Docket No. PMDA-17044-PCT
1
DESCRIPTION
TRANSMITTING APPARATUS, RECEIVING APPARATUS, CONTROL
STATION, COMMUNICATION SYSTEM, AND TRANSMISSION PRECODING
5 METHOD
Field
[0001] The present invention relates to a transmitting apparatus, a receiving apparatus, a communication system, 10 and a transmission precoding method for performing multi¬user MIMO (Multiple-Input Multiple-Output) transmission.
Background [0002] In recent years, as a wireless communication
15 system that realizes high-speed transmission in a limited frequency band, a multi-user MIMO (MU (Multi-User)-MIMO) system obtained by applying a space division multiple access (SDMA) scheme to an MIMO system in which pluralities of antennas are set in both of a transmitter and a receiver
20 has been actively examined. In the MU-MIMO system, a
plurality of terminals including pluralities of antennas are present with respect to a base station including a plurality of antennas. The base station performs simultaneous transmission to the terminals in the same
25 wireless frequency band.
[0003] In communication in a downlink, that is, a direction from the base station to the terminal in the MU-MIMO system, signals are simultaneously transmitted from the base station to the terminals. Therefore, in general,
30 in a received signal in the terminal, signals to the other terminals are included in addition to a desired signal, which is a signal to the own terminal. That is, inter-user interference (IUI), which is interference caused by the

Docket No. PMDA-17044-PCT
2
signals to the other terminals occurs. IUI measures are desirably performed on the base station side, where limitations concerning a processing amount and the number of antennas are small compared with the terminals, as much 5 as possible. Therefore, in the downlink in the MU-MIMO system, the base station carries out processing called precoding as the IUI measures. The precoding indicates processing for forming a beam by weighting a plurality of signals transmitted from a plurality of antennas.
10 [0004] As a representative precoding method performed as the IUI measures in the downlink in the MU-MIMO system, block diagonalization (BD) method has been extensively examined. Please refer to, for example, Non Patent Literatures 1 and 2. The BD method is a precoding method
15 for forming a beam space to direct null to terminals other than a desired terminal, that is, form directivity for reducing received power in the terminals other than the desired terminal to 0. By applying the BD method to all the terminals, it is possible to realize a MU-MIMO system
20 in which the IUI does not occur. Consequently, it is possible to simplify processing and an apparatus configuration in the terminals.
Citation List
25 Non Patent Literature
[0005] Non Patent Literature 1: M. Rim, "Multi-user downlink beamforming with multiple transmit and receive antennas, "Electron. Lett., vol. 38, no. 25, pp. 1725-1726, Dec. 2002.
30 Non Patent Literature 2: L. U. Choi and R. D. Murch, "A Transmit Preprocessing Technique for Multiuser MIMO Systems Using a Decomposition Approach," IEEE Trans. On Wireless Commun., vol. 3, no. 1, pp. 20-24, Jan. 2004.

Docket No. PMDA-17044-PCT
3
Summary
Technical Problem
[0006] When a plurality of transmission antennas are 5 present, a transmission diversity effect is obtained. However, in the BD method, because null steering for directing null to the terminals other than the desired terminal is performed, a degree of freedom of a beam formed by a plurality of signals transmitted from the antennas of
10 the base station is lost. Therefore, in the precoding to which the BD method is applied, it is difficult to form a beam to increase the transmission diversity effect, that is, improve a received signal-to-nose power ratio (SNR: Signal-to-Noise power Ratio) of the terminals. In particular, in
15 an environment in which a larger number of terminals are present, the degree of freedom of the beam formation is greatly lost because of the null steering for the terminals. In this way, in the DB method, there is a problem in that it is difficult to improve a transmission diversity gain.
20 [0007] The present invention has been devised in view of the above, and an object of the present invention is to obtain a transmitting apparatus that can improve a transmission diversity gain compared with the BD method.
25 Solution to Problem
[0008] To solve the problems and achieve the object, a transmitting apparatus according to the present invention includes a plurality of transmission antennas capable of forming a plurality of beams respectively directed to a
30 plurality of receiving apparatuses. A receiving apparatus according to the present invention includes a precoder to perform precoding on signals transmitted from the transmission antennas such that received power in third

Docket No. PMDA-17044-PCT
4
receiving apparatuses, which are the receiving apparatuses excluding a first receiving apparatus serving as a transmission destination of a transmission signal among the receiving apparatuses and two or more second receiving 5 apparatuses among the receiving apparatuses, is equal to or smaller than a threshold.
Advantageous Effects of Invention
[0009] The transmitting apparatus according to the 10 present invention achieves an effect that it is possible to improve a transmission diversity gain compared with the BD method.
Brief Description of Drawings 15 [0010] FIG. 1 is a diagram showing a configuration example of a communication system according to a first embodiment.
FIG. 2 is a diagram showing a configuration example of a base station in the first embodiment. 20 FIG. 3 is a diagram showing a configuration example of a terminal in the first embodiment.
FIG. 4 is a diagram showing a configuration example of a processing circuit in the first embodiment.
FIG. 5 is a diagram showing a configuration example of 25 a control circuit in the first embodiment.
FIG. 6 is a diagram representing a flowchart showing an example of a processing procedure in a precoder unit in the first embodiment.
FIG. 7 is a diagram showing an average SNR during 30 precoding application with respect to an average SNR during precoding non-application in the case of the number of base station transmission antennas T=16, the number of terminals m=8, and the number of terminal reception branches Nw=2.

Docket No. PMDA-17044-PCT
5
FIG. 8 is a diagram representing a flowchart showing an example of a processing procedure of an ordering unit in the first embodiment.
FIG. 9 is a diagram showing a configuration example of 5 a base station not including the ordering unit.
FIG. 10 is a diagram showing the configuration of a control station and a base station controlled by the control station in a second embodiment.
FIG. 11 is a diagram showing a configuration example 10 of the base station in the second embodiment.
FIG. 12 is a diagram showing a configuration example of a base station in a third embodiment.
FIG. 13 is a diagram representing a flowchart showing an example of a processing procedure in a precoder unit in 15 the third embodiment.
Description of Embodiments
[0011] Transmitting apparatuses, receiving apparatuses,
communication systems, and transmission precoding methods
20 according to embodiments of the present invention are
explained below with reference to the drawings. Note that the present invention is not limited by the embodiments. [0012] First Embodiment.
FIG. 1 is a diagram showing a configuration example of
25 a communication system according to a first embodiment of the present invention. As shown in FIG. 1, the communication system in this embodiment includes a base station 1 and a terminal 2-1 to a terminal 2-m; m is an integer equal to or larger than 2. In the following
30 explanation, the terminals 2-1 to 2-m are sometimes
referred to as users. When the terminals 2-1 to 2-m are explained without being distinguished, the terminals 2-1 to 2-m are described as terminals 2. The base station 1

Docket No. PMDA-17044-PCT
6
includes a plurality of antennas. The terminals 2-1 to 2-m include one or more antennas.
[0013] In this embodiment, downlink communication, which is communication from the base station 1 to the user 5 terminals 2, is explained. Therefore, the base station 1 is a transmitting apparatus and the terminals 2 are receiving apparatuses. In the communication system in this embodiment, a MU-MIMO scheme is used in the downlink communication. The base station 1 carries out precoding on
10 transmission signals transmitted from the antennas and is
capable of forming beams directed to the terminals 2. Note that the base station 1 and the terminals 2 can perform communication in which the terminals 2 are transmitting apparatuses and the base station 1 is a receiving apparatus,
15 that is, uplink communication. An uplink communication method can be any communication scheme.
[0014] First, terms in this embodiment are explained. In the following explanation, physical transmission and reception antennas are referred to as "antennas" and an
20 array of a plurality of antennas included in one apparatus, that is, an antenna group is referred to as "array". Alternatively, a plurality of signal arrays corresponding to the array is sometimes also simply referred to as array for convenience. An array of a plurality of transmission
25 antennas is referred to as "transmission array". An array of a plurality of reception antennas is referred to as "reception array". An effective number of antennas observed when the transmission array or the reception array is multiplied with a weight matrix, which is a matrix
30 indicating weight, is referred to as "branch". The number of reception branches, which are branches on a reception side, is the number of data transmitted to the terminals 2, which are the receiving apparatuses, in parallel and is the

Docket No. PMDA-17044-PCT
7
number of rows of a reception weight matrix, which is a weight matrix multiplied in the terminals 2. The number of transmission branches, which are branches on a transmission side, is the number of rows of a transmission weight matrix, 5 which is a weight matrix multiplied in the base station 1, which is the transmitting apparatus, that is, transmission precoding.
[0015] There is no limitation on the number of antennas included in the terminals 2. The present invention is also
10 applicable when the number of antennas is different in each of the terminals 2 and when the number of reception branches is different in each of the terminals 2. However, to simplify the explanation, in the following explanation, the number of antennas included in the terminals 2 is R(R
15 is an integer equal to or larger than one) irrespective of the terminals. In the terminals 2, Nw (NW2), the number of transmission antennas and the number of reception branches satisfy a relation of t>NWftotai_

Docket No. PMDA-17044-PCT
8
LxNw= (m-L) xNw.
[0016] Subsequently, the downlink communication in the communication system in this embodiment, in which the MU-MIMO scheme is adopted, is modeled by a formula. A
5 transmission signal vector transmitted to a terminal 2-i (i=l, ..., and m) is represented as s (bold faced (td a transmission power distribution matrix, which is a matrix indicating power distribution to the terminal 2-i, is represented as P(bold faced, and a precoding matrix, that
0 is, a beam formation matrix corresponding to the terminal 2-i is represented as B(bold face)i. A true transmission line matrix of RxT from an antenna of the base station 1 to an antenna of the terminal 2-i is represented as H(bold face)(hat)i, a reception weight matrix of NwxR of the
5 terminal 2-i is represented as W(bold faced, and a true received signal vector before reception weight multiplication of the terminal 2-i is represented as y(bold face)i(t) . Further, a received signal vector after the reception weight multiplication of the terminal 2-i is
0 represented as r(bold faced(t) • A true reception thermal noise vector in the transmission line from the antenna of the base station 1 to the antenna of the terminal 2-i is represented as n (bold face) (hatd(t) . At this point, a system model obtained by modeling the communication system
5 in this embodiment according to a formula can be defined by the following Expression (1). [0017] [Math. 1]

Docket No. PMDA-17044-PCT
9

rx(t)

Wx

Yi(t)

rjt)_

w

Yjt)

W

Hx
H„

[Bx ••• BJ

0 "s^t)" + "n^tjT
/P7_ _sjt)_ A»(t)Jy

(1)
[0018] Further, NwxT matrix obtained by multiplying together the reception weight matrix W(bold face)i and the true transmission line matrix H(bold face)(hat)i is represented as a new transmission line matrix H(bold face)i and an Nw-th order vector obtained by multiplying together the true reception thermal noise vector n(bold face)(hat)i(t) and the reception weight matrix W(bold face)i is represented as a new reception thermal noise vector n(bold face)i(t) . Then, the system model can be represented by the following Expression (2). [0019]
[Math. 2]

rx(t)
rjt)

Hi
H„

[BI ••• BJ

0 "s^t)" + "iMtf
V?7_ _sm(t)_ njt)_

[2)

[0020] The above Expression (2) can be represented as the following Expression (3).
[0021]
[Math. 3]
r(t) = HBP¥(t) + n(t) (3)
[0022] In the expression, H(bold face)(bar) is a system transmission line matrix of NWf totaixT indicating a transmission line from the antenna of the base station 1 to all branches of all the terminals 2 after the multiplication of the reception weight. B(bold face)(bar)

Docket No. PMDA-17044-PCT
10

10
15

is a system precoding matrix of TxNst for all the terminals 2 in the base station 1. Note that Nst is a total number of signals simultaneously transmitted to all the terminals 2 in parallel. P(bold face)(bar) is a system transmission power matrix, which is a matrix that sets transmission power distribution to all the terminals 2, s(bold face) (bar) (t) is an Nst-th order system transmission vector indicating transmission signals to all the terminals 2, and n (bold face) (bar) (t) is an Nw, totai_th order system noise vector, which is a noise vector for all the terminals 2 after the reception weight multiplication. As indicated by the following Expression (4), a product of H(bold face)(bar) and B(boldface)(bar) can be grasped as an effective system transmission line matrix H(bold face)(bar)e by transmission beam formation. [0023]

[Math . 4]
~HA HiB2 . ■ ■ HA
He = H2B1 H2B2 "" H A
HA HA • II D
mm

20
25
30

[0024] A precoding method for using a precoding matrix for leaving only block diagonal terms, that is, components of H (bold) iB(bold face)i and setting non-block diagonal terms, which are the other components, to a zero matrix 0(bold face) in the effective system transmission line matrix H(bold face)(bar)e shown in Expression (4) is the BD method. In this embodiment, as explained below in detail, a precoding matrix for not setting all of the non-block diagonal terms as the zero matrix 0(bold face) and leaving a component of one terminal 2 other than the transmission target terminal 2 of a transmission signal as an interference component is used. Consequently, it is

Docket No. PMDA-17044-PCT
11
possible to secure a degree of freedom of beam formation by the transmission array and improve a diversity gain in the transmission target terminal 2 compared with the BD method while suppressing IUI. 5 [0025] FIG. 2 is a diagram showing a configuration
example of the base station 1 in this embodiment. The base station 1 includes primary modulating units 11-1 to 11-m, a precoder unit 12, an ordering unit 13, transmission-waveform shaping units 14-1 to 14-T, antennas 15-1 to 15-T,
10 and a receiver 16. A primary modulating units 11-i (i=l, ..., ) performs primary modulation on a transmission signal transmitted to the terminal 2-i and outputs the primarily-modulated transmission signal to the precoder unit 12. The primary modulation performed by the primary modulating unit
15 11-i includes, for example, channel coding and mapping to a primary modulation symbol such as a QAM (Quadrature Amplitude Modulation) symbol. When a single carrier block transmission scheme is used, the primary modulation performed by the primary modulating unit 11-i includes
20 discrete Fourier transform processing as well. The primary modulating units 11-1 to 11-m are signal generating units that generate, for each of the terminals 2, which are receiving apparatuses, a transmission signal transmitted to the terminal 2.
25 [0026] The precoder unit 12 is a precoder that performs precoding on signals transmitted from the antennas 15-1 to 15-T, which are a plurality of transmission antennas, such that received power in third receiving apparatuses, which are the terminals 2 excluding a first receiving apparatus,
30 which is the terminal 2 serving as a transmission
destination of transmission signals output from the primary modulating units 11-1 to 11-m among the terminals 2-1 to 2-m, and second receiving apparatuses, which are two or more

Docket No. PMDA-17044-PCT
12
terminals 2 other than the first receiving apparatus among the terminals 2-1 to 2-m, is 0, that is, equal to or smaller than a threshold. The first receiving apparatus is a desired terminal explained below. The second receiving 5 apparatuses are IUI terminals explained below. Received power in the IUI terminals, which are the second receiving apparatuses, is larger than the threshold. [0027] Specifically, the precoder unit 12 performs precoding by multiplying the transmission signals after the
10 primary modulation output from the primary modulating units 11-1 to 11-m with the precoding matrix in this embodiment and outputs the transmission signals after the precoding respectively to the transmission-waveform shaping units 14-1 to 14-T corresponding to the transmission signals. The
15 ordering unit 13 instructs the precoder unit 12 to perform ordering of the terminals 2 in the precoding and power distribution to the terminals 2. That is, the ordering unit 13 determines the order of the terminals 2 in the precoding. The transmission-waveform shaping units 14-1 to
20 14-T respectively perform secondary modulation, digital-analog (D/A) conversion, conversion from a baseband frequency into a radio frequency (RF), and the like on the signals after the precoding and transmit the signals after the processing respectively via the antennas 15-1 to 15-T.
25 For example, the secondary modulation is multicarrier modulation when a multicarrier scheme such as OFDM (Orthogonal Frequency Division Multiplex) is applied and is single carrier modulation when a single carrier scheme such as single carrier block transmission is applied. There is
30 no limitation on a modulation scheme of the secondary
modulation. Modulation other than the OFDM and the single carrier block transmission can be performed. When block transmission such as the OFDM or the single carrier block

Docket No. PMDA-17044-PCT
13
transmission is applied, the transmission-waveform shaping units 14-1 to 14-T perform, for example, inverse discrete Fourier transform and CP (Cyclic Prefix) addition processing before the D/A conversion. Note that the block 5 transmission indicates a scheme for converting a single into a block through the discrete Fourier transform processing and the CP addition as represented by the OFDM and the single carrier block transmission. The signal processing in the transmission-waveform shaping units 14-1
10 to 14-T can be digital processing or can be analog
processing. Note that the transmission signals input to the precoder unit 12 from the primary modulating units 11-1 to 11-m correspond to s(bold face)(bar)(t) in Expression (3) and the output signals output to the transmission-
15 waveform shaping units 14-1 to 14-T from the precoder unit 12 correspond to B(bold face) (bar)P(bold face) (bar)s(bold face)(bar)(t) in Expression (3).
[0028] Because the precoding by the precoder unit 12 is carried out, the antennas 15-1 to 15-T, which are the
20 transmission antennas, are capable of forming a plurality of beams respectively directed to the terminals 2. [0029] The receiver 16 carries out reception processing on received signals received from the terminals 2 through the antennas 15-1 to 15-T. Note that, in an example
25 explained here, the antennas 15-1 to 15-T are transmission and reception antennas. However, reception antennas can be provided separately from the antennas 15-1 to 15-T. However, when the base station 1 uses an estimation result of an uplink transmission line as downlink transmission
30 line information in a calculation process of a precoding matrix explained below, the antennas 15-1 to 15-T are transmission and reception antennas and the receiver 16 performs estimation of a transmission line on the basis of

Docket No. PMDA-17044-PCT
14
received signals received from the antennas 15-1 to 15-T. Any method can be used as an estimation method for a transmission line. For example, a method of estimating a transmission line using a pilot signal, which is a known 5 signal, can be used. Specifically, pilot signals, which are orthogonal among the antennas of the terminal 2, are transmitted from the terminal 2. The receiver 16 of the base station 1 can identify the antennas of the terminal 2 according to the orthogonal pilots and estimate a
10 transmission line. In the calculation process of the
precoding matrix explained below, when the base station 1 uses transmission line information received from the terminal 2, the receiver 16 outputs the received transmission line information to the precoder unit 12.
15 [0030] FIG. 3 is a diagram showing a configuration
example of the terminal 2 in this embodiment. The terminal 2 includes antennas 21-1 to 21-R, reception-waveform shaping units 22-1 to 22-R, a decoder unit 23, a demodulating unit 24, and a transmitter 25. The reception-
20 waveform shaping units 22-1 to 22-R respectively perform
processing for converting a radio frequency into a baseband frequency, analog-digital (A/D) conversion, signal filter processing, and the like on received signals received by the antennas 21-1 to 21-R and output the received signals
25 after the processing to the decoder unit 23. The signal
filter processing is processing for extracting, for example, a signal in a desired frequency band. When a block transmission scheme is applied, the reception-waveform shaping units 22-1 to 22-R carry out CP removal processing
30 and the discrete Fourier transform processing as well. The decoder unit 23 performs MIMO decode processing, which is processing for extracting a desired signal, that is, a signal to the own terminal, on the received signals input

Docket No. PMDA-17044-PCT
15
from the reception-waveform shaping units 22-1 to 22-R and outputs the signals after the processing to the demodulating unit 24. The decoder unit 23 is a decoder that extracts a desired signal from signals received from 5 the base station 1. The decoder unit 23 carries out
estimation processing for a transmission line in a process of the MIMO decode processing. The demodulating unit 24 performs demapping processing, channel decoding processing, and the like on the signals output from the decoder unit 23
10 and restores the signals transmitted from the base station 1. When the single carrier block transmission scheme is applied, the demodulating unit 24 carries out equalization processing for compensating for frequency distortion and the inverse discrete Fourier transform processing. The
15 signal processing in the reception-waveform shaping units 22-1 to 22-R can be digital processing or can be analog processing.
[0031] The transmitter 25 generates a transmission signal and transmits the transmission signal from the
20 antennas 21-1 to 21-R to the base station 1. Note that, in an example explained here, the antennas 21-1 to 21-R are transmission and reception antennas. However, transmission antennas can be provided separately from the antennas 21-1 to 21-R. However, in the calculation process of the
25 precoding matrix explained below, when the base station 1 uses transmission line information received from the terminal 2, the transmitter 25 acquires transmission line information, which is information concerning a transmission line estimated by the decoder unit 23, from the decoder
30 unit 23 and transmits the transmission line information to the base station 1. In the calculation process of the precoding matrix explained below, when the base station 1 uses an estimation result of an uplink transmission line as

Docket No. PMDA-17044-PCT
16
downlink transmission line information, the antennas 21-1 to 21-R are transmission and reception antennas. The transmitter 25 transmits transmission signals from the antennas 21-1 to 21-R. 5 [0032] Hardware configurations of the base station 1 and the terminal 2 in this embodiment are explained. The components configuring the base station 1 shown in FIG. 1 can be respectively realized as hardware such as electronic circuits and antennas. The primary modulating units 11-1
10 to 11-m are mappers or modulators. When the primary modulation includes the discrete Fourier transform processing, a discrete Fourier transform processing circuit is added. The precoder unit 12 is a processing circuit that carries out precoding. The ordering unit 13 is a
15 processing circuit that performs ordering. The
transmission-waveform shaping units 14-1 to 14-T are transmission-waveform shaping circuits and, specifically, configured by D/A converters, frequency converters, and the like. When the transmission-waveform shaping units 14-1 to
20 14-T perform the CP addition and the inverse discrete
Fourier transform processing, the transmission-waveform shaping units 14-1 to 14-T include CP addition circuits and inverse discrete Fourier transform processing circuits. [0033] The processing circuit that realizes the precoder
25 unit 12 and the ordering unit 13 can be dedicated hardware or can be a control circuit including a memory and a CPU (Central Processing Unit; also referred to as central processing device, processing device, arithmetic device, microprocessor, microcomputer, processor, or DSP (Digital
30 Signal Processor)) that execute programs stored in the memory. For example, a nonvolatile or volatile semiconductor memory such as a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM

Docket No. PMDA-17044-PCT
17
(Erasable Programmable Read Only Memory), or an EEPROM (Electrically Erasable Programmable Read Only Memory), a magnetic disk, a flexible disk, an optical disk, a compact disk, a minidisk, or a DVD (Digital Versatile Disk) 5 corresponds to the memory.
[0034] When the precoder unit 12 and the ordering unit 13 are realized by dedicated hardware, the precoder unit 12 and the ordering unit 13 are, for example, single circuits, composite circuits, programmed processors, parallel-
10 programmed processors, ASICs (Application Specific
Integrated Circuits), FPGAs (Field Programmable Gate Arrays), or combinations of the foregoing. When the processing circuit is realized by dedicated hardware, the processing circuit is, for example, a processing circuit
15 500 shown in FIG. 4.
[0035] when the precoder unit 12 and the ordering unit 13 are realized by a control circuit including a CPU, the control circuits is, for example, a control circuit 400 having a configuration shown in FIG. 5. As shown in FIG. 5,
20 the control circuit 400 includes a processor 401, which is a CPU, and a memory 402. When the precoder unit 12 and the ordering unit 13 are realized by the control circuit 400 as shown in FIG. 5), the precoder unit 12 and the ordering unit 13 are realized by the processor 401 reading out and
25 executing programs corresponding to respective kinds of
processing by the precoder unit 12 and the ordering unit 13 stored in the memory 402. The memory 402 is also used as a temporary memory in processing carried out by the processor 401.
30 [0036] At least a part of the primary modulating units 11-1 to 11-m and the transmission-waveform shaping units 14-1 to 14-T can be realized by the processing circuit, which is the dedicated hardware, or the control circuit 400

Docket No. PMDA-17044-PCT
18
like the precoder unit 12 and the ordering unit 13. [0037] The components configuring the terminal 2 shown in FIG. 2 can be respectively realized as hardware such as electronic circuits and antennas. The reception-waveform 5 shaping units 22-1 to 22-R are reception-waveform shaping circuits and, specifically, configured by A/D converters, filters, frequency converters, and the like. When the reception-waveform shaping units 22-1 to 22-R perform the CP removal and the discrete Fourier transform processing,
10 the reception-waveform shaping units 22-1 to 22-R include CP removal circuits and discrete Fourier transform processing circuits. The decoder unit 23 is a processing circuit. The demodulating unit 24 is a demodulator or a demapper. When the demodulating unit 24 performs the
15 equalization processing, the inverse discrete Fourier
transform processing, and the like, the demodulating unit 24 includes an equalizer, an inverse discrete Fourier transform circuit, and the like. [0038] The processing circuit that realizes the decoder
20 unit 23 can be realized by dedicated hardware or can be
realized by the control circuit 400. When the decoder unit 23 is realized by the control circuit 400 as shown in FIG. 5, the decoder unit 23 is realized by the processor 401 reading out and executing a program corresponding to the
25 processing by the decoder unit 23 stored in the memory 402. At least a part of the reception-waveform shaping units 22-1 to 22-R and the demodulating unit 24 can be realized by the processing circuit, which is the dedicated hardware, or the control circuit 400 like the decoder unit 23.
30 [0039] The precoding processing carried out by the precoder unit 12 in this embodiment is explained. The system model in the communication system in this embodiment is as explained with reference to Expression (1) and

Docket No. PMDA-17044-PCT
19
Expression (2). The precoder unit 12 generates a precoding matrix according to a procedure explained below. Note that, in the following explanation, processing by the transmission-waveform shaping units 14-1 to 14-T and the 5 reception-waveform shaping units 22-1 to 22-R is omitted for explanation in formulas. However, there is no influence of the processing in calculation of a precoding matrix. In the following explanation, a space between an output end of the precoder unit 12 of the base station 1
10 and an input end of the decoder unit 23 of the terminal is represented by an equivalent low-pass system. The precoding processing explained below can be independently carried out at each discrete frequency in the OFDM or the single carrier block transmission or can be collectively
15 carried out in an entire band irrespective of a frequency. [0040] In a process of the precoding matrix calculation explained below, information concerning a transmission line matrix in a downlink direction, that is, transmission line information is necessary. A method in which the precoder
20 unit 12 acquires a transmission line matrix is not
particularly limited. However, for example, when the communication system is a communication system that adopts frequency division duplex (FDD) in which communication is performed at different frequencies in a downlink and an
25 uplink, transmission line information estimated in the
terminal 2 and received from the terminal 2 is used. When the communication system is a communication system in which a downlink and an uplink are performed by time division duplex (TDD), reversibility of transmission and reception
30 can be used. Therefore, in this case, the receiver 16 can estimate a transmission line in an uplink direction on the basis of a signal received from the terminal 2 and use the estimated transmission line as transmission line

Docket No. PMDA-17044-PCT
20
information of the downlink. As a method for transmission line estimation, any method can be used as explained above. For example, a method of estimating a transmission line using a pilot signal can be used. 5 [0041] FIG. 6 is a flowchart showing an example of a processing procedure in the precoder unit 12 in this embodiment. In the following explanation, a terminal set as the terminal 2 at a destination of a transmission signal is referred to as desired terminal. First, the precoder
10 unit 12 determines the desired terminal according to order determined by the ordering unit 13. To calculate a precoding matrix for the desired terminal, the precoder unit 12 selects L terminals, which are the terminals 2 that allow IUI, that is, L IUI terminals, corresponding to the
15 desired terminal (step SI).
[0042] As explained above, in the BD method, for terminals other than the desired terminal, a beam is formed to form null to avoid IUI. On the other hand, in this embodiment, IUI is allowed for one terminal 2 other than
20 the desired terminal. That is, a beam is formed not to form null for the L terminals 2 other than the desired terminal and to have directivity in the directions of the L terminals 2 other than the desired terminal. [0043] As a selection method for an IUI terminal, when
25 the desired terminal is represented as a terminal 2-i,
there are, for example, a method of selecting a terminal including a transmission line matrix having a low correlation with H(bold faced, which is a transmission line matrix of the terminal 2-i, and a method of selecting
30 the terminal 2 in a position away from the desired terminal on the basis of geographical information of the terminals 2. The former is a method of selecting an IUI terminal corresponding to the desired terminal on the basis of a

Docket No. PMDA-17044-PCT
21
correlation between a transmission line matrix between the desired terminal and the base station 1 and a transmission line matrix between the terminal 2 other than the desired terminal and the base station 1. For example, when the 5 terminal 2 other than the desired terminal is represented as a terminal 2-k, the precoder unit 12 calculates, concerning all the terminals 2 other than k=i, a square sum of diagonal terms of a cross-correlation matrix H(bold face) kHH(bold face)i between H(bold face)i and a
10 transmission line matrix H(bold face)k of the terminal 2-k and selects the L terminals 2 in ascending order of square sums of the diagonal terms of H(bold face) kHH(bold face)i. The latter is a method of selecting the terminal 2 on the basis of geographical separation degrees between the
15 desired terminal and the terminals 2 other than the desired terminal. When the latter selection method is used, for example, the base station 1 calculates, on the basis of position information of the terminals 2 and position information of the base station 1, for each of the
20 terminals 2, an azimuth angle of the terminal 2 estimated from the base station 1 and selects the L terminals 2 in order from the terminal 2, the azimuth angle of which is most away from the desired terminal. The base station 1 acquires the position information of the terminal 2 by, for
25 example, receiving, from the terminal 2, position
information calculated by the terminals 2 using a GPS (Global Positioning System). As the position information of the base station 1, for example, the position information calculated using the GPS is used. Note that
30 the number of times a certain terminal 2 is selected as the IUI terminal in one precoding matrix calculation is L times at most. That is, in certain precoding, the same terminal cannot be redundantly selected as the IUI terminal (L+l)

Docket No. PMDA-17044-PCT
22
times or more.
[0044] Note that, as explained below, for example, when ordering is carried out by the ordering unit 13 such that a square sum of diagonal terms of the cross-correlation 5 matrix H(bold face) kHH (bold face)i between the terminal 2 and the adjacent, that is, following terminal 2 decreases, the IUI terminal can be selected by a method of grouping the desired terminal and the L IUI terminals in index order of the terminals.
10 [0045] In an example explained below, the number of IUI terminals, that is, L is set to two. However, not only this, but the present invention is applicable in the case of any number of IUI terminals L>2. [0046] Subsequently, when the IUI terminals are
15 represented as terminals 2-j and 2-k, the precoder unit 12 calculates a matrix H(bold face)(bar)i obtained by excluding transmission line components of the desired terminal and the two IUI terminals from a system transmission line matrix (step S2) and performs singular
20 value decomposition (SVD) of H(bold face)(bar)i (step S3). [0047] As explained above, a system transmission line matrix H(bold face) is a matrix indicating transmission lines from the antennas of the base station 1 to all the branches of all the terminals 2 after the multiplication of
25 the reception weight and can be calculated on the basis of a transmission line matrix of each of the terminals 2. A matrix H (bold faced, which is a (Nw, total- (L+l) xNw) xT matrix obtained by excluding the transmission line components of the desired terminal and the IUI terminals from the system
30 transmission line matrix H(bold face), can be represented by the following Expression (5). As indicated by Expression (5), the singular value decomposition of the H(bold face)(bar)i can be performed. The precoder unit 12

Docket No. PMDA-17044-PCT
23
performs the singular value decomposition of H(bold face) (bar)i (step S3) . [0048] [Math. 5]
Hi_i Hi+i

5 H,.

H,-i Hj+i

U^Vf = ujftf o]

Hi-1
Hi+1
[0049] In the expression, U(bold face)i is a left singular vector matrix of H(bold face)(bar)i, V(bold face)i is a right singular vector matrix, and Z(bold face)i is a singular value matrix having singular values in diagonal
0 terms. In Z (bold faced, when the singular values of the diagonal terms are in descending order of powers according to magnitudes, as indicated by Expression (5), the singular values can be represented by being divided into a partial diagonal matrix Z(bold face)i(s) configured by (Nw, total-
5 (L+l)xNw) non-zero singular values and a zero matrix 0(bold face) corresponding to (T- (Nw, total- (L+l) xNw) ) zero singular values. Right singular vectors V(bold face)i(s) and V(bold face)i(n) respectively corresponding to Z (bold face)i(s) and the zero matrix 0(bold face) are present. When V(bold
0 face)i(n), which is a first matrix, is a precoding matrix of the terminal 2-i, an effective transmission line matrix for the terminal 2-i can be represented by the following Expression (6). When the precoding matrix is used, null-steering is performed for the terminals 2 other than the

Docket No. PMDA-17044-PCT
24
terminal 2-i, the terminal 2-j, and the terminal 2-k. When the precoding is carried out using the first matrix, null is formed for the terminals 2 excluding the terminal 2-i and the terminal 2-j. The null indicates that received power of signals transmitted from the antennas 15-1 to 15-T is equal to or smaller than a threshold, for example, the received power is 0. [0050] [Math. 6]
H V-

H„

HV

H V-

iW
[0051] Referring back to FIG. 6, the precoder unit 12 calculates an eigenvector matrix V(bold face)i(e), which is a second matrix, from a desired component H(bold face) iV (bold face)i(n), which is a component corresponding to the terminal 2-i in Expression (6) (step S4). That is, the precoder unit 12 generates the second matrix for forming a beam space suitable for the terminal 2-i, that is directed to the terminal 2-i. Specifically, that is, the precoder unit 12 performs the singular value decomposition of H (bold face)iV(bold face)i(n) or applies the singular value decomposition to a non-negative value Hermitean
matrix (H(bold face)iV(bold face) i(n)) H (bold face)iV(bold

Docket No. PMDA-17044-PCT
25
face)i(n) of H (bold face)iV(bold face)i(n) and calculates an eigenvector matrix V(bold face)i(e) corresponding to a large eigenvalue. The large eigenvalue is an eigenvalue on a forward side at the time when a plurality of eigenvalues 5 are arranged in descending order of powers.
[0052] Subsequently, the precoder unit 12 calculates a precoding matrix corresponding to the desired terminal, that is, the terminal 2-i (step S5). Specifically, the precoder unit 12 calculates the precoding matrix
10 corresponding to the terminal 2-i according to the
following Expression (7). In this embodiment, as indicated by Expression (7), it is possible to realize beam formation for, by multiplying a transmission signal with V(bold face)i(n), after performing null-steering on a space
15 excluding the terminal 2-i, the terminal 2-j, and the
terminal 2-k, multiplying the space with V(bold face)i(e) to form signal spaces for the terminal 2-i, the terminal 2-j, and the terminal 2-k and then improve a reception gain in the terminal 2-i. That is, in this embodiment, signals
20 transmitted from the transmission antennas are multiplied with the first matrix, which is the precoding matrix for reducing received power in the terminals 2 excluding the desired terminal and the L IUI terminals to the threshold or less. A multiplication result is multiplied with the
25 second matrix, which is the precoding matrix for forming a beam directed to the desired terminal. [0053] [Math. 7]
B2 = Vfvf (7) 30 [0054] The precoder unit 12 determines whether the processing, that is, the calculation processing of the precoding matrix is ended for all the terminals 2 (step S6). When the processing is ended for all the terminals 2 (Yes

Docket No. PMDA-17044-PCT
26
at step S6), the precoder unit 12 calculates a system precoding matrix B(bold face) (bar) (step S7) and ends the processing. The system precoding matrix B (bold face)(bar) is a matrix in which precoding matrixes for each of the 5 terminals 2 are arranged in a column direction. When the processing is not ended for all the terminals 2 (No at step S6), the precoder unit 12 changes the desired terminal and returns to step SI. [0055] First, the precoder unit 12 multiplies
10 transmission signals output from the primary modulating
units 11-1 to 11-m with a power distribution matrix P(bold face)(bar) generated on the basis of power distribution notified from the ordering unit 13, multiplies a multiplication result with the system precoding matrix
15 B(bold face)(bar) calculated by the processing explained above, and outputs a multiplication result to the transmission-waveform shaping units 14-1 to 14-T. That is, the precoder unit 12 multiplies signals transmitted from the transmission antennas with a power distribution matrix
20 corresponding to a result of the power distribution and the system precoding matrix B(bold face)(bar), which is a precoding matrix for carrying out the precoding. As it is seen from Expressions (1), (2), and (3), the power distribution matrix is a matrix having, as a diagonal
25 element, a square root of electric power P(bold face)i
distributed to the terminal 2-i. The transmission-waveform shaping units 14-1 to 14-T perform the processing explained above and transmit the signals after the processing from the antennas 15-1 to 15-T.
30 [0056] For example, m=4. That is, when four terminals 2 are present, it is assumed that the precoder unit 12 selects, as IUI terminals of the desired terminal, the terminal 2 having the next index of the desired terminal

Docket No. PMDA-17044-PCT
27
and the terminal 2 having the next index of the terminal 2. That is, concerning beam formation to the terminal 2-1, interference to the terminal 2-2 and the terminal 2-3 is allowed, concerning beam formation to the terminal 2-2, 5 interference to the terminal 2-3 and the terminal 2-4 is allowed, concerning beam formation to the terminal 2-3, interference to the terminal 2-4 and the terminal 2-1 is allowed, and, concerning beam formation to the terminal 2-4, interference to the terminal 2-1 and the terminal 2-2 is
10 allowed. In this case, an effective system transmission line matrix to which the system precoding matrix B(bold face) (bar) in this embodiment is applied can be represented by the following Expression (8). [0057]
15 [Math. 8]
[0058] Concerning an average received signal power-to-noise power ratio (Signal-to-Noise power Ratio: SNR) in the desired terminal, an improvement effect by the precoding is
20 quantitatively indicated from a simulation result. FIG. 7 shows an average SNR during precoding application with respect to an average SNR during precoding non-application in the case of the number of base station transmission antennas T=16, the number of terminals m=8, and the number
25 of terminal reception branches Nw=2. Therefore, a system
transmission line matrix is a matrix of 16x16. Elements in the system transmission line matrix are independent and similar complex Gaussian random numbers. The number of times of random number trials is set to 10000 times. A
30 broken line in the figure indicates a characteristic

Docket No. PMDA-17044-PCT
28
(conventional) at the time when the precoding by the conventional BD method is applied. A solid line indicates a characteristic (invented) at the time when the precoding in this embodiment is applied. It is seen from FIG. 7 that 5 the precoding in the embodiment of the present invention can improve an average SNR more than the conventional BD method. For example, when compared in a case in which an SNR during precoding non-application is 20 decibels, the average SNR can be improved by 10 decibels with respect to
10 the conventional BD method by applying the precoding in the embodiment of the present invention. This is because a diversity effect is obtained by forming a beam directed to the desired terminal as explained above. [0059] Processing by the ordering unit 13 is explained.
15 To enable the precoder unit 12 to easily select the IUI
terminal, arrangement order of the terminals 2 is important. The ordering unit 13 determines the arrangement order of the terminal 2. The ordering unit 13 determines power distribution to the terminals 2.
20 [0060] FIG. 8 is a flowchart showing an example of a processing procedure of the ordering unit 13 in this embodiment. The ordering unit 13 determines the order of the terminals 2 (step Sll). The ordering unit 13 notifies the determined order to the precoder unit 12. Examples of
25 a method of ordering include a method of ordering the terminals 2 in the order of transmission line gains (squares of a Frobenius norm of a transmission line matrix) of the terminals 2, a method of ordering the terminals 2 in descending order or ascending order of non-negative
30 eigenvalues or non-negative singular values of transmission line matrixes between the base station 1 and the terminals 2, a method of ordering the terminals 2 such that geographical positions of adjacent terminals, for example,

Docket No. PMDA-17044-PCT
29
azimuth angles viewed from the base station 1 are close to each other or different from each other, and a method of ordering the terminals 2 such that a correlation between transmission line matrixes of the adjacent terminals, that 5 is, the magnitude of diagonal terms of a cross correlation matrix of a transmission line matrix between the terminals increases or decreases. However, the method of ordering is not limited to these methods. [0061] The order of the ordering by the ordering unit 13
10 can be set to order decided such that the terminal 2 in the next order of the desired terminal is the IUI terminal corresponding to the desired terminal. Examples of such ordering include a method of ordering the terminals 2 such that geographical positions of the adjacent terminals are
15 close to each other or distant from each other, that is,
ordering the terminals 2 such that the terminals continuous in order are geometrically close to each other or separated from each other and a method of ordering the terminals 2 such that a correlation between transmission line matrixes
20 of the adjacent terminals increases or decreases, that is, ordering the terminals 2 such that a correlation between transmission line matrixes of the terminals 2 continuous in order increases or decreases. When the terminals 2 are ordered such that the azimuth angles viewed from the base
25 station 1 are different from each other, for example, the
terminals 2 corresponding to indexes are selected in such a manner as to select any terminal 2 as a first terminal 2, select, as a second terminal 2, a terminal that is in a most distant geographical position from the first terminal
30 2, and select, as a third terminal 2, the terminal 2 that is in a most distant geographical position from the second terminal 2 and is not ordered. [0062] The ordering unit 13 determines power

Docket No. PMDA-17044-PCT
30
distribution to the terminals 2 (step S12). The ordering unit 13 notifies a result of the power distribution, that is, electric power distributed to the terminals 2 to the precoder unit 12. For example, the precoder unit 12 5 carries out the power distribution according to a water
filling principle on the basis of a transmission line gain of the terminals 2. Alternatively, the precoder unit 12 carries out the power distribution to equalize reception quality of all the terminals 2, that is, such that a
10 product of the transmission line gain and the distributed power is an equivalent value among all the terminals 2. However, the power distribution is not limited to the above. Note that the order of step Sll and step S12 can be opposite.
15 [0063] Processing in the decoder unit 23 of the terminal 2 is explained. A transmission line component observed by the terminal 2, which is the receiving apparatus that receives a beam formed by the base station 1 using the system precoding matrix in this embodiment explained above,
20 is considered. As explained above, it is assumed that the number of IUI terminals L is two. It is assumed that, in the terminal 2-i, a signal transmitted to a terminal 2-p and a signal transmitted to a terminal 2-q by the base station 1 are observed as interference signals. That is,
25 it is assumed that, in the base station 1, the terminal 2-i is selected as an IUI of the terminal 2-k. At this point, a received signal received by the terminal 2-i includes a desired transmission line component H(bold face) iB(bold face)i, an interfered component H(bold face)iB(bold face)p
30 by the signal transmitted to the terminal 2-p, and an interfered component H(bold face) iB(bold face)q by the signal transmitted to the terminal 2-q. Therefore, a received signal r(bold face)i(t) received by the terminal

Docket No. PMDA-17044-PCT
31
2-i can be represented by the following Expression (9). In the expression, s(bold face)i(t) is a transmission signal transmitted from the base station 1 to the terminal 2-i and s(bold face)k(t) is a transmission signal transmitted from 5 the base station 1 to the terminal 2-k. [0064] [Math. 9]
r2(t) = H.B.VpTs^t) + H2BpA/p7sp(t) + H2Bq^sq(t) + n2(t) (9)
[0065] The decoder unit 23 of the terminal 2 detects,
10 from the received signal r(bold face)i(t), the transmission signal s(bold face)i(t) transmitted to the terminal 2-i. The detection of the transmission signal s(bold face)i(t) from the received signal r(bold face)i(t) can be realized by general MIMO decode processing. For example, as
15 described in "T. Ohgane, T. Nishimura, and Y. Ogawa,
"Applications of Space Division Multiplexing and Those Performance in a MIMO Channel, "IEICE Trans. Commun., vol. E88-B, no. 5, pp. 1843-1851, May 2005.", a linear detection method represented by ZF (Zero-Forcing) and a minimum mean
20 square error (MMSE) standard can be applied. Alternatively, a nonlinear detection method represented by maximum likelihood estimation or an interference canceller (IC) can also be applied. Any MIMO decode processing can be used. Note that the MIMO decode processing performed by the
25 decoder unit 23 can be carried out on y(bold face)i(t)
before reception weight multiplication instead of being carried out on r(bold face)i(t) after the reception weight multiplication. MIMO decode processing in this case is also the same as the general MIMO decode processing.
30 [0066] In the above explanation, the number of antennas T of the base station 1 and the number of reception branches Nw of the terminal 2 satisfy the relation of

Docket No. PMDA-17044-PCT
32
T>NWftotal_NW= (m-L) xNw. However, there is no limitation on the number of antennas included in the terminal 2. The present invention is also applicable when the number of antennas is different for each of the terminals 2 and when 5 the number of reception branches is different for each of the terminals 2. For example, when the number of IUI terminals is set to L=2, the number of antennas NRfi and the number of branches NWfi of a terminal 2-1 satisfy a relation of NRfi>NWfi, and the IUI terminals for the terminal 2-i,
10 which is the desired terminal, are the terminal 2-j and the terminal 2-k, the present invention is applicable if T> (Zi=im (NWfi) )-NWf j-NWfk is satisfied in a relation between the base station 1 and all desired terminals. [0067] In FIG. 2 referred to above, the example is shown
15 in which the base station 1 includes the ordering unit 13. However, the base station 1 can include a configuration shown in FIG. 9 without including the ordering unit 13. FIG. 9 is a diagram showing a configuration example of a base station la not including the ordering unit 13. In FIG.
20 9, components having the same functions as the functions of the base station 1 shown in FIG. 2 are denoted by the same reference numerals and signs as those in the base station 1 shown in FIG. 2. In the base station la shown in FIG. 2, rearrangement by the ordering unit 13 is not performed.
25 However, the precoder unit 12 can select two or more IUI
terminals according to the selection method explained above and carry out the operation explained above. Consequently, in the base station la shown in FIG. 9, it is possible to form a beam that forms null for the terminals 2 other than
30 the desired terminal and the IUI terminals as in the beam formation by the base station 1 shown in FIG. 2. [0068] As explained above, in this embodiment, the base

Docket No. PMDA-17044-PCT
33
station 1 decides, for each of the terminals 2, two or more IUI terminals that allow interference and forms a beam that forms null for the terminals 2 other than the desired terminal and the IUI terminals. Therefore, it is possible 5 to improve a diversity gain in the transmission target terminal 2 compared with the BD method while suppressing IUI. [0069] Second Embodiment.
FIG. 10 is a diagram showing the configuration of a
10 control station 3 and base stations lb-1 to lb-z controlled by the control station 3 in a second embodiment according to the present invention. In the figure, z is an integer equal to or larger than two. In the first embodiment, the example is explained in which beams are formed by antennas
15 15-1 to 15-T mounted on the base station 1. However, not only this, but, when T antennas are distributedly mounted on a plurality of base stations, the same system precoding matrix as the system precoding matrix in the first embodiment can be used. When the base stations lb-1 to lb-
20 z are explained without being distinguished, the base
stations lb-1 to lb-z are described as base stations lb. In this embodiment, a total numbers of the numbers of antennas included in the base stations lb-1 to lb-z is T. [0070] The control station 3 shown in FIG. 10 includes a
25 precoder calculation unit 31, an ordering unit 32, and a transceiver 33. The precoder calculation unit 31 carries out the same processing as the processing by the precoder unit 12 in the first embodiment. That is, the precoder calculation unit 31 calculates a precoding matrix for
30 performing precoding such that received power in the
terminals 2 excluding desired terminals, which are the terminals 2 at transmission destinations of transmission signals transmitted by the base stations lb-1 to lb-z, and

Docket No. PMDA-17044-PCT
34
IUI terminals, which are the terminals 2 other than the desired terminals, is equal to or smaller than a threshold. However, the precoder calculation unit 31 receives transmission line information used for calculation of a 5 system precoding matrix from the base stations lb-1 to lb-z through the transceiver 33. A method in which the base stations lb-1 to lb-z acquire the transmission line information is the same as the method in the first embodiment. The ordering unit 32 carries out the same
10 processing as the processing by the ordering unit 13 in the first embodiment. The transceiver 33 performs reception processing for signals received from the base stations lb-1 to lb-z and transmission processing for signals transmitted to the base stations lb-1 to lb-z. The transceiver 33
15 transmits the system precoding matrix, which is the
precoding matrix calculated by the precoder calculation unit 31, and power distribution calculated by the ordering unit 32 respectively to the base stations lb-1 to lb-z. Each of the base stations lb-1 to lb-z includes one or more
20 transmission antennas.
[0071] FIG. 11 is a diagram showing a configuration example of the base station lb in this embodiment. As shown in FIG. 11, the base station lb is the same as the base station 1 in the first embodiment except that a
25 transceiver 17 is added to the base station 1 in the first embodiment and the base station lb includes a precoder unit 12a instead of the precoder unit 12. However, the number of the transmission-waveform shaping units and the number of antennas are respectively c; c is an integer equal to or
30 larger than one. Components having the same functions as the functions in the first embodiment are denoted by the same reference numerals and signs as those in the first embodiment and redundant explanation is omitted.

Docket No. PMDA-17044-PCT
35
[0072] The transceiver 17 performs reception processing for a signal received from the control station 3 and transmission processing for a signal to be transmitted to the control station 3. The transceiver 17 acquires 5 transmission line information from the receiver 16 and
transmits the transmission line information to the control station 3. The transceiver 17 outputs a system precoding matrix and power distribution received from the control station 3 to the precoder unit 12a. The precoder unit 12a
10 multiplies transmission signals output from the primary modulating units 11-1 to 11-m with a power distribution matrix P(bold face)i generated on the basis of the power distribution received from the transceiver 17, further multiplies the transmission signals with the system
15 precoding matrix B(bold face) (bar) received from the
transceiver, 17 and outputs a multiplication result to transmission-waveform shaping units 14-1 to 14-c. [0073] Hardware configurations of the control station 3 and the base station lb are explained. The same components
2 0 as the components in the first embodiment among components of the base station lb can be realized by the hardware configuration explained in the first embodiment. The precoder calculation unit 31 and the ordering unit 32 of the control station 3 are processing circuits. Like the
25 processing circuit that realizes the precoder unit 12 and the ordering unit 13 in the first embodiment, the precoder calculation unit 31 and the ordering unit 32 can be dedicated hardware or can be a control circuit including a memory and a CPU that executes a program stored in the
30 memory. The control circuit that realizes the precoder calculation unit 31 and the ordering unit 32 is, for example, the control circuit 400 shown in FIG. 5. The precoder unit 12a is also a processing circuit. The

Docket No. PMDA-17044-PCT
36
processing circuit can be dedicated hardware or can be a control circuit including a memory and a CPU that executes a program stored in the memory. The control circuit that realizes the precoder unit 12a is, for example, the control 5 circuit 400 shown in FIG. 5.
[0074] The transceiver 33 of the control station 3 is configured by a transmitter and a receiver. The transceiver 17 of the base station lb is also configured by a transmitter and a receiver.
10 [0075] As explained above, in this embodiment, the control station 3 calculates the same system precoding matrix B(bold face)(bar) as the system precoding matrix in the first embodiment and notifies the system precoding matrix B(bold face) (bar) to the base station lb. Therefore,
15 even when the communication system includes a plurality of base stations lb, it is possible to obtain the same effects as the effects in the first embodiment. [0076] Third Embodiment.
FIG. 12 is a diagram showing a configuration example
20 of a base station lc in a third embodiment according to the present invention. The base station lc in this embodiment is the same as the base station 1 in the first embodiment except that the precoder unit 12 of the base station 1 in the first embodiment is replaced with a precoder unit 12b
25 and a nonlinear processing unit 18 is added. The terminals 2-1 to 2-m in this embodiment are the same as the terminals 2-1 to 2-m in the first embodiment. Components having the same functions as the functions in the first embodiment are denoted by the same reference numerals and signs as those
30 in the first embodiment and redundant explanation is omitted.
[0077] In this embodiment, in the terminal 2-1 to the terminal 2-(m-L), as in the first embodiment, one IUI

Docket No. PMDA-17044-PCT
37
terminal is selected for each of desired terminals and a precoding matrix is calculated. L is the number of IUI terminals explained in the first embodiment. Note that it is assumed that branch numbers of the signs of the terminal 5 2-1 to the terminal 2-(m-l) indicate order after ordering by the ordering unit 13 and the ordering unit 13 performs rearrangement such that the terminals 2 can be selected as the IUI terminal in order from the next index of the desired terminal. For example, as explained in the first
10 embodiment, it is assumed that the terminals 2 are ordered such that geographical positions of terminals continuous in order are away or a correlation of transmission line matrixes decreases. In this embodiment, concerning a terminal 2-(m-L+l) to the terminal 2-m, which are L
15 terminals 2 from the last, the number of IUI terminals is set to be smaller than L, that is, the number of IUI terminals is reduced to be smaller than the number of IUI terminals of the terminals from the terminal 2-1 to the terminal 2-(m-L).
20 [0078] FIG. 13 is a flowchart showing an example of a processing procedure in the precoder unit 12b in this embodiment. The precoder unit 12b selects a desired terminal as in the first embodiment and determines whether the desired terminal is the terminals 2-(m-L+l) to the
25 terminal 2-m (step S20). When the desired terminal is not an m-th terminal 2 (No at step S20), the precoder unit 12b proceeds to step SI. Step SI to step S7 are the same as the steps in the first embodiment. [0079] When the desired terminal is the terminal 2-(m-
30 L+l) to the terminal 2-m (Yes at step S20), the precoder unit 12b selects the IUI terminals fewer than L corresponding to the terminal 2-i (i>m-L+l), that is, smaller in number than L terminals (step S21). Concerning

Docket No. PMDA-17044-PCT
38
the terminal 2-i(i>m-L+l), the precoder unit 12b calculates a matrix H(bold face)±< (bar) obtained by excluding a desired transmission line component from the system transmission line matrix indicated by Expression (9) (step
5 S22). As indicated by Expression (10), the following singular value decomposition of the matrix H(bold face) i-(bar) can be performed. The precoder unit 12b performs the singular value decomposition of the matrix H (bold face) i'(bar) (step S23).
0 [0080]
[0081] In the expression, U(bold face) i- is a left singular vector matrix of H(bold face)±< (bar) , V(bold face)i- is a right singular vector matrix, and Z (bold
5 face)i- is a singular value matrix having singular values in diagonal terms. In Z (bold face)i, when the singular values of the diagonal terms are in descending order of powers according to magnitudes, as indicated by Expression (10), the singular values can be represented by being
0 divided into a partial diagonal matrix Z (bold face)i'(s) configured by (Nw, totai_Nw) non-zero singular values and a zero matrix 0 (bold face) corresponding to (T- (Nw, totai_Nw) ) zero singular values. Right singular vectors V(bold face)i'(s) and V(bold face)i'(n) respectively corresponding to
5 Z (bold face)m(s) and the zero matrix 0 (bold face) are
present. When V(bold face)i'(n) is a precoding matrix of the terminal 2-i, an effective transmission line matrix for the terminal 2-i can be represented by the following Expression (11). When the precoding matrix is used, null-

Docket No. PMDA-17044-PCT
39
steering is performed for the terminals 2 other than the terminal 2-i to the terminal 2-m. [0082] [Math. 11]
[0083] The precoder unit 12b calculates V(bold face)i> and thereafter forms a beam space suitable for the terminal 2-i from a component corresponding to the terminal 2-i (i>m-L+l), which is the desired terminal in Expression (11),
0 that is, the desired component H (bold face) iV (bold face)i'(n). That is, the precoder unit 12b applies SVD to H(bold face)iV(bold face)i'(n), applies the singular value decomposition to a non-negative value Hermitean matrix (H(bold face)iV(bold f ace) ±.(n)) HH (bold face)iV(bold face)i.(n),
5 and calculates an eigenvector matrix V(bold face)i'(e)
corresponding to a large eigenvalue (step S24). The large eigenvalue is an eigenvalue equal to or larger than a threshold. It is possible to realize beam formation for multiplying a transmission signal with V(bold face)i'(e)
0 after null steering by V(bold face)i'(n) to form signal
spaces for only the terminal 2-i to the terminal 2-m and then improve a reception gain in the terminal 2-i. Consequently, a precoding matrix for the terminal 2-i in this embodiment can be represented by the following
5 Expression (12).
[0084] Referring back to FIG. 13, the precoder unit 12b calculates, concerning i=(m-L+l) to m, a precoding matrix

Docket No. PMDA-17044-PCT
40
of the terminal 2-i on the basis of Expression (12) (step S25) and proceeds to step S6. [0085] [Math. 12]
5 Bj = V^'VJf (12)
[0086] As explained above, in the case of im-L+l, the precoding matrix for realizing the null steering for the terminals
10 excluding the terminal 2-i to the terminal 2-m, that is,
the terminal 2-1 to the terminal 2-(i-l) is calculated and applied to the system transmission line. Then, an effective system transmission line matrix indicated by Expression (13) is observed. Note that, in an illustration
15 of Expression (13), m=6, L-2, and i=5. [0087] [Math. 13]
[0088] In this embodiment, at step S7, the precoder unit 20 12b generates a system precoding matrix using the precoding matrix calculated at step S5 as the precoding matrix of the terminals 2-1 to 2-(m-L) and using the precoding matrix calculated at step S25 as the precoding matrix of the terminal 2-(m-L+l) to the terminal 2-m. That is, when K is 25 an integer equal to or larger than two (K=m-L) and L is an integer equal to or larger than two, a precoding matrix for setting first to K-th terminals 2 as desired terminals among terminals ordered by the ordering unit 13 is a matrix

Docket No. PMDA-17044-PCT
41
for performing the precoding such that received power for fourth receiving apparatuses, which are the terminals 2 excluding the desired terminals and the L IUI terminals, is equal to or smaller than a threshold. On the other hand, a 5 precoding matrix for setting K+l-th and subsequent
terminals 2 as desired terminals is a matrix for performing the precoding such that received power in third receiving apparatuses, which are the terminals 2 excluding the desired terminals and the IUI terminals fewer than L, is
10 equal to or smaller than the threshold. Note that the number smaller than L includes zero as well. [0089] As it is seen from Expression (13), unlike Expression (8) in the first embodiment, the effective system transmission line matrix in this embodiment is
15 subjected to tridiagonalizaton under blocks. That is,
layering is realized in which a component corresponding to a desired terminal is present in diagonal components and components of IUI terminals are present under the diagonal components, that is, in second and third layers.
20 Consequently, it is possible to apply nonlinear MU-MIMO
processing for performing sequential interference removal on a transmission side as explained below. [0090] The precoder unit 12b outputs the system precoding matrix B(bold face)(bar) calculated by the
25 processing explained above, transmission signals output
from the primary modulating units 11-1 to 11-m, and power distribution to the nonlinear processing unit 18. The nonlinear processing unit 18 carries out processing for removing, on a transmission side, in advance, components to
30 be an interference signal on a reception side as explained below using the tridiagonalizaton under blocks. [0091] The nonlinear processing unit 18 carries out the nonlinear MU-MIMO processing on a signal output from the

Docket No. PMDA-17044-PCT
42
precoder unit 12b. According to Expression (13), when the signal output from the precoder unit 12b is received by the terminal 2-i, the received signal can be represented by the following Expression (14). 5 [0092]
[Math. 14]
r2(t) = HiBi>/^'si(t) + H/ E1 BJ7^sJ(t)| + n2(t) (14)
>vj = max(i-Ari,l) v J
[0093] Note that a second term on the right side of Expression (14) is an interference term. In the case of
10 the terminal 2-1, the term is not generated. Therefore, a transmission signal s(bold face)i(t) is known. Making use of this, transmission signals are sequentially calculated to calculate, for example, s(bold face)2(t) . If transmission signals s (bold face)i_L(t), , and s (bold
15 face)i-i(t) for a terminal 2-(i-L) to a terminal 2-(i-l) are known, it is possible to remove interference on the reception side by correcting s(bold face)i(t) to a signal given by Expression (15). [0094]
20 [Math. 15]
[0095] Therefore, the nonlinear processing unit 18 corrects s(bold face)i(t) according to the above Expression (15). The terminal 2-1 is not set as an IUI terminal.
25 That is, IUI does not occur in a received signal of the terminal 2-1. Therefore, it is unnecessary to apply the correction to the transmission signal s(bold)i(t) to the terminal 2-1. Therefore, by sequentially determining transmission signals to set s(bold)i(t) as known, correct
30 s(bold)2(t) , and correct s(bold)3(t) using s(bold)2(t) after

Docket No. PMDA-17044-PCT
43
the correction, it is possible to remove, on the transmission side, that is, in the base station lc, in advance, IUI that occurs on the reception side. That is, the nonlinear processing unit 18 is an interference 5 removing unit that performs sequential interference removal for sequentially determining transmission signals from a transmission signal of the terminal 2 not set as an IUI terminal and removing interference. In this embodiment, the processing for sequentially determining transmission
10 signals from a transmission signal of the terminal 2 not set as an IUI terminal and removing interference as explained above is referred to as sequential interference removal. By applying the system precoding matrix in this embodiment, unlike the general nonlinear MU-MIMO processing
15 in which the number of transmission interference
cancellations increases in proportion to the number of terminals, it is possible to limit the number of interference cancellations to only the number of interference cancellations for the L terminals. Therefore,
20 compared with the general non-linear MU-MIMO processing, it is possible to suppress deterioration due to a computation amount reduction and a signal subtraction. [0096] The non-linear processing unit 18 applies the sequential interference removal to transmission signals,
25 thereafter multiplies the transmission signals with a power distribution matrix P(bold face)i generated on the basis of power distribution, further multiplies the transmission signals with the system precoding matrix B(bold face)(bar) calculated by the processing explained above, and transmits
30 the transmission signals to the transmission-waveform shaping units 14-1 to 14-T.
[0097] However, an actually transmitted signal is expanded or reduced to be unstable because of transmission

Docket No. PMDA-17044-PCT
44
processing given by Expression (15). Therefore, the nonlinear processing unit 18 can apply processing for stabilizing a transmission signal waveform according to a modulo operation disclosed in "H. Harashima and H. Miyakawa, 5 'Matched-Transmission Technique for Channels With
Intersymbol Interference,' IEEE Trans. Commun., vol. 20, Aug. 1972." or perturbation processing disclosed in "B. M. Hochwald, C. B. Peel, and A. L. Swindlehurst, 'A Vector-Perturbation Technique for Near-Capacity Multiantenna
10 Multiuser Communication-Part II: Perturbation,' IEEE Trans. Commun., vol. 53, no. 3, pp. 537-544, March 2005." [0098] A hardware configuration of the base station lc is explained. The same components as the components in the first embodiment among components of the base station lc
15 can be realized by the hardware configuration explained in the first embodiment. The precoder unit 12b and the nonlinear processing unit 18 in this embodiment are processing circuits. The precoder unit 12b and the nonlinear processing unit 18 can be dedicated hardware or
20 can be a control circuit including a memory and a CPU that executes a program stored in the memory. The control circuit that realizes the precoder unit 12b and the nonlinear processing unit 18 is, for example, the control circuit 400 shown in FIG. 5.
25 [0099] As explained above, in this embodiment, in the
terminals 2-1 to 2-(m-L), the L IUI terminals are set as in the first embodiment. In the terminals 2-(m-L+l) to 2-m, IUI terminals are set to less than L terminals and the system precoding matrix is generated. Transmission signals
30 are corrected to remove, on the transmission side, in
advance, interference that occurs on the reception side by the nonlinear processing unit 18. Therefore, the same effects as the effects in the first embodiment are obtained.

Docket No. PMDA-17044-PCT
45
It is possible to layer a multiuser space and it is possible to realize a nonlinear MU-MIMO scheme in which deterioration due to a computation amount reduction and a signal subtraction is suppressed.
[0100] The configurations explained in the embodiments above indicate examples of the contents of the present invention. The configurations can be combined with other publicly-known technologies. A part of the configurations can be omitted or changed in a range not departing from the spirit of the present invention.
Reference Signs List [0101] 1, la, lb-1 to lb-z, lc base station
2-1 to 2-m terminal
3 control station
11-1 to 11-m primary modulating unit
12, 12a, 12b precoder unit
13, 32 ordering unit
14-1 to 14-T, 14-c transmission-waveform shaping unit
15-1 to 15-T, 15-c, 21-1 to 21-R antenna
16 receiver
17, 33 transceiver
18 nonlinear processing unit
22-1 to 22-R reception-waveform shaping unit
23 decoder unit
24 demodulating unit
25 transmitter
31 precoder calculation unit.

Docket No. PMDA-17044-PCT
46
CLAIMS 1. A transmitting apparatus comprising:
a plurality of transmission antennas capable of forming a plurality of beams respectively directed to a 5 plurality of receiving apparatuses; and
a precoder to perform precoding on signals transmitted from the transmission antennas such that received power in third receiving apparatuses, which are the receiving apparatuses excluding a first receiving apparatus serving 10 as a transmission destination of a transmission signal among the receiving apparatuses and two or more second receiving apparatuses among the receiving apparatuses, is equal to or smaller than a threshold.
15 2. The transmitting apparatus according to claim 1,
wherein received power in the second receiving apparatuses is larger than the threshold.
3. The transmitting apparatus according to claim 1 or 2,
20 wherein signals transmitted from the transmission antennas
are multiplied with a first matrix, which is a precoding matrix for reducing the received power in the third receiving apparatuses to the threshold or less, and a multiplication result is multiplied with a second matrix, 25 which is a precoding matrix for forming a beam directed to the first receiving apparatus.
4. The transmitting apparatus according to claim 1, 2, or
3, wherein the second receiving apparatus corresponding to
30 the first receiving apparatus is selected on the basis of a correlation between a transmission line matrix between the first receiving apparatus and the transmitting apparatus and transmission line matrixes between the receiving

Docket No. PMDA-17044-PCT
47
apparatuses other than the first receiving apparatus and the transmitting apparatus.
5. The transmitting apparatus according to claim 1, 2, or
5 3, wherein the second receiving apparatus corresponding to
the first receiving apparatus is selected on the basis of geographical separation degrees between the first receiving apparatus and the receiving apparatuses other than the first receiving apparatus. 10
6. The transmitting apparatus according to any one of
claims 1 to 5, further comprising an ordering unit to
determine order of the receiving apparatuses in the
precoding.
15
7. The transmitting apparatus according to claim 6,
wherein
the ordering unit carries out power distribution to the receiving apparatuses, and 20 the precoder multiplies signals transmitted from the transmission antennas with a power distribution matrix corresponding to a result of the power distribution and a precoding matrix for carrying out the precoding.
25 8. The transmitting apparatus according to claim 7,
wherein the power distribution is carried out according to a water filling principle.
9. The transmitting apparatus according to claim 7,
30 wherein the power distribution is power distribution for
equalizing reception quality in the receiving apparatuses.
10. The transmitting apparatus according to claim 6,

Docket No. PMDA-17044-PCT
48
wherein the order is order decided such that the receiving apparatus next to the first receiving apparatus in order is the second receiving apparatus corresponding to the first receiving apparatus. 5
11. The transmitting apparatus according to claim 10,
wherein the ordering unit performs the ordering such that
the receiving apparatuses continuous in order are
geographically close to or separated from each other.
10
12. The transmitting apparatus according to claim 10,
wherein the ordering unit performs the ordering such that a
correlation between transmission line matrixes of the
receiving apparatuses continuous in order increases or
15 decreases.
13. The transmitting apparatus according to claim 10, 11,
or 12, further comprising an interference removing unit to
perform sequential interference removal, wherein
20 when K is an integer equal to or larger than two and L is an integer equal to or larger than two, a precoding matrix for setting first to K-th receiving apparatuses among the receiving apparatuses ordered by the ordering unit as the first receiving apparatus is a matrix for
25 performing the precoding such that the received power in the third receiving apparatus is equal to or smaller than the threshold and a precoding matrix for setting K+l-th and subsequent receiving apparatuses among the receiving apparatuses as the first receiving apparatus is a matrix
30 for performing the precoding such that received power in fourth receiving apparatuses, which are the receiving apparatuses excluding the first receiving apparatus and the second receiving apparatuses fewer than L, and

Docket No. PMDA-17044-PCT
49
the interference removing unit sequentially carries out, as the sequential interference removal, processing for removing interference that occurs in the receiving apparatuses next in order. 5
14. The transmitting apparatus according to any one of
claims 6 to 9, wherein the ordering unit sets the order to
order of transmission line gains between the transmitting
apparatus and the receiving apparatuses.
10
15. The transmitting apparatus according to any one of
claims 6 to 9, wherein the ordering unit sets the order to
descending order of non-negative eigenvalues or non-
negative singular values of transmission line matrixes
15 between the transmitting apparatus and the receiving apparatuses.
16. The transmitting apparatus according to any one of
claims 6 to 9, wherein the ordering unit sets the order to
20 ascending order of non-negative eigenvalues or non-negative singular values of transmission line matrixes between the transmitting apparatus and the receiving apparatuses.
17. A receiving apparatus that receives signals
25 transmitted from the transmitting apparatus according to any one of claims 1 to 16, the receiving apparatus comprising a decoder to extract a desired signal from the signals received from the transmitting apparatus.
30 18. The receiving apparatus according to claim 17, wherein the decoder extracts the desired signal according to linear detection by a Zero-Forcing method.

Docket No. PMDA-17044-PCT
50
19. The receiving apparatus according to claim 17, wherein the decoder extracts the desired signal according to linear detection based on a minimum mean square error standard.
5 20. The receiving apparatus according to claim 17, wherein the decoder extracts the desired signal according to nonlinear detection by maximum likelihood estimation.
21. The receiving apparatus according to claim 17, wherein
10 the decoder extracts the desired signal according to
nonlinear detection by an interference canceller.
22. A control station in a communication system capable of
forming a plurality of beams respectively directed to a
15 plurality of receiving apparatuses by a plurality of transmission antennas mounted on a plurality of transmitting apparatuses, the control station comprising: a precoder calculation unit to calculate a precoding matrix for performing precoding such that received power in
20 third receiving apparatuses, which are the receiving
apparatuses excluding a first receiving apparatus serving as a transmission destination of a transmission signal among the receiving apparatuses and two or more second receiving apparatuses among the receiving apparatuses, is
25 equal to or smaller than a threshold, and
a transceiver to transmit the precoding matrix to the transmitting apparatuses.
23. A communication system comprising:
30 the transmitting apparatus according to any one of claims 1 to 16; and
the receiving apparatus according to any one of claims 17 to 21.

Docket No. PMDA-17044-PCT
51
24. A communication system comprising:
a control station; and
a plurality of transmitting apparatuses,
5 the communication system being capable of forming a
plurality of beams respectively directed to a plurality of
receiving apparatuses by a plurality of transmission
antennas mounted on the transmitting apparatuses, wherein
the control station includes:
10 a precoder calculation unit to calculate a
precoding matrix for performing precoding such that received power in third receiving apparatuses, which are the receiving apparatuses excluding a first receiving apparatus serving as a transmission destination of a 15 transmission signal among the receiving apparatuses and two or more second receiving apparatuses among the receiving apparatuses, is equal to or smaller than a threshold, and
a transceiver to transmit the precoding matrix to the transmitting apparatuses, and 20 the transmitting apparatus includes:
a transceiver to receive the precoding matrix; and
a precoder to perform precoding using the precoding matrix received by the transceiver. 25
25. A transmission precoding method in a transmitting
apparatus including a plurality of transmission antennas
capable of forming a plurality of beams respectively
directed to a plurality of receiving apparatuses, the
30 transmission precoding method comprising:
a first step of determining a first receiving apparatus, which is the receiving apparatus serving as a transmission destination of a transmission signal, and two

Docket No. PMDA-17044-PCT
52
or more second receiving apparatuses, which are the receiving apparatuses other than the first receiving apparatus; and
a second step of performing precoding on signals 5 transmitted from the transmission antennas such that
received power in the receiving apparatuses excluding the first receiving apparatus and the second receiving apparatuses is equal to or smaller than a threshold.