JP4188372B2 - Wireless communication apparatus, wireless communication method, and wireless communication system - Google Patents

Description

  The present invention uses the same frequency channel, transmits independent data from a plurality of different transmission antennas, receives signals using a plurality of reception antennas, and receives a reception station based on a transfer function matrix between the transmission and reception antennas. The present invention relates to a reception technique for realizing good transmission characteristics while suppressing the circuit scale in a high-speed wireless access system (or wireless LAN system) that realizes wireless communication by demodulating data on the side. In addition, when using Orthogonal Frequency Division Multiplexing (OFDM) modulation system together, transmission to gain transmission diversity gain on the base station side and to extract good characteristics while suppressing the circuit scale on the wireless terminal side Regarding technology. The present invention is particularly used to increase the transmission speed of a high-speed wireless access system (or wireless LAN system) using the 2.4 GHz band and the 5 GHz band.

In recent years, as a high-speed wireless access system using the 2.4 GHz band or the 5 GHz band, wireless LAN systems compliant with the IEEE802.11g standard, the IEEE802.11a standard, and the like are remarkably widespread. Although these systems achieve a maximum transmission rate of 54 Mbps, further increase in transmission rate is required with the spread of wireless LAN.
For this purpose, MIMO (Multiple-Input Multiple-Output) technology is promising. With this MIMO technology, the transmitting station transmits different independent signals on the same channel from multiple transmitting antennas, and the receiving station receives signals using the same multiple antennas. The transfer function matrix is obtained, the independent signal transmitted from each antenna is estimated on the transmitting station side using this matrix, and the data is reproduced.

Here, a case is considered where 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 h _{j, i.} The matrix of M rows and N columns as the (j, i) component is denoted as H. Further, let t _{i be} the transmission signal from the i-th transmitting antenna, Tx be a column vector whose components are (t ₁ , t ₂ , t ₃ ,... T _N ), and r _{j be} the received signal at the j-th receiving antenna. (R ₁ , r ₂ , r ₃ ,..., R _M ) as a column vector, Rx, and thermal noise of the jth receiving antenna as n _j (n ₁ , n ₂ , n ₃ ,. A column vector whose component is n _M ) is denoted as 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 one of the MIMO techniques, there is a method generally called a ZF (Zero Forcing) method (see, for example, Non-Patent Document 1).
Here, an inverse matrix H ⁻¹ of the transfer function matrix is obtained with respect to the above (Equation 1), and this is multiplied from the left of both sides of the equation. As a result, the following expression is obtained.

In other words, when the signals received by the receiving antennas are combined and processing for removing interference caused by signals other than the desired transmitting antenna is performed, a small 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. Therefore, a hard decision process is performed to search the point on the transmission constellation where H ⁻¹ × Rx is the closest to the Euclidean distance, and a true transmission signal is estimated.

In the above ZF method, good characteristics can be expected when the thermal noise term H ⁻¹ × n is sufficiently small and the components for each transmitting antenna can be assumed to be equal. However, in general, this assumption does not hold, and the expected value of the absolute value of the thermal noise H ⁻¹ × n for each transmission antenna differs for a certain transfer function matrix. Furthermore, if the transfer function matrix H is a matrix whose inverse matrix is zero (or its determinant is very small), the estimation of the transmission signal becomes very unstable. In such a situation, there is a possibility that the reception characteristic is greatly deteriorated. As an example of a method for solving such a problem, for example, a MMSE (Minimum Mean Square Error) method is cited.

In the case of the MMSE method, an operator to be applied from the left of both sides of (Equation 1) is not an inverse matrix of the transfer function matrix H, but another matrix operator F is used. This matrix operator F is expressed as (F × Rx (k) when Tx (k) is the transmission signal vector at the kth symbol of the transmission signal and Tx (k) is the reception signal vector at the kth symbol of the reception signal. −Tx (k)) The matrix operator F is selected so that the expected value over a plurality of symbols of ^H × (F × Rx (k) −Tx (k)) is expected to be minimized. Actually, the matrix operator F is estimated using a known preamble signal part, and the matrix operator F estimated there is used in the demodulation of the subsequent data.

As described above, it is possible to increase the transmission capacity by performing communication using MIMO technology, but separate signals transmitted from a plurality of transmission antennas can interfere with each other. By using a plurality of antennas on the side, a diversity diversity gain is expected, but the diversity gain cannot be maximized in the process of canceling this interference with each other.
That is, in the ZF method, an inverse matrix of the transfer function matrix H is used in order to remove the interference between the sequences of each signal sequence and separate the signals. The purpose is to eliminate it, and it 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, it is necessary to adjust and transmit signals so that the signal sequences are orthogonal at the receiving end on the transmitting side. is there. This method is called an E-SDM (Eigenbeam-Space Division Multiplexing) method (see, for example, Non-Patent Document 2).

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 matrix H ^H and H of this matrix, that is, a unitary matrix U that diagonalizes N × N matrix H ^H × H is obtained, and transmission signal vector Tx (k) is obtained. A signal U × Tx (k) subjected to unitary conversion is transmitted. Here, the signal transmitted from the i-th transmitting antenna is [U × Tx] _i (where [x] _i represents the i-th component of the vector x). 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 diagonal components are zero. When this method is used, (Equation 1) is also converted as follows.

When the Hermitian conjugate matrix of the transfer function matrix H × U estimated from the left of both sides is applied, the equation is converted as follows.

This means that the N signal sequences are completely orthogonal. Further, the matrix appearing on the right side can be directly obtained from (Equation 3) from the estimated transfer function matrix. Therefore, regardless of whether or not the original matrix H has an inverse matrix, the transmission signal Tx is stably estimated for a signal sequence having a non-zero eigenvalue by (Equation 3) and (Equation 5). Is possible.

  It can 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.

  Here, FIG. 17 shows the configuration of the transmission section of the first radio station in the prior art. In the figure, 100 is a data division circuit, 101-1 to 101-3 are preamble assignment circuits, 102-1 to 102-3 are modulation circuits, 103 is a transmission signal conversion circuit, 104-1 to 104-3 are radio units, Reference numerals 105-1 to 105-3 denote transmission antennas, 106 denotes a matrix operation circuit # 1, and 107 denotes a transfer function matrix management circuit. As an example, a case where the transmitting station transmits three systems of data using three transmitting antennas will be described as an example.

First, the transfer function matrix H managed by the transfer function matrix management circuit 107 is input to the matrix operation circuit 106. In the matrix operation circuit # 1 (106), a Hermitian conjugate matrix H ^H , a product H ^H × H thereof, and a unitary matrix U for diagonalizing the matrix H ^H are obtained with respect to the input matrix H, and this is transmitted. Input to the conversion circuit 103. Next, when data is input, the data dividing circuit 100 separates the data into three systems. For example, the first system data is input to the preamble applying circuit 101-1, and is input to the modulation circuit (Ch1) 102-1 with the preamble added.

  The modulation circuit (Ch1) 102-1 performs predetermined modulation, and the modulated signal is input to the transmission signal conversion circuit 103. In the transmission signal conversion circuit 103, unitary conversion is performed on the transmission signal Tx including the preamble signal, and is input to the radio unit 104-1 as U × Tx. Then, it is converted into a radio frequency by the radio unit 104-1 and transmitted from the transmission antenna 105-1. Similarly, the second system data is individually transmitted via 101-2 to 105-2, and the third system data is individually transmitted via 101-3 to 105-3.

  FIG. 18 shows the configuration of the receiving unit of the second radio station in the prior art. In FIG. 18, 111-1 to 111-3 are receiving antennas, 112-1 to 112-3 are radio units, 113 is a channel estimation circuit, 114 is a received signal management circuit, 115 is a transfer function matrix management circuit, and 116 is a matrix. Arithmetic circuits # 2, 117 are matrix arithmetic circuits # 3, 118 are hard decision circuits, and 119 is a data synthesis circuit.

The first reception antenna 111-1 to the third reception antenna 111-3 individually receive the reception signal. The received signal is input to the channel estimation circuit 113 via the radio units 112-1 to 112-3. From the reception status of a predetermined preamble signal given on the transmission side, the channel estimation circuit 113 obtains the transfer function between each transmission antenna and the reception antenna here. The acquired information h _{j, i} of each transfer function is managed as a transfer function matrix H by the transfer function matrix management circuit 115.
Here, the transfer function H is obtained by multiplying the transfer function of the actual transmission path by the unitary transformation matrix.

In the matrix operation circuit # 2 (116), H ^H and H ^H × H (= Λ) are obtained by calculation based on the transfer function matrix H managed by the transfer function matrix management circuit 115. Here, each diagonal component of the matrix Λ is assumed that the (1,1) component is λ ₁ , the (2,2) component is λ ₂ , and the (3,3) component is λ ₃ .
In addition to obtaining eigenvalues λ _{1 to} λ ₃ with H ^H × H, λ _i may be obtained from eh _{i, j} e ² as an approximate value using the transfer function.
This result is input to the matrix operation circuit # 3 (117). On the other hand, the data signal following the preamble signal is input to the received signal management circuit 114 for each symbol. In the reception signal management circuit 114, a reception signal vector Rx having the reception signal (r ₁ , r ₂ , r ₃ ) ^T (T indicates conversion from a row vector to a column vector) as a component is managed once. . The received signal vector takes the product of the matrix ^{H H} at matrix operation circuit # 3 (117). The vector obtained in this way is divided by λ1 for the first component, divided by λ2 for the second component, and divided by λ3 for the third component, and transmitted. Perform a first order estimation of the signal vector. This result is input to the hard decision circuit 118, and the result is input to the data synthesis circuit 119. The data demodulated over all symbols is reproduced by outputting the original data even if the signals for each system are appropriately combined.

  FIG. 19 shows a transmission flow of the first radio station in the prior art. When data is input (step S100), the transmitting station divides the data into N data series (step S101), and a preamble signal is assigned to each of these signals (step S102). Then, the modulation process is performed (step S103). The modulated signal is subjected to unitary conversion (step S104), and the unitary converted signal is converted to a radio frequency by the radio unit and transmitted (step S105).

FIG. 20 shows a reception flow of the second radio station in the prior art. When the receiving station receives a radio packet (step S110), it detects a preamble (step S111) and performs channel estimation (step S112). Here, all transfer functions between the transmission antennas and the reception antennas are acquired. A Hermitian conjugate matrix H ^H and a matrix H ^H × H (= Λ), which is a product of these, are obtained by calculation with respect to the transfer function matrix H having this transfer function as each component (step S113).

On the other hand, a 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 (step S114). On the other hand, the matrix operation H ^H × Rx is performed, and each component of the obtained vector is further divided by the diagonal components λ1 to λ3 of the matrix Λ (step S115). Based on this result, hard decision processing is performed (step S116), and the signal estimation transmitted from each transmitting antenna of the corresponding symbol is determined (step S117). Further, when the received data continues, the process returns to the processing step S114, and the processing steps S114 to S117 are repeated. When the received data is finished (step S118), the series of received data of each system is reconstructed, the data on the transmission side is reproduced, and the data is output (step S119).
S. Kurosaki, Y. Asai, T. Sugiyama and M. Umehira, “A SDM-COFDM Scheme Employing a Simple Feed-Forward Inter-Channel Interference Canceller for MIMO Based Broadband Wireless LANs”, IEICE TRANS. COMMUN., Vol.E86 B No.1, January, 2003 Miyashita et al. “Eigenbeam Space Division Multiplexing (E-SDM) System in MIMO Channel”, IEICE Technical, RCS2002-53, May 2002

The biggest problem with this E-SDM method is that channel estimation must be performed accurately on the transmitting station side, and it must be operated in a situation where the off-diagonal components of U ^H × H ^H × H × U are clearly zero. It is a point that must not be.
For example, consider a case where radio packets are continuously transmitted. The transmitting station first performs channel estimation by some method to obtain a transfer function matrix H, and obtains a unitary matrix U based on this. Of course, the same unitary conversion is applied to radio packets transmitted continuously.

However, since the transfer function matrix H fluctuates with time, the off-diagonal component of U ^H × H ′ ^H × H ′ × U does not clearly become zero for the transfer function matrix H ′ after the change. Since this non-diagonal component becomes inter-channel interference, the characteristics are rapidly deteriorated due to the influence of the interference signal. Therefore, good characteristics cannot be expected unless the calculated unitary transformation U has very high accuracy.

Further, in this E-SDM system, the signals of each signal sequence have characteristics corresponding to eigenvalues λ _{1 to} λ _N (matrix H ^H × H is assumed to be a square matrix of N rows and N columns). In general, individual modulation schemes are adopted, or transmission power distribution is adjusted by a technique called a water injection theorem on the transmission side. In other words, it is assumed that the characteristics of each sequence can be efficiently extracted by handling the signal sequences independently. However, such control is complicated, and negotiation between the transmitting and receiving stations may be required. As a simple method for avoiding these, it is conceivable to transmit signals of all signal sequences with the same output and using a single modulation mode. However, in this case, the variation of the values of the eigenvalues is large, and the characteristics of the signal series having the minimum eigenvalue in particular are greatly deteriorated.

  The above is a case where E-SDM is assumed. Here, let us consider general characteristics of the MIMO technique combined with the OFDM modulation scheme. When OFDM modulation is used, quality variations for each signal sequence occur for each of a plurality of subcarriers. 21 and 22 show the frequency dependence of the characteristics of each signal sequence when OFDM modulation and MIMO technology are used in combination. In each figure, the horizontal axis represents frequency, and effectively means a subcarrier number. The vertical axis represents BER (Bit Error Rate) corresponding to transmission quality.

  For example, when E-SDM is used or when the correlation between frequencies is strong, there is a large difference in the characteristics of each signal series as shown in FIG. For example, a signal sequence # 1 having a high BER in the entire frequency domain has a partial packet error rate (PER) of 20% and a signal sequence having a low BER in the entire frequency domain. Assume that the partial PER with a code error in # 2 is 0.0001%. When the number of MIMO superpositions is 2, the average PER of the entire system is the sum of the PER of each signal series divided by 2, which is approximately 10%. That is, if there is a signal sequence with poor quality, the PER of the entire system is pulled by this.

On the other hand, as shown in FIG. 22, a case is considered in which the characteristics of the signal sequence # 1 and the signal sequence # 2 are reversed at a certain frequency, and each of them is significantly degraded in a specific frequency region. In this way, in the frequency region where the characteristics are degraded, the error before error correction is bursty, so that there is a high probability that the error cannot be corrected even if error correction processing is performed. That is, even if it is partial when viewed from the whole, if the quality continuously deteriorates over a certain interval on the frequency axis, the effect of error correction is diminished, and the partial PER of each signal sequence is reduced. It will appear as deterioration of characteristics.
As described above, when OFDM modulation and MIMO technology are used in combination, particularly when the E-SDM system is used, in order to improve the PER characteristic of the entire system, a code error of a specific signal sequence is considered to be bursty. Rather, it is preferably randomized.

The present invention has been made in view of such circumstances, and when performing wireless communication using MIMO technology, a unitary transformation matrix with low accuracy is realized while realizing good characteristics of the E-SDM method. An object of the present invention is to easily provide a wireless communication device, a wireless communication method, and a wireless communication system that can realize stable characteristics even when used.
In addition, when performing wireless communication using both OFDM modulation and MIMO technology, it is possible to simplify the wireless communication method and the wireless communication apparatus that can randomize the code error of each signal sequence and efficiently improve the PER characteristics of the entire system. There is to provide to. In addition, this device for randomization does not change the air interface rules for communication, but it is still preferable for product differentiation if it can be handled only by changes on the transmitting terminal side. .

In order to solve the above problem, the present invention is a wireless communication apparatus used in a wireless communication system configured by a first wireless station and a second wireless station,
The first radio station includes N _tx (N _tx is an integer greater than 1) or more first antenna groups,
Dividing means for dividing user data to be transmitted into N systems (N _tx ≧ N> 1, N is an integer);
Means for constructing a wireless data packet composed of N series of signal sequences obtained by giving signals of N types of known patterns to the user data divided into the N series;
The process of multiplying each of the N signal series including the known pattern by an individual coefficient is performed a plurality of times with different combinations of coefficients, and the N signal series multiplied by the coefficient is synthesized for each combination of coefficients. Conversion means for converting the N signal series to the N _tx system signal series;
A wireless communication apparatus comprising: a transmission unit configured to transmit the signal converted by the conversion unit from the first antenna group.

As a preferred embodiment,
The second radio station includes M (M is an integer greater than 1) second antenna groups,
Means for obtaining a transfer function h _{j, i} or an approximation thereof between an i-th antenna in the first antenna group and a j-th antenna in the second antenna group;
Means for calculating a Hermitian conjugate matrix H ^H of the matrix H with respect to a matrix H of M rows and N columns with the transfer function h _{j, i} as the (j, i) component;
Means for calculating a matrix product of the two matrices, that is, a matrix H ^H × H of N rows and N columns;
Means for calculating a unitary matrix U for diagonalizing the matrix H ^H × H,
The converting means includes
If the information of the k-th (k is an integer of 1 or more) symbol of the N signal series is {x ₁ (k), x ₂ (k),..., X _N (k)}, means for calculating a column vector given by the product of the column vector x (k) and the matrix U, each element of which is k symbol information, that is, U × x (k);
The transmission means includes
Means for transmitting the i-th row component [U × x (k)] _i of the column vector U × x (k) from the i-th antenna in the first antenna group;
In this case, in the application of the MIMO technique, it is possible to provide a simple means for realizing the transmission side technique for extracting the reception diversity gain on the receiving station side to the maximum while using a simple ZF method or the like.

As a suitable example,
Prior to transmitting the wireless data packet, means for transmitting a first wireless control packet containing control information;
Means for receiving a second radio control packet that is a response to the first radio control packet using N or more antennas;
The acquisition means uses a plurality of known patterns of signals assigned to the second radio control packet, and uses the jth (1 ≦ 1) used by the second radio station for transmission from the received signal at each antenna. j ≦ M, j is an integer) includes means for calculating a transfer function h _{j, i} between the antenna and the i-th (1 ≦ i ≦ N, i is an integer) antenna used by the first radio station for reception it can.
This function on the first radio station side provides a simple implementation method for acquiring a transfer function matrix on the transmission side.

As another preferred example, prior to transmission of the wireless data packet, means for transmitting a signal sequence including N known patterns as a first wireless control packet using the N first antenna groups;
Means for receiving a second radio control packet that is a response to the first radio control packet;
Obtain information on the transfer function h _{j, i} between the i-th antenna of the first antenna group and the j-th antenna of the second antenna group accommodated in the second radio control packet. Means to
Can be included.

In the communication between the radio stations, an orthogonal frequency division multiplexing (OFDM) modulation method using K subcarriers (K is an integer greater than 1) may be used.
As a preferred example in this case, the second radio station accommodates a radio control packet and / or user data that contains control information composed of only one signal sequence in which a plurality of signal sequences are not superimposed. When a radio data packet is received, the second radio station of the ks-th subcarrier (1 ≦ ks ≦ K, ks is an integer) from the reception state of the signal of the known pattern received by each receiving antenna is used for transmission. Transfer function h _{j, i} ^[ks] between the _{j th} antenna (1 ≦ j ≦ M, j is an integer) and the i th (1 ≦ i ≦ N, i is an integer) antenna used for reception by the first radio station ^] Is included.
As another preferred example, the second radio station in the ks ′ (1 ≦ ks ′ ≦ K, ks ′ is an integer) subcarrier that has not been transmitted by the jth antenna of the second radio station. The transfer function h _{j, i} ^{[ks ′]} between the j antenna and the i-th antenna of the first radio station is expressed as ks ₁ ≠ ks ′ and ks ₂ ≠ transmitted by the j-th antenna of the second radio station. _{ks' (1 ≦ ks 1 ≦} K, 1 ≦ ks 2 ≦ K, ks 1, ks 2 is an integer) becomes the transfer function h _j for the first ks ₁ subcarrier and the ks _{2 subcarrier, i} ^[ks1] and h including _{j, i} ^[ks2] h _j by interpolation or extrapolation value of _the transfer function acquisition unit for obtaining _i ^[ks'].
This provides a simple implementation method for obtaining a transfer function matrix on the transmission side. In addition, when using this method, even when extending the MIMO technology to an existing wireless LAN system, it is necessary to change the existing frame format or add a new special control message. In addition, it is possible to obtain the effect that it can be realized while maintaining backward compatibility.
In these preferred examples, there may be provided means for notifying the second wireless station side that the first wireless station side has the transfer function obtaining means.

The present invention is also a wireless communication device for receiving a wireless signal from the wireless communication device in the second wireless station,
The second radio station includes M (M is an integer of 1 or more) second antenna groups,
Means for individually receiving radio signals using the second antenna group;
Demodulating means for separating and demodulating signals of each received signal sequence using a known pattern signal given to the received signal as a reference signal;
There is also provided a wireless communication device comprising: output means for synthesizing all demodulated signal sequences and outputting them as user data.
The present invention also provides a wireless communication system having the above-described transmission-side wireless communication device and the above-described reception-side wireless communication device.

As a typical example, the demodulation means includes:
A transfer function h _j, between the i-th antenna in the first antenna group and the j-th antenna in the second antenna group, with a known pattern signal given to the received signal as a reference signal means for obtaining _i ;
Arithmetic means for performing a predetermined operation on the matrix H of M rows and N columns using the transfer function h _{j, i} as the (j, i) component, and obtaining a vector corresponding to the signal points of the N transmission signals; Including
The output means includes
Means for synthesizing N-system transmission signals given by the elements of the vector obtained by the calculation for all received symbols and outputting the synthesized data as the user data;

The computing means is
Means for calculating a Hermitian conjugate matrix H ^H of the matrix H with respect to a matrix H of M rows and N columns with the transfer function h _{j, i} as the (j, i) component;
Means for calculating a matrix product of the two matrices, that is, a matrix H ^H × H of N rows and N columns;
Means for calculating an inverse matrix of the matrix H ^H × H, ie, (H ^H × H) ⁻¹ ;
Furthermore, a means for calculating a matrix (H ^H × H) ⁻¹ × H ^H using these,
When r _m (k) is a received signal of the k-th symbol actually received by the m-th antenna of the second antenna group, (r ₁ (k), r ₂ (k),..., R _M (k)) For a received signal vector Rx (k) represented by ^T (T represents conversion from a row vector to a column vector), (H ^H × H) ⁻¹ × H ^H × Rx (k) Means for computing
May be included.

Further, the calculation means includes
Means for calculating an inverse matrix H ⁻¹ of the matrix H for an N-row N-column matrix H with the transfer function h _{j, i} as the (j, i) component;
When r _m (k) is the received signal of the k-th symbol actually received by the m-th antenna of the second antenna group, (r ₁ (k), r ₂ (k),..., R _M (k)) means for calculating H ⁻¹ × Rx (k) for a received signal vector Rx (k) represented by ^T (T represents conversion from a row vector to a column vector);
May be included.
The function of the second radio station is to provide means for easily performing the ZF method on the receiving station side using the fact that the matrix is a square matrix.

Further, the calculation means includes
The transmission signal of the k-th symbol transmitted from the n-th antenna of the first antenna group is set to t _n (k), and the reception of the k-th symbol actually received by the m-th antenna of the second antenna group. When the signal is expressed as r _m (k), (t ₁ (k), t ₂ (k),..., T _N (k)) ^T (T represents conversion from a row vector to a column vector) , (R ₁ (k), r ₂ (k),..., R _M (k)) received signal vector Rx (k) and matrix represented by ^T For the operator F, a vector F × Rx (k) −Tx (k) and a Hermitian conjugate vector of the vector (F × Rx (k) −Tx (k)) A vector product of ^H (F × Rx (k) It means for selecting a matrix operator F as is expected to minimize the expected value across multiple symbols of ^{-Tx (k)) H × (} F × Rx (k) -Tx (k)),
Means for computing F × Rx (k) for each symbol;
May be included.
The function on the second wireless station side is to provide means for using the MMSE method instead of the ZF method on the receiving station side.
The matrix operator F may be obtained by a MMSE (Minimum Mean Square Error) method.

As a preferred example of the wireless communication device,
On the second radio station side,
When receiving the first radio control packet from the first radio station,
As a response to the first radio control packet, there is means for transmitting a second radio control packet including a plurality of known patterns.

As another preferred example,
On the second radio station side,
When receiving a first radio control packet including known patterns of N systems from the first radio station,
Means for calculating the transfer function h _{j, i} using signals of known patterns of the N systems;
Means for transmitting a second radio control packet containing information on the transfer function as a response to the first radio control packet;
Have

In the communication between the radio stations, an orthogonal frequency division multiplexing (OFDM) modulation method using K subcarriers (K is an integer greater than 1) may be used.
In this case, on the second radio station side,
When transmitting the radio control packet or the radio data packet to the first radio station without superimposing a plurality of signal sequences, the second antenna corresponding to the subcarrier number ks is used as the ks subcarrier signal. You may have a transmission means to transmit using predetermined one antenna in a group.
Here, a means for notifying the first wireless station side that the second wireless station side has the transmission means may be provided.

The present invention also provides a wireless communication method for performing communication between a first wireless station and a second wireless station,
The first radio station includes N _tx (N _tx is an integer greater than 1) or more first antenna groups,
A division step of dividing user data to be transmitted into N systems;
Constructing a wireless data packet composed of N system signal sequences obtained by giving signals of N types of known patterns to the user data divided into the N systems;
The process of multiplying each of the N signal series including the known pattern by an individual coefficient is performed a plurality of times with different combinations of coefficients, and the N signal series multiplied by the coefficient is synthesized for each combination of coefficients. A conversion step for converting the N system signal sequences into N _tx system signal sequences;
And a transmission step of transmitting the signal converted in the conversion step from the first antenna group.

As a typical example, the second radio station includes M (M is an integer greater than 1) second antenna groups,
Obtaining a transfer function h _{j, i} or an approximation thereof between an i-th antenna in the first antenna group and a j-th antenna in the second antenna group;
Calculating an Hermitian conjugate matrix H ^H of the matrix H with respect to an M-row N-column matrix H having the transfer function h _{j, i} as the (j, i) component;
Calculating a matrix product of the two matrices, i.e., an N-by-N matrix H ^H × H;
Calculating a unitary matrix U for diagonalizing the matrix H ^H × H,
The converting step includes
If the information of the k-th (k is an integer of 1 or more) symbol of the N signal series is {x ₁ (k), x ₂ (k),..., X _N (k)}, calculating a column vector given by the product of a column vector x (k) having each element of k symbols as an element and the matrix U, that is, U × x (k),
The transmitting step includes
Transmitting the i-th row component [U × x (k)] _i of the column vector U × x (k) from the i-th antenna in the first antenna group.

As a preferred example, prior to transmitting the wireless data packet, transmitting a first wireless control packet containing control information;
Receiving a second radio control packet that is a response to the first radio control packet using N or more antennas, and
The acquisition step uses a plurality of known patterns of signals assigned to the second radio control packet, and uses a received signal at each antenna for the jth (1 ≦ 1) used by the second radio station for transmission. j ≦ M, j is an integer) and includes a step of calculating a transfer function h _{j, i} between the antenna and the i-th (1 ≦ i ≦ N, i is an integer) antenna used by the first radio station.

As another preferred example, prior to transmission of the wireless data packet, a signal sequence including N known patterns as a first wireless control packet is transmitted using N first antenna groups;
Receiving a second radio control packet that is a response to the first radio control packet;
Obtain information on the transfer function h _{j, i} between the i-th antenna of the first antenna group and the j-th antenna of the second antenna group accommodated in the second radio control packet. And steps to
including.

In the communication between the radio stations, an orthogonal frequency division multiplexing (OFDM) modulation method using K subcarriers (K is an integer greater than 1) may be used.
As a preferred example in this case, the second radio station accommodates a radio control packet and / or user data that contains control information composed of only one signal sequence in which a plurality of signal sequences are not superimposed. When a radio data packet is received, the second radio station of the ks-th subcarrier (1 ≦ ks ≦ K, ks is an integer) from the reception state of the signal of the known pattern received by each receiving antenna is used for transmission. Transfer function h _{j, i} ^[ks] between the _{j th} antenna (1 ≦ j ≦ M, j is an integer) and the i th (1 ≦ i ≦ N, i is an integer) antenna used for reception by the first radio station ^] Is included.
As another preferred example, the second radio station in the ks ′ (1 ≦ ks ′ ≦ K, ks ′ is an integer) subcarrier that has not been transmitted by the jth antenna of the second radio station. The transfer function h _{j, i} ^{[ks ′]} between the j antenna and the i-th antenna of the first radio station is expressed as ks ₁ ≠ ks ′ and ks ₂ ≠ transmitted by the j-th antenna of the second radio station. _{ks' (1 ≦ ks 1 ≦} K, 1 ≦ ks 2 ≦ K, ks 1, ks 2 is an integer) becomes the transfer function h _j for the first ks ₁ subcarrier and the ks _{2 subcarrier, i} ^[ks1] and h _It includes a transfer function acquisition step of acquiring h _{j, i} ^[ks'] by interpolation or extrapolation value of _{j, i} ^[ks2] .
In these preferred embodiments, the method may further include a step of notifying the second radio station side that the transfer function obtaining step is executed on the first radio station side.

The present invention is also a wireless communication method for receiving at the second wireless station a wireless signal transmitted by the wireless communication method,
The second radio station includes M (M is an integer of 1 or more) second antenna groups,
Individually receiving radio signals using the second antenna group;
A demodulation step for separating and demodulating signals of each received signal sequence using a known pattern signal given to the received signal as a reference signal;
An output step of combining all demodulated signal sequences and outputting as user data is provided.

As a typical example, the demodulation step includes:
A transfer function h _j, between the i-th antenna in the first antenna group and the j-th antenna in the second antenna group, using a known pattern signal given to the received signal as a reference signal obtaining _i ;
An operation step of performing a predetermined operation on the matrix H of M rows and N columns using the transfer function h _{j, i} as the (j, i) component, and obtaining a vector corresponding to signal points of N transmission signals; Including
The output step includes
The method includes a step of synthesizing N-system transmission signals given by each element of the vector obtained by the calculation for all received symbols and outputting the synthesized data as the user data.

As a preferred example, the calculating step includes:
Calculating an Hermitian conjugate matrix H ^H of the matrix H with respect to an M-row N-column matrix H having the transfer function h _{j, i} as the (j, i) component;
Calculating a matrix product of the two matrices, i.e., an N-by-N matrix H ^H × H;
Calculating an inverse matrix of the matrix H ^H × H, ie, (H ^H × H) ⁻¹ ;
Further, using these, a matrix (H ^H × H) ⁻¹ × H ^H is calculated,
When r _m (k) is a received signal of the k-th symbol actually received by the m-th antenna of the second antenna group, (r ₁ (k), r ₂ (k),..., R _M (k)) For a received signal vector Rx (k) represented by ^T (T represents conversion from a row vector to a column vector), (H ^H × H) ⁻¹ × H ^H × Rx (k) A step of calculating
including.

As another preferable example, the calculation step includes:
Calculating an inverse matrix H ⁻¹ of the matrix H for an N-row N-column matrix H having the transfer function h _{j, i} as the (j, i) component;
When r _m (k) is the received signal of the k-th symbol actually received by the m-th antenna of the second antenna group, (r ₁ (k), r ₂ (k),..., R _M (k)) calculating H ⁻¹ × Rx (k) for a received signal vector Rx (k) represented by ^T (T represents conversion from a row vector to a column vector);
including.

As another preferable example, in the calculating step, a transmission signal of the k-th symbol transmitted from the n-th antenna of the first antenna group is set to t _n (k), and further, the m- _th symbol of the second antenna group. When the received signal of the k-th symbol actually received by the antenna is expressed as r _m (k), (t ₁ (k), t ₂ (k),..., T _N (k)) ^T (T Represents a transmission signal vector Tx (k), (r ₁ (k), r ₂ (k),..., R _M (k)) ^T Received signal vector Rx (k) and matrix operator F, vector F × Rx (k) −Tx (k) and Hermitian conjugate vector (F × Rx (k) −Tx (k) of the vector) ) ^H vector product of (F × Rx (k) -Tx (k)) to be expected to minimize the expected value across multiple symbols of ^{H × (F × Rx (k} ) -Tx (k)) Selecting a matrix operator F for
Calculating F × Rx (k) for each symbol;
including.
In this case, the matrix operator F may be obtained by an MMSE (Minimum Mean Square Error) method.

As another preferred example, on the second radio station side,
When receiving a first radio control packet including known patterns of N systems from the first radio station,
Calculating the transfer function h _{j, i} using signals of the N patterns of known patterns;
As a response to the first radio control packet, transmitting a second radio control packet containing information on the transfer function;
Run

As another typical example, on the second radio station side,
When receiving the first radio control packet from the first radio station,
As a response to the first radio control packet, a step of transmitting a second radio control packet including a plurality of known patterns is executed.

As another typical example, an Orthogonal Frequency Division Multiplexing (OFDM) modulation scheme using K subcarriers (K is an integer greater than 1) is used in communication between the wireless stations.
In this case, on the second radio station side,
When transmitting the radio control packet or the radio data packet to the first radio station without superimposing a plurality of signal series, the second antenna corresponding to the subcarrier number ks is used as the ks subcarrier signal. You may perform the transmission step which transmits using predetermined one antenna in a group.
Here, there may be a step of notifying the first radio station side that the transmission step is executed on the second radio station side.

As another typical aspect of the wireless communication apparatus of the present invention,
A wireless communication device used in a wireless communication system that performs wireless communication using an orthogonal frequency division multiplexing (OFDM) modulation scheme using a plurality of subcarriers,
The dividing means divides user data to be transmitted into N systems individually for each subcarrier,
The conversion means is configured so that information of n _s (n _s is an integer of 1 or more) symbols of N signal sequences in a certain subcarrier is {x ₁ ( _ns ), x ₂ ( _ns ),. .., X _N (n _s )} and the column vector of N rows having these as each component is x (n _s ), the unit matrix of N rows and N columns or the column of the unit matrix is appropriately replaced. N _R (N _R > 1: N _R is an integer) type rotation matrix group obtained as {R _k } (N _R ≧ k ≧ 1: k is an integer) corresponds to each subcarrier. with _{R k} for a given k, comprising at least means for converting the vector x a _{(n s)} in _{R k × x (n s)} ,
The transmission means transmits the i-th component of the converted column vector in each subcarrier from the i-th antenna of the first antenna group.

As a preferred example, the conversion means converts the signal of the n _s symbol transmitted from the i-th antenna of the first antenna group in each subcarrier so as to be [R _k × x ( _ns )] _i. To do.
With such a configuration, when using the MIMO technology, the antenna used for transmitting each signal sequence is replaced for each subcarrier, and as a result, errors for each signal sequence are averaged and randomized, and coding for error correction is performed. Improve gain. In the conventional scheme, for a predetermined k corresponding to each subcarrier, a rotation matrix R _k is used for a signal of the n _s symbol transmitted from the i-th antenna of the first antenna group in the sub-carrier. [R _k × x (n _s )] _i and the converted signal [R _k × x (n _s )] _i in each subcarrier is converted into the _{i th} of the first antenna group. It differs in that it transmits from the antenna.
As a result, when performing highly efficient wireless communication using the MIMO technique, it is possible to equalize reception characteristics for each of a plurality of superimposed signal sequences and to randomize a code error on the transmission path. In particular, when combined with the E-SDM system, the MIMO technique has a tendency that a large variation in reception characteristics occurs between superimposed signal sequences. As a result, it is possible to improve the error correction gain and improve the packet error rate characteristics as a whole.

As another preferable example, the second radio station includes M (M is an integer of 1 or more) second antenna groups,
For each subcarrier individually,
Means for obtaining a transfer function h _{j, i} or an approximate value thereof between an i-th antenna in the first antenna group and a j-th antenna in the second antenna group;
Means for calculating a Hermitian conjugate matrix H ^H of the matrix H with respect to a matrix H of M rows and N columns with the transfer function h _{j, i} as the (j, i) component;
Means for calculating a matrix product of the two matrices, ie, a square matrix H ^H × H of N rows and N columns;
Means for calculating an N-by-N unitary matrix U for diagonalizing the matrix H ^H × H;
Further comprising
The converting means converts the signal of the _nsth symbol transmitted from the i-th antenna of the first antenna group in each subcarrier so as to be [U × R _k × x ( _ns )] _i. To.
As a result, when using the E-SDM scheme in combination with the MIMO technique, a large difference occurs in the characteristics of the superimposed signal sequences, whereas this characteristic is made uniform, and as a result, errors for each signal sequence are averaged and It is possible to provide a simple implementation method for randomizing and improving the coding gain in error correction.

As another preferable example, the second radio station includes M (M is an integer of 1 or more) second antenna groups,
N _tx > N,
For each subcarrier individually,
Means for obtaining a transfer function h _{j, i} or an approximate value thereof between an i-th antenna in the first antenna group and a j-th antenna in the second antenna group;
Means for calculating a Hermitian conjugate matrix H ^H of the matrix H with respect to a matrix H of M rows and N _tx columns, wherein the transfer function h _{j, i} is the (j, i) component;
Means for calculating a matrix product of the two matrices, that is, a square matrix H ^H × H of N _tx rows and N _tx columns;
Means for calculating a unitary matrix U of N _tx rows and N _tx columns for diagonalizing the matrix H ^H × H;
Further comprising
When the conversion means represents a matrix of N _tx rows and N columns and an integer j where N ≧ j ≧ 1, only the (j, j) component is 1 and the other components are 0, T the in each subcarrier first antenna group of the i antenna: [U × T × a signal of the _{n s} symbols to be transmitted from the _{(n tx} ≧ i ≧ 1 i integer) _{R k} × x _{(n s} )] Convert to _i .
With such a configuration, the signal of the n _sth symbol transmitted from the i-th antenna (N _tx ≧ i ≧ 1: i is an integer) of the first antenna group for a predetermined k corresponding to each subcarrier. , A unitary matrix U for diagonalizing H ^H × H of the transfer function matrix H, a rotation matrix R _k for exchanging antennas, and a matrix T for expanding the number of antennas, [U × T × R _k × x (n _s )] _i , and the converted signal [U × T × R _k × x (n _s )] _i is transmitted from the i-th antenna of the first antenna group.
As a result, the number of transmission antennas is increased over the number of signal sequences to be superimposed, and the above-mentioned processing is performed using only the channels with good characteristics by discarding the poor-quality channels from the MIMO channels existing by the number of transmission antennas. A simple realization method can be provided.

As a typical example, the rotation matrix group {R _k } is an integer satisfying N> j ≧ 1, where R ₁ is one matrix selected from matrices obtained by appropriately exchanging N _× N unit matrix columns. A matrix of N rows and N columns where only the (j + 1, j) and (1, N) components are 1 and the other components are 0 with respect to j is an integer such that N ≧ k ≧ 2. It is composed of a total of N matrices given as R _k = P ^k−1 × R ₁ for ^k ,
The rotation matrix R _k used for the subcarriers in which user information is accommodated corresponds to the rotation matrix group {R _k } appropriately rearranged in order of N subcarrier periods.
Thereby, a simple realization method for generating a matrix used as the above rotation matrix can be provided.

As another typical example, in the rotation matrix group {R _k }, one matrix selected from matrices obtained by appropriately exchanging columns of unit matrices of N rows and N columns is R _1, and N> j ≧ 1 A matrix of N rows and N columns in which only the (j + 1, j) and (1, N) components are 1 and the other components are 0 is P, and N ≧ k ≧ 2 It is composed of a total of N matrices given as R _k = P ^k−1 × R ₁ for an integer k
When mapping an OFDM-modulated signal on subcarriers, interleave processing is performed so that the bit sequences of the signal sequences divided into the N systems are adjacent to each other in a _Nil subcarrier period ( _Nil is an integer greater than 1). In the case of applying, the rotation matrix R _k used for the subcarriers in which user information is accommodated is appropriately rearranged by preparing _Nil matrixes of the rotation matrix group {R _k }. Correspond in order with N _il × N subcarrier periods.
This also provides a simple implementation method for generating a matrix used as the above-described rotation matrix.

The present invention is also a wireless communication system having the above-described wireless communication device in the first wireless station,
The second radio station includes M (M is an integer of 1 or more) second antenna groups,
Means for individually receiving radio signals using the second antenna group;
Means for separating and demodulating the signals of each of the signal series, using a known pattern signal given to the received signal as a reference signal;
A radio communication system comprising: means for synthesizing all demodulated signal sequences and outputting as user data.

As another typical aspect of the wireless communication method of the present invention,
A wireless communication method used in a wireless communication system that performs wireless communication using an orthogonal frequency division multiplexing (OFDM) modulation scheme using a plurality of subcarriers,
In the dividing step, user data to be transmitted is divided into N systems individually for each subcarrier,
In the conversion step, the information of the n _s symbols (n _s is an integer equal to or greater than 1) of the N signal sequences in a certain subcarrier are {x ₁ ( _ns ), x ₂ ( _ns ),. .., X _N (n _s )} and the column vector of N rows having these as each component is x (n _s ), the unit matrix of N rows and N columns or the column of the unit matrix is appropriately replaced. N _R (N _R > 1: N _R is an integer) type rotation matrix group obtained as {R _k } (N _R ≧ k ≧ 1: k is an integer) corresponds to each subcarrier. with _{R k} for a given k, comprises at least the step of converting the vector x a _{(n s)} in _{R k × x (n s)} ,
The transmission means transmits the i-th component of the converted column vector in each subcarrier from the i-th antenna of the first antenna group.

As a preferred example, in the conversion step, the signal of the _nsth symbol transmitted from the i-th antenna of the first antenna group in each subcarrier is converted to be [R _k × x ( _ns )] _i. To do.

As another preferable example, the second radio station includes M (M is an integer of 1 or more) second antenna groups,
For each subcarrier individually,
Obtaining a transfer function h _{j, i} or an approximation thereof between the i-th antenna of the first antenna group and the j-th antenna of the second antenna group;
Calculating an Hermitian conjugate matrix H ^H of the matrix H with respect to an M-row N-column matrix H having the transfer function h _{j, i} as the (j, i) -th component;
Calculating a matrix product of the two matrices, that is, a N × N square matrix H ^H × H;
Calculating an N-row N-column unitary matrix U for diagonalizing the matrix H ^H × H;
Further comprising
In the conversion step, the signal of the n _s symbol transmitted from the i-th antenna of the first antenna group in each subcarrier is converted to be [U × R _k × x ( _ns )] _i .

As another preferred example, the second radio station includes M (M is an integer of 1 or more) second antenna groups,
N _rx > N,
For each subcarrier individually,
Obtaining a transfer function h _{j, i} or an approximation thereof between the i-th antenna of the first antenna group and the j-th antenna of the second antenna group;
Calculating a Hermitian conjugate matrix H ^H of the matrix H with respect to a matrix H of M rows and N _tx columns having the transfer function h _{j, i} as the (j, i) component;
Calculating a matrix product of the two matrices, that is, a square matrix H ^H × H of N _tx rows and N _tx columns;
Calculating an N _tx row N _tx column unitary matrix U for diagonalizing the matrix H ^H × H;
Further comprising
In the conversion step, when a matrix of N _tx rows and N columns and an integer j where N ≧ j ≧ 1, only the (j, j) component is 1 and the other components are 0 is expressed as T. the in each subcarrier first antenna group of the i antenna: [U × T × a signal of the _{n s} symbols to be transmitted from the _{(n tx} ≧ i ≧ 1 i integer) _{R k} × x _{(n s} )] Convert to _i .

Typically, the second radio station includes M (M is an integer of 1 or more) second antenna groups,
Receiving wireless signals individually using the second antenna group; and
A step of separating and demodulating the signals of each of the signal series using a known pattern signal given to the received signal as a reference signal;
And synthesizing all demodulated signal sequences and outputting them as user data.

As described above in detail, according to the present invention, when performing a highly efficient wireless communication using the MIMO technology, if the transfer function matrix can be estimated with high accuracy, a good equivalent to the E-SDM method is obtained. While realizing the characteristics, it is possible to obtain an effect that it is possible to show stable characteristics even when the transfer function matrix cannot be accurately estimated.
In addition, when using the MIMO technique for the OFDM modulation scheme, by replacing the antenna used for transmitting each signal sequence for each subcarrier, the error for each signal sequence is averaged and randomized, and the coding gain in error correction is increased. Can be improved.
For example, when performing highly efficient wireless communication using MIMO technology, it is possible to equalize reception characteristics for each of a plurality of superimposed signal sequences and to randomize code errors on the transmission path. In particular, when combined with the E-SDM system, the MIMO technique has a tendency that a large variation in reception characteristics occurs between superimposed signal sequences. As a result, it is possible to improve the error correction gain and improve the packet error rate characteristics as a whole.
Further, when using the E-SDM system in combination with the MIMO technique, a large difference is usually generated in the characteristics of the superimposed signal sequences, but this characteristic is made uniform, and as a result, errors for each signal sequence are averaged and It is possible to provide a simple implementation method for randomizing and improving the coding gain in error correction.
Also, the number of transmission antennas is increased from the number of signal sequences to be superimposed, the channel having poor characteristics is cut out from the MIMO channels existing by the number of transmission antennas, and the above processing is performed using only the channel having good characteristics. You can also

The block diagram which shows the structural example of the receiving part of the 2nd radio station in the radio | wireless communications system which concerns on 1st Embodiment of this invention. The block diagram which shows the structural example of the transmission part of the 1st and 2nd radio station in the radio | wireless communications system which concerns on the embodiment. FIG. 3 is a block diagram showing a configuration example of a transmission unit of the first and second radio stations in the radio communication system according to the embodiment. 9 is a flowchart showing the contents of a reception process of a second radio station in the radio communication system according to the embodiment. 7 is a flowchart showing the contents of channel estimation processing in a first wireless station in the wireless communication method according to the embodiment. 9 is a flowchart showing processing contents when a radio control packet is received by a second radio station in the radio communication method according to the embodiment. 7 is a flowchart showing the contents of channel estimation processing in a first wireless station in the wireless communication method according to the embodiment. 9 is a flowchart showing processing contents when a radio control packet is received by a second radio station in the radio communication method according to the embodiment. The block diagram which shows the structural example of the transmission / reception part of the radio station in the radio | wireless communications system which concerns on the same embodiment. The figure which shows the structural example of the transmission part of the 1st radio station which concerns on 2nd Embodiment of this invention. The figure which shows the structural example of the receiving part of the 2nd radio station which concerns on the same embodiment. The figure which shows the 2nd structural example of the transmission part of the 1st radio station which concerns on the same embodiment. The figure which shows the 3rd structural example of the transmission part of the 1st radio station which concerns on the embodiment. The figure which shows the transmission flow of the 1st radio station which concerns on the same embodiment. The figure which shows the 2nd transmission flow of the 1st radio station which concerns on the same embodiment. The figure which shows the 3rd transmission flow of the 1st radio station which concerns on the same embodiment. The figure which shows the structure of the transmission part of the 1st radio station in a prior art. The figure which shows the structure of the receiving part of the 2nd radio station in a prior art. The figure which shows the transmission flow of the 1st radio station in a prior art. The figure which shows the reception flow of the 2nd radio station in a prior art. FIG. 1 shows the frequency dependence of the characteristics of each signal sequence when OFDM modulation and MIMO technology are used together. FIG. 2 shows the frequency dependence of the characteristics of each signal sequence when OFDM modulation and MIMO technology are used together.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Data division circuit 2-1 to 2-3 ... Preamble provision circuit 3-1 to 3-3 ... Modulation circuit 4 ... Transmission signal conversion circuit # 1
5-1 to 5-4... Wireless units 6-1 to 6-4... Antenna 7... Channel estimation circuit 8.
10: Matrix operation circuit # 2
11: Transmission signal conversion circuit # 2
12 ... Transmission signal conversion circuit # 3
13: Transmission signal conversion circuit # 4
21-1 to 21-3 ... antennas 22-1 to 22-3 ... radio unit 23 ... channel estimation circuit 24 ... received signal management circuit 25 ... transfer function matrix management circuit 26 ... matrix operation circuit (reception) # 1
27: Matrix operation circuit (reception) # 2
28 ... Hard decision circuit 29 ... Data synthesis circuit 31-a to 31-b ... Reception antenna 32-a to 32-b ... Radio unit 33-a to 33-b ... FFT circuit 34 ... Channel estimation circuit 35 ... Channel separation circuit 36-1-a to 36-Ka, 36-1-b to 36-Kb ... subcarrier demodulation circuits 37-a to 37-b ... P / S conversion circuit 38 ... data synthesis circuit 39 ... transfer function Complement circuit 40 ... transfer function management circuit # 2
41 ... Matrix operation circuit 42 ... Data division circuit 43-a to 43-b ... S / P conversion circuit 44-a to 44-b ... Preamble applying circuit 45-1-a to 45-Ka, 45-1- b to 45-Kb ... subcarrier modulation circuit 46 ... transmission signal conversion circuits 47-a to 47-b ... IFFT circuits 48-a to 48-b ... radio units 49-a to 49-b ... transmission antenna 51- 1 to 51-3 ... receiving antennas 52-1 to 52-3 ... wireless unit 53 ... channel estimation circuit 54 ... received signal management circuit 55 ... transfer function matrix management circuit 56 ... matrix operation circuit # 2
57. Matrix operation circuit # 3
58 ... Hard decision circuit 59 ... Data synthesis circuit 60 ... Data division circuits 61-1 to 61-3 ... Preamble applying circuits 62-1 to 62-3 ... Modulation circuit 63 ... Transmission signal conversion circuits 64-1 to 64-3 ... Radio unit 65-1 to 65-3 ... Transmitting antenna 66 ... Matrix operation circuit 67 ... Transfer function matrix management circuit 68 ... Channel estimation circuit 69 ... Control information termination circuit 70 ... Control information generation circuit

Embodiments of the present invention will be described below with reference to the drawings.
<First Embodiment>
FIG. 1 is a diagram illustrating a configuration of a receiving unit of a second radio station in the radio communication system according to the first embodiment of the present invention. A wireless communication system according to an embodiment of the present invention includes a first wireless station including N (N is an integer greater than 1) or more first antenna groups, and M (M is an integer greater than 1). And a second radio station having a second antenna group.
As an example, a case where the transmitting station transmits three systems of data using three transmitting antennas will be described as an example. In the figure, 51-1 to 51-3 are receiving antennas, 52-1 to 52-3 are radio units, 53 is a channel estimation circuit, 54 is a received signal management circuit, 55 is a transfer function matrix management circuit, and 56 is a matrix operation Circuits # 2 and 57 are matrix operation circuits # 3 and 58, a hard decision circuit, and 59 is a data synthesis circuit. The basic circuit configuration is the same as the conventional receiver configuration shown in FIG. 18, but the processing contents of the matrix calculation circuit # 2 (56) and the matrix calculation circuit # 3 (57) are different.

For example, as an example, the processing performed in the matrix operation circuit # 2 (56) is processing for obtaining the following matrix with respect to the estimated transfer function matrix H.
(1) H ^H : Hermitian conjugate matrix of transfer function matrix H
(2) H ^H × H
(3) (H ^H × H) ^-1 : Inverse matrix of matrix (2)
(4) (H ^H × H) ⁻¹ × H ^H : Product of the matrix (3) and the matrix (1) As a result of these calculations, the matrix (H ^H × H) is given to the matrix calculation circuit # 3 (57). ^-1 × H Enter ^H. Also. In the matrix operation circuit # 3 (57), a product (H ^H × H) ⁻¹ × H ^H × Rx of this matrix and the received signal vector Rx is obtained by calculation.

Similarly, as an example, the matrix calculation circuit # 2 (56) performs processing for ^obtaining an inverse matrix H ⁻¹ of the transfer function matrix H. The matrix operation circuit # 2 (56) inputs a matrix H- ¹ to the matrix operation circuit # 3 (57), and the matrix operation circuit # 3 (57) receives the matrix and the received signal vector Rx. The product H ⁻¹ × Rx is obtained by calculation.
Note that these operations ((H ^H × H) ⁻¹ × H ^H × Rx or H ⁻¹ × Rx) are solutions of the transmission signal vector Tx when the thermal noise term in (Equation 1) is negligible. This corresponds to the operation for obtaining. Therefore, it is naturally possible to obtain the transmission signal vector Tx by decomposing these operations or operations equivalent thereto into a plurality of steps and sequentially processing them.

Similarly, as an example, in the matrix operation circuit # 2 (56), for the transmission signal vector Tx (k), the reception signal vector Rx (k), and the matrix operator F, (F × Rx (k) −Tx ( k)) A process of obtaining a matrix operator F that is expected to minimize an expected value over a plurality of symbols of a physical quantity given by ^H × (F × Rx (k) −Tx (k)). The matrix calculation circuit # 2 (56) inputs the matrix F to the matrix calculation circuit # 3 (57), and the matrix calculation circuit # 3 (57) obtains the product F of this matrix and the received signal vector Rx. XRx is calculated.

This matrix operator F is known as a matrix used in the MMSE method in the conventional method, and (F × Rx (k) −Tx (k)) ^H for the preamble signal by using the preamble signal of the received signal. It is possible to obtain it by minimizing × (F × Rx (k) −Tx (k)).
Here, the hard decision circuit 58 is selectable when each component of the transmission signal vector estimated from the transfer function matrix and the reception signal vector does not coincide with the transmission signal point that originally takes a discrete value. This is for performing processing for determining a signal at a transmission signal point. This includes, for example, processing for performing error correction using a signal that has been subjected to soft decision once, and as a result, determining a signal point when performing error correction encoding / decoding.

  2 and 3 are diagrams illustrating a configuration example of the transmission unit of the first and second radio stations. In the figure, 60 is a data division circuit, 61-1 to 61-3 are preamble assignment circuits, 62-1 to 62-3 are modulation circuits, 63 is a transmission signal conversion circuit, 64-1 to 64-3 are radio units, Reference numerals 65-1 to 65-3 denote transmission antennas, 66 denotes a matrix operation circuit, 67 denotes a transfer function matrix management circuit, 68 denotes a channel estimation circuit, 69 denotes a control information termination circuit, and 60 denotes a control information generation circuit. The configuration on the transmission side may be the same as that of the conventional method, but in this figure, in order to clarify the method of obtaining the transfer function matrix H, the channel estimation circuit 68, or the control information termination circuit 69 and the control information generation circuit 70 has been added.

  In FIG. 2, when the first radio station transmits a radio data packet, a control signal for requesting transmission of a transfer function matrix estimation signal is transmitted by the control information generation circuit 70. Generate and send this. The signal at this time is not necessarily transmitted using a plurality of transmission antennas, and may be transmitted from a single antenna. In the second radio station, when the control information termination circuit 19 recognizes that the signal has been received, the control information generation circuit 20 generates a control signal, and three signals with three preamble signals are added. Transmit from transmitting antennas 65-1 to 65-3. The first radio station that has received this signal inputs the preamble signal to the channel estimation circuit 68 via the radio units 64-1 to 64-3, and acquires the transfer function matrix H by the channel estimation circuit 68. This is input to the transfer function matrix management circuit 67.

  In FIG. 3, when the first radio station transmits a radio data packet, a control signal is generated by the control information generation circuit 70 prior to the transmission of the radio data packet, and three signals with three preamble signals are added. Are transmitted from the transmitting antennas 65-1 to 65-3. The second radio station inputs the preamble signal to the channel estimation circuit 68 via the radio units 64-1 to 64-3, and acquires the transfer function matrix H by the channel estimation circuit 68. This information is input to the control information generation circuit 70, and a control signal including information related to this transfer function matrix is generated by the control information generation circuit 70 and transmitted.

  In the first radio station, when the control information termination circuit 69 recognizes that this signal has been received, information relating to the accommodated transfer function matrix is extracted and input to the transfer function matrix management circuit 67. Note that the control signal including information on the transfer function matrix is not necessarily transmitted using a plurality of transmission antennas, and may be transmitted from a single antenna.

Further, it is not always necessary to exchange the signal defined as the “first control packet” or the “second control packet” before transmitting the wireless data packet.
As an example, in the case where data is regularly transmitted and received in both directions using the MIMO technique, the transfer function matrix can be acquired every time data is received.

As another example, a case of a wireless LAN compliant with IEEE802.11a in the 5 GHz band (or IEEE802.11g in the 2.4 GHz band) will be described. In this wireless LAN system, MIMO technology is not used, and MIMO technology is expected as future expansion, but from the viewpoint of backward compatibility, even if it is an extended system, Transmission / reception of signals compliant with IEEE802.11a (or IEEE802.11g) is possible.
In this IEEE802.11a (or IEEE802.11g) compliant system, when a wireless data packet containing user data is transmitted, a receiving station that has successfully received it is relatively reliable as an ACK (Acknowledgement) signal indicating successful reception. The wireless control packet is returned using the transmission mode. Even when wireless data packets are transmitted using MIMO technology, the MIMO technology is not used for transfer of wireless control packets from the viewpoint of stable transmission and reception of control signals, that is, multiple signal sequences are transmitted on the same frequency channel. It is expected to be transferred without being superimposed on. That is, when data communication is continuously performed, it is considered that data (MIMO application) → ACK (MIMO non-application) → data (MIMO application) → ACK (MIMO non-application) →.

In this case, it is ideal that the transfer function matrix H can be estimated in the radio control packet for ACK return without using the MIMO technique. Further, it is preferable that the signal for this purpose does not change the air interface of the existing IEEE802.11a (or IEEE802.11g) compliant wireless LAN.
In order to realize this, an OFDM modulation technique employed in an IEEE802.11a (or IEEE802.11g) compliant system is used. In the OFDM modulation technique, a plurality of subcarriers are continuously arranged so as to be orthogonal on the frequency axis, and communication is performed using a fast Fourier transform technique.

When MIMO technology is applied, a transfer function matrix H is obtained for each subcarrier, and channel separation processing is performed within the same subcarrier. At this time, the value of the transfer function matrix H differs between different subcarriers, but the correlation between adjacent subcarriers becomes stronger. Using this characteristic, for example, when the transfer function between the i-th transmitting antenna and the j-th receiving antenna at subcarrier number n is h _{j, i} (n), the following is obtained by interpolation from information on adjacent subcarriers: An approximation can be made.

Here, equation (6) can be used to obtain a transfer function between h _{j, i} (n) every other subcarrier (that is, two subcarrier cycles). Equations (7) and (8) can be used to obtain a transfer function between h _{j, i} (n) every two subcarriers (that is, three subcarrier periods).

In order to use this, when transmitting a non-MIMO signal on the transmitting side, for example, even-numbered subcarriers use the first transmitting antenna, odd-numbered subcarriers use the second transmitting antenna, or subcarriers that are multiples of 3. Is a first transmission antenna, a subcarrier that is a multiple of 3 + 1 is a second transmission antenna, a subcarrier that is a multiple of 3 + 2 is transmitted using a third transmission antenna, and so on. Can be obtained, and the remaining elements can be estimated using Equation (6) to Equation (8).
The same processing can be applied even when the period is 4 subcarrier periods or more (that is, when 4 or more signal sequences are superimposed).

Further, the end subcarriers (for example, the first subcarrier, etc.) that cannot be obtained by interpolation may be approximated by other methods such as extrapolation.
The remarkable point of this technology is that there have been proposals to use this subcarrier by interpolation in the past. Jan Boer et al., “Jan Boer et. Al.,“ Backwards compatibility -How to make a MIMO- OFDM system backwards compatible and coexistence with 11a / g at the link level.- ", IEEE802.11-03 / 714r0, September, 2003".

However, these are for channel estimation at a receiving station that receives a signal on which N signal sequences are superimposed, and are used for demodulation processing of the received signal itself.
In general, since the correlation of transfer functions between adjacent subcarriers is not so strong, demodulation processing performed using this greatly deteriorates reception characteristics and is not a very practical technique. However, if this is limited to use for obtaining a unitary transformation matrix performed on the transmission side, there will be no problem in the demodulation processing even if the estimation accuracy of the unitary transformation matrix is low.

In other words, the poor estimation accuracy is merely a slight reduction in the large gain by applying the present invention, and does not fall below the characteristics when the present invention is not applied.
If not all radio stations implement this function, it is preferable to negotiate between radio stations at the start of communication and notify each other how to transmit signals when MIMO is not applied. In accordance with the negotiation result, channel estimation processing is adaptively performed using the correlation of transfer functions between adjacent subcarriers.

FIG. 4 is a diagram showing a reception flow of the second radio station in the radio communication system according to the embodiment of the present invention. Also in this figure, the difference from FIG. 20 of the conventional method is only the contents of the matrix calculation process # 2 performed in the process S74 and the matrix calculation process # 3 in the process S76.
As an example, the processing performed in the matrix operation circuit # 2 (56) sequentially obtains H ^H , H ^H × H, (H ^H × H) ⁻¹ and finally (H ^H × H) ⁻¹ × H ^H Ask for. In the matrix operation circuit # 3 (57), (H ^H × H) ⁻¹ × H ^H × Rx is obtained by calculation.

As another example, the matrix calculation circuit # 2 (56) obtains an inverse matrix H ⁻¹ of the transfer function matrix H, and the matrix calculation circuit # 3 (57) obtains H ⁻¹ × Rx by calculation. Further, in the matrix operation circuit # 2 (56), expectation over a plurality of symbols of a physical quantity given by (F × Rx (k) −Tx (k)) ^H × (F × Rx (k) −Tx (k)) A matrix operator F that is expected to minimize the value can be obtained, and F × Rx can be obtained by calculation in the matrix operation circuit # 3 (57).
On the other hand, the transmission flow of the transmission unit of the first wireless station may be basically the same as the content of FIG. 19, but as an example when channel estimation is performed on the transmission side, explain.

  FIG. 5 is a diagram showing channel estimation processing in the first radio station in the radio communication method according to the embodiment of the present invention. When a wireless data packet to be transmitted is input in the first wireless station (S21), a wireless control packet is transmitted prior to transmission of this wireless data packet (S22). This requests the second radio station to transmit a radio control packet to which a predetermined preamble signal is added from each of the plurality of transmission antennas. After the radio control packet standby processing (S23), if the radio control packet is normally received (S24), the signal is received by multiple receiving antennas (S25), and the transfer function matrix H is obtained by channel estimation. Calculate (S26). If the radio control packet cannot be normally received in the process S24, the process returns to the process S22 to retransmit the radio control packet.

FIG. 6 is a diagram illustrating a flow when a radio control packet is received by the second radio station in the radio communication method according to the embodiment of the present invention. In this wireless communication method, a first wireless station including N (N is an integer greater than 1) or more first antenna groups and M (M is an integer greater than 1) second antenna groups are provided. Communication is performed with the second radio station provided.
In FIG. 6, when a signal is received (S31), it is determined whether it is a radio control packet (S32). If it is a radio data packet, normal reception processing is performed (S37).

On the other hand, if the packet is a radio control packet (S32), the signal type is confirmed (S33), and transmission of a radio control packet with a predetermined preamble signal is requested from a plurality of transmission antennas. (S34), a plurality of transmission antennas each transmit a radio control packet to which a predetermined preamble signal is added (S35).
On the other hand, if it is normal control information (S34), normal control information processing is performed (S36).

  FIG. 7 is a diagram showing channel estimation processing in the first radio station in the radio communication method according to the embodiment of the present invention. In FIG. 7, when a wireless data packet to be transmitted is input in the first wireless station (S41), prior to transmission of this wireless data packet, a known preamble signal is assigned using each of N transmission antennas. The wireless control packet is transmitted (S42). After the radio control packet standby process (S43), if the radio control packet is normally received (S44), the radio control packet is terminated (S45), and the transfer function matrix accommodated therein Information on H is extracted and set in the transfer function matrix management circuit on the transmission side (S46). If the radio control packet cannot be normally received in the process S44, the process returns to the process S42 to retransmit the radio control packet.

FIG. 8 is a diagram illustrating a flow when a radio control packet is received by the second radio station in the radio communication method according to the embodiment of the present invention. In FIG. 8, when a signal is received (S51), it is determined whether it is a radio control packet (S52). If it is a radio data packet, normal reception processing is performed (S58). On the other hand, if it is a radio control packet (S52), the signal type is confirmed (S53), and if it is a signal requesting information on the transfer function matrix (S54), it is received by multiple receiving antennas. A transfer function matrix H is obtained by channel estimation using a known preamble pattern for each received signal (S55). Thereafter, a radio control packet containing the acquired transfer function matrix information is generated and transmitted (S56).
On the other hand, if it is normal control information in process S54 (S54), normal control information processing is performed (S57).
As described above, in the description from FIG. 5 to FIG. 8, the first wireless station does not necessarily have to transmit a wireless control packet every time data is transmitted. For example, when a certain time or more has elapsed since the transfer function was last acquired, transmission may be performed as necessary.

As an application area of MIMO technology, the expansion of high-speed wireless LAN systems using 5 GHz band and 2.4 GHz band is currently attracting attention. In such a wireless LAN system, the function of the first wireless station and the function of the second wireless station are generally implemented in one wireless station, and the OFDM modulation method is used.
FIG. 9 shows the configuration of the transceiver unit of the radio station in the radio communication system according to the embodiment of the present invention. In this figure, for the sake of simplicity, the case where there are two transmission / reception antennas (N = M = 2) is used as an example.

  9, 31-a to 31-b are receiving antennas, 32-a to 32-b are radio units, 33-a to 33-b are FFT circuits, 34 is a channel estimation circuit, 35 is a channel separation circuit, 36 -1-a to 36-Ka and 36-1-b to 36-Kb are subcarrier demodulation circuits, 37-a to 37-b are P / S conversion circuits, 38 is a data synthesis circuit, and 39 is a transfer function complement circuit. , 40 is a transfer function management circuit # 2, 41 is a matrix operation circuit, 42 is a data division circuit, 43-a to 43-b are S / P conversion circuits, 44-a to 44-b are preamble assignment circuits, 45- 1-a to 45-Ka and 45-1-b to 45-Kb are subcarrier modulation circuits, 46 is a transmission signal conversion circuit, 47-a to 47-b are IFFT circuits, and 48-a to 48-b are wireless Reference numerals 49-a to 49-b denote transmission antennas.

First, when a signal is received by the receiving antennas (31-a to 31-b), an FFT circuit (33-a to 33-) is sent to each receiving antenna via the radio unit (32-a to 32-b). In b), the signal for each subcarrier is separated on the frequency axis. This separated signal is input to the channel estimation circuit 34, and each transfer function information is obtained from a known preamble signal in the received signal.
If the received signal is a signal in which two signal sequences are superimposed, the channel separation circuit 35 performs signal separation processing for each sequence based on the transfer function information obtained by the channel estimation circuit 34. The result is input to subcarrier demodulation circuits (36-1-a to 36-Ka and 36-1-b to 36-Kb).

The processing in the subcarrier demodulation circuits (36-1-a to 36-Ka and 36-1-b to 36-Kb) estimates the transmission signal for each subcarrier and performs error correction processing etc. as appropriate. It is. The demodulated signal, which has been separated for each subcarrier, is subjected to parallel / serial conversion by the P / S conversion circuit (37-a to 37-b), and the data synthesis circuit 38 reproduces the data. Is output.
On the other hand, if the signal sequence is not superimposed, the channel separation circuit 35 performs processing such as maximum ratio combining for each subcarrier on the signal of each reception antenna, and subcarrier demodulation circuits (36-1-a to 36). -Ka). At this time, the channel estimation circuit 34 inputs information that becomes a part of the MIMO transfer function matrix to the transfer function complement circuit 39. The transfer function complementing circuit 39 generates and complements a toothless transfer function matrix component by an interpolation operation represented by (Equation 6) and the like.

The information of the transfer function matrix H obtained here is recorded in the transfer function matrix management circuit # 2 (40). This matrix is updated every time a signal is received, and only the latest information is recorded. The matrix calculation circuit 41 sequentially obtains H ^H and H ^H × H for the transfer function matrix H managed by the transfer function matrix management circuit 40, and finally obtains an eigenvector for H ^H × ^H. Find a unitary transformation matrix U that diagonalizes xH.
The processing when this radio station transmits a signal is also divided into a case where two signal sequences are superimposed and transmitted and a case where one signal is transmitted.

  When superimposing and transmitting two signals that transfer user data at high speed, when the data is input to the data dividing circuit 42, the data is divided into two signal sequences and the S / P converter circuit ( In 43-a to 43-b), a signal is further distributed for each subcarrier. A known preamble signal is added to each subcarrier signal by a preamble adding circuit (44-a to 44-b), and subcarrier modulation circuits (45-1-a to 45-Ka and 45-1-) are added. b-45-Kb), predetermined modulation is applied to the transmission signal conversion circuit 46.

  In the transmission signal conversion circuit 46, a signal converted using the unitary conversion matrix for each subcarrier generated by the matrix calculation circuit 41 is generated and input to the IFFT circuits (47-a to 47-b). In the IFFT circuit (47-a to 47-b), the signal separated on the frequency axis is converted into a signal on the time axis, and the transmitting antenna (49-49) is transmitted via the radio unit (48-a to 48-b). -a to 49-b).

On the other hand, when transmitting a signal of one system, for example, the same signal is input from the data dividing circuit 42 to the S / P conversion circuit (43-a) and the S / P conversion circuit (43-b). In the transmission signal conversion circuit 46, the signal transmitted from the transmission antenna 49-a is an odd subcarrier, and the signal transmitted from the transmission antenna 49-b is an even subcarrier. -a, 45-3-a, 45-5-a ..., 45-2-b, 45-4-b, 45-6-b ... only signals are valid (ie 45-2- a, 45-4-a, 45-6-a ..., 45-1-b, 45-3-b, 45-5-b ... signals are discarded), IFFT circuit (47-a ~ Output to 47-b). Other processes are the same as conventional processes.
In the above processing, if there is a predetermined difference between the transfer function when a signal is transmitted from the first radio station to the second radio station and the transfer function in the opposite direction, the transmission is performed. A process for correcting this difference can also be added to the transfer function used to acquire the unitary transformation on the side.

Second Embodiment
Hereinafter, a second embodiment of the present invention will be described. The present embodiment relates to a technique for exchanging antennas used for transmitting each signal sequence for each subcarrier when using the MIMO technique for the OFDM modulation scheme.
Here, in order to simplify the explanation, a case where transmission is performed by superimposing three signal sequences will be described using an example in which the number of antennas is three or four.
FIG. 14 is a diagram showing a transmission flow of the first wireless station (transmitting station) in the second embodiment of the present invention.
When data is input (step S1), the input data is divided into N data series (step S2), and a preamble signal is assigned to each of these signals (step S3). The modulation process is individually performed (step S4). For the modulated signal, as a transmission signal conversion process # 1, for each subcarrier, a conversion process for shuffling (replacement) the correspondence between the signal sequence and the transmission antenna using the matrix R _k corresponding to each subcarrier. (Step S5), the converted signal is converted into a radio frequency by the radio unit, and the signal is transmitted (step S6). The reception flow of the second radio station (reception station) is the same as that in the conventional system shown in FIG. 20, and there is no difference.

Here, in the conversion process in the transmission signal conversion process # 1 (step S5), for example, when the number of signal sequences to be superimposed is 3, the following three matrices R ₁ , R ₂ , and R ₃ are used. .

(9)

···(Ten)

(11)

As an example, R represents an _{R 1} with respect to the sub-carrier # 1, the _{R 2} with respect to the sub-carrier # 2, the _{R 3} with respect to the sub-carrier # 3, the sub-carrier # 4 ₁ is converted while substituting R ₂ in order for subcarrier # 5. For example, the conversion R ₂ corresponds to the process of replacing # 1 of the signal sequence in MIMO with # 2, replacing # 2 with # 3, and replacing # 3 with # 1. That is, antennas that transmit each signal sequence are different between adjacent subcarriers such as subcarrier # 1 and subcarrier # 2.

This rotation matrix is basically obtained by the process of appropriately replacing each row with respect to R ₁ shown in (Equation 9). As a method of generating such a matrix for a square matrix of N rows and N columns, only the (j + 1, j) component and the (1, N) component are 1 for an integer j where N> j ≧ 1. And an N-by-N matrix with other components being 0, and for an integer k such that N ≧ k ≧ 2

... (12)

Can be obtained as a total of N matrices.
Note that R ₁ does not need to be a unit matrix as shown in Equation (9), and may be one obtained by appropriately replacing the rows (or columns) of the unit matrix.

  Here, when an IEEE802.11a or IEEE802.11g-compliant system is considered as an existing wireless LAN system, the order of the data bit strings is switched in three subcarrier cycles with respect to 48 data transmission subcarriers of OFDM modulation. Interleave processing is performed.

That is, subcarrier # 1 → subcarrier # 4 → subcarrier # 7 →... → subcarrier # 2 → subcarrier # 5 → subcarrier # 8 →... → subcarrier # 3 → subcarrier # 6 → The bit string of data is arranged in the order of subcarrier # 9 →... Subcarrier # 48 → subcarrier # 1. In the above example, when the rotation matrixes of (Equation 9) to (Equation 11) are used, the same rotation matrix is used for subcarrier # 1 and subcarrier # 4. In this case, since the effect of shuffling by the rotation matrix cannot be obtained, a contrivance is necessary in such a case. For example,
Subcarrier # 1: using matrix R ₁
Subcarrier # 2: using matrix R ₂
Subcarrier # 3: matrix _{R 3} use,
Subcarrier # 4: using matrix R ₂
Subcarrier # 5: Matrix _{R 3} used,
Subcarrier # 6: using matrix R ₁
Subcarrier # 7: Matrix _{R 3} used,
Subcarrier # 8: matrix _{R 1} subcarriers # 9: Matrix _{R 2} used,
Subcarrier # 10: using matrix R ₁
Subcarrier # 11: using matrix R ₂
Subcarrier # 12: Matrix _{R 3} used,

In this order, the three rotation matrices are used and adjusted so as to be 9 subcarrier cycles, and the rotation matrices are adjusted so as not to overlap in the 3 subcarrier cycles that are interleave cycles.
Thereby, even when interleaving is performed, the effect of the present invention can be obtained.

  FIG. 15 shows a transmission flow of the first radio station (transmission station) in one specific example. The difference from the transmission flow shown in FIG. 14 is that transmission signal conversion processing # 2 (step S8) is performed between processing step S5 and processing step S6. Here, after the rotation matrix corresponding to each subcarrier is applied to the transmission signal vector x for each subcarrier in the processing step S5, processing for further applying an individual unitary transformation matrix for each subcarrier is performed (step S8). As a result, the transmission signal vector becomes U × R × x. A signal given by each component of this vector is transmitted from each radio unit and antenna for each subcarrier (step S6).

  FIG. 16 shows a transmission flow of the first wireless station in one specific example. The difference from the transmission flow shown in FIG. 14 is that transmission signal conversion processing # 3 (step S9) and transmission signal conversion processing # 4 (step S10) are performed between processing step S5 and processing step S6. Here, after the individual rotation matrix for each subcarrier is applied to the transmission signal vector x in processing step S5, the vector R × x is applied to the matrix given by (Equation 13) (step S9).

···(13)

Thereby, for example, in order to transmit three signal sequences, a column vector of 3 rows and 1 column is converted into a column vector of 4 rows and 1 column. Thereafter, transmission signal conversion processing # 4 is further performed in which an individual unitary conversion matrix U is further applied to each of the 4 × 4 subcarriers (step S10). As a result, the transmission signal vector becomes U × T × R × x. A signal given by each component of this vector is transmitted from each radio unit and antenna (step S6).
Further, although steps S5 and S8 in FIG. 15 and steps S5, S9 and S10 in FIG. 16 are performed in order here, a conversion matrix in which these are combined is separately created and converted in a batch. Processing may be performed.

  Note that the above-described rotation matrix is an example, and other rotation matrices may be used, or the subcarriers may be associated with a different order from that described above. Furthermore, since the transfer function matrix H is different for each subcarrier, the unitary transformation matrix corresponding thereto is also individual for each subcarrier. This unitary transformation matrix is obtained separately from the processing flow shown in FIG. 16, as in the conventional method.

A configuration example for realizing the above method as a wireless communication apparatus will be described below with reference to the drawings.
FIG. 10 is a diagram illustrating a configuration example of the transmission unit of the first wireless station in the present embodiment. In FIG. 10, 1 is a data division circuit, 2-1 to 2-3 are preamble assignment circuits, 3-1 to 3-3 are modulation circuits, 4 is transmission signal conversion circuits # 1, 5-1 to 5-3 are Radio units 6-1 to 6-3 denote antennas.

  When user data is input to the data dividing circuit 1, it is divided into three signal sequences in this figure, and each is input to the preamble assigning circuits 2-1 to 2-3. Here, a different predetermined preamble signal is given for each signal sequence. The three signal sequences generated by adding the preamble signal are independently modulated by the modulation circuits 3-1 to 3-3. Here, since OFDM modulation is performed, each signal sequence is modulated for each subcarrier.

  These signals are subjected to predetermined conversion for each subcarrier in the transmission signal conversion circuit # 1 (4). The converted signal is transmitted from the antennas 6-1 to 6-3 via the radio units 5-1 to 5-3.

Here, for example, three matrices R ₁ , R ₂ , and R ₃ shown in (Expression 9) to (Expression 11) are used. As an example, R represents an _{R 1} with respect to the sub-carrier # 1, the _{R 2} with respect to the sub-carrier # 2, the _{R 3} with respect to the sub-carrier # 3, the sub-carrier # 4 ₁ is converted while substituting R ₂ in order for subcarrier # 5.

For example, the conversion R ₂ corresponds to the process of replacing # 1 of the signal sequence in MIMO with # 2, replacing # 2 with # 3, and replacing # 3 with # 1. That is, antennas that transmit each signal sequence are different between adjacent subcarriers such as subcarrier # 1 and subcarrier # 2.

  The signals converted by the transmission signal conversion circuit 4 in this way are transmitted from the antennas 6-1 to 6-3 via the radio units 5-1 to 5-3, respectively.

  When OFDM modulation is used, pilot subcarriers containing known signals that do not contain user data may be included in a plurality of subcarriers, but the numbering here excludes pilot subcarriers. It was explained as being.

FIG. 11 shows a configuration example of the receiving unit of the second radio station in the radio communication method of the present embodiment.
In FIG. 11, 21-1 to 21-3 are antennas, 22-1 to 22-3 are radio units, 23 is a channel estimation circuit, 24 is a received signal management circuit, 25 is a transfer function matrix management circuit, and 26 is a matrix operation. Circuits (reception) # 1 and 27 are matrix operation circuits (reception) # 2, 28 are hard decision circuits, and 29 is a data synthesis circuit.

The radio units 22-1 to 22-3 individually perform data reception processing via the antennas 21-1 to 21-3. Each received signal is input to the channel estimation circuit 23. Here, an individual transfer function is calculated for each path and each subcarrier from a known signal portion included in the received signal, such as a preamble signal. This information is input to the transfer function matrix management circuit 25 and managed as a transfer function matrix H. The matrix operation circuit (reception) # 1 (26) performs a matrix operation for preparation for demodulation as necessary. For example, if the matrix is a square matrix, only the inverse matrix of the matrix H is obtained, otherwise the matrix H ^H that is a Hermitian conjugate matrix of the matrix ^H is further (H ^H × H), sequentially computing the H ^H product ^{(H H × H) -1 ×} H H. The following description will proceed assuming a non-square matrix.

Data subsequent to the preamble signal or the like is once managed by the reception signal management circuit 24, and the product of the matrix (H ^H × H) ⁻¹ × H ^{H and} the reception signal Rx is calculated as a matrix operation circuit (reception) #. 2 (27). The hard decision circuit 28 performs hard decision processing on the obtained signal and estimates a transmission signal. The data synthesizing circuit synthesizes the reproduction signal sequences separated into the respective signal systems, reproduces user data on the transmission side, and outputs the data. The above description has been made by taking the ZF method as an example, but other methods including the MMSE method, the MLD method, and a combination method thereof may be used.
Furthermore, in the embodiment of the present invention, the description is focused on the case where it is applied to the OFDM modulation system, but it can be applied in units of subcarriers, and of course, it can also be applied to the single carrier modulation system. Is possible.

  FIG. 12 shows a second configuration example of the transmission unit of the first wireless station in the wireless communication method of the present embodiment. In the example shown in FIG. 12, the data division circuit 1, preamble assignment circuits 2-1 to 2-3, modulation circuits 3-1 to 3-3, transmission signal conversion circuit # 1 (4), and radio units 5-1 to 5-1 5-3 and antennas 6-1 to 6-3 are the same as those in FIG. 10, and in addition, a channel estimation circuit 7, a transfer function matrix management circuit 8, a matrix operation circuit # 1 (9), a matrix operation circuit # 2 (10), a transmission signal conversion circuit # 2 (11) is provided.

  Even in the first wireless station that intends to transmit a signal including user data, when receiving a signal from the second wireless station, the signal received by the antennas 6-1 to 6-3 is transmitted to the wireless unit 5. Input to the channel estimation circuit 7 through -1 to 5-3. Here, the transfer function for each path and for each subcarrier is calculated from a known signal portion included in the received signal, for example, a preamble signal, in the same manner as the receiving unit shown in FIG.

This information is input to the transfer function matrix management circuit 8 and managed as a transfer function matrix H for each subcarrier. In the matrix operation circuit # 1 (9), matrices H ^H and H ^H × H which are Hermitian conjugate matrices of the matrix H are sequentially calculated. Based on this result, the matrix operation circuit # 2 (10) calculates a unitary matrix U that can diagonalize the matrix H ^H × H for each subcarrier.
At this time, the eigenvalue of the (i, i) component of the matrix diagonalized by the unitary transformation matrix is adjusted so that the absolute value of the eigenvalue increases (or decreases) as the value of i decreases. Using the unitary transformation matrix U for each subcarrier obtained as described above, the transmission signal conversion circuit # 2 (11) uses the unitary transformation matrix U for the signal output from the transmission signal conversion circuit 4 in the case of FIG. A conversion process is performed, and this is transmitted via the radio units 5-1 to 5-3 and the antennas 6-1 to 6-3.
The processing on the receiving side is the same as in FIG.

  FIG. 13 shows a third configuration example of the transmission unit of the first wireless station in the wireless communication method of this embodiment. In the configuration example shown in FIG. 13, the data division circuit 1, preamble assignment circuits 2-1 to 2-3, modulation circuits 3-1 to 3-3, transmission signal conversion circuit # 1 (4), and radio unit 5-1. ˜5-3, antennas 6-1 to 6-3, channel estimation circuit 7, transfer function matrix management circuit 8, matrix operation circuit # 1 (9), and matrix operation circuit # 2 (10) are shown in FIG. This is common with the configuration example. In addition, a transmission signal conversion circuit # 3 (12) and a transmission signal conversion circuit # 4 (13) are provided.

  In the configuration example shown in FIGS. 10 and 12, the wireless units 5-1 to 5-3 and the antennas 6-1 to 6-3, which are the 3 systems, are expanded to 4 systems, and the wireless unit 5-4 and the antennas are expanded. 6-4 has been added. Accordingly, the order of the matrix processed from the channel estimation circuit 7 in the matrix calculation circuit # 2 (10) is also partially changed.

In the configuration example illustrated in FIG. 13, the transmission signal conversion circuit # 3 (12) integrates the conversion matrix represented by the above (Formula 13) with respect to the signal output from the transmission signal conversion circuit 4. This calculation means that a transmission vector formed by three components is simply converted into a transmission vector of four components. The reason why the matrix is 4 rows and 3 columns is that the number of superimpositions of the MIMO signal sequence is 3, whereas the number of antennas for signal transmission is 4. In general, if the number of superpositions is M and the number of antennas for signal transmission is N, the matrix of N rows and M columns, the matrix of the upper M rows is the unit matrix of M rows and M columns, and the lower (N−M) columns Is a matrix of all zeros. As a result, a transmission signal vector having components equivalent to the number of transmission antennas (that is, the same order as the unitary conversion matrix) is obtained, and the transmission signal conversion circuit # 4 (13) performs unitary conversion processing on the transmission signal vector. Are transmitted via the radio units 5-1 to 5-4 and the antennas 6-1 to 6-4. The processing on the receiving side is the same as that in FIG.
Furthermore, also in the configuration example shown in this figure, a series of processing is performed individually for each subcarrier. Also, transmission signal conversion circuit # 1 (4) and transmission signal conversion circuit # 2 (11) in FIG. 12, and transmission signal conversion circuit # 1 (4), transmission signal conversion circuit # 3 (12) and transmission in FIG. The signal conversion circuit # 4 (13) is a separate functional block, but a separate conversion matrix may be created separately to perform a batch conversion process in one functional block.

  As described above in detail, according to the present invention, when performing highly efficient wireless communication using MIMO technology, reception characteristics for each of a plurality of superimposed signal sequences are made uniform, and a code error on a transmission path is randomized. Can be realized. In particular, when combined with the E-SDM scheme using the OFDM modulation scheme, the MIMO technique has a tendency that large variations in reception characteristics occur between superimposed signal sequences. By averaging the reception characteristics for each signal sequence and randomizing the errors, it is possible to improve the error correction gain and improve the overall packet error rate characteristics.

  Each circuit in the transmission unit shown in FIG. 10, each circuit in the reception unit shown in FIG. 11, each circuit in the transmission unit shown in FIG. 12, and each circuit in the transmission unit shown in FIG. 13 are realized by dedicated hardware. The program may be configured by a memory and a CPU (central processing unit), and a program (not shown) for realizing the function of each of these circuits is loaded into the memory and executed to execute the function. It may be realized.

  Further, a program for realizing the functions of each circuit in the transmission unit shown in FIG. 10, each circuit in the reception unit shown in FIG. 11, each circuit in the transmission unit shown in FIG. 12, and each circuit in the transmission unit shown in FIG. Each circuit in the transmission unit shown in FIG. 10 and each circuit in the reception unit shown in FIG. 11 are recorded on a computer-readable recording medium, and the program recorded on the recording medium is read and executed by the computer system. The processing necessary for each circuit in the transmission unit shown in FIG. 12 and each circuit in the transmission unit shown in FIG. 13 may be performed. Here, the “computer system” includes an OS and hardware such as peripheral devices.

  The “computer-readable recording medium” refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM, a DVD-ROM, and a CD-ROM, and a storage device such as a hard disk built in the computer system. . Furthermore, the “computer-readable recording medium” dynamically holds a program for a short time like a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line. It also includes a device (transmission medium or transmission wave) and a device that holds a program for a certain period of time, such as a volatile memory inside a computer system serving as a server or client in that case. The program may be for realizing a part of the functions described above, and further, a program that can realize the functions described above in combination with a program already recorded in a computer system, a so-called difference file (difference). Program).

  Moreover, all the embodiments described above are illustrative of the present invention and are not intended to limit the present invention, and the present invention can be implemented in various other variations and modifications. Therefore, the scope of the present invention is defined only by the claims and their equivalents.

According to the present invention, when highly efficient wireless communication using MIMO technology is performed, if the transfer function matrix can be estimated with high accuracy, the transmission function matrix can be realized while realizing good characteristics equivalent to the E-SDM method. Even when the function matrix cannot be estimated with high accuracy, it is possible to obtain an effect that stable characteristics can be exhibited.
In addition, when using the MIMO technique for the OFDM modulation scheme, by replacing the antenna used for transmitting each signal sequence for each subcarrier, the error for each signal sequence is averaged and randomized, and the coding gain in error correction is increased. Can be improved. That is, when performing highly efficient wireless communication using MIMO technology, it has the effect of uniformizing reception characteristics for each of a plurality of superimposed signal sequences and randomizing code errors on the transmission path. The present invention is used to increase the transmission speed of a high-speed wireless access system (or wireless LAN system) using the 2.4 GHz band or the 5 GHz band.

Claims (53)

A wireless communication device used in a wireless communication system configured by a first wireless station and a second wireless station,
The first radio station includes N _tx (N _tx is an integer greater than 1) or more first antenna groups,
Dividing means for dividing user data to be transmitted into N systems (N _tx ≧ N> 1, N is an integer);
Means for constructing a wireless data packet composed of N series of signal sequences obtained by giving signals of N types of known patterns to the user data divided into the N series;
Conversion means for converting into a signal sequence of the signal system string or al N _tx system including the known pattern of N systems,
Transmitting means for transmitting the signal converted by the converting means from the first antenna group;
When the second radio station includes M ( M is an integer greater than 1) second antenna groups, the i-th antenna in the first antenna group and the second antenna group Acquisition means for acquiring a transfer function h _{j, i} between the j-th antenna and an approximate value thereof;
The relative transfer functions h _{j, i} the first (j, i) matrix H of M rows and N columns whose components, means for calculating a Hermitian conjugate matrix H ^H of the matrix H,
Means for calculating a matrix product of the two matrices, that is, an N- row N- column matrix H ^H × H ;
Means for calculating the unitary matrix U for diagonalizing the matrix H ^H × H ,
The converting means includes
When the information of the k-th symbol (k is an integer equal to or greater than 1) of the N signal series is { x ₁ (k), x ₂ (k), ... , X _N (k) }, means for calculating a column vector given by a product of a column vector x (k) having each information of k symbols as an element and the unitary matrix U , that is, U × x (k) ;
The transmission means includes
A wireless communication apparatus comprising: means for transmitting the i- th row component [U xx (k)] _i of the column vector U xx (k) from the i-th antenna in the first antenna group . Prior to transmitting the wireless data packet, means for transmitting a first wireless control packet containing control information;
Means for receiving a second radio control packet that is a response to the first radio control packet using N or more antennas;
The acquisition means uses a plurality of known patterns of signals assigned to the second radio control packet, and uses the jth (1 ≦ 1) used by the second radio station for transmission from the received signal at each antenna. j ≦ M, j is an integer) including means for calculating a transfer function h _{j, i} between the antenna and the i-th (1 ≦ i ≦ N, i is an integer) antenna used by the first radio station for reception. The wireless communication apparatus according to claim 1 , wherein: Prior to transmission of the wireless data packet, means for transmitting a signal sequence including N known patterns as the first wireless control packet using the N first antenna groups;
Means for receiving a second radio control packet that is a response to the first radio control packet;
Obtain information on the transfer function hj, i between the i-th antenna of the first antenna group and the j-th antenna of the second antenna group accommodated in the second radio control packet. Means,
The wireless communication apparatus according to claim 1 , comprising: The orthogonal frequency division multiplexing (the K 1 integer greater than) K subcarriers of this in communication between radio stations using (OFDM: Orthogonal Frequency Division Multiplexing) according to claim 1, characterized by using a modulation scheme Wireless communication device. When receiving a radio control packet containing control information and / or user data containing user data from the second radio station, in which a plurality of signal sequences are not superimposed, that is, only one signal sequence is received. The jth (1 ≦ j) used by the second radio station of the ks (1 ≦ ks ≦ K, ks is an integer) subcarrier from the reception state of the signal of the known pattern received by each receiving antenna. ≦ M, j is an integer) Transfer function acquisition for acquiring the transfer function h _{j, i} ^[ks] between the antenna and the i-th (1 ≦ i ≦ N, i is an integer) antenna used for reception by the first radio station The wireless communication apparatus according to claim 4 , further comprising: means. In the ks ′ (1 ≦ ks ′ ≦ K, ks ′ is an integer) subcarrier not transmitted by the jth antenna of the second radio station, the jth antenna of the second radio station and the first The transfer function h _{j, i} ^{[ks ′ ]} between the i-th antennas of the radio station is expressed as ks ₁ ≠ ks ′ and ks ₂ ≠ ks ′ (1 ≦ ks ₁₎ transmitted from the j-th antenna of the second radio station. _{≦ K, 1 ≦ ks 2 ≦} K, ks 1, ks 2 is an integer) becomes the transfer function _h ^j for the first ks ₁ subcarrier and the ks ₂ ^{subcarrier, i [ks1]} and h ^{_j, i} of ^[ks2] The wireless communication apparatus according to claim 4 , further comprising transfer function acquisition means for acquiring h _{j, i} ^{[ ks ′ ]} by interpolation or extrapolation value. 6. The wireless communication apparatus according to claim 5 , further comprising means for notifying the second wireless station side that the transfer function acquiring means is provided on the first wireless station side. 7. The wireless communication apparatus according to claim 6 , further comprising means for notifying the second wireless station side that the transfer function obtaining means is provided on the first wireless station side. A wireless communication device for receiving at the second wireless station a wireless signal from the wireless communication device according to claim 1,
The second radio station includes M (M is an integer of 1 or more) second antenna groups,
Means for individually receiving radio signals using the second antenna group;
Demodulating means for separating and demodulating signals of each received signal sequence using a known pattern signal given to the received signal as a reference signal;
A wireless communication apparatus comprising: output means for combining all demodulated signal sequences and outputting as user data. The demodulating means includes
A transfer function h _j between the i-th antenna in the first antenna group and the j-th antenna in the second antenna group, using a known pattern signal given to the received signal as a reference signal _. means for obtaining _i ;
A calculation means for performing a predetermined calculation on a matrix H of M rows and N columns using the transfer function h _{j, i} as the (j, i) component, and obtaining a vector corresponding to a signal point of N transmission signals; Including
The output means includes
According to claim 9, characterized in that it comprises a means for outputting a transmission signal of N lines given in each element of the resulting vector in the operation, were synthesized for every symbol received as the user data Wireless communication device. The computing means is
Means for calculating a Hermitian conjugate matrix H ^H of the matrix H with respect to a matrix H of M rows and N columns with the transfer function h _{j, i} as the (j, i) component;
Means for calculating a matrix product of the two matrices, that is, an N-row N-column matrix H ^H × H;
Means for calculating an inverse matrix of the matrix H ^H × H, ie, (H ^H × H) ⁻¹ ;
Further means for calculating matrix ^{(H H × H) -1 ×} H H Using these,
When r _m (k) is the received signal of the k-th symbol actually received by the m-th antenna of the second antenna group, (r ₁ (k), r ₂ (k),. _M (k)) For a received signal vector Rx (k) represented by T (T indicates conversion from a row vector to a column vector), (H ^H × H) ⁻¹ × H ^H × Rx (k) Means for computing
The wireless communication apparatus according to claim 10 , comprising: The computing means is
Means for calculating an inverse matrix H ⁻¹ of the matrix H for a matrix H of N rows and N columns with the transfer function h _{j, i} as the (j, i) component;
When r _m (k) is the received signal of the k-th symbol actually received by the m-th antenna of the second antenna group, (r ₁ (k), r ₂ (k),. _M (k)) means for calculating H ⁻¹ × Rx (k) for a received signal vector Rx (k) represented by T (T indicates conversion from a row vector to a column vector);
The wireless communication apparatus according to claim 10 , comprising: The computing means is
The transmission signal of the k-th symbol transmitted from the n-th antenna of the first antenna group is defined as t _n (k), and the reception of the k-th symbol actually received by the m-th antenna of the second antenna group. When a signal is expressed as r _m (k), (t ₁ (k), t ₂ (k),..., T _N (k)) T (T indicates conversion from a row vector to a column vector) , (R ₁ (k), r ₂ (k),..., R _M (k)) T, a received signal vector Rx (k), and a matrix For the operator F, a vector product (F × Rx (k)) of a vector F × Rx (k) −Tx (k) and a Hermitian conjugate vector (F × Rx (k) −Tx (k)) H of the vector -Tx (k)) means for selecting the matrix operator F so as to be expected to minimize the expected value over a plurality of symbols of ^H * (F * Rx (k) -Tx (k));
Means for computing F × Rx (k) for each symbol;
The wireless communication apparatus according to claim 10 , comprising: The wireless communication apparatus according to claim 13 , wherein the matrix operator F is obtained by a MMSE (Minimum Mean Square Error) method. On the second radio station side,
When receiving the first radio control packet from the first radio station,
The radio communication apparatus according to claim 9 , further comprising means for transmitting a second radio control packet including a plurality of known patterns as a response to the first radio control packet. On the second radio station side,
When receiving a first radio control packet including known patterns of N systems from the first radio station,
Means for calculating the transfer function h _{j, i} using signals of known patterns of the N systems;
Means for transmitting a second radio control packet containing information on the transfer function as a response to the first radio control packet;
The wireless communication apparatus according to claim 10 , further comprising: The orthogonal frequency division multiplexing (the K 1 integer greater than) K subcarriers of this in communication between radio stations using (OFDM: Orthogonal Frequency Division Multiplexing) according to claim 9, characterized by using a modulation scheme Wireless communication device. On the second radio station side,
When transmitting the radio control packet or the radio data packet to the first radio station without superimposing a plurality of signal series, the second antenna corresponding to the subcarrier number ks is used as the ks subcarrier signal. The wireless communication apparatus according to claim 17 , further comprising: a transmitting unit that transmits using a predetermined one antenna in the group. 19. The wireless communication apparatus according to claim 18 , further comprising means for notifying the first wireless station side that the second wireless station side has the transmission means. A wireless communication system comprising the wireless communication device according to claim 1 on a transmission side and the wireless communication device according to claim 9 on a reception side. In a wireless communication method for performing communication between a first wireless station and a second wireless station,
The first radio station includes N _tx (N _tx is an integer greater than 1) or more first antenna groups,
A division step of dividing user data to be transmitted into N systems;
Constructing a wireless data packet composed of N signal series obtained by giving N types of known pattern signals to the user data divided into the N systems;
A conversion step of converting a signal sequence including the N known patterns into an N _tx signal sequence;
A transmission step of transmitting the signal converted by the conversion step from the first antenna group;
When the second radio station includes M ( M is an integer greater than 1) second antenna groups, the i-th antenna in the first antenna group and the second antenna group An acquisition step of acquiring a transfer function h _{j, i} between the j-th antenna and an approximate value thereof;
The relative transfer functions h _{j, i} the first (j, i) matrix H of M rows and N columns whose components, calculating a Hermitian conjugate matrix H ^H of the matrix H,
Calculating a matrix product of the two matrices, that is, a matrix H ^H × H of N rows and N columns ;
Calculating a unitary matrix U for diagonalizing the matrix H ^H × H ,
The converting step includes
When the information of the k-th symbol (k is an integer equal to or greater than 1) of the N signal series is { x ₁ (k), x ₂ (k), ... , X _N (k) }, calculating a column vector given by the product of a column vector x (k) having each information of k symbols as an element and the unitary matrix U , that is, U × x (k) ,
The transmitting step includes
Transmitting from the first antenna group a signal for transmitting the i- th row component [U × x (k)] _i of the column vector U × x (k) from the i- th antenna in the first antenna group. wireless communication method, which comprises a. Prior to transmission of the wireless data packet, transmitting a first wireless control packet containing control information;
Receiving a second radio control packet that is a response to the first radio control packet using N or more antennas, and
The acquisition step uses a plurality of known patterns of signals assigned to the second radio control packet, and the jth (1 ≦ 1) used by the second radio station for transmission from the received signal at each antenna. j ≦ M, j is an integer) calculating a transfer function h _{j, i} between the antenna and the i-th (1 ≦ i ≦ N, i is an integer) antenna used by the first radio station for reception. The wireless communication method according to claim 21 , wherein: Prior to transmission of the wireless data packet, transmitting a signal sequence including N known patterns as a first wireless control packet using the N first antenna groups;
Receiving a second radio control packet that is a response to the first radio control packet;
Obtain information on the transfer function h _{j, i} between the i-th antenna of the first antenna group and the j-th antenna of the second antenna group accommodated in the second radio control packet. And steps to
The wireless communication method according to claim 21 , further comprising: The orthogonal frequency division multiplexing (the K 1 integer greater than) K subcarriers of this in communication between radio stations using (OFDM: Orthogonal Frequency Division Multiplexing) according to claim 21, characterized by using a modulation scheme Wireless communication method. When receiving a radio control packet containing control information and / or user data containing user data from the second radio station, in which a plurality of signal sequences are not superimposed, that is, only one signal sequence is received The jth (1 ≦ j) used by the second radio station of the ks-th (1 ≦ ks ≦ K, ks is an integer) subcarrier from the reception state of the signal of the known pattern received by each receiving antenna. ≦ M, j is an integer) Transfer function acquisition for acquiring the transfer function h _{j, i} [ks] between the antenna and the i-th (1 ≦ i ≦ N, i is an integer) antenna used for reception by the first radio station The wireless communication method according to claim 24 , further comprising: In the ks ′ (1 ≦ ks ′ ≦ K, ks ′ is an integer) subcarrier not transmitted by the jth antenna of the second radio station, the jth antenna of the second radio station and the first The transfer function h _{j, i} ^{[ks ′]} between the i-th antennas of the radio station is expressed as ks ₁ ≠ ks ′ and ks ₂ ≠ ks ′ (1 ≦ ks ₁₎ transmitted from the j-th antenna of the second radio station. _{≦ K, 1 ≦ ks 2 ≦} K, ks 1, ks 2 is an integer) becomes the transfer function _h ^j for the first ks ₁ subcarrier and the ks ₂ ^{subcarrier, i [ks1]} and h ^{_j, i} of ^[ks2] The wireless communication method according to claim 24, further comprising ^: a transfer function acquisition step of acquiring h _{j, i} ^{[ks ′]} by interpolation or extrapolation value. It said to run first of said transfer function acquisition step with the radio station, the radio communication method according to claim 25, characterized in that it comprises a step of notifying the second radio station. 27. The wireless communication method according to claim 26 , further comprising a step of notifying the second wireless station side that the transfer function obtaining step is executed on the first wireless station side. A wireless communication method for receiving at the second wireless station a wireless signal transmitted by the wireless communication method according to claim 22,
The second radio station includes M (M is an integer of 1 or more) second antenna groups,
Individually receiving radio signals using the second antenna group;
A demodulation step for separating and demodulating signals of each received signal sequence using a known pattern signal given to the received signal as a reference signal;
An output step of combining all demodulated signal sequences and outputting as user data. The demodulation step includes
A transfer function h _j between the i-th antenna in the first antenna group and the j-th antenna in the second antenna group, using a known pattern signal given to the received signal as a reference signal _. and the step of acquiring the _i,
A calculation step of performing a predetermined calculation on a matrix H of M rows and N columns using the transfer function h _{j, i} as a (j, i) component, and obtaining a vector corresponding to signal points of N transmission signals; Including
The output step includes
Claim 29, characterized in that it comprises a step of transmission signals of N systems given by each element of the resulting vector in the operation, were synthesized for every received symbol, and outputs it as the user data Wireless communication method. The calculation step includes:
Calculating a Hermitian conjugate matrix H ^H of the matrix H with respect to an M-row N-column matrix H having the transfer function h _{j, i} as the (j, i) component;
Calculating a matrix product of the two matrices, that is, an N × N matrix H ^H × H;
Calculating an inverse matrix of the matrix H ^H × H, ie, (H ^H × H) ⁻¹ ;
Furthermore, using these, a step of calculating a matrix (H ^H × H) ⁻¹ × H ^H ,
If the received signal of the k-th symbol was actually received at the m antenna of the second antenna group was _{r m (k), (r1} (k), r2 (k), ···, rM (k )) Calculate (H ^H × H) ⁻¹ × H ^H × Rx (k) for the received signal vector Rx (k) represented by T (T indicates conversion from a row vector to a column vector). Steps,
The wireless communication method according to claim 30 , further comprising: The calculation step includes:
Calculating an inverse matrix H ⁻¹ of the matrix H for a matrix H of N rows and N columns with the transfer function h _{j, i} as the (j, i) component;
When r _m (k) is the received signal of the k-th symbol actually received by the m-th antenna of the second antenna group, (r ₁ (k), r ₂ (k),. _M (k)) calculating H ⁻¹ × Rx (k) for a received signal vector Rx (k) represented by T (T indicates conversion from a row vector to a column vector);
The wireless communication method according to claim 30 , further comprising: The calculation step includes:
The transmission signal of the k-th symbol transmitted from the n-th antenna of the first antenna group is defined as t _n (k), and the reception of the k-th symbol actually received by the m-th antenna of the second antenna group. When the signal is expressed as r _m (k), (t ₁ (k), t ₂ (k),..., T _N (k)) ^T (T indicates conversion from a row vector to a column vector) , (R ₁ (k), r ₂ (k),..., R _M (k)) received signal vector Rx (k) and matrix represented by ^T to operators F, vector F × Rx (k) -Tx ( k) and Hermitian conjugate of a vector of the vector (F × Rx (k) -Tx (k)) H vector product of (F × Rx (k) a step of minimizing the expected value across multiple symbols of ^{-Tx (k)) H × (} F × Rx (k) -Tx (k)) to select a matrix operator F, as expected,
Calculating F × Rx (k) for each symbol;
The wireless communication method according to claim 30 , further comprising: The wireless communication method according to claim 33 , wherein the matrix operator F is obtained by a MMSE (Minimum Mean Square Error) method. On the second radio station side,
When receiving the first radio control packet from the first radio station,
30. The radio communication method according to claim 29 , wherein a step of transmitting a second radio control packet including a plurality of known patterns as a response to the first radio control packet is executed. On the second radio station side,
When receiving a first radio control packet including known patterns of N systems from the first radio station,
Calculating the transfer function h _{j, i} using signals of the N patterns of known patterns;
As a response to the first radio control packet, transmitting a second radio control packet containing information on the transfer function;
The wireless communication method according to claim 30 , wherein the wireless communication method is executed. The orthogonal frequency division multiplexing (the K 1 integer greater than) K subcarriers of this in communication between radio stations using (OFDM: Orthogonal Frequency Division Multiplexing) according to claim 29, characterized by using a modulation scheme Wireless communication method. On the second radio station side,
When transmitting the radio control packet or the radio data packet to the first radio station without superimposing a plurality of signal sequences, the second antenna corresponding to the subcarrier number ks is used as the ks subcarrier signal. The wireless communication method according to claim 37 , wherein a transmission step of transmitting using a predetermined one antenna in the group is executed. The wireless communication method according to claim 38 , further comprising a step of notifying the first wireless station side that the transmission step is executed on the second wireless station side. The first radio station and the second radio station are configured by a first radio station and a second radio station, and the first radio station and the second radio station communicate using an orthogonal frequency division multiplexing (OFDM) modulation scheme using a plurality of subcarriers. A wireless communication device used in a wireless communication system for performing
The first radio station is N _tx ( N _tx Is an integer greater than 1) with at least a first antenna group,
Dividing means for dividing user data to be transmitted individually for each subcarrier into N systems ( N _tx ≧ N > 1, N is an integer);
It means for constructing a wireless data packet made up of a signal sequence of N lines obtained by applying a signal of a known pattern of N type user data divided into the N systems,
Conversion means for converting a signal sequence including the N known patterns into an N _tx system signal;
Transmitting means for transmitting the signal converted by the converting means from the first antenna group;
Have
The conversion means is configured so that information of n _s (n _s is an integer of 1 or more) symbols of N signal sequences in a certain subcarrier is {x ₁ ( _ns ), x ₂ ( _ns ),. .., X _N (n _s )} and the column vector of N rows having these as each component is x (n _s ), the unit matrix of N rows and N columns or the column of the unit matrix is appropriately replaced. N _R (N _R > 1: N _R is an integer) type rotation matrix group obtained as {R _k } (N _R ≧ k ≧ 1: k is an integer) corresponds to each subcarrier. with _{R k} for a given k, comprising at least means for converting the vector x a _{(n s)} in _{R k × x (n s)} ,
The transmission unit, radio communications device you characterized in that in each sub-carrier is to transmitted from the i-th antenna of the first antenna group and the i component of the transformed column vector. The conversion means converts the signal of the n _s symbol transmitted from the i-th antenna of the first antenna group in each subcarrier so as to be [R _k × x ( _ns )] _i. The wireless communication apparatus according to claim 40 . The second radio station includes M (M is an integer of 1 or more) second antenna groups,
For each subcarrier individually,
Means for obtaining a transfer function h _{j , i} or an approximation thereof between an i-th antenna of the first antenna group and a j-th antenna of the second antenna group;
Means for calculating a Hermitian conjugate matrix H ^H of the matrix H with respect to a matrix H of M rows and N columns with the transfer function h _{j , i} as the (j, i) component;
Means for calculating a matrix product of the two matrices, ie, a square matrix H ^H × H of N rows and N columns;
Means for calculating an N-by-N unitary matrix U for diagonalizing the matrix H ^H × H;
Further comprising
And the converting means, converting the signal of the n _s symbols to be transmitted from the i-th antenna of the first antenna group in each sub-carrier _{[U × R k × x (} n s)] As the _i the wireless communications apparatus of claim 40, wherein. The second radio station includes M (M is an integer of 1 or more) second antenna groups,
Ntx> N,
For each subcarrier individually,
Means for obtaining a transfer function h _{j , i} or an approximation thereof between an i-th antenna in the first antenna group and a j-th antenna in the second antenna group;
Means for calculating a Hermitian conjugate matrix H ^H of the matrix H with respect to a matrix H of M rows and N _tx columns with the transfer function h _{j , i} as the (j, i) component;
Means for calculating a matrix product of the two matrices, that is, a square matrix H ^H × H of N _tx rows and N _tx columns;
Means for calculating a unitary matrix U of N _tx rows and N _tx columns for diagonalizing the matrix H ^H × H;
Further comprising
When the conversion means represents a matrix of N _tx rows and N columns and an integer j where N ≧ j ≧ 1, only the (j, j) component is 1 and the other components are 0, T the in each subcarrier first antenna group of the i antenna: [U × T × a signal of the _{n s} symbols to be transmitted from the _{(n tx} ≧ i ≧ 1 i integer) _{R k} × x _{(n s 41} ] The wireless communication apparatus according to claim 40 , wherein conversion is performed so that _i is obtained. In the rotation matrix group {R _k }, R ₁ is one matrix selected from the matrix obtained by appropriately replacing the columns of the N _× N unit matrix, and the rotation matrix group {R _k } is an integer j for N> j ≧ 1. A matrix of N rows and N columns in which only the (j + 1, j) component and the (1, N) component are 1 and the other components are 0 is P, and further, R for an integer k such that N ≧ k ≧ 2. It is composed of a total of N matrices given as _k = P ^k−1 × R ₁ ,
The rotation matrix Rk used for the subcarriers in which user information is accommodated is made to correspond to the rotation matrix group {R _k } appropriately rearranged in order with N subcarrier periods. The wireless communication device according to any one of claims 40 to 43 . In the rotation matrix group {R _k }, R ₁ is one matrix selected from the matrix obtained by appropriately replacing the columns of the N _× N unit matrix, and the rotation matrix group {R _k } is an integer j for N> j ≧ 1. A matrix of N rows and N columns in which only the (j + 1, j) component and the (1, N) component are 1 and the other components are 0 is P, and further, R for an integer k such that N ≧ k ≧ 2. It is composed of a total of N matrices given as _k = P ^k−1 × R ₁ ,
When mapping an OFDM-modulated signal on subcarriers, interleave processing is performed so that the bit sequences of the signal sequences divided into the N systems are adjacent to each other in a _Nil subcarrier period ( _Nil is an integer greater than 1). In the case of applying, the rotation matrix R _k used for the subcarriers in which user information is accommodated is appropriately rearranged by preparing _Nil matrixes of the rotation matrix group {R _k }. _44. The wireless communication apparatus according to claim 40 , wherein the wireless communication apparatuses correspond in order of N _il × N subcarrier periods. The first radio station and the second radio station are configured by a first radio station and a second radio station, and the first radio station and the second radio station communicate using an orthogonal frequency division multiplexing (OFDM) modulation scheme using a plurality of subcarriers. A wireless communication method used in a wireless communication system for performing
The first radio station is N _tx ( N _tx Is an integer greater than 1) with at least a first antenna group,
A division step of dividing user data to be transmitted separately for each subcarrier into N systems ( N _tx ≧ N > 1, N is an integer);
A step of constructing a wireless data packet made up of a signal sequence of the N systems in the divided N lines in the user data obtained by applying a signal of N kinds of known pattern,
A converting step of converting a signal sequence including the N known patterns into an N _tx system signal;
Have a transmission step of transmitting the converted signal by the conversion step from the first antenna group,
The conversion step, the _n s of the signal sequence of the N lines in a certain sub-carrier _{(n s} is an integer of 1 or more) information symbols, each _{{x 1 (n s),} x 2 (n s), ·· .., X _N (n _s )} and the column vector of N rows having these as each component is x (n _s ), the unit matrix of N rows and N columns or the column of the unit matrix is appropriately replaced. N _R (N _R > 1: N _R is an integer) type rotation matrix group obtained as {R _k } (N _R ≧ k ≧ 1: k is an integer) corresponds to each subcarrier. with _{R k} for a given k, comprising at least means for converting the vector x to _{(n s)} in _{R k × x (n s)} ,
The transmitting step transmits the i-th component of the converted column vector in each subcarrier from the i-th antenna of the first antenna group.
A wireless communication method. In the conversion step, the signal of the _nsth symbol transmitted from the i-th antenna of the first antenna group in each subcarrier is converted so as to be [R _k × x ( _ns )] _i. The wireless communication method according to claim 46. The second radio station includes M (M is an integer of 1 or more) second antenna groups,
For each subcarrier individually,
Obtaining a transfer function h _{j , i} or an approximation thereof between an i-th antenna of the first antenna group and a j-th antenna of the second antenna group;
Calculating a Hermitian conjugate matrix H ^H of the matrix H with respect to a matrix H of M rows and N columns with the transfer function h _{j , i} as the (j, i) component;
Calculating a matrix product of the two matrices, that is, a N × N square matrix H ^H × H;
Calculating an N-row N-column unitary matrix U for diagonalizing the matrix H ^H × H;
Further comprising
In the conversion step, the signal of the _nsth symbol transmitted from the i-th antenna of the first antenna group in each subcarrier is converted so as to be [U × R _k × x ( _ns )] _i. The wireless communication method according to claim 46 , wherein: The second radio station includes M (M is an integer of 1 or more) second antenna groups,
N _rx > N,
For each subcarrier individually,
Obtaining a transfer function h _{j , i} or an approximation thereof between an i-th antenna of the first antenna group and a j-th antenna of the second antenna group;
Calculating a Hermitian conjugate matrix H ^H of the matrix H for a matrix H of M rows and N _tx columns, wherein the transfer function h _{j , i} is the (j, i) -th component;
Calculating a matrix product of the two matrices, that is, a square matrix H ^H × H of N _tx rows and N _tx columns;
Calculating an N _tx row N _tx column unitary matrix U for diagonalizing the matrix H ^H × H;
Further comprising
Wherein the converting step, the (j, j) with respect to N _tx × N and N ≧ j ≧ 1 becomes an integer j in the matrix of column components is all that is other ingredients 0 1 matrix upon denoted as T the in each subcarrier first antenna group of the i antenna: [U × T × a signal of the _{n s} symbols to be transmitted from the _{(n tx} ≧ i ≧ 1 i integer) _{R k} × x _{(n s} 47 ) The wireless communication method according to claim 46 , wherein conversion is performed so as to be _i . In the rotation matrix group {R _k }, R ₁ is one matrix selected from the matrix obtained by appropriately replacing the columns of the N _× N unit matrix, and the rotation matrix group {R _k } is an integer j for N> j ≧ 1. A matrix of N rows and N columns in which only the (j + 1, j) component and the (1, N) component are 1 and the other components are 0 is P, and further, R for an integer k such that N ≧ k ≧ 2. It is composed of a total of N matrices given as _k = P ^k−1 × R ₁ ,
The rotation matrix R _k used for the subcarriers in which user information is accommodated is made to correspond to the rotation matrix group {R _k } appropriately rearranged in order with N subcarrier periods. A wireless communication method according to any one of claims 46 to 49 . In the rotation matrix group {R _k }, R ₁ is one matrix selected from the matrix obtained by appropriately replacing the columns of the N _× N unit matrix, and the rotation matrix group {R _k } is an integer j for N> j ≧ 1. A matrix of N rows and N columns in which only the (j + 1, j) component and the (1, N) component are 1 and the other components are 0 is P, and further, R for an integer k such that N ≧ k ≧ 2. _k = is composed of ^{P k-1} × N total matrix given as _{R 1,}
When mapping an OFDM-modulated signal on subcarriers, interleave processing is performed so that the bit sequences of the signal sequences divided into the N systems are adjacent to each other in a _Nil subcarrier period ( _Nil is an integer greater than 1). when applying the said rotation matrix Rk using the sub-carrier by the user information is received, the rotation of each of the matrix group {Rk} respectively N _il or by instead appropriately arranging those prepared N _il The wireless communication method according to any one of claims 46 to 49 , wherein the correspondence is made in order with × N subcarrier cycles. A wireless communication device for receiving at the second wireless station a wireless signal from the wireless communication device according to claim 40,
The second radio station is M ( M Is an integer greater than or equal to 1) second antenna group,
Means for individually receiving radio signals using the second antenna group;
Demodulating means for separating and demodulating signals of each received signal sequence using a known pattern signal given to the received signal as a reference signal;
Output means for synthesizing all demodulated signal sequences and outputting them as user data;
A wireless communication apparatus comprising: The wireless communication device according to claim 40, on the transmission side, and the wireless communication device according to claim 52, A wireless communication system having a communication device on a receiving side.