JP2006340265A - Radio communication device and radio communication method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provode a radio communication method easily performing a transformation processing for a transmitting signal on a transmitter station side in a MIMO (Multiple-Input Multiple-Output) transmission, and to improve receiving characteristics on a receiving side. <P>SOLUTION: A channel estimating circuit 7 acquires a transport function matrix H from a receiving state of a preamble signal that is a known pattern. A base vector generating circuit 14 generates a base vector constituting a partial space including row vector groups (h<SB>1</SB>, h<SB>2</SB>) constituting the transport function matrix H and generates a transformation matrix U using an Hermite-conjugate column vector. A transmitting signal transforming circuit 4 operates the transformation matrix U upon a signal corresponding to two transmission systems to transform the column vector of two rows into the column vector of four rows. Each of radio units 5-1 to 5-4 transmits the transmitting signal transformed into the column vector of four rows via each of antennas 6-1 to 6-4. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  In the present invention, signals of a plurality of different signal systems are transmitted using a plurality of transmission antennas on the same frequency channel, signals are received using a plurality of reception antennas, and a known signal assigned to each signal system is transmitted. In a high-speed wireless access system that realizes wireless communication by separating and demodulating each signal sequence on the receiving side using a signal pattern as a reference signal, a transmission diversity gain is obtained on the base station side, and reception characteristics on the wireless terminal side are increased. More particularly, the present invention relates to a wireless communication method for increasing the transmission speed of a high-speed wireless access system using a 2.4 GHz band or a 5 GHz band, and a wireless communication apparatus using the method.

  In recent years, the IEEE802.11g standard, the IEEE802.11a standard, and the like are remarkable as high-speed wireless access systems using the 2.4 GHz band or the 5 GHz band. In these systems, a maximum transmission rate of 54 Mbps is realized. However, with the spread of wireless LAN, further increase in transmission rate is required.

  As a technology for that purpose, MIMO (Multiple-Input Multiple-Output) technology is effective. This MIMO technology is such that different independent signals are transmitted on the same channel from a plurality of transmitting antennas on the transmitting station side, and signals are received using the same plurality of antennas on the receiving station side, between each transmitting antenna / receiving antenna. The transfer function matrix is obtained, the independent signal transmitted from each antenna on the transmitting station side is estimated on the receiving station side using this matrix, and the data is reproduced.

Here, consider a case in which N signals are transmitted using N transmission antennas and signals are received using M antennas. First, there are N × M transmission paths between the antennas of the transmitting and receiving stations, and the transfer function when transmitted from the i-th transmitting antenna and received by the j-th receiving antenna is defined as h _{j, i.} The matrix of M rows and N columns as the (j, i) component is denoted as H. Further, a transmission signal from the i-th transmitting antenna is denoted by t _i , a column vector having components (t ₁ , t ₂ , t ₃ ,..., T _N ) as Tx, and a received signal at the j-th receiving antenna as r _j. A column vector whose components are (r ₁ , r ₂ , r ₃ ,..., R _M ) is Rx, and the thermal noise of the j-th receiving antenna is n _j (n ₁ , n ₂ , n ₃ ,..., N _M ). A column vector whose component is is denoted by n. In this case, the following relational expression holds.

Therefore, there is a need for a technique for estimating the transmission signal Tx based on the signal Rx received on the receiving station side. As the most basic of the MIMO technology, there is a method generally called a ZF (Zero Forcing) method. Here, an inverse matrix H ⁻¹ of the transfer function matrix is obtained with respect to the above formula (1), and this is multiplied from the left of both sides of the formula. As a result, the following expression is obtained.

That is, when the signals received by the respective receiving antennas are combined and processing for removing interference caused by signals from other than the desired transmitting antenna is performed, a minute thermal noise term H ⁻¹ × n is added to the actual transmitted signal vector Tx. Signal points are obtained. Here, when a signal subjected to multi-level modulation such as BPSK, QPSK, 16QAM, and 64QAM is used as a transmission signal, signal points that can be taken as the transmission signal are discontinuous. Accordingly, performs soft decision or hard decision (H ^-1 Search × Rx and Euclidean distance to the nearest point on the transmission constellation) processes for H ^-1 × Rx, estimates the true transmission signal.

  In the ZF method described above, an inverse matrix of the transfer function matrix H is used in order to separate signals by removing interference between sequences of each signal sequence. This is only for the purpose of removing the mutual interference between the signal sequences, and does not maximize the diversity gain by combining the receiving antennas. Originally, it is preferable to perform signal separation processing with the policy of maximizing the S / N ratio (signal-to-noise level ratio) of each signal sequence after signal separation. In order to maximize this S / N ratio, that is, to realize the maximum ratio combining process on the receiving side, there is a method in which the transmitting side adjusts and transmits signals so that the signal sequences are orthogonal at the receiving end. Conceivable. This method is called an E-SDM (Eigenbeam-Space Division Multiplexing) method (see, for example, Non-Patent Document 1).

In this E-SDM system, it is assumed that the transfer function matrix H is known on the transmission side. Based on this matrix H, a product of Hermitian conjugate matrixes H ^H and H of this matrix, that is, a unitary matrix U for diagonalizing an N × N matrix H ^H × H is obtained, and a unitary matrix U is obtained for the transmission signal vector Tx. The converted signal U × Tx is transmitted. If a unitarily transformed signal including a preamble signal is transmitted, even if the actual transfer function matrix is H on the receiving station side, the transfer function matrix obtained by channel estimation is H × U. Here, the following relational expressions hold for the unitary transformation matrix U and the matrix H.

The matrix Λ on the right side of Equation (3) is an N × N diagonal matrix, and the diagonal components of the matrix are non-zero (each component is called an eigenvalue, and λ ₁ , λ ₂ , λ ₃ ... Other off-diagonal components are zero). When this method is used, Equation (1) is also converted as follows.

  When a unitary conjugate matrix of transfer function matrix H × U estimated from the left of both sides is multiplied, Equation (4) is converted as follows.

  This means that the N signal sequences are completely orthogonal. Further, the matrix Λ appearing on the right side can be directly obtained from the estimated transfer function matrix by Equation (3). For this reason, it is possible to stably estimate the transmission signal Tx using the formulas (3) and (5) regardless of whether the original matrix H has an inverse matrix or not.

In the above description, the matrix U is basically assumed to be a square matrix, but may be a non-square matrix. For example, consider a case where two transmission signal systems are spatially multiplexed using four transmission antennas, and reception processing is performed using two reception antennas on the reception side. In this case, the transfer function matrix H is a 2 × 4 matrix, and H ^H × H is a 4 × 4 matrix.

  However, since the transfer function matrix H is originally a matrix having only two rows, the number of non-zero eigenvalues is two. As a result, it is possible to obtain two eigenvectors corresponding to this eigenvalue, but the remaining two eigenvectors in the four-dimensional vector are indefinite. The intention is that the remaining two vectors need to be orthogonal to the two eigenvectors, but if they are two orthogonal vectors belonging to a subspace that is orthogonal to the subspace spanned by the two eigenvectors Any vector can be chosen.

  Even if a signal is transmitted in that direction, since the eigenvalue is zero, the reception side cannot receive the signal at all. In such a case, the unitary transformation matrix U is treated as a 4-by-2 matrix consisting only of eigenvectors corresponding to non-zero eigenvectors, and only the non-zero eigenvalues with the right side of Equation (3) being 2 rows and 2 columns are extracted. It is also possible to treat it as a diagonal matrix. In Equation (4), two signal systems are virtually converted into four systems by the conversion matrix U, signals are transmitted from the four antennas, and the signal is transmitted via the MIMO channel characterized by the transfer function matrix H. Thus, the signal is received by the receiving antenna. In the following description, a matrix obtained by extracting only eigenvectors corresponding to nonzero eigenvalues from a unitary matrix that is a square matrix will be referred to as a unitary matrix (unitary transformation) for the sake of convenience without particular distinction.

  As a supplement, it can also be understood that the E-SDM system is equivalent to the phased array antenna technology that gives the transmission signal directivity by converting the signal on the transmission side. As a result, it is possible to obtain a diversity gain while avoiding interference between the transmitting antennas, so that very good characteristics can be expected.

  In the above description, the processing on the receiving station side is assumed to correspond to Equation (5). However, if the preamble signal itself also performs unitary transformation, the channel estimation result of the MIMO channel performed at the receiving station is not the original transfer function matrix H but the transfer function as H × U which is a MIMO channel including unitary transformation. A matrix is obtained. In this case, it can be obtained by applying an inverse matrix of H × U to Equation (4) as shown in Equation (2).

  In other words, the expression (6) means that the signal detection process on the receiving side is the same as the expression (wherever the preamble signal and the data part are processed) regardless of whether or not the transmission side performs unitary conversion. It is not necessary to be based on 5), and in addition to the ZF method described above, an MLD (Maximum Liklihood Detection) method, an MMSE (Minimum Mean Square Error) method, or the like may be used. As described above, the effect of maximum ratio combining cannot be expected with the ZF method alone. However, when unitary conversion is performed on a transmission signal by the E-SDM method on the transmission side, the transmission signal is orthogonalized. As a result, there is an effect that the gain of the maximum ratio synthesis can be obtained.

  In this E-SDM system, the number of transmission signal systems (the number of spatially multiplexed signal systems) does not necessarily need to match the number of transmission antennas.

  Hereinafter, an example of the configuration of the transmitting station and the receiving station in the above-described conventional method will be described with reference to the drawings. FIG. 10 is a block diagram showing a configuration of a transmission unit of a transmission station in the prior art. First, as an example, a case where a transmitting station transmits two systems of data using four transmission antennas will be described as an example. For convenience of explanation, the transmission function and the reception function are described separately. However, in general, a wireless communication apparatus having a transmission / reception function is used. In the control on the transmitting station side, it is necessary to acquire the transfer function matrix of the MIMO channel from the transmitting station to the receiving station. However, the acquisition method may be any general method. Here, a case will be described as an example where a signal in the direction of the transmitting station is received from the receiving station, and a transfer function matrix in the direction of the receiving station is estimated therefrom.

  In FIG. 10, 1 is a data division circuit, 2-1 to 2-2 are preamble assignment circuits, 3-1 to 3-2 are modulation circuits, 4 is a transmission signal conversion circuit, and 5-1 to 5-4 are radio units. 6-1 to 6-4 are antennas, 7 is a channel estimation circuit, 8 is a transfer function matrix management circuit, 9 is a matrix operation circuit, and 10 is an eigenvector generation circuit.

  First, the antennas 6-1 to 6-4 and the radio units 5-1 to 5-4 are used for transmission and reception. When the signal to which the preamble signal for estimating the transfer function matrix of the MIMO channel is received via the antennas 6-1 to 6-4 and the radio units 5-1 to 5-4, the channel estimation circuit 7 A transfer function matrix is acquired from the reception state of the preamble signal which is a pattern.

  In the MIMO channel, the transfer function matrix from the transmitting station to the receiving station and the transfer function matrix from the receiving station to the transmitting station are in a relationship in which the matrix is transposed for the part depending on the propagation space and the transmission / reception antenna. However, due to the characteristics of the radio unit in transmission / reception, there are cases where it is actually necessary to correct by multiplying by a predetermined coefficient in addition to transposition.

The channel estimation circuit 7 also performs correction processing as necessary, and outputs the result to the transfer function matrix management circuit 8. The transfer function matrix management circuit 8 manages this information, and the matrix calculation circuit 9 calculates the Hermitian conjugate matrix H ^H of the transfer function matrix and the respective products H ^H · H. The eigenvector generation circuit 10 obtains ^HH · H eigenvectors. Here, the transfer function matrix H is a matrix of 2 rows and 4 columns, and H ^H · H is 4 rows and 4 columns regardless of the number of antennas on the receiving side.

  Therefore, the eigenvector generation circuit 10 generates two eigenvectors corresponding to a large eigenvalue among the four eigenvalues. Further, two generated eigenvectors (column vectors) are arranged, and a 4-by-2 conversion matrix U is generated in advance. At the time of data transmission, the data input to the data division circuit 1 is divided into the number of systems of signals to be spatially multiplexed (here, two systems), and the preamble providing circuits 2-1 to 2-2 for each system. A preamble signal having a known pattern is applied, and predetermined modulation is performed by the modulation circuits 3-1 to 3-2.

  The transmission signal conversion circuit 4 receives the previous conversion matrix U, regards the signals corresponding to the two transmission systems as a column vector having two components, and applies the conversion matrix U to the signal to be transmitted. Convert from a column vector of rows to a column vector of 4 rows. The respective signals are sent to the radio units 5-1 to 5-4, and after being converted to radio frequencies, signal amplification, etc., are output via the antennas 6-1 to 6-4.

  FIG. 11 is a block diagram showing a configuration of a receiving unit of a receiving station in the prior art. Here, a case where the number of receiving antennas is two will be described as an example. In the figure, 121-1 to 121-2 are receiving antennas, 122-1 to 122-2 are radio units, 123 is a channel estimation circuit, 124 is a received signal management circuit, 125 is a transfer function matrix management circuit, and 126 is signal detection. A circuit 127 indicates a data synthesis circuit.

  Signals received by the receiving antennas 121-1 to 121-2 are subjected to processing such as signal amplification and frequency conversion by the radio units 122-1 to 122-2 independently in each receiving system. The MIMO channel estimation signal added to the head portion of the received signal is cut out and input to the channel estimation circuit 123. The channel estimation circuit 123 acquires MIMO channel transfer function information from a known signal reception state.

  In general, in the case of a wireless communication apparatus having a transmission function in addition to a reception function, this channel estimation circuit 123 can be shared with the channel estimation circuit 7 shown in FIG. The transfer function information estimated here is input to the transfer function matrix management circuit 125 and managed as a transfer function matrix.

  On the other hand, the signal of the data part following the preamble signal is input to the received signal management circuit 124. In the signal detection circuit 126, a signal transmitted from the transmission station side in units of symbols from the reception signal of each signal system managed by the reception signal management circuit 124 and the transfer function matrix managed by the transfer function matrix management circuit 125. Is estimated. The estimated result is output to the data synthesizing circuit 127, and the original data is reproduced by synthesizing the signal systems and output. As the processing in the signal detection circuit 126, any processing such as Equation (2), Equation (5), Equation (6) described above may be used.

  Next, FIG. 12 is a block diagram illustrating a configuration example of a wireless communication apparatus having functions of both a transmitting station and a receiving station in the prior art. Note that portions corresponding to those in FIG. 10 are denoted by the same reference numerals and description thereof is omitted. In the figure, 11 is a received signal management circuit, 12 is a signal detection circuit, and 13 is a data synthesis circuit. The basic operation is a process combining FIG. 10 and FIG. 11, and the channel estimation circuit 7, the channel estimation circuit 123, the transfer function matrix management circuit 108, and the transfer function matrix management circuit 125 perform transmission and reception. The only difference is that they are shared by both systems.

  FIG. 13 is a flowchart showing processing at the time of signal transmission in the prior art. When data is input (S101), the transmitting station divides the data into N data series (S102), and each of these signals is assigned a preamble signal (S103), and is modulated individually for each series. Processing is performed (S104). The modulated signal is subjected to unitary conversion (S105), and the signal after unitary conversion is converted into a radio frequency by the radio unit and transmitted (S106). If the transmission data continues (S107), Steps S104 to S106 are repeated, and after the transmission data ends (S107), the process is completed (S108).

Next, FIG. 14 is a flowchart showing processing for generating a unitary transformation matrix in the conventional method. When the transmitting station acquires information on the transfer function matrix by some means (S111), a product H ^H · H of the transfer function matrix H and its Hermitian conjugate matrix is obtained (S112), and an eigenvector of H ^H · H is obtained ( In step S113, a unitary transformation matrix including the eigenvector as a component is generated (S114). This unitary transformation matrix is temporarily stored in the memory (S115), and the process ends (S116). This stored unitary transformation matrix is used for unitary transformation in step S105 shown in FIG. 13 during signal transmission.

Next, FIG. 15 is a flowchart showing the receiving operation of the receiving station in the prior art. When receiving the wireless packet (S121), the receiving station detects the preamble (S122) and performs channel estimation (S123). Here, all transfer functions between the transmission antennas and the reception antennas are acquired. On the other hand, the signal received subsequent to the preamble signal is managed as a received signal vector Rx having the received signal r _j at each receiving antenna as a component for each symbol (S124). On the other hand, the transmission signal of each transmission system is estimated from the relationship of Rx≈H × Tx (S125). If the received data continues (NO in S126), the process returns to step S124, and steps S124 to S125 are repeated. When the received data is completed (YES in S126), the series of received data of each system is reconstructed, the data on the transmission side is reproduced and output (S127), and the process is completed (S128).

  Next, FIG. 16 is a flowchart showing processing at the time of signal transmission in the prior art using the OFDM modulation scheme. The steps corresponding to those in FIG. 13 are given the same reference numerals. Although FIG. 13 has been described on the premise of a single carrier, when the OFDM modulation method is used, processing from step S102 to step S103 and step S104 to step S105 are individually performed for a plurality of subcarriers. become. The difference from FIG. 13 is that data is divided for each subcarrier in step S131, IFFT processing is performed on the signal of each subcarrier in step S132, and a guard interval is inserted in step S133. It is a point to do.

Next, FIG. 17 is a flowchart showing processing at the time of signal reception in the prior art using the OFDM modulation method. The steps corresponding to FIG. 15 are denoted by the same reference numerals. In FIG. 15, the description has been given on the assumption of a single carrier. However, when the OFDM modulation method is used, Steps S122 to S123 and Steps S124 to S125 are individually processed for a plurality of subcarriers. become. The difference from FIG. 15 is that the guard interval in the received signal is removed in step S141 and step S143, the fast Fourier transform is performed on the received signal in step S142 and step S144, and the signal is received for each subcarrier. It is a point to separate.
Miyashita et al. "Eigenbeam space division multiplexing (E-SDM) system in MIMO channel", IEICE Tech., RCS2002-53, May 2002.

By the way, in the above-described E-SDM system, the signal of each signal sequence is separated on the transmission side so as to be separated into subchannels corresponding to eigenvalues λ _{1 to} λ _N for the square matrix H ^H × H of N rows and N columns. Process. Therefore, an eigenvector corresponding to each eigenvalue is obtained, and the unitary transformation described above is performed by combining the eigenvectors. If the number of transmitting antennas is up to 2, eigenvalues and eigenvectors can be obtained by performing deterministic processing using a formula, but especially when the number of antennas is 3 or more, The operation for obtaining eigenvalues and eigenvectors is not easy.

Assuming computation on a computer, computation for obtaining eigenvalues and eigenvectors of the matrix M can be obtained by the following means, for example. First, taking N = 3 as an example, a matrix M of 3 rows and 3 columns, and three eigenvalues λ ₁ , λ ₂ , λ ₃ (where λλ ₁ ‖ ≧ ‖λ ₂ ‖ ≧ ‖λ ₃ ‖, x‖ is the absolute value of the complex number x), and the respective eigenvectors are denoted as u ₁ , u ₂ , u ₃ . An arbitrary three-dimensional vector v can be expressed by a linear combination of these eigenvectors (coefficients are [A ₁ , A ₂ , A ₃ ]).

  When the matrix M is applied to this, the following expression is obtained.

  Therefore, when M is acted k times, it becomes as follows.

Here, if ‖λ ₁ ‖> ‖λ ₂ ‖> ‖λ ₃ ‖, an approximation of M ^k · v≈A ₁ · λ ₁ ^k · u ₁ is possible if k⇒∞. When this is sequentially used, each eigenvalue and eigenvector can be approximately obtained.

  However, such an operation requires multiplying a vector by a matrix a plurality of times, and the amount of calculation becomes enormous with processing delay. This impact may be negligible on a computer, but if it is necessary to perform processing in real time, it has many individual circuits for performing parallel processing to reduce delay, It is necessary to perform operations in parallel. In particular, the complex multiplier circuit is generally large in circuit scale, and considering the circuit scale that can be practically implemented, it is very difficult to obtain an eigenvector (or a unitary transformation matrix obtained by synthesizing it) by the above-described method.

  In particular, in the case of two-way communication, when a signal from a communication partner station is received and immediately returned to the station, processing in a short time is essential. If it takes time from acquisition of the transfer function matrix to generation of the eigenvector, unitary transformation must be performed using past information for at least that time, but since the transfer function matrix varies with time, the accuracy is increased accordingly. Must be used, and as a result, the characteristics on the receiving side are degraded.

  As described above, if any conversion processing depending on the transfer function matrix is performed on the transmission signal on the transmission side, the generation of the conversion matrix is configured only by deterministic processing, and as much as possible. A simple process is preferred.

  The present invention has been made in view of such circumstances, and an object of the present invention is to easily perform conversion processing on a transmission signal on the transmission station side in MIMO transmission, and to achieve reception characteristics on the reception side. An object of the present invention is to provide a wireless communication apparatus and a wireless communication method that can improve the performance.

In order to solve the above-described problem, the present invention provides a plurality of integer M's with M ≧ 2, N with N ≧ 3, and N ′ with N> N ′ ≧ 2 on the same frequency channel. A transmission station equipped with N transmission antennas that multiplex and transmit signal sequences in space, and M or more receptions that receive transmitted radio signals and perform reception processing by separating them into the plurality of signal sequences A wireless communication apparatus configured to perform communication using a MIMO channel between a plurality of antennas between the transmission and reception stations, wherein the transmission station includes N transmission antennas of the transmission station; Channel estimation means for acquiring each transfer function between M receiving antennas of the receiving station, and transfer function matrix generating means for generating a transfer function matrix of the MIMO channel having the acquired transfer function as each element And the entered user Data dividing means for dividing the data into N ′ systems, and a transmission signal sequence for generating a first signal sequence of the N ′ system by giving a signal of an individual known pattern to the data divided into the N ′ system M, N M-dimensional row vectors constituting the matrix are defined as h ₁ , h ₂ ,..., H _M with respect to the transfer function matrix H of M rows and N columns generated by the generation means and the transfer function matrix generation means. When expressed, a row vector selecting means for selecting N ′ of the row vectors and N ′ number of orthogonally belonging to a subspace that can be expressed as a linear combination of the selected N ′ number of row vectors A basis vector extracting means for extracting a basis vector as a row vector and a transformation matrix given as a matrix of N rows and N 'columns each having a column vector given by a Hermite conjugate of the N' basis vectors Transformation matrix generation means for 'Transmission signal conversion means for converting using a conversion matrix a transmission signal of the system, characterized by comprising a transmitting means for transmitting the converted transmission signal by using the transmitting antennas N present.

According to the present invention, in the above invention, the row vector selection means selects N ′ from the _M N-dimensional row vectors h ₁ , h ₂ ,. It is characterized by doing.

According to the present invention, in the above invention, the row vector selecting means is a vector h _k (1 ≦ 1) having the largest vector size among the _M N-dimensional row vectors h ₁ , h ₂ ,. k ≦ M, k is an integer), and the basis vector extracting means extracts a basis vector so as to include a basis vector obtained by normalizing h _k .

According to the present invention, in the above invention, the basis vector extracting means is configured to generate a gram for the row vector selected from N ′ of the _M N-dimensional row vectors h ₁ , h ₂ ,. A basis vector is extracted using the Schmidt orthogonalization method.

  According to the present invention, in the above invention, the transmitting station uses an orthogonal frequency division multiplexing modulation system, the conversion matrix generation unit generates the conversion matrix for each subcarrier, and the transmission signal conversion unit The N ′ system transmission signal is converted for each subcarrier using the conversion matrix, and the transmission means uses the orthogonal frequency division multiplexing modulation scheme, and uses the N transmission antennas for the converted transmission signal. And transmitting.

In order to solve the above-described problem, the present invention is based on the same frequency channel for an integer M where M ≧ 2, an integer N where N ≧ 3, and an integer N ′ where N> N ′ ≧ 2. A transmission station having N transmission antennas that multiplex and transmit a plurality of signal sequences in space, and M or more that perform reception processing by receiving a transmitted radio signal and separating it into the plurality of signal sequences In a wireless communication method for performing communication using a MIMO channel between a plurality of antennas between the transmitting and receiving stations with a receiving station provided with a receiving antenna, the transmitting station includes N transmitting antennas of the transmitting station and the Acquiring each transfer function between M receiving antennas of the receiving station, generating a transfer function matrix of the MIMO channel having the acquired transfer function as each element, and input user data Is divided into N 'systems Generating a first signal sequence of the N ′ system by giving a signal of an individual known pattern to the data divided into the N ′ system, and the generated transfer function matrix of M rows and N columns When M N-dimensional row vectors constituting the matrix with respect to H are denoted as h ₁ , h ₂ ,..., H _M , selecting N ′ of the row vectors; A step of extracting basis vectors which are N ′ mutually orthogonal row vectors belonging to a subspace that can be expressed as a linear combination of N ′ row vectors, and Hermitian conjugate of the N ′ basis vectors. Generating a transformation matrix given as an N-row N′-column matrix having a column vector in each column; transforming an N′-system transmission signal using the transformation matrix; and converting the transformed transmission signal N said transmit antennas Characterized by a step of transmitting using.

According to the present invention, in the above invention, with respect to the M N-dimensional row vectors h ₁ , h ₂ ,..., H _M , an absolute value of each vector or an approximate value thereof is obtained; The method further includes a step of comparing the magnitudes of the absolute values and a step of selecting N ′ from the M row vectors having the larger vector size.

According to the present invention, in the above invention, with respect to the M N-dimensional row vectors h ₁ , h ₂ ,..., H _M , an absolute value of each vector or an approximate value thereof is obtained; A step of comparing the magnitudes of absolute values; a step of selecting a vector having the largest vector size from the M row vectors; and a vector obtained by normalizing the selected vector is extracted as one basis vector And a step of selecting a total of N ′ including the basis vectors.

According to the present invention, in the above invention, the step of extracting the basis vectors may be performed on the row vectors selected from N ′ out of the _M N-dimensional row vectors h ₁ , h ₂ ,. The Gramschmitt orthogonalization method is used.

  The present invention is the above invention, wherein the transmitting station transmits a transmission signal using an orthogonal frequency division multiplexing modulation scheme, and the step of generating the conversion matrix generates a conversion matrix for each subcarrier, The step of converting the transmission signal converts N ′ transmission signals for each subcarrier using a conversion matrix, and the step of transmitting the transmission signal is a transmission converted using an orthogonal frequency division multiplexing modulation scheme. A signal is transmitted using N transmission antennas.

According to the present invention, in the transmitting station, each transfer function between the N transmitting antennas of the transmitting station and the M receiving antennas of the receiving station is obtained by channel estimation means, and a transfer function matrix is generated. Means for generating the transfer function matrix of the MIMO channel having the acquired transfer function as each element, dividing the input user data into N ′ systems by the data dividing means, and transmitting signal sequence generating means, A signal of an individual known pattern is given to the data divided into the N ′ system to generate a first signal sequence of the N ′ system, and the generated transfer function matrix H of M rows and N columns is When M N-dimensional row vectors constituting the matrix are expressed as h ₁ , h ₂ ,..., H _M by the row vector selection means, N ′ of the row vectors are selected, and a basis vector is selected. The selected by the extraction means N ′ basis vectors that are orthogonal row vectors belonging to a subspace that can be expressed as a linear combination of N ′ row vectors are extracted, and Hermitian conjugate of the N ′ basis vectors is obtained by transform matrix generation means. A conversion matrix given as a matrix of N rows and N ′ columns having the column vector given by each column is generated, and the transmission signal is converted by the transmission signal conversion means using the conversion matrix, and the transmission means. The converted transmission signal is transmitted using the N transmission antennas. Therefore, it is not necessary to obtain the eigenvector itself, it is only necessary to obtain the orthogonal basis vector that exists in the partial space spanned by a part of the row vector group constituting the transfer function matrix, and the transformation matrix can be obtained easily and deterministically. The advantage of being able to do is obtained. As a result, the processing delay required to obtain the transformation matrix can be shortened, and the application of the transfer function matrix can be performed immediately after obtaining the information of the transfer function matrix. In addition, there is a secondary effect that it is possible to cope even in a propagation environment with a fast time variation.

Further, according to the present invention, by the row vector selecting means, the M N-dimensional row vector h _1, h 2, _..., selects the N 'pieces from the larger magnitude of the vector from the h _M . Therefore, there is an advantage that the directivity of the transmission signal can be selected more efficiently when the transmission diversity gain is increased by increasing the number of antennas on the transmission side and increasing the redundancy.

Further, according to the present invention, the row vector selection means causes the vector h _k (1 ≦ k) having the largest vector size from among the _M N-dimensional row vectors h ₁ , h ₂ ,. ≦ M and k are integers), and the basis vector extraction means extracts the basis vectors so as to include the basis vectors obtained by normalizing h _k . Therefore, there is an advantage that the directivity of the transmission signal can be selected more efficiently when the transmission diversity gain is increased by increasing the number of antennas on the transmission side and increasing the redundancy.

Further, according to the present invention, Gram Schmidt is selected for the row vector selected from the _M N-dimensional row vectors h ₁ , h ₂ ,..., H _M by the basis vector extraction means. Base vectors are extracted using the orthogonalization method. Therefore, there is an advantage that the orthogonal basis vector that is the basis of the transformation matrix can be easily generated by deterministic processing. In particular, in the conventional E-SDM system, when the number of multiplexed signals is 3 or more, there is no formula for directly obtaining the eigenvector, but by performing the prescribed procedure, the transformation matrix can be obtained reliably. it can. As a result, the processing delay can be shortened and the amount of calculation can be reduced.

  Further, according to the present invention, the transmitting station employs an orthogonal frequency division multiplexing modulation scheme, the conversion matrix generation unit generates the conversion matrix for each subcarrier, and the transmission signal conversion unit converts each subcarrier. The N ′ system transmission signal is converted for each carrier using the conversion matrix, and the transmission means uses the orthogonal frequency division multiplexing modulation method to convert the converted transmission signal using the N transmission antennas. Send. Therefore, it is possible to obtain an advantage that signals of a plurality of signal systems can be efficiently transmitted in combination with an OFDM modulation scheme having high resistance to frequency selective fading in a multipath environment.

According to the present invention, the step of acquiring each transfer function between the N transmitting antennas of the transmitting station and the M receiving antennas of the receiving station, and the acquired transfer function as each element Generating a transfer function matrix of the MIMO channel, dividing the input user data into N ′ systems, and assigning signals of individual known patterns to the data divided into the N ′ systems A step of generating a first signal sequence of the N ′ system, and M N-dimensional row vectors constituting the matrix for the generated transfer function matrix H of M rows and N columns, h ₁ , h ₂ ,. h When denoted as _M , a step of selecting N ′ of the row vectors and N ′ orthogonal to each other belonging to a subspace that can be represented as a linear combination of the selected N ′ row vectors Extract the basis vector that is the row vector Generating a transformation matrix given as an N-row N'-column matrix having column vectors given by Hermitian conjugation of the N 'basis vectors in each column, and N'-system transmission signals The transmitting station executes a step of converting using the conversion matrix and a step of transmitting the converted transmission signal using the N transmitting antennas. Therefore, an advantage is obtained in that a conversion matrix for efficiently transmitting a signal can be easily obtained as compared with a conventional method using a unitary conversion matrix using eigenvectors.

In addition, according to the present invention, for the M N-dimensional row vectors h ₁ , h ₂ ,..., H _M , a step of obtaining an absolute value of each vector or an approximate value thereof, and an absolute value of the obtained vector A step of comparing the magnitudes of the values and a step of selecting N ′ of the vector vectors having the larger vector size from the M row vectors are further executed at the transmitting station. Therefore, there is an advantage that the directivity of the transmission signal can be selected more efficiently when the transmission diversity gain is increased by increasing the number of antennas on the transmission side and increasing the redundancy.

In addition, according to the present invention, for the M N-dimensional row vectors h ₁ , h ₂ ,..., H _M , a step of obtaining an absolute value of each vector or an approximate value thereof, and an absolute value of the obtained vector A step of comparing the magnitudes of values, a step of selecting a vector having the largest vector size from the M row vectors, and a vector obtained by normalizing the selected vector is extracted as one basis vector. This is executed in the step, the step of selecting a total of N ′ including the basis vectors, and the transmitting station. Therefore, there is an advantage that the directivity of the transmission signal can be selected more efficiently when the transmission diversity gain is increased by increasing the number of antennas on the transmission side and increasing the redundancy.

In addition, according to the present invention, in the step of extracting the basis vector, the N ′ row vectors selected from the _M N-dimensional row vectors h ₁ , h ₂ ,. The Gramschmitt orthogonalization method is used. Therefore, there is an advantage that the orthogonal basis vector that is the basis of the transformation matrix can be easily generated.

  According to the present invention, the transmitting station transmits a transmission signal using an orthogonal frequency division multiplexing modulation scheme, and generates a conversion matrix for each subcarrier in the step of generating the conversion matrix. In the step of converting the signal, the transmission signal of N ′ system is converted for each subcarrier using a conversion matrix, and in the step of transmitting the transmission signal, the transmission signal converted using the orthogonal frequency division multiplexing modulation method Are transmitted using N transmission antennas. Therefore, it is possible to obtain an advantage that signals of a plurality of signal systems can be efficiently transmitted in combination with an OFDM modulation scheme having high resistance to frequency selective fading in a multipath environment.

  Hereinafter, a wireless communication apparatus according to an embodiment of the present invention will be described with reference to the drawings.

A. First, the principle of the present invention will be described. In general, the effect of applying E-SDM that forms and transmits an eigen beam on the transmission side is expressed by Expression (6) in addition to the case where the E-SDM original processing expressed by Expression (5) is performed on the reception side. The characteristics are improved when a method such as the ZF method or the MMSE method is used on the receiving side. On the other hand, when the MLD method is used, since the MLD method itself can take in the nonlinear effect well, sufficient characteristics can be obtained only on the reception side without forming an appropriate eigen beam on the transmission side. It was possible to get. In other words, when the MLD method is used on the receiving side or when various approaches based on the MLD method are used, the presence or absence of eigenbeam formation (application of the E-SDM method) on the transmitting side is directly determined. Did not affect the reception characteristics.

The basic principle of the MLD method is shown below. First, it is assumed that the relationship of Equation (1) is established between the transfer function matrix of the MIMO channel and the transmission signal and the reception signal. Here, Tx is an N-dimensional transmission signal vector, and each element represents a transmission signal of each signal system. For example, when multi-level modulation such as 64QAM is used, the number of candidates as transmission signals of each signal system matches the multi-level number m of the modulation mode (m = 64 in the case of 64QAM). When transmitting N ′ signal systems, there are m ^{N ′} candidates as combinations of transmission signals of each system. These candidates are denoted as Tx ^[n] , where n is the identification number. Here, when Tx ^[n] having a plurality of options is substituted for Tx in Expression (1), an expected value of the received signal when the transmission signal vector is Tx ^[n] (hereinafter referred to as a received replica signal). ) Is obtained. Here, the difference between the actual received signal and the received replica signal is obtained, and it is expected that the one with the smaller geometric distance is closer to the actual transmitted signal. As one example, when the Euclidean distance is used as the geometric distance, this Euclidean distance is given by the following equation.

Here, if Tx ^[n] that minimizes the Euclidean distance is selected, it can be said that the transmission signal is most likely.

  The reason why this MLD method shows good characteristics is that the signal received on the receiving station side does not cause a loss due to linear operation such as adding signals of multiple systems in opposite phases in the middle of processing. is there. In other words, as long as the transmission signal energy is transmitted between the transmitting and receiving stations with a small loss, the reception energy can be used efficiently.

  Thus, how the transmitted signal is received on the receiving side will be described by taking an example in which the number of transmission antennas is four, the number of reception antennas is two, and the number of signal systems to be transmitted is two. As described above, in general, the number of transmission antennas and the number of signal systems to be transmitted (spatial multiplexing) do not necessarily match. Transmitting two signal systems by appropriately linearly combining them to four antennas means transmitting using the four antenna degrees of freedom with respect to the degrees of freedom of the two antennas. A transmission diversity gain is expected.

  First, a two-dimensional transmission signal vector is Tx. On the other hand, U is a conversion matrix that is appropriately weighted and assigned to the four transmission antennas for signals of each signal system. Let H be the transfer function matrix between the four transmitting antennas and the two receiving antennas.

Here, the row vectors h ₁ , h ₂ , u ₁ , u ₂ constituting each matrix are each a four-dimensional complex vector. Therefore, a subspace that can be expressed by a linear combination of h ₁ and h _{2 in} a four-dimensional space is considered, and two base vectors are extracted as two orthogonal vectors belonging to this subspace, and these are extracted as u ₁ , u ₂ . Gram Schmidt's orthogonalization method is known as a simple method for extracting such basis vectors, and is given by the following equation.

That is, u ₁ is a unit vector in the h ₁ direction, and u ₂ is obtained by subtracting the projected part from h ₂ to u ₁ from h ₂ and normalizing it. In the case where the subspace is spanned by three or more vectors, the same processing is sequentially repeated. In this way, a transformation matrix U for transmitting signals of two transmission systems with four antennas is obtained. Using this conversion matrix, the transmission signal vector Tx is converted as U × Tx and transmitted. As a result, the relational expression of Formula (4) is established between the received signal and the received signal.

Here, only the _two vectors u ₁ and u ₂ are mentioned with respect to the original four-dimensional space, but the remaining two basis vectors orthogonal to the two basis vectors are designated as u ₃ and u ₄ , Try to understand this property. Therefore, consider the following transformation matrix U ′.

At this time, if it is noted that h ₁ ⊥u ₃ , h ₁ ⊥u ₄ , h ₂ ⊥u ₃ , and h ₂ ⊥u ₄ , the inner product of these vectors is zero. Therefore, when this conversion is used, the signal received on the receiving side is always a zero vector for any transmission signal [t ₁ , t ₂ ] as follows.

The intent of this equation is that even if some signal is transmitted on the transmitting side, those signals cancel each other in space, and the received signal strength at the receiving antenna on the receiving side is always zero. . In other words, if a signal is transmitted in this direction in a four-dimensional space, it means that the signal is completely lost at the time of reception. In other words, if transmission is performed avoiding a space that can be expressed by linear combination of u ₃ and u ₄ , signal transmission with less loss is possible. In other words, at least the received signal strength at the receiving antenna should be stronger by a smaller loss.

Mathematically, for a 2 × 4 transfer function matrix H, H ^H · H is a 4 × 4 square matrix. Since the eigenvalues of this matrix were originally generated from a matrix of 2 rows and 4 columns, the smaller of the two non-zero eigenvalues (the number of transmit antennas “4” and the number of receive antennas “2”) In addition to the number of values, there will be non-zero eigenvalues), and there will be two eigenvalues with zero values. The two eigenvectors corresponding to the non-zero eigenvalues belong to the space spanned by u ₁ and u ₂ , and the eigenvectors corresponding to the remaining zero eigenvalues belong to the space spanned by u ₃ and u ₄ . Of course, note that u ₁ and u ₂ are not guaranteed to match the eigenvectors for nonzero eigenvalues.

For the above reason, when the transformation matrix U is configured using the vectors u ₁ and u ₂ obtained by the mathematical formulas (13) and (14) and the signal is transmitted by performing this transformation, at least the reception antenna is lost. The signal transmission can be performed by minimizing. Of course, since the vector is directed in a different direction from the eigenvector, the E-SDM method given by Equation (5) cannot be applied using this conversion. Further, in the ZF method represented by the formula (6), there is no guarantee that the maximum ratio composition is possible. However, if it is possible to use the MLD method or a method obtained by simplifying the MLD method on the receiving side, sufficient characteristics can be obtained by converting the transmission signal using the basis vectors obtained as described above. .

  In the above description, the number of transmission antennas is “4”, the number of reception antennas is “2”, and the number of signal systems to be spatially multiplexed is “2”, but the number of transmission antennas is more redundant than the number of spatially multiplexed signals. Therefore, it is possible to apply to other values.

  In general, a wireless communication system includes a base station (or access point) connected to a wired network and a plurality of terminal stations wirelessly connected to the base station. At this time, in order to suppress the circuit scale on the terminal station side, an approach of adding an additional function to the base station side is a common technique. In the present invention, even when the number of antennas on the terminal side is two or more, the number of antennas on the base station side is increased, and transmission diversity gain is obtained as a result of the redundancy.

  Various embodiments for realizing the above principle of the present invention will be described below with reference to the drawings. Here, for the sake of simplicity, description will be made assuming that transmission is performed by superimposing two signal sequences using four transmission antennas.

B. First embodiment B-1. Configuration of First Embodiment FIG. 1 is a block diagram showing a configuration of a transmission unit of a transmission station according to the first embodiment of the present invention. As an example, a case where the transmitting station transmits two systems of data using four transmitting antennas as in FIG. 10 will be described. Here, for convenience of explanation, the transmission function and the reception function are described separately, but in general, a wireless communication apparatus having a transmission / reception function is used. Note that portions corresponding to those in FIG. 10 are denoted by the same reference numerals and description thereof is omitted.

In FIG. 1, the matrix operation circuit 9 and the eigenvector generation circuit 10 shown in FIG. In the basis vector generation circuit 14, when the transfer function matrix H is expressed as in Expression (11), a basis vector constituting a partial space including the row vector group (h ₁ , h ₂ ) constituting this is generated. . The generation method may be any method, but, for example, Gramschmitt orthogonalization methods such as Equation (13) and Equation (14) can be used. For the basis vector generated here, a vector that is Hermitian conjugate as shown in Equation (12) is generated, and a conversion matrix U expressed by this combination is output to the transmission signal conversion circuit 4, and here this conversion is performed. A transmission signal is converted using a matrix.

  In the present invention, the configuration on the receiving station side is not changed from that shown in FIG.

  Next, FIG. 2 is a block diagram showing a configuration of the basis vector generation circuit 14 in the first embodiment. In the figure, 21 is a row vector extraction circuit, 22 is a selection circuit, 23 is an orthogonal basis generation circuit, 24 is a transformation matrix generation circuit, 25 is an absolute value calculation circuit, and 26 is a comparison circuit.

  When the transfer function matrix information is input, the row vector extraction circuit 21 decomposes the transfer function matrix in the direction from the transmitting station to the receiving station into row vectors, and notifies the absolute value calculation circuit 25 of the information. The absolute value calculation circuit 25 calculates the absolute value of each row vector or its approximate value by calculation, and notifies the comparison circuit 26 of the result. The selection circuit 22 receives information on each row vector from the row vector extraction circuit 21 and information on the magnitude relationship between the absolute values of each vector from the comparison circuit 26, and a predetermined number of orthogonal basis generation circuits 23 from the larger absolute value. To enter.

  For example, if the transfer function matrix has 2 rows and 4 columns, there are two types of row vectors. If the number of systems of signals to be spatially multiplexed during transmission is two, these two types of row vectors are used. If the transfer function matrix is 3 rows and 4 columns (when there are three reception antennas on the receiving side), there are three types of row vectors, but the number of systems of signals that are spatially multiplexed during transmission is two. For example, the selection circuit 22 selects two of the three types of row vectors having a large absolute value and inputs them to the orthogonal basis generation circuit 23.

  For example, when two row vectors are specified for a four-dimensional vector, the orthogonal base generation circuit 23 generates two types of orthogonal bases as shown in Equation (13) and Equation (14). At this time, any one of them may be substituted into the formula (13), or the larger absolute value may be selectively associated with the formula (13). In addition, Equations (13) and (14) assume the Gram Schmidt orthogonalization method, but the orthogonal basis may be obtained using other methods. The obtained orthogonal basis (row vector) is input to the conversion matrix generation circuit 24, and a conversion matrix U is generated using column vectors that are Hermitian conjugates, respectively, and is output.

The selection circuit 22 selects the row vectors so that the absolute value includes the largest one, and the remaining vectors do not necessarily have to be selected in descending order of the absolute values. In this case, the one having the maximum absolute value is made to correspond to Equation (13), and the orthogonal bases are obtained in order.
Although the absolute value calculation circuit 25 and the comparison circuit 26 are assumed here, these are not essential and can be omitted. In this case, the selection circuit 22 selects a desired number of row vectors by any means.

Further, the operation performed by the absolute value operation circuit 25 is generally obtained by u · u ^H (or this square root) using the Hermitian conjugate column vector u ^H for the row vector u. As the approximate solution, the sum of the absolute value of the real part and the absolute value of the imaginary part of each component of the vector can be used.

B-2. Operation of the First Embodiment Next, the operation of the first embodiment will be described. Here, FIG. 3 is a flowchart showing an operation at the time of signal transmission of the transmitting station according to the first embodiment. When data is input (S1), the transmitting station divides the data into N data series (S2). Preamble signals are assigned to these signals (S3), and each of the series is individually modulated. (S4). The modulated signal is subjected to signal conversion (S5), and the converted signal is converted to a radio frequency by the radio unit and transmitted (S6). When the transmission data continues (S7), the processes S4 to S6 are repeated, and after the transmission data is completed (S7), the process is completed (S8). The difference from the conventional method of FIG. 13 is that the signal conversion performed in step S5 is replaced with a different conversion matrix from the unitary conversion based on the conventional eigenvector.

  Next, FIG. 4 is a flowchart showing an operation for generating a transformation matrix according to the first embodiment. When the transmitting station acquires information on the transfer function matrix by some means (S11), a predetermined number of row vectors are selected from the row vectors of the transfer function matrix H (S12), and the subspace spanned by the row vector group is expanded. The orthogonal basis vector to be constructed is calculated (S13). A transformation matrix having the orthogonal basis vector as a component is generated (S14), temporarily stored in the memory (S15), and the process is terminated (S16). This stored transformation matrix is used in step S5 shown in FIG. 3 during signal transmission.

  Next, FIG. 5 is a flowchart showing another operation for generating a transformation matrix according to the first embodiment. The steps corresponding to those in FIG. After obtaining the transfer function matrix (S11), the absolute value of each row vector constituting the transfer function matrix or an approximate value of the absolute value is obtained (S21), and the magnitude comparison of each row vector is performed (S22). In the selection of the row vector performed in step S12, the selection is performed with reference to the magnitude comparison result in step S22. For example, from the M row vectors, N ′ items may be selected in order from the largest absolute value, or one having the largest absolute value may be selected at random. Other processes are the same as those in FIG.

C. Second Embodiment Next, a second embodiment of the present invention will be described. Here, FIG. 6 is a block diagram showing a configuration of a wireless communication apparatus having functions of both a transmitting station and a receiving station according to the second embodiment of the present invention. The parts corresponding to those in FIG. 1 or FIG. In the figure, 30 is a modulation unit, and 31 is a demodulation unit.

  Similar to the configuration described with reference to FIG. 12 of the conventional system, a wireless communication apparatus having both a transmitting station function and a receiving station function is common. As for processing, the processing in FIG. 1 and the processing of the receiving station in FIG. 11 are combined, and the channel estimation circuit 7 and the channel estimation circuit 123, the transfer function matrix management circuit 8 and the transfer function matrix management circuit 125 transmit and The only difference is that it is shared by both receiving systems.

D. Third Embodiment Next, a third embodiment of the present invention will be described. Here, FIG. 7 is a block diagram showing a configuration of a wireless communication apparatus having functions of both a transmitting station and a receiving station according to the third embodiment of the present invention. The difference from the configuration of the second embodiment shown in FIG. 6 is that a control information separation circuit 15 is added after the data synthesis circuit 13. In the above description, as a method for obtaining a transfer function matrix related to the MIMO channel in the transmitting station, information on the transfer function matrix in the direction from the receiving station to the transmitting station is acquired, and the direction transfer from the transmitting station to the receiving station is obtained from this value. Although the function matrix has been obtained, other methods may be used.

  As an example, there is a method in which the information of the transfer function matrix acquired on the receiving station side is accommodated in the control information as it is and returned to the transmitting station side. On the transmitting station side, when the information of the transfer function matrix of the MIMO channel is acquired as the control information, this is terminated by the control information separation circuit 15 and transferred to the transfer function matrix management circuit 8 or directly as the information of the transfer function matrix. This is input to the basis vector generation circuit 14. On the transmission side, a basis vector is generated based on this information.

  Furthermore, as a method for notifying the transmission function matrix information from the receiving station to the transmitting station, a line different from the wireless line used for wireless communication may be used. For example, it is possible to use a wire, another standard, a radio standard with another frequency, or the like.

F. Fourth Embodiment Next, a fourth embodiment of the present invention will be described. Here, FIG. 8 is a block diagram showing a configuration of a wireless communication apparatus using the OFDM modulation scheme according to the fourth embodiment of the present invention. In the figure, 41 is a data division circuit, 42-1 to 42-K are modulation units, 43-1 to 43-4 are IFFT circuits, 44-1 to 44-4 are GI addition circuits, 45-1 to 45-4. Is a radio unit, 46-1 to 46-4 are antennas, 47-1 to 47-4 are GI removal circuits, 48-1 to 48-4 are FFT circuits, 49-1 to 49-K are demodulation units, and 50 is A data synthesis circuit, 51 is an OFDM modulator, and 52 is an OFDM demodulator.

  Basically, the operation is the same as that of the configuration shown in FIG. 6 except that the modulation unit 30 and the demodulation unit 31 that are one system in the case of a single carrier are provided for each subcarrier. 1 to 42-K and demodulation units 49-1 to 49-K, and further, on the transmission side, IFFT circuits 43-1 to 43-4 are provided for each signal system, where all subcarriers are provided. The GI addition circuits 44-1 to 44-4 insert a guard interval GI into this signal at the OFDM symbol period.

  On the receiving side, the GI removal circuits 47-1 to 47-4 remove the guard intervals in the OFDM symbol period for the received signals, and the removed signals are sent to the FFT circuits 48-1 to 48-4. Then, the signals are separated into subcarriers by fast Fourier transform processing, and the separated signals are input to the demodulation units 49-1 to 49-K for each subcarrier. In the data separation circuit 41 and the data synthesis circuit 50, signal division and synthesis processing is performed on the number of signal systems to be multiplexed in the case of a single carrier in FIG. 6, but when using the OFDM modulation method, Furthermore, it is different in that it is divided for each subcarrier and combined processing is performed for all of them.

  Here, the case where the OFDM modulation scheme is applied to FIG. 6 has been described. However, even when the transfer function information is notified by the control information as shown in FIG. 7, the modulation unit 30 and the demodulation unit shown in FIG. Naturally, it is possible to achieve the same correspondence as in FIG. 8 by replacing 31 with the OFDM modulation unit 51 and the OFDM demodulation unit 52 in FIG. In addition, the same OFDM modulation scheme can be adopted even when only the transmission station function shown in FIG. 1 is provided.

  Next, the operation at the time of signal transmission of the transmitting station according to the fourth embodiment will be described. Here, FIG. 9 is a flowchart showing an operation at the time of signal transmission of the transmitting station according to the fourth embodiment. The steps corresponding to those in FIG. Basically, it is the same as FIG. 16 in the conventional method using the OFDM modulation method, but the implementation of unitary conversion in step S105 in FIG. 16 is replaced with the signal conversion in step S5 in FIG. In the fourth embodiment, the processing on the receiving side is basically unchanged from the conventional method.

  As described in detail above, according to the first to fourth embodiments, when performing highly efficient wireless communication using MIMO technology, the loss of signal transmission from the transmitting station to the receiving station is minimized, It becomes possible to transmit a signal efficiently.

  Further, in the E-SDM system as the conventional system, it is necessary to obtain eigenvectors related to the transfer function matrix for such efficient signal transmission. For this reason, particularly when the size of the transfer function matrix is large, it is difficult to obtain the eigenvector, and there is a tendency that the amount of calculation and the processing delay become enormous. In the E-SDM system, it is necessary to obtain eigenvectors (or transformation matrices) in a short time from acquisition of information related to the transfer function, but a large processing delay hinders this.

  In the present invention, even when the size of a matrix is large, a transformation matrix can be generated by deterministic processing, and the amount of computation can be greatly reduced. As a result, even immediately after acquiring information related to the transfer function matrix, an efficient signal transmission can be performed immediately using an optimal transformation matrix.

  In particular, if this technology is used on the base station or access point (AP) side, it is possible to improve reception characteristics in each terminal station having a simple configuration without changing the configuration of a large number of terminal stations.

  The above-described first to fourth embodiments are all illustrative and do not limit the present invention, and the present invention can be implemented in various other modifications and changes. Therefore, the scope of the present invention is defined only by the claims and their equivalents.

  In the above-described embodiment, each circuit is basically mounted on hardware, but may be executed in a computer system. In this case, a series of processing steps by the above-described basis vector generation circuit 14, control information separation circuit 15, and the like are stored in a computer-readable recording medium in the form of a program, and the computer reads and executes this program. Thus, the above process is performed. That is, each processing means and processing unit in the basis vector generation circuit 14 and the control information separation circuit 15 is processed by a central processing unit such as a CPU by reading the above program into a main storage device such as a ROM or a RAM. It may be realized by executing arithmetic processing.

  Here, the computer-readable recording medium means a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, a semiconductor memory, or the like. Alternatively, the computer program may be distributed to the computer via a communication line, and the computer that has received the distribution may execute the program.

It is a block diagram which shows the structure of the transmission part of the transmission station by 1st Embodiment of this invention. FIG. 2 is a block diagram showing a configuration of a basis vector generation circuit 14 in the first embodiment. It is a flowchart which shows the operation | movement at the time of the signal transmission of the transmitting station by this 1st Embodiment. It is a flowchart which shows the operation | movement for the conversion matrix production | generation by this 1st Embodiment. It is a flowchart which shows another operation | movement for the conversion matrix production | generation by this 1st Embodiment. It is a block diagram which shows the structure of the radio | wireless communication apparatus provided with the function of both the transmission station by 2nd Embodiment of this invention, and a receiving station. It is a block diagram which shows the structure of the radio | wireless communication apparatus provided with the function of both the transmission station by 3rd Embodiment of this invention, and a receiving station. It is a block diagram which shows the structure of the radio | wireless communication apparatus using the OFDM modulation system by 4th Embodiment of this invention. It is a flowchart which shows the operation | movement at the time of the signal transmission of the transmitting station by this 4th Embodiment. It is a block diagram which shows the structure of the transmission part of the transmission station in a prior art. It is a block diagram which shows the structure of the receiving part of the receiving station in a prior art. It is a block diagram which shows the structural example of the radio | wireless communication apparatus provided with the function of both the transmitting station and receiving station in a prior art. It is a flowchart which shows the process at the time of the signal transmission in a prior art. It is a flowchart which shows the process for the unitary transformation matrix production | generation in a conventional system. It is a flowchart which shows the receiving operation of the receiving station in a prior art. It is a flowchart which shows the process at the time of the signal transmission in the prior art using an OFDM modulation system. It is a flowchart which shows the process at the time of the signal reception in the prior art using an OFDM modulation system.

Explanation of symbols

1 Data division circuit (data division means)
2-1-2-2 Preamble applying circuit 3-1-2-2 Modulation circuit 4 Transmission signal conversion circuit (transmission signal conversion means)
5-1 to 5-4 Wireless unit (transmission means)
6-1 to 6-4 Antenna (Transmission means)
7 Channel estimation circuit (channel estimation means)
8 Transfer function matrix management circuit (transfer function matrix generation means)
DESCRIPTION OF SYMBOLS 11 Reception signal management circuit 12 Signal detection circuit 13 Data composition circuit 14 Base vector generation circuit 15 Control information separation circuit 21 Row vector extraction circuit (transmission signal sequence generation means)
22 Selection circuit (row vector selection means)
23 Orthogonal basis generation circuit (basis vector extraction means)
24 Conversion matrix generation circuit (conversion matrix generation means)
25 Absolute value calculation circuit 26 Comparison circuit 30 Modulation unit 31 Demodulation unit 41 Data division circuit 42-1 to 42-K Modulation unit 43-1 to 43-4 IFFT circuit 44-1 to 44-4 GI addition circuit 45-1 45-4 Radio unit 46-1 to 46-4 Antenna 47-1 to 47-4 GI removal circuit 48-1 to 48-4 FFT circuit 49-1 to 49-K Demodulation unit 50 Data synthesis circuit 51 OFDM modulation unit 52 OFDM demodulation unit 121-1 to 121-2 reception antenna 122-1 to 122-2 radio unit 123 channel estimation circuit 124 reception signal management circuit 125 transfer function matrix management circuit 126 signal detection circuit 127 data synthesis circuit

Claims (10)

For an integer M satisfying M ≧ 2, an integer N satisfying N ≧ 3, and an integer N ′ satisfying N> N ′ ≧ 2, N signals are transmitted by multiplexing a plurality of signal sequences on the same frequency channel. A transmission station having a plurality of transmission antennas and a reception station having M or more reception antennas that receive transmitted radio signals and perform reception processing by separating the signals into a plurality of signal sequences. In a wireless communication apparatus that performs communication using a MIMO channel between a plurality of antennas between stations,
The transmitting station is
Channel estimation means for obtaining respective transfer functions between the N transmitting antennas of the transmitting station and the M receiving antennas of the receiving station;
Transfer function matrix generating means for generating a transfer function matrix of the MIMO channel having the acquired transfer function as each element;
Data dividing means for dividing the input user data into N ′ systems;
Transmission signal sequence generating means for generating a first signal sequence of the N ′ system by giving a signal of an individual known pattern to the data divided into the N ′ system;
When M N-dimensional row vectors constituting the matrix are expressed as h ₁ , h ₂ ,..., H _M with respect to the transfer function matrix H of M rows and N columns generated by the transfer function matrix generating means. Row vector selection means for selecting N ′ of the row vectors;
Basis vector extraction means for extracting basis vectors which are N ′ mutually orthogonal row vectors belonging to a subspace that can be expressed as a linear combination of the selected N ′ row vectors;
Transformation matrix generation means for generating a transformation matrix given as a matrix of N rows and N 'columns each having a column vector given by Hermitian conjugate of the N' basis vectors;
Transmission signal conversion means for converting the transmission signal of the N ′ system using a conversion matrix;
Transmitting means for transmitting the converted transmission signal using the N transmitting antennas. A wireless communication apparatus comprising: 2. The row vector selection means selects N ′ pieces from the _M N-dimensional row vectors h ₁ , h ₂ ,..., H _M from the larger vector size. The wireless communication device described. The row vector selection means selects a vector h _k (1 ≦ k ≦ M, k is an integer) having the largest vector size from among the _M N-dimensional row vectors h ₁ , h ₂ ,. Select to include,
The basis vector extracting means, radio communication apparatus according to claim 1, wherein the extracting the basis vectors to include base vector obtained by normalizing h _k. The basis vector extraction means uses the Gram Schmidt orthogonalization method for the row vectors selected from the _M N-dimensional row vectors h ₁ , h ₂ ,. The wireless communication apparatus according to claim 1, wherein a vector is extracted. The transmitting station uses an orthogonal frequency division multiplexing modulation scheme,
The transformation matrix generation means generates the transformation matrix for each subcarrier,
The transmission signal conversion means converts N ′ system transmission signals for each subcarrier using the conversion matrix,
2. The radio communication apparatus according to claim 1, wherein the transmission unit transmits the converted transmission signal using N transmission antennas using an orthogonal frequency division multiplexing modulation system. For an integer M satisfying M ≧ 2, an integer N satisfying N ≧ 3, and an integer N ′ satisfying N> N ′ ≧ 2, N signals are transmitted by multiplexing a plurality of signal sequences on the same frequency channel. Between the transmitting and receiving stations, and a receiving station having M or more receiving antennas that receive a transmitted radio signal and perform reception processing by separating the plurality of signal sequences. In a wireless communication method for performing communication using a MIMO channel between a plurality of antennas,
The transmitting station is
Obtaining each transfer function between the N transmitting antennas of the transmitting station and the M receiving antennas of the receiving station;
Generating a transfer function matrix of the MIMO channel with each of the acquired transfer functions as elements;
Dividing the input user data into N ′ systems;
Providing a signal of an individual known pattern to the data divided into the N ′ systems to generate a first signal sequence of the N ′ system;
When M N-dimensional row vectors constituting the matrix are expressed as h ₁ , h ₂ ,..., H _M with respect to the generated transfer function matrix H of M rows and N columns, N in the row vector 'Step to select pieces,
Extracting base vectors which are N ′ mutually orthogonal row vectors belonging to a subspace that can be expressed as a linear combination of the selected N ′ row vectors;
Generating a transformation matrix given as a matrix of N rows and N 'columns with column vectors given by Hermitian conjugates of the N' basis vectors in each column;
Converting N ′ system transmission signals using the transformation matrix;
Transmitting the converted transmission signal using the N transmission antennas. A wireless communication method, comprising: Obtaining absolute values or approximate values of the respective vectors for the _M N-dimensional row vectors h ₁ , h ₂ ,..., H _M ;
Comparing the magnitudes of the absolute values of the acquired vectors;
The wireless communication method according to claim 6, further comprising: selecting N ′ of the M row vectors having a larger vector size. Obtaining absolute values or approximate values of the respective vectors for the _M N-dimensional row vectors h ₁ , h ₂ ,..., H _M ;
Comparing the magnitudes of the absolute values of the acquired vectors;
Selecting a vector having the largest vector size from the M row vectors;
Extracting a standardized vector of the selected vector as one basis vector;
The wireless communication method according to claim 6, further comprising: selecting a total of N ′ including the basis vectors. The step of extracting the basis vectors uses Gram Schmidt orthogonalization for the row vectors selected from the _M N-dimensional row vectors h ₁ , h ₂ ,..., H _M. The wireless communication method according to claim 6. The transmitting station transmits a transmission signal using an orthogonal frequency division multiplexing modulation scheme,
The step of generating the transformation matrix generates a transformation matrix for each subcarrier,
The step of converting the transmission signal is performed by converting a transmission signal of N ′ system using a conversion matrix for each subcarrier,
7. The radio communication method according to claim 6, wherein the step of transmitting the transmission signal includes transmitting the transmission signal converted by using an orthogonal frequency division multiplexing modulation scheme, using the N transmission antennas.

Images