N of antennas increases,

... (1)
(See Non-Patent Document 3). Note that the superscript H represents a conjugate transpose of a matrix, and I _N represents a unit matrix of N rows and N columns. For this reason, in Non-Patent Document 3, for a sufficiently large N, the optimal precoding matrix is H ^H. That is, when a transmission data vector composed of N pieces of transmission data is s, a transmission signal vector having elements transmitted from N transmission antennas as elements is formed so that x = H ^H s. Is optimal. In fact, by doing this, if the noise vector is n and the received signal vector whose elements are signals received by the N receiving antennas is y,
y = Hx + n = HH ^H s + n = s + n
Thus, the influence of interference between antennas is eliminated.

・・・（２）
としたとき、チャネル行列ＨのＭ行分の要素を、

・・・（３）
のように推定することができる。なお、ここでは、上からＭ行分の成分が、ユーザ端末において送信にも使用されるアンテナにおいて受信される成分となるようにチャネル行列Ｈを定義している。 On the other hand, in this embodiment, a part of the downlink channel matrix H can be estimated from the uplink channel matrix G by using the symmetry of the channel using TDD. That is, the downlink channel matrix H and the uplink channel matrix G are

... (2)
, The elements for M rows of the channel matrix H are

... (3)
It can be estimated as follows. Here, the channel matrix H is defined so that components for M rows from the top are components received by an antenna also used for transmission in the user terminal.

・・・（４）
とする。すると、チャネル行列Ｈは、上りリンクのチャネル行列Ｇと残りの部分のチャネル成分の行列Ｆとを連結して、

・・・（５）
と表される。なお、ここでは、上りリンクのチャネル行列Ｇに対してそのまま共役転置行列とすることによって、下りリンクのチャネル行列Ｈの一部を推定しているが、これに限られない。例えば、時間平均を取るなど、上りリンクのチャネル行列Ｇに所定の処理を施した結果の共役転置行列を、下りリンクのチャネル行列Ｈの一部の推定値としてもよい。 Here, the remaining components of the channel matrix H (components for NM rows from the bottom) are represented by a matrix F. That is,

... (4)
And Then, the channel matrix H connects the uplink channel matrix G and the matrix F of the remaining channel components, and

... (5)
It is expressed. Here, a part of the downlink channel matrix H is estimated by using the conjugate transpose matrix as it is for the uplink channel matrix G, but the present invention is not limited to this. For example, a conjugate transposition matrix obtained by performing predetermined processing on the uplink channel matrix G, such as taking a time average, may be used as an estimated value of a part of the downlink channel matrix H.

・・・（６）
となる。 At this time, the optimal precoding matrix P can be expressed as P = H ^H = [G F ^H ]. In this case, HH ^H / N becomes asymptotic to I _{N as N} increases,

... (6)
It becomes.

・・・（７）
となる。ここで、本行列の左下の部分行列ＦＧは、式（６）に示されるように、ゼロ行列である。一方、本行列の右上の部分行列は、Ｇが平均０のガウス分布に従うため、Ｇによらずに行列Ｔを選択すれば、やはりゼロ行列に収束する。したがって、ＦＴ／Ｎ＝Ｉ_N-Mに十分近くなるような行列Ｔを選択することができれば、最適又は準最適なプリコーディング行列を用いて信号を送信することが可能となる。 However, in this embodiment, the base station cannot know the matrix F. For this reason, consider a case where the base station uses the matrix T instead of the matrix F as a partial precoding matrix. In this case, the matrix product of the channel matrix H and the precoding matrix P obtained by concatenating the partial matrix and the partial precoding matrix for which the channel can be estimated is

... (7)
It becomes. Here, the lower left submatrix FG of this matrix is a zero matrix as shown in Equation (6). On the other hand, since the submatrix at the upper right of this matrix follows a Gaussian distribution with an average of 0, if the matrix T is selected regardless of G, it will converge to a zero matrix. Therefore, if a matrix T that is sufficiently close to FT / N = I _NM can be selected, a signal can be transmitted using an optimal or sub-optimal precoding matrix.

  However, just as the channel matrix G follows a mean 0 Gaussian distribution, the matrix F follows a mean 0 Gaussian distribution. For this reason, if the partial precoding matrix T is selected regardless of F, FT / N also converges to a zero matrix. Therefore, in the present embodiment, a user terminal that can know the matrix F that is a part of the channel matrix H estimates this partial precoding matrix T, and sends information about the partial precoding matrix to the base station. Notice. Hereinafter, the content of the specific process which a base station and a user terminal perform, and the example structural aspect of a base station and a user terminal are demonstrated.

(Processing executed by base station and user terminal)
First, as the simplest procedure, the user terminal can feed back the matrix F itself, which is a part of the channel matrix H, or its conjugate transpose matrix to the base station. That is, the user terminal can feed back a channel matrix or its conjugate transpose matrix when signals transmitted from N antennas of the base station are received by NM antennas not used for transmission. . Thereby, the base station can know the conjugate transpose matrix F ^H of the matrix F, and by using this as a precoding matrix, optimal signal transmission by MIMO can be performed.

  On the other hand, this method may cause a problem that the feedback cost is high. That is, since the elements (channel values) themselves included in the matrix F are fed back to the base station, the amount of information to be fed back becomes very large, which may impose a transmission capacity.

Therefore, in the following, the user terminal selects the partial precoding matrix T by selecting from a plurality of predetermined partial precoding matrix candidates T _i (i is an index, for example, an integer of 0 ≦ i <n). presume. Note that the user terminal can significantly reduce the amount of information by feeding back an index for a selected one of the precoding matrix candidates. For example, when the number of candidates for the partial precoding matrix T is four, feedback information can be transmitted with 2 bits.

  First, as a preliminary preparation, the base station and the user terminal previously store a partial precoding matrix candidate and an index in association with each other. That is, the base station can know the partial precoding matrix selected by the user terminal by designating one index. Here, if the number of antennas of the base station is constant, according to the number of columns of the partial precoding matrix T, that is, according to the value obtained by subtracting the number of transmission antennas from the number of reception antennas at the user terminal, The size of the partial precoding matrix changes. Therefore, the base station and the user terminal classify the partial precoding matrix for each number of columns of the matrix, store the partial precoding matrix in association with the index, and specify one partial precoding matrix based on the number of columns and the index. You may do it. Even in this case, if the base station can know in advance the difference between the number of reception antennas and the number of transmission antennas in the user terminal, the column number information need not be fed back.

  First, the user terminal acquires a downlink channel matrix H based on signals (for example, pilot signals) transmitted from each of a plurality of antennas by the base station. Here, the user terminal may acquire only the channel matrix F with an antenna not used for transmission. On the other hand, the base station acquires an uplink channel matrix G by receiving a signal (for example, a pilot signal) transmitted from each of one or more antennas by the user terminal. Note that either the acquisition of the uplink channel matrix in the base station or the acquisition of the downlink channel matrix in the user terminal may be performed first.

・・・（８）
のように、部分プリコーディング行列Ｔを選択する。なお、式中、Ωは、部分プリコーディング行列の候補の集合である。すなわち、取得した下りリンクのチャネル行列の一部である行列Ｆとの行列積を基地局のアンテナの本数Ｎで割った結果の行列と単位行列との差が最小となるような候補が、部分プリコーディング行列として選択される。式（８）の例では、ｉ０番目の候補が、部分プリコーディング行列Ｔとして選択された例を示している。 Based on the matrix F that is part of the channel matrix H, the user terminal selects one candidate as a feedback target from among a plurality of stored partial precoding matrix candidates T _i . Here, as described above, the partial precoding matrix T should be selected to be sufficiently close to FT / N = I _NM . Therefore, the user terminal

... (8)
The partial precoding matrix T is selected as follows. In the equation, Ω is a set of partial precoding matrix candidates. That is, candidates for which the difference between the matrix obtained by dividing the matrix product with the matrix F, which is a part of the acquired downlink channel matrix, by the number N of antennas of the base station and the unit matrix is minimized Selected as precoding matrix. In the example of Expression (8), the i0th candidate is selected as the partial precoding matrix T.

  In the example of Expression (8), a candidate having the smallest difference between the matrix obtained by dividing the matrix product with the matrix F by the number N of antennas of the base station and the unit matrix is selected as the partial precoding matrix T. However, it is not limited to this. For example, a candidate such that the difference between the matrix obtained by dividing the matrix product with the matrix F by the number N of base station antennas and the unit matrix is equal to or smaller than a predetermined value may be selected as the partial precoding matrix T. . In this case, in a plurality of candidates, when the difference between the matrix obtained by dividing the matrix product with the matrix F by the number N of antennas of the base station and the unit matrix is equal to or less than a predetermined value, One may be selected at random.

  When the user terminal finishes selecting the partial precoding matrix T, the user terminal feeds back the index value associated with the selected candidate to the base station. In the example of Expression (8), i0 is fed back to the base station.

  The base station extracts a partial precoding matrix T to be used from the stored partial precoding matrix candidates based on the fed back index. Then, the uplink channel matrix G and the extracted partial precoding matrix T are concatenated to obtain a precoding matrix P = [GT].

  The base station transmits each element of the transmission signal vector obtained as a result of multiplying the precoding matrix P obtained in this way by a transmission data vector including transmission data as an element from each of the N antennas. . As a result, the result obtained by dividing the matrix product of the channel matrix H and the precoding matrix P by the number N of antennas is sufficiently close to the unit matrix, so that the user terminal can be transmitted without much interference between antennas. Data can be extracted.

・・・（９）
のように表すことができる。ここでＧ₁は、第１のユーザ端末からの信号から得られる上りリンクのチャネル行列であり、Ｆ₁は、基地局の各アンテナと、第１のユーザ端末の送信には使用されないアンテナとの間の下りリンクのチャネル行列である。同様に、Ｇ₂は、第２のユーザ端末からの信号から得られる上りリンクのチャネル行列であり、Ｆ₂は、基地局の各アンテナと、第２のユーザ端末の送信には使用されないアンテナとの間の下りリンクのチャネル行列である。 Note that the same processing can be applied when there are a plurality of user terminals, that is, in the case of multi-user MIMO. For example, a base station has 2N antennas, all of them are used for both transmission and reception, two user terminals have N antennas, and all of them are used for reception, A case where only M (M <N) are used for transmission will be described. In this case, for example, the downlink channel H is

... (9)
It can be expressed as Here, G ₁ is an uplink channel matrix obtained from the signal from the first user terminal, and F ₁ is the relationship between each antenna of the base station and an antenna that is not used for transmission of the first user terminal. It is a downlink channel matrix. Similarly, G ₂ is an uplink channel matrix obtained from a signal from the second user terminal, and F ₂ is an antenna that is not used for transmission of each antenna of the base station and the second user terminal. Is a downlink channel matrix.

・・・（１０）
が成り立つ。 Also in this case, since HH ^H / 2N = I _2N , if the precoding matrix P = H ^H is used,

... (10)
Holds.

・・・（１１）
となる。ここで、上述の説明と同様に、Ｇ及びＦが、平均がゼロのガウス分布であるとすれば、Ｇ及びＦと関係なく部分プリコーディング行列Ｕ及びＶを選択すると、これらの行列積はゼロ行列となる。すなわち、例えば、ＵをＦ₁のみに基づいて選択すれば、Ｇ₁Ｕ、Ｇ₂ ^HＵ、Ｆ₂Ｕはそれぞれ高い確率でゼロ行列に漸近する。また、例えば、ＶをＦ₂のみに基づいて選択すれば、Ｇ₁ ^HＶ、Ｆ₁Ｖ、Ｇ₂Ｖはそれぞれ高い確率でゼロ行列に漸近する。 However, the base station does not know the matrices F ₁ and F ₂ . For this reason, instead of these, the base station uses another partial precoding matrix U and V and sets the precoding matrix P to [G ₁ U G ₂ V]. In this case, the matrix product of the channel matrix H and the precoding matrix P is

(11)
It becomes. Here, as in the above description, if G and F are Gaussian distributions with an average of zero, if the partial precoding matrices U and V are selected regardless of G and F, these matrix products are zero. It becomes a matrix. That is, for example, if U is selected based only on F ₁ , G ₁ U, G ₂ ^H U, and F ₂ U each asymptotically approach a zero matrix with high probability. For example, if V is selected based only on F ₂ , G ₁ ^HV , F ₁ V, and G ₂ V each asymptotically approach the zero matrix with high probability.

Therefore, in this case as well, it is sufficient to select U and F ₂ V / 2N that minimize the difference between F ₁ U / 2N and the unit matrix as the partial precoding matrix, and V that minimizes the difference between the unit matrix. . For this reason, for example, the first user terminal can select the partial precoding matrix U without knowing F ₂ or G ₂ . Similarly, the second user terminal can select the partial precoding matrix V without knowing F ₁ or G ₁ . Note that the first user terminal may select a partial precoding matrix U such that the difference between F ₁ U / 2N and the unit matrix is a predetermined value or less. Further, the second user terminal may select a partial precoding matrix V such that the difference between F ₂ V / 2N and the unit matrix is equal to or less than a predetermined value. Further, these partial precoding matrices may be selected from among a plurality of candidates stored in each user terminal, and in this case, feedback of the partial precoding matrix to the base station uses a corresponding index. It may be done.

  As described above, the method according to the present embodiment can be easily applied to multi-user MIMO.

  In the case of multi-user MIMO, some user terminals use all antennas for transmission and reception, and the remaining some user terminals use all antennas for reception but not for transmission. In some cases, it may not occur. For example, some user terminals have only one antenna. In such a case, the estimation and feedback of the partial precoding matrix as described above need only be performed by user terminals having different numbers of antennas for transmission and reception. This is because, for user terminals that use all antennas for transmission and reception, the channel matrix G obtained from signals transmitted in the uplink can be used as it is as a precoding matrix, and other information is unnecessary.

(Configuration example of base station and user terminal)
A configuration example of the base station and the user terminal that performs the above-described processes is shown in FIG. In FIG. 2, only one user terminal is shown. However, in the case of multi-user MIMO, there are a plurality of user terminals. In the example of FIG. 2, the base station 200 includes a transmission / reception unit 201, a channel estimation unit 202, a partial precoding matrix extraction unit 203, a partial precoding matrix candidate storage unit 204, and a precoding matrix acquisition unit 205. The base station 200 also has a function as another general base station, but is not shown here for simplicity. In addition, the user terminal 220 includes, for example, a transmission / reception unit 221, a channel estimation unit 222, a partial precoding matrix estimation unit 223, and a partial precoding matrix candidate storage unit 224. The user terminal 220 also has a function as another general base station, but is not shown here for simplicity.

  In the base station 200, the transmission / reception unit 201 is a MIMO transmission / reception unit that uses a plurality of antennas for transmission and reception. Channel estimation section 202 estimates uplink channel matrix G when signals transmitted from some of the plurality of antennas of user terminal 220 are received by each antenna of base station 200. The channel estimation unit 202 inputs the estimated channel matrix G to the precoding matrix acquisition unit 205.

  The partial precoding matrix extraction unit 203 extracts the partial precoding matrix T selected by the user terminal from the partial precoding matrix candidate storage unit 204 according to the index fed back from the user terminal 220. The extracted partial precoding matrix T is input to the precoding matrix acquisition unit 205. The partial precoding matrix candidate storage unit 204 stores a plurality of partial precoding matrix candidates. Note that the partial precoding matrix candidate storage unit 204 may store the precoding matrix candidates separately for each matrix size, as described above. For example, when the number of antennas used by the base station 200 is changed depending on the case, the number of rows of the precoding matrix may vary, and the difference in the number of antennas used for transmission and reception at the user terminal may vary. is there. The partial precoding matrix candidate storage unit 204 stores partial precoding matrices of various sizes corresponding to a plurality of rows and a plurality of columns in advance, so that appropriate partial precoding can be performed in various situations. A matrix can be selected.

  The precoding matrix acquisition unit 205 acquires the precoding matrix P by concatenating the input uplink channel matrix G and the partial precoding matrix T in the column direction. The precoding matrix P is input to the transmission / reception unit 201. The transmission / reception unit 201 obtains a transmission signal vector by multiplying the input precoding matrix P from the left by a transmission data vector including transmission data as an element, and transmits each element of the transmission signal vector from a corresponding antenna.

  In the user terminal 220, for example, the transmission / reception unit 221 transmits and receives signals using only one antenna for transmission while using two antennas for reception. Channel estimation section 222 estimates at least a part F of downlink channel matrix H based on signals transmitted from a plurality of antennas of base station 200 and received by a plurality of antennas of a user terminal. Note that the channel estimation unit 222 generally estimates the entire channel matrix H for normal data reception, but in this embodiment, an antenna that is not used for signal transmission in the channel matrix. It is sufficient to estimate only the component (matrix F) for the signal received in (1). After the base station 200 acquires the precoding matrix P, since the matrix product of the channel matrix H and the precoding matrix P is sufficiently close to the unit matrix, the transmitted data can be obtained without estimating the channel. It is because it becomes possible to acquire.

  The partial precoding matrix estimation unit 223 estimates the partial precoding matrix T from the estimated value of the matrix F that is a part of the channel matrix. For example, among the plurality of candidates stored in the partial precoding matrix candidate storage unit 224, the difference between the matrix obtained by dividing the matrix product with the estimated matrix F by the number of antennas of the base station 200 and the unit matrix is The smallest candidate is estimated as a partial precoding matrix. In addition, for example, among a plurality of candidates, a candidate in which a difference between a matrix obtained by dividing a matrix product with the estimated matrix F by the number of antennas of the base station 200 and a unit matrix is equal to or less than a predetermined value is partial precoding. Estimated as a matrix. The partial precoding matrix candidate storage unit 224 is the same as the partial precoding matrix candidate storage unit 204, and the stored partial precoding matrix candidate and the corresponding index value are common.

  Note that the base station 200 and the user terminal 220 have a hardware configuration as shown in FIG. 3, for example, and include, for example, a CPU 301, a ROM 302, a RAM 303, an external storage device 304, and an input / output device 305. In the base station 200 and the user terminal 220, for example, the CPU 301 executes a program that realizes the functions of the base station 200 and the user terminal 220 recorded in any of the ROM 302, the RAM 303, and the external storage device 304. Then, the base station 200 and the user terminal 220 use the input / output device 305 to acquire information from the transmission / reception units 201 and 221 and output information to the transmission / reception units 201 and 221, for example. Note that the CPU 301 may have a function of controlling the transmission / reception units 201 and 221 for signal transmission / reception.

  Note that the base station 200 and the user terminal 220 may include dedicated hardware for executing the above-described functions, or a part of the base station 200 and the user terminal 220 may be executed by the computer and the other part may be executed by a computer that operates the program. May be. Further, all the following functions may be executed by a computer and a program.

  Thus, according to the present embodiment, the base station is able to estimate an appropriate precoding matrix P even in a situation where it is not possible to estimate all values of the downlink channel matrix from the uplink channel matrix. Can be obtained. In addition, the feedback cost can be lowered by feeding back only the index of the partial precoding matrix. Note that, unlike estimating the precoding matrix for the entire downlink channel matrix, the precoding matrix is estimated for only a part of the channel matrix, so that the size of the matrix to be calculated is reduced. Therefore, estimation of the partial precoding matrix does not require calculation cost as much as estimation of the whole precoding matrix. Therefore, the partial precoding matrix can be estimated even in an apparatus such as a user terminal that may have limited calculation capability.

Claims (13)

A communication device that uses a first plurality of antennas to communicate with a counterpart device that has a second plurality of antennas and uses only a part of them for transmission,
Each of the first plurality of antennas and the second plurality of antennas obtained by receiving signals transmitted from the part of the second plurality of antennas by each of the first plurality of antennas. Estimating means for estimating a first partial precoding matrix from a first channel matrix whose elements are the values of channels between each of said portions of antennas;
In a second channel matrix having a channel value as an element when a signal transmitted from each of the first plurality of antennas is received by at least one of the remaining portions of the second plurality of antennas. Receiving means for receiving information identifying the second partial precoding matrix estimated at the counterpart device based on the counterpart device;
Obtaining means for concatenating the first partial precoding matrix and the second partial precoding matrix to obtain a precoding matrix;
Transmitting means for transmitting each element of a transmission signal vector obtained by multiplying the transmission data vector including transmission data as an element by the precoding matrix from each of the first plurality of antennas;
A communication apparatus comprising: When the communication device communicates with a plurality of counterpart devices,
The estimation means is configured to obtain a signal transmitted from a part of the plurality of antennas for a part of the partner apparatuses having a plurality of antennas and using only a part of the plurality of antennas for transmission. The first channel having a value of a channel between each of the first plurality of antennas and the part of the plurality of antennas obtained by reception at each of the plurality of antennas as an element Estimating the first partial precoding matrix for each of the partial counterpart devices from a matrix;
The estimation means is further obtained by receiving a signal transmitted from at least one antenna with respect to the remaining counterpart devices of the plurality of counterpart devices by each of the first plurality of antennas. Estimating the first partial precoding matrix for each of the remaining counterpart devices from the first channel matrix having a channel value between each of the plurality of antennas and the at least one antenna as an element. ,
The receiving unit is configured to receive a signal transmitted from each of the some of the counterpart devices from each of the first plurality of antennas in at least one of the remaining part of the plurality of antennas. Receiving information identifying the second partial precoding matrix estimated in each of the partial counterpart devices based on the second channel matrix having the value of
The communication apparatus according to claim 1. Storage means for associating and storing precoding matrix candidates and indices estimated in the counterpart apparatus with respect to the second channel matrix;
The counterpart device knows the correspondence between the index and the candidate, and selects the second partial precoding matrix from the candidates;
The receiving means receives an index corresponding to one of the candidates selected as the second partial precoding matrix from the counterpart device as information specifying the second partial precoding matrix. ,
The communication apparatus according to claim 1. The second partial precoding matrix is a matrix that minimizes a difference between a matrix obtained by dividing a matrix product with the second channel matrix by the number of the first plurality of antennas and a unit matrix. Presumed,
The communication device according to any one of claims 1 to 3, wherein The second partial precoding matrix is such that a difference between a matrix obtained by dividing a matrix product with the second channel matrix by the number of the first plurality of antennas and a unit matrix is equal to or less than a predetermined value. Estimated as a matrix,
The communication device according to any one of claims 1 to 3, wherein A first plurality of antennas are used, and only a part of the first plurality of antennas is used for signal transmission, and the part of the first plurality of antennas and the rest are used for signal reception. A communication device that communicates with a counterpart device having a second plurality of antennas using at least one part of
A first channel matrix whose elements are channel values when signals transmitted from each of the second plurality of antennas are received by at least one of the remaining portions of the first plurality of antennas; An estimation means for estimating a first partial precoding matrix based on
Transmitting means for transmitting information identifying the first partial precoding matrix to the counterpart device;
Have
In the counterpart apparatus, a second channel whose element is a value of a channel when signals transmitted from each of the part of the first plurality of antennas are received by each of the second plurality of antennas Transmission of a result obtained by multiplying a precoding matrix obtained by concatenating the second partial precoding matrix obtained based on the matrix and the first partial precoding matrix on a transmission data vector including transmission data as an element Each of the elements of the signal vector is transmitted from each of the second plurality of antennas;
A communication device. Storage means for storing a precoding matrix candidate estimated with respect to the first channel matrix and an index in association with each other;
The association between the index and the candidate is also known to the counterpart device,
The estimation means selects the first partial precoding matrix from the candidates,
The transmission means transmits an index corresponding to the candidate selected as the first partial precoding matrix to the counterpart apparatus as information for specifying the first partial precoding matrix. The communication apparatus according to claim 6. The estimating means sets the first channel matrix as a matrix that minimizes a difference between a matrix obtained by dividing a matrix product with the first channel matrix by the number of the second plurality of antennas and a unit matrix. Estimate the partial precoding matrix,
The communication apparatus according to claim 6 or 7, wherein The estimating means is configured as a matrix such that a difference between a matrix obtained by dividing a matrix product with the first channel matrix by the number of the second plurality of antennas and a unit matrix is equal to or less than a predetermined value. Estimate a partial precoding matrix of 1;
The communication apparatus according to claim 6 or 7, wherein A method for controlling a communication device that uses a first plurality of antennas to communicate with a counterpart device that has a second plurality of antennas and uses only a part of them for transmission,
Each of the first plurality of antennas and the first plurality of antennas obtained by the estimation means receiving the signals transmitted from the part of the second plurality of antennas by each of the first plurality of antennas. An estimation step of estimating a first partial precoding matrix from a first channel matrix having a value of a channel between each of the two antennas as a part of the plurality of antennas;
A second means having a value of a channel when a signal transmitted from each of the first plurality of antennas is received by at least one of the remaining portions of the second plurality of antennas as an element; Receiving from the partner device information identifying the second partial precoding matrix estimated in the partner device based on the channel matrix of
An obtaining step for obtaining a precoding matrix by concatenating the first partial precoding matrix and the second partial precoding matrix;
A transmitting step of transmitting each element of a transmission signal vector obtained by multiplying the precoding matrix by a transmission data vector including transmission data as an element from each of the first plurality of antennas;
A control method characterized by comprising: A first plurality of antennas are used, and only a part of the first plurality of antennas is used for signal transmission, and the part of the first plurality of antennas and the rest are used for signal reception. A communication device control method for communicating with a counterpart device having a second plurality of antennas using at least one part of
The estimation means uses a value of a channel when a signal transmitted from each of the second plurality of antennas is received by at least one of the remaining portions of the first plurality of antennas as an element. An estimation step of estimating a first partial precoding matrix based on one channel matrix;
A transmitting step of transmitting information identifying the first partial precoding matrix to the counterpart device;
Have
In the counterpart apparatus, a second channel whose element is a value of a channel when signals transmitted from each of the part of the first plurality of antennas are received by each of the second plurality of antennas Transmission of a result obtained by multiplying a precoding matrix obtained by concatenating the second partial precoding matrix obtained based on the matrix and the first partial precoding matrix on a transmission data vector including transmission data as an element Each of the elements of the signal vector is transmitted from each of the second plurality of antennas;
A control method characterized by that. Using a first plurality of antennas, a computer provided in a communication device that has a second plurality of antennas and performs communication with a counterpart device that uses only a part thereof for transmission,
Each of the first plurality of antennas and the second plurality of antennas obtained by receiving signals transmitted from the part of the second plurality of antennas by each of the first plurality of antennas. An estimation step of estimating a first partial precoding matrix from a first channel matrix whose elements are the values of channels between each of the portions of the antenna;
In a second channel matrix having a channel value as an element when a signal transmitted from each of the first plurality of antennas is received by at least one of the remaining portions of the second plurality of antennas. Controlling the receiving means to receive from the counterpart device information identifying the second partial precoding matrix estimated in the counterpart device based on;
Obtaining a precoding matrix by concatenating the first partial precoding matrix and the second partial precoding matrix;
Controlling transmission means to transmit each element of a transmission signal vector obtained by multiplying the transmission data vector including transmission data as an element by the precoding matrix from each of the first plurality of antennas; ,
A program for running A first plurality of antennas are used, and only a part of the first plurality of antennas is used for signal transmission, and the part of the first plurality of antennas and the rest are used for signal reception. A computer provided in a communication device that performs communication with a partner device having a second plurality of antennas using at least one of a part of
A first channel matrix whose elements are channel values when signals transmitted from each of the second plurality of antennas are received by at least one of the remaining portions of the first plurality of antennas; An estimation step of estimating a first partial precoding matrix based on
Controlling transmission means to transmit information identifying the first partial precoding matrix to the counterpart device;
A program for executing
In the counterpart apparatus, a second channel whose element is a value of a channel when signals transmitted from each of the part of the first plurality of antennas are received by each of the second plurality of antennas Transmission of a result obtained by multiplying a precoding matrix obtained by concatenating the second partial precoding matrix obtained based on the matrix and the first partial precoding matrix on a transmission data vector including transmission data as an element Each of the elements of the signal vector is transmitted from each of the second plurality of antennas;
A program characterized by that.

Images