JP4668080B2 - Channel information feedback method and radio 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 signals based on a transfer function matrix between the transmission and reception antennas. Channel information feedback method for performing efficient transmission by feeding back the state of the propagation path in a high-speed wireless access system that realizes wireless communication by demodulating data in the wireless communication device of And a wireless communication apparatus.

  In recent years, the IEEE802.11g standard, the IEEE802.11a standard, and the like are remarkable as high-speed wireless access systems using the 2.4 GHz band or the 5 GHz band. In these systems, an orthogonal frequency division multiplexing (OFDM) modulation scheme, which is a technique for stabilizing characteristics in a multipath fading environment, is used, and a transmission rate of 54 Mbps at the maximum is realized.

  However, the transmission rate here is a transmission rate on the physical layer, and the transmission efficiency in the MAC (Medium Access Control) layer is actually about 50 to 70%. It is about 30 Mbps. On the other hand, in the world of wired LANs, the provision of 100 Mbps high-speed lines has spread due to the widespread use of Ethernet (registered trademark) 100Base-T interfaces and FTTH (Fiber to the home) using optical fibers in homes. In the world of wireless LAN, further increase in transmission speed is demanded.

  As a technology for that purpose, MIMO (Multiple-Input Multiple-Output) technology is effective. With this MIMO technology, a transmitting-side wireless communication device transmits different independent signals on the same channel from a plurality of transmitting antennas, and a receiving-side wireless communication device receives signals using the same plurality of antennas. A transfer function matrix between the transmitting antenna and the receiving antenna is obtained, and by using this matrix, an independent signal transmitted from each antenna is estimated by the radio communication device on the transmitting side, and data is reproduced.

Here, consider a case in which N signals are transmitted using N transmission antennas and signals are received using M antennas. First, there are M × N transmission paths between the antennas of the wireless communication device that transmits and receives, and the transfer function when the signal is transmitted from the i-th transmitting antenna and received by the j-th receiving antenna is h _{j, i.} A matrix of M rows and N columns having this as the (j, i) th component is denoted as H. Further, a transmission signal from the i-th transmission antenna is denoted by t _i , a column vector having components (t ₁ , t ₂ , t ₃ ,... T _N ) as Tx, and a reception signal at the j-th antenna as r _j (r ₁ , r ₂ , r ₃ ,... R _M ) as components, Rx, and the jth receiving antenna's thermal noise as n _j , and (n ₁ , n ₂ , n ₃ ,... N _M ) as components. The column vector is denoted as n. In this case, the following relational expression holds.

  Therefore, a technique for estimating the transmission signal Tx based on the signal Rx received by the radio communication device on the receiving side is required for transmission based on the relational expression of Expression (1).

  In the above-mentioned MIMO communication, communication can be most efficiently performed by using a propagation path information and transmitting a signal in an optimum situation with respect to the propagation path. For example, in MIMO transmission using an eigenmode SDM (Space Division Multiplexing) method, when the transmission function matrix H of the MIMO channel in the signal transmission direction can be acquired by the wireless communication device on the transmission side, this transfer function matrix H Optimize the transmission signal corresponding to.

Specifically, a unitary matrix U that can diagonalize the product of the transfer function matrix H and the Hermitian conjugate matrix H ^H is generated, and the transmission signal is converted by this unitary matrix and the signal is transmitted. The following relational expression holds between the unitary transformation row U and the transfer function matrix H.

  In Expression (2), the right-side matrix Λ is a diagonal matrix in which only the diagonal component has a value and the other components are zero. By transmitting a signal by applying the unitary matrix U having such characteristics to the column vector Tx, Expression (1) is converted as follows.

  By the conversion to Equation (3), the transmission signal is orthogonalized for each MIMO channel, and even if a simple ZF (Zero Forcing) method is used in processing on the reception side, each transmission signal is converted to each MIMO channel. Can be adjusted so that the SNR (Signal Noise Ratio) characteristic is good. In addition, each column vector of this unitary matrix gives a coefficient for multiplying each antenna when the column vector Tx as a transmission signal is distributed to each transmission antenna, that is, a transmission weight vector (each component is called a transmission weight).

  By using this transmission weight, orthogonal beam forming is performed for each MIMO channel, and the gain of each beam, that is, the channel corresponding to the eigen beam becomes the eigenvalue of the eigenvector. Therefore, the upper limit of the channel capacity C of all MIMO channels is given by the following equation (4).

In Equation (4), B is prestige, P _i is the total transmission power of the i-th MIMO channel, and σ ² is the variance of noise power. From equation (4), it is possible to estimate what transmission mode of the transmission rate can be applied and how many MIMO channels can be multiplexed.

  Here, the transmission mode is a mode defined by a combination of a modulation scheme such as QPSK (Quadrature Phase Shift Keying) and 64QAM (Quadrature Amplitude Modulation) and an error correction coding rate.

Incidentally, not necessarily a common value to the transmission power P _i to all of the MIMO channels in the formula (4), also may be changed transmission mode for each MIMO channel. In general, we are possible to optimize the value of the P _i by using a technique called water filling. Among these, if there is a MIMO channel with P _i = 0, it means that it is more efficient to allocate power to other MIMO channels without using that channel for actual propagation. . That is, the number of MIMO multiplexing is set to be smaller than the original upper limit value. In this way, it is possible to determine the optimum value of the MIMO channel to be multiplexed.

  In any case, what is important here is that the transmission side can grasp the information of the MIMO channel, that is, the transfer function matrix H. By acquiring this information by the wireless communication device on the transmission side, it is possible to optimize the transmission weight at the time of signal transmission, and to optimize the transmission mode and the MIMO multiplexing number. As a result, it is possible to maintain a good communication state and avoid unnecessary transmission errors and accompanying retransmissions. In the case of a system using the OFDM modulation scheme, the transfer function matrix differs for each subcarrier, and it is necessary to acquire the transfer function matrix of all or some of the subcarriers.

  As described above, there are various utilization methods when the transfer function matrix of the MIMO channel can be acquired in the wireless communication apparatus on the transmission side, but the problem is how to acquire the information. As the simplest example, in the case of bidirectional MIMO transmission, there is a method in which a transfer function matrix estimated at the time of receiving a MIMO signal is stored and used for the next transmission. However, in general wireless communication, downstream transmission from a base station device to a wireless communication device as a terminal is dominant, and MIMO transmission is rarely used in the upstream in the opposite direction. Also, normal traffic is bursty, and transmission in one direction continues continuously, and data in the reverse direction generally flows after a while, and the propagation path conditions change greatly with time. Is expected to.

  In the general wireless communication as described above, the accuracy is too low to be used for the above-described efficient transmission. Still further, for example, even in a system capable of up to 4 multiplexing, if the multiplexing number in the received signal of the MIMO transmission is 2 multiplexing, information for performing 3 multiplexing or 4 multiplexing is obtained. Is insufficient.

  A technique for solving such a problem is proposed in Patent Document 1. The technique disclosed in Patent Document 1 proposes a method for realizing the eigenmode SDM method, and mentions a method for obtaining the transfer function matrix H therein.

  FIG. 21 shows an example of a transfer function information collection method in the conventional method. Here, the first wireless communication device is disposed on the upper side and the second wireless communication device is disposed on the lower side, and the horizontal axis indicates the time transmitted on the information transmitted between these wireless communication devices. In FIG. 21, reference numeral 101 is a radio control packet # 1, reference numeral 102 is a radio control packet # 2, and reference numeral 103 is a radio data packet.

  In transmitting data, the first wireless communication apparatus first transmits wireless control packet # 1 (101) to the second wireless communication apparatus. This radio control packet # 1 (101) mainly includes two types of options and includes a transmission request for a MIMO preamble for estimating a transfer function of the MIMO channel, or transmits a MIMO preamble and transmits a channel estimation result. As a transfer function information return request.

  In the second wireless communication apparatus, in the former case of the above two types of options, a signal including the MIMO preamble is transmitted as the wireless control packet # 2 (102). Thereby, it becomes possible to acquire the transfer function matrix in the direction of the first wireless communication device from the second wireless communication device.

  On the other hand, in the latter case, the second radio communication apparatus accommodates transfer function information as a channel estimation result and returns radio control packet # 2 (102) to the first radio communication apparatus. This makes it possible to acquire a transfer function matrix in the direction of the second wireless communication device from the first wireless communication device. Based on the information related to the transfer function thus obtained, the first wireless communication apparatus transmits a wireless data packet (103) containing user data.

Here, the transfer function matrix H _FW from the first wireless communication device to the second wireless communication device, and the transfer function matrix from the second wireless communication device to the first wireless communication device in the opposite direction. the H _BW, can be related by performing a predetermined conversion process. The transfer function matrix itself between the antenna end on the transmitting side and the antenna end on the receiving side can be correlated by matrix transposition processing, but the transfer function matrix H expressed by the above equation (1) or the like is Physical quantity equivalent to the rotation of the amplification factor and phase for each radio unit (including transmission power amplifier and frequency converter (upconverter)) and radio unit (low noise amplifier and frequency converter (downconverter)) ) Etc.) and the physical quantity corresponding to the rotation of the phase. For example, the complex physical quantities corresponding to the amplification factor and phase rotation of the radio unit (transmission side) and the radio unit (reception side) in the i-th antenna of the k-th (k = 1 or 2) radio communication device are represented by α _{k, Let i} , β _{k, i} . Similarly, let H _sp be a transfer function matrix that depends only on the space between the transmitting and receiving antenna ends. Assuming that a diagonal matrix having {α _{k, i} } as a diagonal component is A _k , and a diagonal matrix having {β _{k, i} } as a diagonal component is B _k , an effective transfer function matrix is It is represented by (5) and formula (6).

The matrix ^t H _sp represents a transposed matrix of H _sp . From the above formulas (5) and (6), the following relational expression is obtained.

In Equation (7), the matrix B ₁ ⁻¹ · A ₁ is a diagonal matrix having {α _{1, i} / β _{1, i} } as a diagonal component. Similarly, the matrix B ₂ ⁻¹ · A ₂ is a diagonal matrix having {α _{2, i} / β _{2, i} } as a diagonal component.

If these pieces of information are notified to each other when communication is started, based on the transfer function matrix from the second wireless communication device to the first wireless communication device using Equation (7). Thus, it is possible to easily calculate a transfer function matrix from the first wireless communication device in the opposite direction to the second wireless communication device. Therefore, as a method of using the radio control packet # 1 (101), it is possible to obtain a desired transfer function matrix H _FW by using either the former or the latter option.

  However, the efficiency at that time is slightly different. For example, let us consider a case where information on the transfer function matrix is returned in the former radio control packet # 2 (102). As an extreme example in this case, consider a case where the first wireless communication apparatus includes eight antennas and the second wireless communication apparatus includes four antennas. In general, a base station device in a wireless system is designed to have higher functionality than a wireless communication device serving as a terminal, and the wireless communication device in the wireless system can have a hardware configuration in order to reduce the cost of many wireless communication devices. Since it is desirable to be as light as possible, such a configuration can be envisaged.

  In such a case, there are 8 × 4 = 32 transmission paths for the MIMO channel. Not only is transfer function information defined for each path, but when an OFDM modulation scheme is used as in IEEE 802.11a / g, it is necessary to individually notify information for 48 data subcarriers. When the transfer function information of one path of a certain subcarrier is expressed by assigning 12 bits to the real part and 12 bits to the imaginary part, as a whole, 48 subcarriers x 32 paths x 3 bytes (2 x 12 bits) = 4068 bytes of information.

  Since the transfer function information is control information, it is expected to use a transmission mode with a certain degree of reliability. However, assuming a transmission mode of 24 Mbps of 160QAM R = 1/2 in IEEE802.11a / g, Since the capacity that can be transmitted in one OFDM symbol is 12 bytes, 384 OFDM symbols are also required. This is about 1.5 ms in terms of time. This time is an excessively heavy load, and is actually unacceptable in actual communication considering that various overheads such as a preamble are actually added thereto.

  On the other hand, consider the case where the latter radio control packet # 2 (102) accommodates and transmits a MIMO preamble. As the preamble for the four antennas of the second wireless communication apparatus, for example, when the preamble is transmitted by simply switching the antenna for each symbol, four symbols are sufficient. Compared to the above, this corresponds to an overhead of 1/96 times. Thus, the advantage of the latter becomes conspicuous as the application of the OFDM modulation scheme and the number of antennas to be used increase.

  Next, conventional techniques for channel estimation techniques for MIMO channels will be described below. First, FIG. 22 shows a configuration example of a wireless data packet in the case of SISO (Single-Input Single-Output) transmission without performing spatial multiplexing. In FIG. 22, reference numeral 111 denotes a short training signal, reference numeral 112 denotes a long training signal, reference numeral 113 denotes PLCP control information (PLCP: Physical Layer Convergence Protocol), and reference numeral 114 denotes data.

  In wireless communication, a known preamble signal is given prior to data transmission. The role of this preamble signal is to acquire transfer function information of the propagation path in addition to timing detection, AFC (Automatic Frequency Control), and AGC (Automatic Gain Control). For this purpose, actual data and control are used. Sent prior to the transmission of information. Supplementally, this type of signal may be referred to by various names such as a preamble signal, a training signal, and a pilot signal. Further, in the following description, “orthogonal pilot signal” indicates a plurality of long training signals or preamble signals that are in an orthogonal relationship. The preamble signal includes a short training signal.

  For example, in IEEE802.11a / g, there are signals given for the purpose of timing detection, AFC, AGC, etc. and acquisition of transfer function information. The signal given for the former purpose is the short training signal 111. The signal given for the latter purpose is called the long training signal 112. Of course, in addition to the above purpose, the long training signal 112 may be used to detect timing with the long training signal 112 and draw AFC, AGC, etc. roughly adjusted with the short training signal 111 with higher accuracy. Such a preamble signal is followed by a PLCP control signal 113, and control information regarding data 114 following the PLCP control signal 113 is defined in this information. Further, the PLCP control information 113 also includes the data length, the modulation mode applied to the data, and the like.

  Based on the PLCP control information 113, the receiving side demodulates the data 114 and reproduces the user data transmitted on the transmitting side. In this example, packet-based wireless communication such as IEEE802.11a / g is assumed. However, for example, a TDMA (Time Division Multiple Access) method such as HiperLAN / 2 (in Japan called HisWANa: High Speed Wireless Access System) is used. In the case of the communication used, the PLCP control information 113 may be transmitted separately. Generally speaking, the preamble signal of the long training signal 112 necessary for channel estimation of transfer function information is transmitted prior to the data 114.

  FIG. 23 shows a configuration example of a wireless data packet in the case of MIMO communication in contrast to the case of the above SISO communication. 23, reference numerals 121-1 and 121-2 are short training signals, reference numerals 122-1 and 122-2 are long training signals, reference numerals 123-1 and 123-2 are PLCP control information, and reference numerals 124-1 and 124-. 2 represents a MIMO long training signal, and reference numerals 125-1 and 125-2 represent data. Basically, the short training signals 121-1 and 121-2, the long training signals 122-1 and 122-2, and the PLCP control information 123-1 and 123-2 are the same signal.

  First, the PLCP control information 123-1 and 123-2 are demodulated using the short training signals 121-1 and 121-2 and the long training signals 122-1 and 122-2. The PLCP control information 123-1 and 123-2 include control information related to MIMO transmission.

  Specifically, in addition to the data length and modulation mode described in FIG. 22, information such as the number of multiplexed MIMO channels is also defined. Using this control information, the frame configuration of subsequent information is determined. Subsequent to the PLCP control information 123-1 and 123-2, there are MIMO long training signals 124-1 and 124-2. The MIMO long training signals 124-1 and 124-2 are used to obtain a transfer function matrix between the antennas, and the data 125-1 and 125-2 are then channel-separated and demodulated. As a point, data # 1 (125-1) and data # 2 (125-2) contain separate signals, respectively, and MIMO long training signal # 1 (124-1) is used for channel separation. In addition, the MIMO long training signal # 2 (124-2) also includes different signals.

  Various methods have been proposed for selecting the MIMO long training signals 124-1 and 124-2. For example, in the discussion of task group n (TGn) of the IEEE 802.11 committee, methods shown in Non-Patent Document 1, Non-Patent Document 2, Non-Patent Document 3, and the like are proposed.

  Here, the technique disclosed in Non-Patent Document 1 is a method of transmitting a preamble signal from all antennas in order at different times while transmitting a preamble signal from only one antenna at the same time. An outline of the MIMO long training signal will be described with reference to FIGS. 24 and 25, reference numerals 131-1 to 136-1 and reference numerals 131-2 to 136-2 are MIMO long training signals for each subcarrier.

  Further, reference numerals 137-1 and 137-2 are MIMO long training signals output from the first transmission antenna, and reference numerals 138-1 and 138-2 are MIMO long training signals output from the second transmission antenna. Represents a signal. FIG. 24 and FIG. 25 show the MIMO preamble signals of the first symbol and the second symbol of the MIMO preamble signal configured over a plurality of symbols, respectively. The horizontal axis of each figure represents a subcarrier number.

  As can be seen from FIGS. 24 and 25, in each symbol, a signal is transmitted from one transmission antenna on the same subcarrier, and the remaining transmission antennas have no signal (Null). Therefore, the transfer function between the corresponding transmitting and receiving antennas can be acquired from the long training signal of each subcarrier of the symbol.

 For the combination of Null subcarrier and antenna, a signal is transmitted in the next symbol. Therefore, the entire transfer function matrix can be obtained by combining the channel estimation results of each symbol.

  Next, the technique shown in Non-Patent Document 2 is a phase shift equivalent to 1600 ns between the MIMO long training signal # 1 (124-1) and the MIMO long training signal # 2 (124-2) in FIG. This is a method with a difference in minutes. In this method, on the receiving side, for example, the long training signal # 2 (124-2 for MIMO) together with the cross-correlation between the known signal having the same pattern as the long training signal # 1 (124-1) for MIMO and the received signal. ) To obtain the cross-correlation between the known signal and the received signal having the same pattern. The phase shift corresponding to 1600 ns here is set so that the cross-correlation between the two in the range where the time is shifted by 0 to 1600 ns is zero. That is, when each signal is shifted by time δ and cross-correlation over one symbol is taken, in the time domain where δ is 0 to 1600 ns, each is orthogonal (correlation value is zero). As a result, the impulse response between the antennas can be measured separately. The delay profile measured here can be converted into a transfer function in the frequency domain, that is, in each subcarrier, by performing an inverse Fourier transform.

  In the case of the technique shown in Non-Patent Document 1, a two-symbol length preamble is necessary to obtain a transfer function related to two transmission antennas. In the technique shown in Non-Patent Document 2, one symbol is used. It is possible to obtain a transfer function for two antennas.

  The technique shown in Non-Patent Document 3 corresponds to a method of using only the configuration shown in FIG. 24 and omitting the configuration shown in FIG. 25 among the techniques shown in Non-Patent Document 1. Of course, a transfer function corresponding to a combination of subcarriers and antennas for which no signal is transmitted cannot be obtained directly. However, in the example of FIG. 24, since the transfer function can be obtained for odd subcarriers for the first transmitting antenna, the transfer function for even subcarriers of the first transmitting antenna is based on the transfer function for odd subcarriers. It can be obtained by insertion or extrapolation processing. The second transmission antenna can be obtained in the same manner.

  The above description of Non-Patent Documents 1, 2, and 3 is an example in the case where there are two transmission antennas, but it can be extended to a case of three or more antennas. The explanation is also made in. What is common to all systems is that a MIMO preamble corresponding to the number of spatial multiplexing of the subsequent data area is employed. In other words, when transmitting a SISO signal that has not been spatially multiplexed, a MIMO preamble is not added.

  The technique described in Patent Document 1 described above directly collects transfer function information by defining a dedicated radio control packet for the purpose of channel estimation. These wireless control packets are different from conventional wireless control packets for transmitting various control signals, and wireless control packet # 1 (101) addressed from the first wireless communication device to the second wireless communication device. ) And the radio control packet # 2 (102) addressed to the first radio communication device from the second radio communication device.

  Next, an outline of the conventional wireless communication will be described. In general radio communication, in addition to transmission of radio data packets containing user data, radio control packets containing various control information are frequently transmitted and received between the base station apparatus and the radio communication apparatus. The most frequently used information is ACK (Acknowledgement) information for confirming normal reception of data. Based on this information, retransmission control of a radio data packet in which a code error has occurred is performed. FIG. 26 shows the flow of data transfer in the conventional system. In FIG. 26, reference numerals 141 and 143 denote wireless data packets, and reference numerals 142 and 144 denote wireless control packets # 3 including ACK information. The horizontal axis represents time, and the upper side represents a first wireless communication apparatus that is a wireless data packet transmitting side, and the lower side represents a second wireless communication apparatus that is a wireless data packet receiving side. When the first wireless communication apparatus transmits a wireless data packet 141 including user data, the second wireless communication apparatus transmits a wireless control packet # 3 (142) including ACK information for notifying the reception state. To do. Normally, the wireless data packet 141 and the wireless data packet 143 to be transmitted are given a sequence number which is a serial number of user data, and it is determined whether or not the communication status is normal via this number.

  Examples of information included in the radio control packet # 3 (142) include, for example, a message indicating that the radio data packet 141 is simply received normally, and a sequence number assigned to the radio data packet 141. It may be. In addition, in order to perform selective retransmission control which is more efficient retransmission control, a plurality of sequence numbers that have been successfully received so far may be included. The first wireless communication apparatus side receives this wireless control packet # 3 (142), and if the wireless data packet 141 has been transmitted, the next user data is detected and the code error of the wireless data packet 141 is detected. If the wireless data packet 141 is received, the wireless data packet 141 itself is transmitted as the wireless data packet 143.

  Further, the second wireless communication device returns a wireless control packet # 3 (144) for the wireless data packet 143. The transmission timing of the radio control packet # 3 (142) and the radio control packet # 3 (144) may be immediately after the radio data packet 141 and the radio data packet 143 or after a predetermined time has elapsed. Absent. IEEE802.11a / g stipulates that radio control packet # 3 (142) and radio control packet # 3 (144) be transmitted at the timing when 16 μsec has elapsed after reception of radio data packet 141 and radio data packet 143. However, in HyperLAN / 2 using TDMA control, transmission is performed after a certain amount of time under the control of the base station apparatus.

  In addition, as an example of a radio control packet containing the control information as described above, a base station apparatus in a base station centralized system determines whether there is data scheduled to be transmitted to the radio communication apparatus side. Polling control packets and RTS / CTS (Request to Send / Clear to Send) packets for avoiding the hidden terminal problem in the autonomous distributed system.

  FIG. 27 shows an outline of data transfer using RTS / CTS in packet-based communication using CSMA (Career Sense Multiple Access) control. In FIG. 27, reference numeral 151 is a radio control packet # 4 corresponding to RTS, reference numeral 152 is a radio control packet # 5 corresponding to CTS, reference numerals 153 and 155 are radio data packets, reference numerals 154 and 156 are radio control including ACK information. Packet # 3 is shown. When the first wireless communication apparatus starts communication, wireless control packet # 4 (151) is transmitted in advance. In this, in addition to the identification ID on the destination device side, the time for continuously occupying the band is set as a NAV (Network Allocation Vector). All wireless communication devices that have received this wireless control packet # 4 (151) determine whether or not they are addressed to their own devices based on the identification ID of the destination device. To do. The second radio communication apparatus determines that the destination of this radio control packet # 4 (151) is addressed to the own apparatus, and transmits radio control packet # 5 (152) corresponding to CTS. At this time, the remaining bandwidth occupation time is calculated and set in the NAV of the radio control packet # 5 (152).

  As a result, all the wireless communication apparatuses that can receive the wireless control packet # 4 (151) or the wireless control packet # 5 (152) refrain from transmission within this band occupation time. The first wireless communication apparatus that has received the wireless control packet # 5 (152) transmits a wireless data packet 153, and the second wireless communication apparatus transmits a wireless control packet # 3 (154) as ACK information. Within the time of NAV, the wireless data packet 155 and the wireless control packet # 3 (156) may be sequentially transmitted after that.

  Since these radio control packets contain control information, there is a need to reliably notify the other party using a more reliable transmission mode. In addition, in RTS / CTS, it is necessary to notify not only the communication partner but also all other users sharing the same frequency channel. The transmission is performed using a transmission mode that is mandatory for all wireless communication devices, not the transmission mode that is assumed to be used.

  The new standard IEEE802.11n, which is expected to succeed the existing IEEE802.11a / g system, is currently in the process of standardization, but it is required to be backward compatible as the same 802.11 family system. When / CTS is used, this radio control packet is not converted to MIMO but is defined to be transferred in the SISO mode, so that the conventional IEEE 802.11a / g system can also detect this packet. Further, it is expected that a radio control packet including ACK information is communicated in the SISO mode for the same reason.

  Next, processing contents in the first wireless communication device (transmission side) and the second wireless communication device (reception side) in the conventional method will be described with reference to flowcharts.

  FIG. 28 shows a flow of a radio control packet transmission process in the case of transmitting a signal of the format shown in FIG. Although an example relating to a radio control packet is described here, it corresponds to an example in which spatial multiplexing is not generally performed, and also corresponds to a case of a radio data packet of SISO transmission. First, information to be transmitted, that is, control information or user data is input (step S101). First, in order to form a packet, a short training signal is given (step S102), a long training signal is given (step S103), and PLCP control information is given (step S104). Then, information to be transmitted is added (step S105), a radio control packet or a radio data packet is formed, and this is subjected to predetermined modulation processing and transmitted from one antenna (step S106). The process ends (step S107).

  Although transmission may be performed from a plurality of antennas, basically the same processing is performed. If PLCP control information is unnecessary, step S104 may be omitted.

  FIG. 29 shows a wireless data packet transmission processing flow in the case of transmitting a signal of the format shown in FIG. Here, a method for transmitting wireless data packets in the case of MIMO transmission in which a plurality of data streams are spatially multiplexed will be described. When data is input (step S111), the input data is subjected to S / P (Serial / Parallel) conversion into signals of a plurality of systems (step S112). Next, a short training signal is assigned in each system (steps S113-1 and S113-2), a long training signal is assigned (steps S114-1 and S114-2), and PLCP control information is assigned. (Steps S115-1 and S115-2). Then, different long training signals for MIMO # 1 and # 2 are assigned for each signal system (steps S116-1 and S116-2), and data to be transmitted is individually assigned (steps S117-1 and S117-2). ), Forming a wireless data packet of each signal system. This is subjected to a predetermined modulation process and transmitted from each antenna (steps S118-1 and S118-2), and the process ends (step S119).

If PLCP control information is not necessary, steps S115-1 and S115-2 may be omitted. The MIMO preamble provided in steps S116-1 and S116-2 may be any preamble as long as the transfer function can be estimated individually, as well as those introduced in Non-Patent Documents 1, 2, 3, etc. .
International Publication No. 2005/055484 Pamphlet “Proposal for 802.11n”, IEEE802.11 standardization contribution, 11-04-0938-00-00n, August, 2004 “WWiSE IEEE802.11n Proposal”, IEEE802.11 standardization contribution, 11-04-0935-00-00n, August, 2004 “Backwards compatibility, -How to make a MIMO-OFDM system backwards compatible and coexistence with lla / g at the link level.- ', IEEE802.11 standardization contribution, 11-03-0714-00-OOn, September, 2003

  In the conventional method described above, the radio control packet # 1 (101) addressed to the second radio communication device from the first radio communication device and the radio control addressed to the first radio communication device from the second radio communication device. It was necessary to exchange two radio control packets of packet # 2 (102) when performing MIMO channel information feedback.

  However, the time required to exchange these pieces of information is a loss in data communication, and the time length cannot be ignored. In environments with fast channel fluctuations, such as public hotspots and outdoors, the channel fluctuates on the order of tens of ms, and frequently transmitting radio control packets for MIMO channel estimation frequently reduces transmission efficiency. Leads to.

  In addition, before the first wireless communication apparatus transmits a wireless data packet, it is possible to acquire the transfer function matrix of the MIMO channel from the wireless data packet including the MIMO preamble received from the second wireless communication apparatus. However, if the second wireless communication apparatus is provided with four antennas and the spatial multiplexing number of the wireless data packet in MIMO is 2, the wireless data packet includes the MIMO antenna for two antennas. Since only the preamble is included, it is not possible to collect all four pieces of information.

  In this way, it is possible to collect information on all antennas without adding new overhead or minimizing the addition of feedback for acquiring MIMO channel information at the wireless communication device on the transmitting side. Establishment of a new method has been an issue.

  The present invention has been made to solve the above-mentioned problems, and its object is to realize feedback of all antenna information with less overhead when the transfer function information of the MIMO channel is fed back between the transmitting and receiving apparatuses. The present invention provides a channel information feedback method and a wireless communication apparatus.

In order to solve the above problem, the present invention includes first and second wireless communication apparatuses having a plurality of transmission / reception antennas, and a plurality of signal sequences in the same frequency channel between the plurality of transmission / reception antennas. Channel information feedback method in a wireless communication system that performs spatial multiplexing and transmission in the second wireless communication apparatus, wherein each of a plurality of orthogonal known pattern signals is used as a first orthogonal pilot signal. Performing a first orthogonal pilot signal giving process to be given to the head region of the radio control packet used for each radio control transmitted from the transmission antenna of the radio control packet, and an information region following the head region of the radio control packet On the other hand, a step of performing signal virtual orthogonalization processing for performing the same processing as the orthogonalization processing of the orthogonal known pattern signals is performed. And performing signal transmission processing for transmitting the radio control packet without spatial multiplexing using the plurality of transmission antennas, and in the first radio communication apparatus, Performing signal reception processing for reception using a plurality of reception antennas; performing orthogonal pilot signal extraction processing for extracting a signal in a region to which the first orthogonal pilot signal is added from the received radio control packet; Channel estimation processing for generating transfer function matrix information between the transmitting and receiving antennas for communication in the direction of the first wireless communication device from the second wireless communication device based on the relationship between the extracted signal and the known pattern signal And a direction opposite to the communication direction corresponding to the transfer function matrix information based on the generated transfer function matrix information Performing a transfer function calculation process for calculating transfer function matrix information between the transmitting and receiving antennas for communication in the direction from the first wireless communication apparatus to the second wireless communication apparatus. Is the method.

The present invention includes first and second wireless communication apparatuses having a plurality of transmission / reception antennas, and performs transmission by spatially multiplexing a plurality of signal sequences on the same frequency channel between the plurality of transmission / reception antennas. A channel information feedback method in a wireless communication system, wherein each of a plurality of orthogonal known pattern signals is transmitted from the plurality of transmission antennas as a first orthogonal pilot signal in the second wireless communication apparatus. Performing a first orthogonal pilot signal adding process to be added to a leading area of a radio control packet used for radio control of the radio control packet, and a known pattern orthogonal to the information area following the leading area of the radio control packet and performing a signal virtualization orthogonalization process applied the same processing as straight交化processing of the signal, the first orthogonal pie Either Tsu bets signal, or each of the known pattern signal in which a plurality of orthogonal obtained by multiplying a predetermined coefficient to the first orthogonal pilot signal as a second orthogonal pilot signals, from the plurality of transmit antennas Performing a second orthogonal pilot signal providing process to be added to each radio control packet to be transmitted, and performing a signal transmission process for transmitting the radio control packet without spatial multiplexing using the plurality of transmission antennas And a step of performing signal reception processing for receiving the radio control packet using the plurality of reception antennas in the first radio communication device, and from the received radio control packet, the first radio communication device Orthogonal pilot signal extraction processing for extracting a signal in a region to which the orthogonal pilot signal and the second orthogonal pilot signal are added Steps and, the transfer function between the transmitting and receiving antennas the extracted signal from said second radio communication device based on the relationship between the linear addition signal the known pattern signal for communication of the first wireless communications device a direction to perform A step of performing channel estimation processing for generating matrix information, and a second direction from the first wireless communication apparatus that is in a direction opposite to the direction of communication corresponding to the transfer function matrix information based on the generated transfer function matrix information. Performing a transfer function calculation process for calculating transfer function matrix information between the transmitting and receiving antennas for communication in the direction of a wireless communication device.

According to the present invention, in the invention described above, the first radio communication device and the second radio communication device each use orthogonal frequency division multiplexing modulation using K (K> 1: K is an integer) subcarriers. The first wireless communication apparatus has N (N> 1: N is an integer) transmission / reception antennas, and the second wireless communication apparatus has M (M> 1: M is an integer). Are transmitted from the plurality of transmission / reception antennas in the step of performing the first orthogonal pilot signal providing process or the second orthogonal pilot signal providing process of the second wireless communication apparatus. said first orthogonal pilot signals applied to each of the radio control packet or the second orthogonal pilot signals, when decomposed into components for each of the sub-carrier from each of the transmitting and receiving antenna Each of the received signals is set so as not to include at least one signal component of the subcarrier, and in the step of performing the channel estimation process, the kth (1) of the K subcarriers ≦ k ≦ K: k is an integer) The j-th (1 ≦ j ≦ M: j is an integer) th transmission / reception antenna of the second wireless communication device of the subcarrier and the first wireless communication device of the first wireless communication device i (1 ≦ i ≦ N: i is an integer) when performing acquisition of the transfer function matrix information between th receiving antenna, the first sent from the second of the j-th reception antenna of the wireless communication device When the component of the k-th subcarrier is included in one orthogonal pilot signal or the second orthogonal pilot signal, the first orthogonal pilot signal, the second orthogonal pilot signal, or these Linear addition Signal and by calculating a cross-correlation between the signal of the known pattern, between the second of the i-th antenna of the said j-th reception antenna of the wireless communication device first wireless communication device perform direct transfer function and acquires the transfer function matrix information directly, the first orthogonal pilot signals transmitted from the j-th reception antenna of the second radio communication device or the second, If there are no components of the k-th subcarrier in the orthogonal pilot signals, the acquired in the direct transfer function acquisition process at least one subcarrier close to the k-th subcarrier first interpolation about transfer function matrix information between the i-th reception antenna of the said j-th reception antenna of the second radio communication device a first wireless communication device Alternatively, transfer function complement acquisition processing is performed in which transfer function matrix information of the subcarrier is acquired by complementation by extrapolation calculation.

According to the present invention, in the above-described invention, the first wireless communication device includes a wireless unit for each of the transmission / reception antennas, and the wireless unit is connected to the h- th antenna of the own device. When the complex physical quantity indicating the amplification factor and phase rotation amount of the transmission signal is α _h and the complex physical quantity indicating the amplification factor and phase rotation amount of the reception signal in the h- th radio unit is β _h , the first In the wireless communication device, performing a calibration coefficient acquisition process for acquiring α _h / β _h that is a ratio of the complex physical quantity of the own device at a predetermined period for each of the transmission and reception antennas;
In the predetermined cycle, the amplification ratio and phase rotation amount of the transmission signal in the radio unit, or reception in the radio unit so that the ratio of the complex physical quantity takes the same value across all the transmission / reception antennas of its own device Performing a calibration process for adjusting a signal amplification factor and a phase rotation amount for each of the transmission / reception antennas, and performing the transfer function correction process of the first wireless communication device. The transmission / reception antenna for communication from the first wireless communication device to the second wireless communication device as a matrix obtained by transposing a transfer function matrix between the transmission / reception antennas for communication from the wireless communication device to the first wireless communication device. The transfer function matrix information is given.

The present invention includes first and second wireless communication apparatuses having a plurality of transmission / reception antennas, and performs transmission by spatially multiplexing a plurality of signal sequences on the same frequency channel between the plurality of transmission / reception antennas. a wireless communication system, a radio which the second wireless communication device is used for each radio control sent from the plurality of transmit antennas each of a plurality of orthogonal known pattern signal as Cartesian pilot signal Orthogonal pilot signal adding means for adding to the head region of the control packet, and signal virtual for performing the same processing as the orthogonal processing of the orthogonal known pattern signal on the information region following the head region of the radio control packet includes a quadrature means, a signal transmitting means for transmitting the wireless control packet without spatial multiplexing using the plurality of transmitting antennas, wherein the 1 radio communication apparatus that receives the radio control packet using the plurality of reception antennas, and an orthogonal pilot that extracts a signal in a region to which the orthogonal pilot signal is added from the received radio control packet. Signal estimation means, and channel estimation means for generating transfer function matrix information between the transmitting and receiving antennas for communication in the direction of the own apparatus from the second wireless communication apparatus based on the relationship between the extracted signal and the known pattern signal Based on the generated transfer function matrix information, transfer function matrix information between the transmitting and receiving antennas for communication in the direction of the second wireless communication device from the own device that is in the opposite direction to the communication direction corresponding to the transfer function matrix information. A wireless communication system comprising: a transfer function matrix calculating means for calculating .

The present invention includes first and second wireless communication apparatuses having a plurality of transmission / reception antennas, and performs transmission by spatially multiplexing a plurality of signal sequences on the same frequency channel between the plurality of transmission / reception antennas. a wireless communication system, said second wireless communication apparatus, used in each radio control transmitted each of the plurality of orthogonal known pattern signal as a first orthogonal pilot signals from the plurality of transmit antennas Same as the orthogonal processing of the orthogonal known pattern signal with respect to the information area subsequent to the head area of the radio control packet; Signal virtual orthogonalization means for performing processing, and one of the first orthogonal pilot signals, or multiplying the first orthogonal pilot signal by a predetermined coefficient Second orthogonal pilot signal providing means for assigning each of the plurality of orthogonal known pattern signals obtained as a second orthogonal pilot signal to each of the radio control packets transmitted from the plurality of transmission antennas; Signal transmitting means for transmitting the radio control packet without spatial multiplexing using the plurality of transmission antennas , wherein the first radio communication device transmits the radio control packet using the plurality of reception antennas. Signal receiving means for receiving, orthogonal pilot signal extracting means for extracting a signal of a region to which the first orthogonal pilot signal and the second orthogonal pilot signal are added from the received radio control packet, and the extracted signal On the basis of the relationship between the signal obtained by linear addition and the known pattern signal, Channel estimation means for generating transfer function matrix information between the transmitting and receiving antennas for transmission, and the second apparatus from the own device that is in a direction opposite to the direction of communication corresponding to the transfer function matrix information based on the generated transfer function matrix information. And a transfer function calculating means for calculating transfer function matrix information between the transmitting and receiving antennas for communication in the direction of the wireless communication device .

According to this invention, the first and second wireless communication apparatuses having a plurality of transmission / reception antennas are provided, and a plurality of signal sequences are spatially multiplexed and transmitted between the plurality of transmission / reception antennas on the same frequency channel. Radio control packet used for radio control transmitted from a plurality of transmitting antennas by using each of a plurality of orthogonal known pattern signals as a first orthogonal pilot signal in a second radio communication device of the radio communication system to perform The first orthogonal pilot signal providing process to be applied to the head region of the first region is performed, the same processing as the orthogonal processing of the known pattern signal orthogonal to the information region subsequent to the head region is performed, and a plurality of radio control packets are transmitted. Transmission is performed without spatial multiplexing using a transmission antenna, and a radio control packet is received using a plurality of reception antennas at the first wireless communication apparatus, and received. It said first orthogonal pilot signals extracts the signal of a given area from the wireless control packet, the extracted signal and the first radio communication apparatus direction from the second radio communication device based on the relationship between the known pattern signal Channel estimation processing for generating a transfer function between the transmitting and receiving antennas for the communication of the first communication from the first wireless communication device in the direction opposite to the direction of communication corresponding to the transfer function based on the generated transfer function to the second wireless The transfer function between the transmitting and receiving antennas for communication in the communication device direction is calculated.
As a result, in the feedback for acquiring the channel information on the transmitting side, all the transmitting and receiving antennas can be exchanged by exchanging the existing control information between the transmitting and receiving stations without adding new overhead or minimizing the addition. It becomes possible to collect information about.

In addition, according to the present invention, the first and second wireless communication apparatuses having a plurality of transmission / reception antennas are provided, and a plurality of signal sequences are spatially multiplexed on the same frequency channel between the plurality of transmission / reception antennas. In the second wireless communication apparatus of the wireless communication system that performs transmission, the plurality of orthogonal known pattern signals are used as the first orthogonal pilot signals for the respective wireless controls transmitted from the plurality of transmission antennas. The first orthogonal pilot signal adding process to be added to the head area of the radio control packet is performed, and the same process as the orthogonal processing of the known pattern signal orthogonal to the information area following the head area of the radio control packet is performed. , One of the first orthogonal pilot signals, or a plurality of orthogonal known patterns obtained by multiplying the first orthogonal pilot signal by a predetermined coefficient. Each chromatography tone signal as the second quadrature pilot signals, performs a second orthogonal pilot signals are given processing to be applied to each of the radio control packets transmitted from a plurality of transmitting antennas, a plurality of transmitting antennas a radio control packet The first radio communication apparatus receives a radio control packet using a plurality of receiving antennas, and receives the first orthogonal pilot signal and the second orthogonal from the received radio control packet. A transmission / reception antenna for communication in the direction from the second wireless communication apparatus to the first wireless communication apparatus based on a relationship between a signal obtained by linearly adding the extracted signals and a known pattern signal. The channel estimation process that generates the transfer function between them is performed, and based on the generated transfer function, the direction opposite to the direction of communication corresponding to the transfer function And configured to calculate a transfer function between the transmitting and receiving antennas from a certain first wireless communication device to the communication of the second wireless communication device direction.
Thereby, even when the number of transmission / reception antennas for performing channel estimation increases by adding the second orthogonal pilot signal to the radio control packet transmitted from the transmission / reception antenna, the first orthogonal pilot signal and the second orthogonal pilot are increased. It is possible to provide a means for configuring more orthogonal pilot signals by combining signals.

Further, according to the present invention, the first radio communication apparatus and the second radio communication apparatus each adopt an orthogonal frequency division multiplexing modulation scheme using K (K> 1: K is an integer) subcarriers. The first wireless communication apparatus has N (N> 1: N is an integer) transmission / reception antennas, and the second wireless communication apparatus has M (M> 1: M is an integer) transmission / reception antennas. When the first radio pilot signal providing process or the second orthogonal pilot signal providing process of the second radio communication apparatus is performed, the first radio control packet transmitted from the plurality of transmission / reception antennas is assigned to the first radio control packet. When the orthogonal pilot signal or the second orthogonal pilot signal is decomposed into components for each subcarrier, each signal transmitted from each of the transmission / reception antennas includes at least one signal component of the subcarrier. When the channel estimation is performed, the jth (1) of the second wireless communication apparatus of the kth (1 ≦ k ≦ K: k is an integer) th subcarrier among the K subcarriers. ≦ j ≦ M: j is an integer) transmission / reception antenna and the i-th (1 ≦ i ≦ N: i is an integer) th transmission / reception antenna of the first wireless communication apparatus, When the first orthogonal pilot signal transmitted from the jth transmission / reception antenna of the second wireless communication apparatus or the component of the kth subcarrier is included in the second orthogonal pilot signal, the first orthogonal pilot signal By calculating the cross-correlation between the pilot signal, the second orthogonal pilot signal, or a signal obtained by linearly adding them, and a signal having a predetermined known pattern, the j-th signal of the second wireless communication apparatus is obtained. Transmit / receive antenna and first wireless communication The first orthogonal pilot signal transmitted from the j-th transmission / reception antenna of the second wireless communication device, performing direct transfer function acquisition processing for directly acquiring a transfer function between the i-th antennas of the second wireless communication device, Alternatively, when the second orthogonal pilot signal does not include the component of the kth subcarrier, the first acquired in the direct transfer function acquisition process with at least one subcarrier close to the kth subcarrier. The interpolation function for the transfer function between the j-th transmission / reception antenna of the second wireless communication device and the i-th transmission / reception antenna of the first wireless communication device is supplemented by interpolation or extrapolation to obtain the transfer function of the subcarrier. It was set as the structure to do.
In the present invention, the orthogonality of each subcarrier in OFDM is used, and a simple implementation for orthogonalizing pilot signals to be assigned to radio control packets for each antenna while maintaining consistency with existing radio communication systems. Means can be provided. Also, even when the channel transfer function matrix information acquired using the orthogonalized pilot signal has only an incomplete number of information for all subcarriers and all paths, all subcarriers are maintained while maintaining the estimation accuracy. In addition, it is possible to provide a simple realization method for obtaining transfer function matrix information of all paths.

Further, according to the present invention, the first wireless communication device includes a wireless unit for each transmission / reception antenna, and the wireless unit transmits the amplification factor and phase of the transmission signal in the wireless unit connected to the h- th antenna of the own device. When the complex physical quantity indicating the rotation amount is α _h and the complex physical quantity indicating the amplification factor and the phase rotation amount of the received signal in the h- th radio unit is β _h , the first wireless communication apparatus uses a predetermined period. Calibration coefficient acquisition processing for acquiring α _h / β _h that is the ratio of the complex physical quantity of the own device is performed for each of the transmission / reception antennas, and the ratio of the complex physical quantity is determined for all of the transmission / reception of the own device at a predetermined period. Calibration processing for adjusting the amplification factor and phase rotation amount of the transmission signal in the radio unit or the amplification factor and phase rotation amount of the reception signal in the radio unit so as to take the same value over the antenna The transfer function correction process of the first wireless communication apparatus is performed, and the transfer function matrix between the transmission and reception antennas for the communication from the second wireless communication apparatus to the first wireless communication apparatus is transposed as the first matrix. The transmission function matrix of the transmission / reception antenna for communication in the direction from the wireless communication device to the second wireless communication device is provided.
Thereby, in the environment where the objectivity of the transfer function matrix information between the forward link and the return link is not secured, conversion between the two can be facilitated. As a result, even when the characteristics of the power amplifier, the low noise amplifier, and the like change due to time fluctuation or temperature-dependent fluctuation, it is possible to follow the fluctuation in real time.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
(First embodiment)
First, the operation principle of the wireless communication apparatus according to the first embodiment of the present invention will be described. FIG. 1 is a configuration example of a radio control packet in the first embodiment. Here, for example, a case where the second wireless communication apparatus has two antennas will be described. In FIG. 1, reference numerals 11-1 and 11-2 are short training signals, reference numerals 12-1 and 12-2 are long training signals, reference numerals 13-1a and 13-2a are PLCP control signals, reference numerals 14-1a and 14- 2a indicates control information or data. The purpose of use of these signals is the same as that of the prior art described above.

  In the first embodiment, the transferred control information or data 14-1a and 14-2a are not spatially multiplexed by MIMO, and the contents of the stored information are the same. The long training signals 12-1 and 12-2 are preambles capable of MIMO channel estimation, and are orthogonalized with each other, and the contents of processing related to orthogonalization are shown as pattern # 1 and pattern # 2. Each of the long training signals 12-1 and 12-2, PLCP control signals 13-1a and 13-2a, control information or data 14-1a and 14-2a is subjected to a common process for each antenna, Among these, at least the long training signals 12-1 and 12-2 are in an orthogonal relationship (or orthogonal relationship in a certain time domain).

  FIG. 2 is another configuration example of the radio control packet in the first embodiment. In FIG. 2, the short training signals 11-1 and 11-2 and the long training signals 12-1 and 12-2 are the same as those in FIG. In the radio control packet of FIG. 2, the PLCP control signals 13-1a and 13-2a of FIG. 1 are omitted, and the control information or data 14-1b and 14-2b follows the long training signals 12-1 and 12-2. Has been placed. Also in FIG. 2, similar to the case of FIG. 1, common processing is performed for each antenna, and the long training signal 12-1 and the long training signal 12-2 are in an orthogonal relationship.

  FIG. 1 assumes packet-based control such as IEEE802.11a / g, whereas FIG. 2 assumes centralized control using TDMA. The PLCP control signals 13-1a and 13-2a in FIG. 1 contain the subsequent control information or the transmission mode and data length of the data 14-1a and 14-2a. In the case of the TDMA control shown in FIG. It is also possible to notify these information in other packets. Therefore, what is important in FIGS. 1 and 2 is that long training signals 12-1 and 12-2 that can estimate the transfer function information of the MIMO channel are provided, and in FIG. 1 that follows. Each of the control information or data 14-1a and 14-2a is subjected to the same processing. In FIG. 2, the same processing is performed on each of the control information or data 14-1b and 14-2b. Alternatively, the information accommodated in the data 14-1a and 14-2a is exactly the same, and the information accommodated in the control information or the data 14-1b and 14-2b is also exactly the same.

  FIG. 3 shows an orthogonalization method for each antenna in the first embodiment. In FIG. 1, the content of the processing related to orthogonalization is shown as pattern # 1 and pattern # 2. However, in FIG. 3, as a specific method of the processing, the non-patent document 2 described above is used for orthogonalizing long training signals. The addition of a cyclic shift of a phase corresponding to a predetermined time used as a means is used.

  For example, when an OFDM modulation scheme is used, for example, assuming IEEE802.11a / g, an OFDM symbol having a period of 4 μsec is set, 800 nsec among them is used as a guard interval, and the remaining 3. Information was transferred by cutting out a 2 μsec portion. In this case, there is a feature that the correlation between the signal obtained by adding a phase shift corresponding to a delay of 1600 nsec corresponding to half of this value to the long training signal and the original signal becomes zero.

  The long training signal for MIMO in Non-Patent Document 2 uses this condition. However, in Non-Patent Document 2, the subsequent data is spatially multiplexed, so the cyclic shift is not performed in the data part transmitted from each antenna. Not granted. On the other hand, in the first orthogonalization method for each antenna in the first embodiment shown in FIG. 3, not only the long training signal 22-2 of the antenna # 2, but also the PLCP control signal 23-2 and the control information or data 24 -2 is also subjected to a cyclic shift of 1600 nsec. Note that any other means may be used as long as the orthogonalization condition in the long training signal between the plurality of signals is satisfied.

  When there are three or more patterns for performing the orthogonalization as described above, the second wireless communication apparatus can be easily expanded even when the number of antennas is three or more. As shown in FIG. The processing of # 3 and pattern # 4 may be performed similarly for antenna # 3 and antenna # 4.

(Second Embodiment)
FIG. 5 shows a configuration example of a radio control packet in the second embodiment of the present invention. 5 is different from FIG. 4 according to the first embodiment in that long training signals 35-1 to 35-4 are added following the radio control packet described in FIG. 4.

  At this time, the long training signal 32-1 and the long training signal 35-1 and the long training signal 32-2 and the long training signal 35-2 have the same contents. The long training signal 32-3 and the long training signal 35-3, and the long training signal 32-4 and the long training signal 35-4 have exactly the same contents except that the signs are inverted (that is, the phase is shifted by π). Signal.

  When the signal obtained by combining the long training signals 32-1 to 32-4 in space and the signal obtained by combining the long training signals 35-1 to 35-4 in space are added, the long training signal 32-3 and the long training signal are added. 35-3, and the long training signal 32-4 and the long training signal 35-4 cancel each other, and only the long training signals of the antenna # 1 and the antenna # 2 remain.

  Similarly, when the signal obtained by synthesizing the long training signals 32-1 to 32-4 in space and the signal obtained by synthesizing the long training signals 35-1 to 35-4 in space are subtracted, the long training signal 32-1 The long training signal 35-1, the long training signal 32-2 and the long training signal 35-2 cancel each other, and only the long training signals of the antenna # 3 and the antenna # 4 remain.

  This makes it possible to obtain transfer functions related to all antennas from antenna # 1 to antenna # 4. In FIG. 4 of the first embodiment, the pattern # 3 and the pattern # 4 are different from the pattern # 1 and the pattern # 2. However, in the second embodiment, the pattern # 3 and the pattern # are not necessarily processed. 4 need not be different from pattern # 1 and pattern # 2. Specifically, even if the pattern # 1 and pattern # 3 and the pattern # 2 and pattern # 4 have the same processing content, the long training signal can be separated for each antenna. In the present specification, such a signal that can be separated by linear addition of the long training signal region is also called “orthogonal” in a broad sense. Therefore, when a good orthogonalization method cannot be selected for the desired number of antennas, it can be dealt with by arranging a pattern orthogonal to the long training signal for each head antenna at the end of the radio control packet. Is possible.

  The above description is an embodiment under conditions that do not necessarily use the OFDM modulation scheme and can be extended to a single carrier system. On the other hand, when the OFDM modulation method is used, other embodiments can be used, which will be described below.

(Third embodiment)
FIG. 6 shows a method for orthogonalizing a long training signal for each antenna in the third embodiment of the present invention. In the OFDM modulation scheme, a signal is transmitted using a plurality of subcarriers. In FIG. 6, the horizontal axis represents the subcarrier number. Here, for example, in the OFDM modulation scheme used in IEEE 802.11a / g, a pilot subcarrier for transmitting a known signal is used in addition to a subcarrier for transmitting data.

  The subcarrier number in FIG. 6 may be either a serial number assigned only to data subcarriers or a serial number assigned to subcarriers including pilot subcarriers. Here, for example, a case where serial numbers are assigned only to data subcarriers will be described. In the following description, the case where the number of antennas of the second wireless communication apparatus is two is taken as an example, but the same extension is possible when there are three or more antennas.

  As shown in FIG. 6, in the third embodiment, paying attention to a certain subcarrier, a long training signal of the subcarrier is transmitted from only one antenna, and the antenna is switched for each subcarrier. In FIG. 6, long training signals 41a, 43a, and 45a in the first antenna 47a are used for odd subcarriers, and long training signals 42a, 44a, and 46a in the second antenna 47a are used for even subcarriers.

  Similarly, when there are three transmission antennas, the cycle may be three, and when there are four transmission antennas, the cycle may be four. Further, it is not necessary to be completely periodic, and the antennas can be switched at random. Further, the long training signal of each subcarrier may be the same for all subcarriers or may be different for each subcarrier.

  As shown in FIG. 7, a signal subsequent to this long training is transmitted by selecting the antenna used for the long training transmission of FIG. 6 for each subcarrier. Specifically, odd-numbered subcarriers use data # 1a (41b), # 1b (43b), # 1c (45b) in the first antenna 47b, and even-numbered subcarriers are data # 2a in the second antenna 48b. (42b), data # 2b (44b), and data # 2c (46b) are used. In FIG. 7, the subsequent signal is described as data, but various control information may be included or PLCP control information may be included.

  In the technique shown in Non-Patent Document 3 described above, the same transmission method is used for the long training signal, but the subsequent data is not the configuration shown in FIG. 7, and individual signal sequences are transmitted from all antennas. Different in that it is.

  In addition, in the non-patent document 3, the purpose of using the long training signal is to apply the long training signal in order to perform signal separation of the subsequent data portion in the MIMO communication, whereas in the third embodiment, the subsequent data is a space. Do not multiplex.

  On the receiving side in the third embodiment, this long training signal is used for two purposes. One purpose of use is to demodulate the subsequent data portion shown in FIG. Another purpose of use is to perform channel estimation of a MIMO channel that was not used for subsequent data transmission.

  For example, taking the case of FIGS. 6 and 7 as an example, no long training signal is transmitted from the first antenna (47a, 47b) in the second subcarrier. Therefore, the transfer function related to the first antenna of the second subcarrier cannot be obtained directly. However, for the first antennas (47a, 47b), the transfer functions of the first and third subcarriers can be acquired. It is possible to estimate the transfer function value by a process such as interpolation using the transfer functions of these two or other neighboring subcarriers.

  Similarly, for subcarrier # 3, the transfer function for the second antenna (48a, 48b) cannot be obtained directly, but can be estimated from the transfer functions of subcarrier # 2 and # 4. This estimation process can be performed by extrapolation in addition to interpolation. For example, the first subcarrier is estimated from the information on subcarriers No. 2 and No. 4 regarding the second antenna (48a, 48b). It is also possible. Furthermore, any estimation means such as spline interpolation other than linear estimation may be used for the interpolation and extrapolation processing.

  As a supplement, in Non-Patent Document 3 described above, the transfer function of the MIMO channel is estimated by the same complementary processing. However, in Non-Patent Document 3, the estimated transfer is directly performed for signal separation of the MIMO channel. A function matrix was used. For this reason, when the number of antennas increases, the density of information used when performing the supplement processing in the subcarrier direction decreases, and the estimation error of the transfer function matrix increases accordingly. Therefore, in the technique described in Non-Patent Document 3, it is difficult to adapt in a region where the number of antennas is large, but in the third embodiment, the estimated MIMO channel information is used for signal separation of subsequent data. Rather than being used to perform beam forming during transmission of a wireless data packet to be transmitted after reception of a wireless control packet or wireless data packet, the estimation error shown in Non-Patent Document 3 is an error within a sufficiently allowable range. is there.

(Fourth embodiment)
FIG. 8 is a diagram illustrating a method for orthogonalizing long training signals for each antenna according to the fourth embodiment of the present invention. As in the third embodiment, the fourth embodiment is based on the OFDM modulation scheme. In the fourth embodiment, as described in the first embodiment above, when long training signals are simultaneously transmitted from a plurality of antennas on the same subcarrier, the long training signals are orthogonal to each other. As long as they can be separated from each other, it is utilized that a transfer function can be estimated appropriately even if long training signals are simultaneously transmitted from a plurality of antennas on the same subcarrier.

  As shown in FIG. 8, the pattern is changed so that the long training signals are orthogonal between the first and second antennas and between the third and fourth antennas while alternately changing the transmission antennas with even and odd subcarriers. ing. This pattern may be, for example, a cyclic shift for a predetermined time as described with reference to FIG. 3, or may be another method. Once these patterns are determined, the same processing as the previous pattern is performed on the data or control information following the long training signal. The subsequent data or control information may include PLCP control information.

  Furthermore, as shown in the second embodiment, it is also possible to add these long training signals or their inverted sign to the end of the radio control packet or radio data packet. When the number of antennas that perform channel estimation increases, the number of MIMO channels that can be estimated can be increased by a combination of such embodiments.

  Up to this point, a specific pilot signal has been described as an example. However, the intention of “orthogonal” in the long training signal is slightly different, and will be described below.

  In this invention and embodiment, what corresponds to either of the following conditions (1)-(4) is described as orthogonal (each long training signal is orthogonal).

(1) For example, when the OFDM modulation method is applied, the signal for each subcarrier can be separated by performing FFT processing. When the FFT processing can separate and acquire the amplitude and phase information (that is, transfer function) regarding the channel training long training signal of each antenna in the received signal, “each long training signal is orthogonal” And

(2) The transfer function can be obtained by obtaining an impulse response for each antenna. In order to obtain an impulse response, a predetermined pattern in which the correlation is zero when the time is shifted by δ in a certain time region equal to or less than the symbol length and autocorrelation is obtained over one symbol is used as a long training signal. The impulse response is obtained by cross-correlating the training signal with the long training signal. There are multiple types of predetermined patterns with such characteristics. However, if the cross-correlation patterns are zero in the above time domain, the impulse response of the signal of each pattern can be obtained even if they are simultaneously spatially multiplexed. It can be obtained separately. In this way, when the impulse responses can be separated, it is assumed that “the long training signals are orthogonal”. (For example, in the case of a long training signal such as IEEE802.11a, the correlation between a signal shifted by 1600 ns and a signal without shift is zero in the region from 0 to 1600 ns. This is because the signal of IEEE802.11a (Because it has periodicity, it should be noted that in other systems with different symbol lengths, the amount of time for phase shift, etc. will be different.)

(3) By linearly adding the first orthogonal pilot signal and the second orthogonal pilot signal, the signal component of a certain antenna that is spatially multiplexed in the first orthogonal pilot signal and second orthogonal pilot signal regions is removed. Or a signal component related to a specific antenna can be extracted. In this way, when a transfer function related to a certain antenna can be obtained by linear synthesis of signals of a plurality of symbols, it is assumed that “each long training signal is orthogonal”.

(4) Even when a transfer function related to a certain antenna can be acquired by the combination of (1) to (3) above, it is assumed that “long training signals are orthogonal”.

  Next, a flowchart of processing of a MIMO channel information feedback method in the first wireless communication device and the second wireless communication device that realizes the operation principle of the first to fourth embodiments described above is shown.

(Fifth embodiment)
FIG. 9 is a diagram illustrating a transmission process flow of the radio control packet according to the fifth embodiment of the present invention. The fifth embodiment is a transmission process of a wireless communication system based on the operation principle described in the first, third, and fourth embodiments. Here, for the sake of simplicity, the case where signals are transmitted using two antennas is shown. Further, a system based on IEEE802.11a / g is taken as a specific example, and it is assumed that PLCP control information is included in a transmission signal in the system.

  First, when control information to be transmitted is input (step S1), the control information is copied and the same contents are output to two systems (step S2). A short training signal is given to each system (steps S3-1 and S3-2). Thereafter, a long training signal is applied in each system (steps S4-1 and S4-2). However, the long training signal given here is obtained by processing each of the other patterns. That is, pattern # 1 is performed in the process of step S5-1, and pattern # 2 is performed in the process of step S5-2. Here, the “pattern” corresponds to the content corresponding to the first, third, and fourth embodiments, and indicates the content of the process for orthogonalizing the long training signal for each stream.

  Next, in the first system, PLCP control information is assigned while performing the process of pattern # 1 (step S5-1), control information to be transmitted is assigned (step S6-1), and the radio control packet is transmitted. The formed radio control packet is subjected to a predetermined modulation process and transmitted from each antenna (step S7-1). Similarly, in the second system, PLCP control information is assigned while performing pattern # 2 (step S5-2), control information to be transmitted is assigned (step S6-2), and the radio control packet is transmitted. The formed radio control packet is subjected to predetermined modulation processing and transmitted from each antenna (step S7-2). After the above processing, the processing is terminated (step S8).

  The above is an example of a system based on IEEE802.11a / g, but if PLCP control information is not required, steps S5-1 and S5-2 can be omitted.

(Sixth embodiment)
FIG. 10 is a diagram showing a radio control packet transmission flow according to the sixth embodiment of the present invention. The sixth embodiment is a transmission process of a wireless communication system based on the operation principle of the second embodiment. In FIG. 10, steps corresponding to the processing shown in FIG. 9 are assigned step numbers obtained by changing the antenna sequence numbers.

  The difference from FIG. 9 is that processing (steps S10-1, S10-2, S10-3, and S10-4) that provides a long training signal after the processing that provides control information (steps S6-1 and S6-2). ). In addition, here, as an example, a case where a signal is transmitted using four antennas is shown. Therefore, in step S9, a control signal is copied instead of step S2 in FIG. Yes. If there are two antennas, the series of steps S3-2 to S7-2 and the series of steps S3-4 to S7-4 can be omitted. In step S7-3, a signal obtained by reversing the sign of the long training signal given in step S7-1 is given. Similarly, in step S7-4, a signal obtained by reversing the sign of the long training signal given in step S7-2 is given.

(Seventh embodiment)
FIG. 11 is a diagram showing a radio control packet transmission flow according to the seventh embodiment of the present invention. The seventh embodiment is a transmission process of a wireless communication system based on the operation principle of the third embodiment.

  The main difference from FIG. 9 is that the processing corresponding to steps S3-1 to S7-1 and steps S3-2 to S7-2 in FIG. This is the point that is implemented. Therefore, in step S13-1 and step S13-2, S / P conversion for distributing control information for each subcarrier is performed, and step S14-1 to step S17-1 (this is generally referred to as step S20-1). And step S14-2 to step S17-2 (referred to as step S20-2 as a whole), and based on the result, step S18-1 and step S18-2 are performed. Do. The processing in step S18-1 and step S18-2 corresponds to the processing in step S7-1 and step S7-2 in FIG. 9, except that IFFT (Inverse Fast Fourier Transfer) processing in the OFDM modulation scheme is included. This is the point to implement.

  Note that the OFDM modulation scheme can also be applied to the sixth embodiment shown in FIG. 10, and steps S3-1 to S10-1, steps S3-2 to S10-2, steps in FIG. For the processes of S3-3 to S10-3 and steps S3-4 to S10-4, a process corresponding to the OFDM modulation scheme such as the processes of steps S20-1 and S20-2 in FIG. 11 can be applied. That's fine.

The fifth to seventh embodiments described above are processes on the side of transmitting a radio control packet. Next, the eighth to tenth embodiments showing the reception process in the first radio receiving apparatus will be described. explain.
(Eighth embodiment)
FIG. 12 is a diagram illustrating a reception process flow of a radio control packet in the first radio communication apparatus according to the eighth embodiment of the present invention. In FIG. 12, a plurality of antennas included in the first wireless communication apparatus and processes in the reception system connected to the antennas are shown as step S30-1, step S30-2,..., Step S30-3, respectively.

 When a signal is received by each receiving system (steps S22-1 to S22-3), timing detection processing is performed using a correlation with a known pattern (steps S23-1 to S23-3). Based on the detected timing, an orthogonal pilot signal for channel estimation is extracted for each sequence (steps S24-1 to S24-3), and channel estimation is performed individually (steps S25-1 to S25-3). When receiving control information subsequent to the orthogonal pilot signal, demodulation processing is performed based on the channel estimation result on the assumption that they are not spatially multiplexed (step S26). Thereafter, a transfer function matrix is generated by converting the channel estimation result obtained in steps S25-1 to S25-3 into a channel estimation result based on the MIMO channel according to the orthogonality of the orthogonal pilot signal. (Step S27).

  Thereafter, a transfer function matrix correction process is performed according to the difference between the transfer functions of the uplink and downlink (or forward link and return link), and the radio from the second radio communication device to the first radio communication device is performed. The transfer function matrix of the MIMO channel in the direction of the second wireless communication device is calculated from the first wireless communication device in the opposite direction from the transfer function matrix of the channel estimation result with the control packet (step S28), and the process is terminated. (Step S29).

  Here, the timing detection processing in steps S23-1 to S23-3 may be performed individually in each reception system, or the timings may be extracted by combining the reception timings detected in each reception system. Further, the demodulation processing in step S26 may be combined as appropriate, for example, by combining the signals received by the respective receiving systems with a maximum ratio to improve the reception quality, or may be performed individually. A feature of the eighth embodiment is that the channel estimation results of steps S25-1 to S25-3 are used also in the demodulation step S26 of the control information, and also used to generate the transfer function matrix of the MIMO channel in step S27. It is.

  Furthermore, in the second and sixth embodiments described above, further orthogonal pilot signals are added and transmitted following the control information, but these orthogonal pilot signals are transmitted in steps S24-1 to S24-3. If it is understood that both are extracted together, it can be expanded to embodiments corresponding to the second and sixth embodiments. However, at this time, the channel estimation information used when demodulating the control information in step S26 is acquired from the orthogonal pilot signal transmitted prior to the control information.

(Ninth embodiment)
FIG. 13 shows a processing flow on the reception side of the first wireless communication apparatus of the wireless control packet using the OFDM modulation method as the ninth embodiment of the present invention. The system according to the ninth embodiment corresponds to a system extended by applying the OFDM modulation scheme to the eighth embodiment.

  When a signal is received by each receiving system (steps S32-1 to S32-3), timing detection processing is performed using a correlation with a known pattern (steps S33-1 to S33-3). Based on the detected timing, orthogonal pilot signals for channel estimation are extracted from each sequence (steps S34-1 to S34-3), and FFT (Fast Fourier Transfer) processing is performed individually (steps). S35-1 to S35-3). Based on the result of the FFT processing, channel estimation in each subcarrier is performed (steps S36-1 to S36-3). When receiving control information subsequent to the orthogonal pilot signal, demodulation processing is performed based on the channel estimation result on the assumption that they are not spatially multiplexed (step S37). Thereafter, according to the channel estimation results of steps S36-1 to S36-3 and the orthogonality of the orthogonal pilot signals, a transfer function matrix for each subcarrier is generated by converting into channel estimation results assuming a MIMO channel ( Step S38). Thereafter, the transfer function matrix is corrected in accordance with the difference between the uplink and downlink (or forward link and return link) transfer functions, and the radio from the second radio communication device to the first radio communication device is performed. The transfer function matrix of the MIMO channel in the direction of the second wireless communication device is calculated from the first wireless communication device in the opposite direction from the transfer function matrix of the channel estimation result with the control packet (step S39), and the process is terminated. (Step S40).

  Here, the processing of steps S41-1 to S41-3 indicates processing performed for each reception system connected to each reception antenna. Further, the processing related to the region enclosed in step S42 indicates a region where processing for each subcarrier is performed in the reception system of the system using the OFDM modulation scheme. In the reception processing of the OFDM modulation system, usually, the signal is cut out in symbol units, and after the guard interval (GI) is removed, FFT processing is performed. Of course, in order to perform FFT processing, A / D conversion is performed on the received signal. These processing are collectively described as FFT processing in steps S35-1 to S35-3. Further, the FFT processing is not performed only on the orthogonal pilot signal, but is performed on the entire received signal including the control information, and is representatively represented as steps S35-1 to S35-3. That is, the flowchart shown in FIG. 13 is a flowchart showing the overall flow of each symbol, rather than showing processing in units of OFDM symbols.

(Tenth embodiment)
FIG. 14 shows a processing flow on the reception side of the first wireless communication apparatus of the wireless control packet using the OFDM modulation method as the tenth embodiment of the present invention. This corresponds to an extension of the ninth embodiment to the case where the orthogonal pilot signal shown in the third and fourth embodiments, that is, a long training signal is used. In FIG. 14, steps corresponding to the antenna sequence numbers are assigned to the processing corresponding to the processing shown in FIG.

  The difference from FIG. 13 is that when calculating the transfer function matrix of the MIMO channel, predetermined components of the transfer function matrix directly obtained from the long training signal having the configuration shown in FIG. 7 are determined in steps S43-1 to S43-3. In step S46, the remaining components not obtained here are obtained as transfer function matrix supplement acquisition processing. As the transfer function matrix complement acquisition processing here, processing by interpolation, extrapolation calculation or the like is applied as described above.

(Eleventh embodiment)
FIG. 15 shows a calibration processing flow of the first wireless communication apparatus in the eleventh embodiment of the present invention. In the above-described eighth to tenth embodiments, the calculation shown in Expression (7) is performed in the transfer function matrix correction processing (steps S28, S39, and S47). However, if the matrix B ₁ ⁻¹ · A ₁ in Equation (7) is a matrix given by a constant multiple of the unit matrix, such processing can be omitted.

That is, if the physical quantities of complex numbers corresponding to the amplification factor and phase rotation of the radio unit (transmission side and reception side) in the i-th antenna of the first radio communication device are α _{1, i} and β _{1, i} , respectively, {Α _{1, i} / β _{1, i} } may be a constant with the antenna of FIG.

Therefore, when the calibration process is started at a predetermined cycle (step S51), α _{1, i} / β _{1, i} is obtained as calibration coefficients for each receiving antenna (steps S52-1 to S52-3).

  The radio unit is operated so that this value becomes a common constant for all antennas, the phase and amplitude are adjusted, calibration processing is performed (steps S53-1 to S53-3), and the processing is terminated (step S54). ). Similar to the description so far, the area surrounded by the dotted line in steps S55-1 to S55-3 represents that the process is performed for each reception system. However, the constants in steps S53-1 to S54-3 may always be fixed constants, for example, values that change as appropriate using one value of a plurality of antenna systems may be used. Absent. What is important is that this constant is a common value for all antenna systems.

(Twelfth embodiment)
FIG. 16 is a processing flow at the time of wireless data packet transmission of the first wireless communication apparatus in the twelfth embodiment of the present invention. Here, the description will proceed on the assumption that the multiplexing number N that is already spatially multiplexed and the transmission weight matrix when each signal sequence is spread and transmitted to a plurality of antennas are known. Such information can be obtained based on the transfer function information of the MIMO channel described above.

  When data is first input to the transmission side of the first wireless communication device (step S61), the signal is divided into N series by S / P conversion (step S62). A known preamble signal is assigned to each series of signals. This preamble signal includes a short training signal and a long training signal as an orthogonal pilot signal. Subsequent to this preamble signal, each series of signals is modulated (step S64), and a transmission signal vector having the modulation signal of each series as a component is obtained as Tx (step S65). If the transmission weight matrix obtained in advance is represented as U, the matrix U is multiplied by the transmission signal vector Tx (step S66), and the result is transmitted from the radio unit of each series via the antenna (step S67). . The above steps S64 to S67 are repeated until the transmission data ends in symbol units (step S68).

(13th Embodiment)
FIG. 17 shows an example of a processing flow of transmission weight matrix generation of the first wireless communication apparatus in the thirteenth embodiment of the present invention. Here, the case where the above-described MIMO channel information feedback method is applied to the eigenmode SDM transmission shown in equations (2) and (3) is shown as an example. When the transfer function matrix H is acquired (step S71), a square matrix H ^H · H is generated using this matrix and a Hermitian conjugate matrix (step S72), and eigenvectors of this matrix are obtained (step S73). A unitary matrix U is generated by combining eigenvectors (step S74). The unitary matrix U is temporarily stored (step S75), and when transmitting a wireless data packet as shown in FIG. 16, this matrix is read and used as a transmission weight matrix. Here, the case where the eigenmode SDM is applied is shown as an example, but the transmission weight matrix may be obtained by another simpler method.

(Fourteenth embodiment)
FIG. 18 shows an example of a processing flow for selecting a transmission mode that is a combination of the number of MIMO spatial multiplexing transmissions and the modulation scheme / coding rate of the first wireless communication apparatus in the fourteenth embodiment of the present invention.

  There are various methods for determining the transmission mode, such as using information on the PHY layer such as transfer function information, information on the MAC layer regarding past communication success / failure history, and both. Among them, the transfer function matrix H obtained by using the MIMO channel information feedback method described above provides useful information for selecting an optimum transmission mode.

For example, when the eigenvalue of H or the eigenvalue of ^HH · H, which is the product of this matrix and the Hermitian matrix, is calculated, the eigenvalue becomes a physical quantity corresponding to the signal-to-noise ratio SNR of each subchannel in MIMO transmission. Actually, as shown in Equation (4), the upper limit of the transmission capacity can be grasped from these eigenvalues. Alternatively, assuming a certain modulation mode and coding rate for error correction, the risk can be reduced by not using a subchannel whose eigenvalue is not equal to or greater than a predetermined dark value. In some cases, it is also possible to allocate the transmission power allocated for the substream to other substreams.

For this purpose, when a transfer function matrix is obtained (step S81), H ^H · H is calculated (step S82), and its eigenvalue is obtained (step S83). From the distribution of the eigenvalues, the transmission quality for each transmission mode, which is a combination of the number of MIMO spatial multiplexing transmissions and the modulation scheme / coding rate, is estimated (step S84), and the optimum transmission mode is selected and stored (step S85). ). The optimum transmission mode stored here is used in the transmission of the next wireless data packet.

In the process of the fourteenth embodiment, ^HH · H is used as the eigenvalue calculation in the processes of steps S82 and S83, but the transfer function matrix H itself may be used. In this case, step S82 is omitted, and the eigenvalue of H is obtained in step S83. Further, in the processing of steps S84 and S85, it is also possible to perform batch processing by directly associating the distribution of eigenvalues with the optimum transmission mode by some table or arithmetic expression. The transmission power allocated to each substream and the number of signal systems to be spatially multiplexed can be determined using the water injection theorem.

  In the above description, when performing MIMO spatial multiplexing transmission, one-to-one communication in which there is one transmitter and one receiver is assumed. However, as an extension of MIMO spatial multiplexing transmission, multi-user MIMO communication in which a plurality of wireless communication devices and one wireless communication device communicate simultaneously is also possible. This multi-user MIMO communication system includes one base station apparatus and a plurality of mobile stations, that is, radio communication apparatuses, and the base station apparatus and the plurality of mobile stations communicate simultaneously. In this case, the base station device detects the MIMO channel information (transfer function matrix information) between each mobile station and the base station device, and the base station device (or the base station device and the mobile station cooperate). ) Send / receive weight control and wireless transmission.

(Fifteenth embodiment)
FIG. 19 shows a processing flow of transmission / reception weight matrix generation of the first wireless communication apparatus in the fifteenth embodiment of the present invention. The base station apparatus in the multiuser MIMO communication system has the function of the first wireless communication apparatus in the above-described embodiment.

  When performing communication (step S91), first, in the base station apparatus, spatial multiplexing is simultaneously performed based on the transfer function matrix information collected in advance using the MIMO channel information feedback method of the present embodiment described above. A mobile station to be performed, that is, a wireless communication device is selected (step S92). A transfer function matrix for the selected mobile station is read (step S93), and a transmission / reception weight matrix for each mobile station is calculated (step S94). In the calculation of the transmission / reception weight matrix, weights such as directing nulls in a predetermined direction in the antenna directivity control so that the signals of the respective mobile stations do not interfere with each other, rather than the eigenvectors described with reference to FIG. Form. The calculated transmission / reception weight matrix is temporarily stored (step S95) and used for subsequent transmission of wireless data packets.

(Device configuration)
In addition to the above description of the system, the apparatus configuration will be described below. FIG. 20 is a configuration example of a wireless communication apparatus having functions of both the transmission side and the reception side in the above-described embodiment. 20, reference numeral 401 denotes a data division circuit, reference numerals 402-1 to 402-4 denote preamble assignment circuits, reference numerals 403-1 to 403-4 denote modulation circuits, reference numeral 404 denotes a transmission signal conversion circuit, and reference numerals 405-1 to 405. -4 is a radio unit, 406-1 to 406-4 are antennas, 407 is a channel estimation circuit, 408 is a transfer function matrix management circuit, 409 is a signal detection circuit, 410 is a unitary matrix generation circuit, 411 Is a transfer function matrix correction circuit, reference numeral 412 is a data synthesis circuit, reference numeral 413 is an error detection circuit, reference numeral 414 is a control information separation circuit, reference numeral 415 is a MAC control circuit, reference numeral 416 is a MAC frame generation circuit, reference numeral 417 is a modulation unit, Reference numeral 418 denotes a demodulation unit.

  As for the device configuration, the prior art and the present invention have the same configuration, but the processing contents in each internal functional block are slightly different. First, the signal flow in the prior art will be described. When data is input, a MAC frame is formed by the MAC frame generation circuit 416 and input to the data division circuit 401. In the data division circuit 401, when spatial multiplexing is performed as MIMO communication, a signal is divided into a plurality of systems and input to the preamble assignment circuits 402-1 to 402-4. When spatial multiplexing is not performed, one system is selected (for example, the preamble assigning circuit 402-1 is fixedly selected) and input. Preamble applying circuits 402-1 to 402-4 apply a preamble signal to the input signal. In each signal system, the assigned preamble signals are different.

  In addition, control information related to the PHY layer, for example, the above-described PLCP control information is provided for a wireless LAN based on IEEE 802.11, for example. Thereafter, the signals are input to the modulation circuits 403-1 to 403-4 and subjected to modulation processing. Here, for example, modulation schemes such as BPSK (Binary Phase Shift Keying) and 64QAM, and error correction coding before and after the modulation processing are performed as necessary. These signals are input to the transmission signal conversion circuit 404, and predetermined linear conversion as shown in the above equation (3) is performed as necessary. Note that no conversion is performed when spatial multiplexing is not performed. The signals thus converted are input to the radio units 405-1 to 405-4, subjected to frequency conversion and amplification processing, and transmitted via the antennas 406-1 to 406-4.

  On the other hand, when signals are received via the antennas 406-1 to 406-4, the signals are input to the radio units 405-1 to 405-4, subjected to frequency conversion and amplification processing, and are input to the channel estimation circuit 407. Entered. The channel estimation circuit 407 extracts a preamble signal added to the head of the received signal and performs channel estimation. The signals after the preamble signal are input to the signal detection circuit 409 as they are. The transfer function matrix obtained by estimation by the channel estimation circuit 407 is input to the transfer function matrix management circuit 408, and this information is used in the signal detection process until a series of data packets are received. The signal detection circuit 409 receives the transfer function matrix from the transfer function matrix management circuit 408 and the received signal from the channel estimation circuit 407 and performs signal detection processing. If spatial multiplexing is used, signal detection processing may be performed using the ZF method or any other method. If spatial multiplexing is not performed, signals received by the antennas 406-1 to 406-4 are appropriately combined, and signal detection processing is performed so as to obtain diversity gain. Note that the signal detection circuit 409 can also perform error correction processing as necessary.

  The signal received in this manner is input to the data synthesis circuit 412 and, when spatial multiplexing is performed, the divided signal series is synthesized. The synthesized data is input to the error detection circuit 413, and the presence / absence of a code error is determined. If no error is detected, the signal is input to the control information separation circuit 414. The error determination result is also input to the MAC control circuit 415. The control information separation circuit 414 refers to the MAC header or the like of the received signal, and first determines whether the signal is addressed to the own station. When addressed to the own station, the fact is input to the MAC control circuit 415, and if it is control information, the data is output to the MAC control circuit 415. Is output.

  The MAC control circuit 415 analyzes the control information separated by the control information separation circuit 414, acquires the result from the error detection circuit 413 and the report from the control information separation circuit 414 (whether it is addressed to the own station), and is necessary. Control information is generated. For example, in the case of an IEEE802.11 wireless system, control information is output to the MAC frame generation circuit 416 in order to send back an ACK signal when a received signal addressed to itself is detected without a code error. If any control information is requested to be returned, the corresponding control information is output to the MAC frame generation circuit 416. In some cases, transfer function matrix information is read as control information and input to the transfer function matrix management circuit 408. If information on the transfer function of the forward link is calculated from the transfer function matrix of the return link, it is necessary to perform the correction process shown in the above equation (7). In this case, the transfer function matrix correction circuit 411 performs this processing. Therefore, when the transfer function matrix information of the forward link is directly contained in the control information and fed back, this transfer function matrix correction circuit 411 becomes unnecessary.

  The forward link transfer function matrix obtained in this way is input to the unitary matrix generation circuit 410 at the time of data transmission. The number of signal systems to be spatially multiplexed is instructed from the data division circuit 401 or the MAC control circuit 415, and a unitary matrix corresponding to the instructed number of systems is generated. For example, in the case of a system composed of one base station device and a plurality of wireless communication devices, the communication partner of each wireless communication device is basically only the base station device. In this case, the transfer function matrix management circuit 408 may manage only the transfer function matrix with the base station apparatus. On the other hand, in the case of the base station apparatus, the transfer function matrix of the subordinate radio communication apparatus is managed, read out as necessary, and the unitary matrix generation circuit 410 generates a unitary matrix.

  Next, a case where control information such as an ACK signal is generated by the MAC control circuit 415 and transmitted via the MAC frame generation circuit 416 will be described in more detail. As described above, an ACK signal or the like is usually transmitted without performing spatial multiplexing. Therefore, the ACK signal input to the data dividing circuit 401 is input to the preamble adding circuit 402-1 without being divided, a predetermined modulation process is performed by the modulation circuit 403-1, and the transmission signal converting circuit 404 is A signal is transmitted from the antenna 406-1 via the wireless unit 405-1 without passing through the process.

  When an ACK signal is received, the signals received by the antennas 406-1 to 406-4 are received via the radio units 405-1 to 405-4, and the channel estimation circuit 407 Get the transfer function matrix of the received signal. Here, since spatial multiplexing is not performed, the transfer function information is not a matrix format but a vector format. This information is recorded in the transfer function matrix management circuit 408, and the signal detection circuit 409 performs signal detection of subsequent data based on this information. In this signal detection, based on the transfer function information acquired by the channel estimation circuit 407, it is generally adjusted so that the phases of the received signals of the respective systems become the same phase, and they are combined to perform signal detection. It is. As a result, reception diversity gain can be obtained, and even if the reception characteristics of a specific antenna are poor, if the reception characteristics of other antennas are good, good communication is possible, and fluctuations in reception quality are suppressed and communication is stabilized. I can do it.

  The above-described apparatus configuration is the same as that of the prior art. However, in the embodiment of the present invention, in the processing in the case of transmitting a radio control packet (or a radio data packet) as a plurality of signal sequences without spatial multiplexing, The signal flow is different.

  For example, a case will be described in which the first embodiment described above is applied in the return of control information such as ACK. For example, when four antennas 406-1 to 406-4 are provided, the data dividing circuit 401 copies the data corresponding to the four systems and inputs them to the preamble assignment circuits 402-1 to 402-4. Here, preambles that are orthogonal to each other are formed. For example, assuming an OFDM modulation scheme, assuming that a preamble as shown in FIG. 6 is used, the preamble assigning circuit 402-1 uses only the subcarriers # 1, # 5, # 9. -2 is only for subcarriers # 2, # 6, # 10..., Only for subcarriers # 3, # 7, # 11... , # 8, # 12... Are regarded as valid signals, and other signals are discarded and the processing after the modulation circuits 403-1 to 403-4 is performed (corresponding to the long training signal transmission processing in FIG. 6). Corresponding to the data signal transmission processing of FIG. The preamble signal and subsequent data (including the PLCP control signal and the like) are all processed as shown in FIG.

  The above processing is based on the assumption that the preamble signal shown in FIG. 6 is used. However, in general, the signal processing corresponds to, for example, processing obtained by extending the processing shown in FIG. 11 to four systems. Can be made.

  On the other hand, the signal detection process of the received signal can be performed in the same manner as in the conventional technique, but the transfer function matrix management circuit 408 is different from the conventional technique. Since the MAC control circuit 415 knows that an ACK signal (or a wireless control packet for response) is received after data transmission, the transfer function matrix is obtained from the ACK signal or the like to the transfer function matrix management circuit 408. Give instructions to get. When this instruction is input, apart from the processing in the signal detection circuit 409, for example, in addition to the direct transfer function information acquisition shown in steps S43-1 to 43-3 in FIG. Completion processing for missing transfer function matrix information.

  In general, the same processing can be applied to the configuration of the wireless packet shown in FIG. 4. Specifically, in addition to FIG. 6, the configuration may be as shown in FIG. In this case, it is possible to correspond as shown in FIG. For example, in the case of FIG. 3, if the signal of the preamble adding circuit 402-1 is used as a reference for the IEEE 802.11 system signal, the preamble adding circuit 402-2 applies a phase rotation corresponding to a delay of 1600 ns. is doing. Similarly, in the modulation circuit 403-2, a rotation of a phase corresponding to a delay of 1600 ns is given to a signal subsequent to the preamble signal.

  At this time, the transfer function matrix management circuit 408 can acquire the transfer function matrix directly by the channel estimation circuit 407, unlike the example in which FIGS. 6 and 7 are applied. Further, for the processing of the signal detection circuit 409, the channel estimation circuit 407 determines that the received signal is a signal that has not been subjected to spatial multiplexing, and uses the result of normal channel estimation. However, regarding the feedback of the transfer function matrix information, processing is performed on the assumption that the transmission signal has the configuration shown in FIG. 3, unlike the example provided with FIGS. 6 and 7. This corresponds to a process of acquiring a transfer function necessary for signal detection for each signal sequence as if the received signal was spatially multiplexed. In this way, two types of channel estimation processes are performed individually, one is used by the signal detection circuit 409, and the other is used for acquiring a transfer function matrix managed by the transfer function matrix management circuit 408.

  In this manner, for example, it is possible to acquire the transfer function information of the MIMO channel between the transmitting and receiving antennas while using the existing ACK signal or the like without spatial multiplexing. In the description of FIG. 20, the OFDM modulation scheme is not intended. However, for example, the processing of the transmission signal conversion circuit 404 from the data division circuit 401 is performed for each subcarrier, and the transmission signal conversion circuit 404 If an IFFT circuit is inserted in the subsequent stage, it can be expanded to a transmission unit compatible with the OFDM modulation system.

  Also in the reception system, an FFT circuit is provided in the channel estimation circuit 407, the transfer function matrix management circuit 408 and the signal detection circuit 409 are individually processed for each subcarrier, and the signals of all subcarriers are processed by the data synthesis circuit. If combined, it can support the OFDM modulation system. In the case of the configuration of the radio control packet shown in FIG. 5, the long training signals 35-1 to 35-4 are given by the preamble grant circuits 402-1 to 402-4 in a form following the data, and at the time of reception. It is also possible to acquire transfer function matrix information by reading this portion with the channel estimation circuit 407 and performing channel estimation with the long training signals 32-1 to 32-4 and the long training signals 35-1 to 35-4. . Further, as a combination of these, the configuration of FIG. 8 may be used.

  According to the above-described embodiment, in feedback for acquiring MIMO channel information in the radio communication device on the transmission side, it is possible to add a new overhead without adding a new overhead, to minimize the addition, and to all antennas. The effect that information can be collected can be obtained. In addition, based on the MIMO channel information obtained in this way, it is also expected to select an optimum transmission mode, calculate an appropriate transmission / reception weight matrix, and use this to realize efficient MIMO transmission. it can.

  In the above-described embodiment, a wireless packet that accommodates and transmits user information is referred to as a “wireless data packet”, and a packet that accommodates and transmits various types of control information is referred to as a “wireless control packet”. The former is premised on MIMO transmission, and the latter is premised on SISO transmission (meaning that MIMO spatial multiplexing transmission is not performed, and a plurality of antennas are used for transmission and reception).

  However, the above-described embodiment intends to maintain a predetermined orthogonal relationship in the channel estimation pilot signal (preamble signal and training signal) for SISO used when transmitting a packet that does not perform MIMO spatial multiplexing transmission. However, MIMO channel information between a plurality of transmission / reception antennas is estimated by transmitting different patterns from the plurality of antennas. Accordingly, when a wireless data packet is transmitted via SISO without MIMO spatial multiplexing transmission, the transfer function of the MIMO channel can be estimated by the same method.

  Furthermore, the radio control packet is a packet for transmitting an ACK signal or an RTS / CTS signal, for example, a health check for confirming that a mobile station is continuously present in the area. It includes control information and widely general control information such as responses to polling.

  Also, some of the embodiments of the MIMO channel information feedback method described above are available in existing wireless systems and can coexist with them. That is, a device that implements these embodiments and an existing device can communicate with each other without particular awareness, and when both wireless communication devices that perform communication implement these embodiments at the same time, the technology of the present invention. Can be used.

  In recent wireless communication systems, attribution processing such as authentication is performed at the beginning of communication for establishing a wireless connection. Furthermore, in this, the function of each wireless communication device is negotiated. By including information such as whether or not the MIMO channel information feedback method of the present invention is implemented and which of the various feedback methods is supported as one of the function items, the present invention can be used in subsequent communications. .

  The above-described embodiments are all illustrative of the present invention and are not limited to 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.

  In addition, the orthogonal pilot signal provision means, the first orthogonal pilot signal provision means, and the second orthogonal pilot signal provision means of the present invention correspond to the preamble provision circuits 402-1 to 402-4, and the signal orthogonalization means includes: It is inherent in the modulation circuits 403-1 to 403-4. The signal transmission unit corresponds to the wireless units 405-1 to 405-4. Further, the orthogonal pilot extraction means and the channel estimation means of the present invention correspond to the channel estimation circuit 407, and the transfer function matrix calculation means corresponds to the transfer function management circuit 408 and the transfer function matrix correction circuit 411.

It is a figure which shows the structural example of the radio | wireless control packet in the 1st Embodiment of this invention. It is a figure which shows the other structural example of the radio | wireless control packet in the 1st Embodiment of this invention. It is a figure which shows the orthogonalization method for every antenna in the 1st Embodiment of this invention. It is a figure which shows the other orthogonalization method for every antenna in the 1st Embodiment of this invention. It is a figure which shows the structural example of the radio | wireless control packet in the 2nd Embodiment of this invention. It is a figure which shows the orthogonalization method (the 1) of the long training signal for every antenna in the 3rd Embodiment of this invention. It is a figure which shows the orthogonalization method (the 2) of the long training signal for every antenna in the 3rd Embodiment of this invention. It is a figure which shows the orthogonalization method of the long training signal for every antenna in the 4th Embodiment of this invention. It is a figure which shows the transmission processing flow of the radio | wireless control packet in the 5th Embodiment of this invention. It is a figure which shows the transmission processing flow of the radio | wireless control packet in the 6th Embodiment of this invention. It is a figure which shows the transmission processing flow of the radio | wireless control packet in the 7th Embodiment of this invention. It is a figure which shows the reception processing flow of the radio | wireless control packet of the 1st radio | wireless communication apparatus in the 8th Embodiment of this invention. It is a figure which shows the reception processing flow of the radio | wireless control packet of the 1st radio | wireless communication apparatus in the 9th Embodiment of this invention. It is a figure which shows the reception processing flow of the radio | wireless control packet of the 1st radio | wireless communication apparatus in the 10th Embodiment of this invention. It is a figure which shows the calibration processing flow of the 1st radio | wireless communication apparatus in the 11th Embodiment of this invention. It is a processing flow at the time of the radio | wireless data packet transmission of the 1st radio | wireless communication apparatus in the 12th Embodiment of this invention. It is a figure which shows the processing flow of the transmission weight matrix production | generation of the 1st radio | wireless communication apparatus in the 13th Embodiment of this invention. It is a figure which shows the processing flow of transmission mode selection of the 1st radio | wireless communication apparatus in 14th Embodiment of this invention. It is a figure which shows the processing flow of the transmission / reception weight matrix production | generation of the 1st radio | wireless communication apparatus in 15th Embodiment of this invention. It is the block diagram which showed the apparatus structure of the radio | wireless communication apparatus in embodiment of this invention. It is a figure which shows the example of the collection method of the transfer function information in a conventional system. It is a figure which shows the structural example of the radio | wireless data packet in the case of SISO transmission which does not perform the spatial multiplexing in a conventional system. It is a figure which shows the structural example of the radio | wireless data packet in the case of MIMO communication. It is a figure which shows the outline | summary (the 1) of the long training signal for MIMO in a conventional system. It is a figure which shows the outline | summary (the 2) of the long training signal for MIMO in a conventional system. It is a figure which shows the flow of the data transfer in a conventional system. It is a figure which shows the outline | summary of the data transfer using RTS / CTS in a conventional system. It is a figure which shows the transmission processing flow of the radio | wireless control packet in a conventional system. It is a figure which shows the transmission processing flow of the radio | wireless data packet in a conventional system.

Explanation of symbols

401 Data division circuit 402-1 to 402-4 Preamble provision circuit 403-1 to 403-4 Modulation circuit 404 Transmission signal conversion circuit 405-1 to 405-4 Radio unit 406-1 to 406-4 Antenna 407 Channel estimation circuit 408 Transfer function matrix management circuit 409 signal detection circuit 410 unitary matrix generation circuit 411 transfer function matrix correction circuit 412 data synthesis circuit 413 error detection circuit 414 control information separation circuit 415 MAC control circuit 416 MAC frame generation circuit

Claims (6)

A wireless communication system comprising first and second wireless communication devices having a plurality of transmission / reception antennas, wherein a plurality of signal sequences are spatially multiplexed and transmitted between the plurality of transmission / reception antennas on the same frequency channel. A channel information feedback method,
In the second wireless communication device,
1st orthogonal pilot signal provision which assign | provides each of several orthogonal known pattern signal to the head area | region of the radio | wireless control packet used for each radio | wireless control transmitted from the said several transmission antenna as 1st orthogonal pilot signal Processing steps;
Performing a signal virtual orthogonalization process for performing the same processing as the orthogonal processing of the orthogonal known pattern signal on the information area subsequent to the head area of the radio control packet;
Performing a signal transmission process of transmitting the radio control packet without spatial multiplexing using the plurality of transmission antennas,
In the first wireless communication device,
Performing signal reception processing for receiving the radio control packet using the plurality of receiving antennas;
Performing an orthogonal pilot signal extraction process for extracting a signal in a region to which the first orthogonal pilot signal is added from the received radio control packet;
Channel estimation processing for generating transfer function matrix information between the transmitting and receiving antennas for communication in the direction of the first wireless communication device from the second wireless communication device based on the relationship between the extracted signal and the known pattern signal Steps to do,
Based on the generated transfer function matrix information, transmission between the transmitting and receiving antennas for communication in the direction of the second wireless communication device from the first wireless communication device in the direction opposite to the direction of communication corresponding to the transfer function matrix information. Performing a transfer function calculation process for calculating function matrix information;
A channel information feedback method comprising: A wireless communication system comprising first and second wireless communication devices having a plurality of transmission / reception antennas, wherein a plurality of signal sequences are spatially multiplexed and transmitted between the plurality of transmission / reception antennas on the same frequency channel. A channel information feedback method,
In the second wireless communication device,
1st orthogonal pilot signal provision which assign | provides each of several orthogonal known pattern signal to the head area | region of the radio | wireless control packet used for each radio | wireless control transmitted from the said several transmission antenna as 1st orthogonal pilot signal Processing steps;
A step of performing said orthogonal straight交化signal virtualization orthogonalization process applied the same processing as the processing of the known pattern signal for subsequent information area to the head area of the wireless control packet,
Each one, or a known pattern signal in which a plurality of orthogonal obtained by multiplying a predetermined coefficient to the first orthogonal pilot signals of said first quadrature pilot signals as the second orthogonal pilot signals, the plurality Performing a second orthogonal pilot signal providing process to be applied to each of the radio control packets transmitted from the transmission antennas of
Performing a signal transmission process of transmitting the radio control packet without spatial multiplexing using the plurality of transmission antennas,
In the first wireless communication device,
Performing signal reception processing for receiving the radio control packet using the plurality of receiving antennas;
Performing an orthogonal pilot signal extraction process of extracting a signal in a region to which the first orthogonal pilot signal and the second orthogonal pilot signal are added from the received radio control packet;
Generating a transfer function matrix information between the transmitting and receiving antennas for the communication of the first wireless communication device direction the extracted signal from said second radio communication device based on the relationship between linear addition signal and the known pattern signal Performing channel estimation processing,
Based on the generated transfer function matrix information, transmission between the transmitting and receiving antennas for communication in the direction of the second wireless communication device from the first wireless communication device in the direction opposite to the direction of communication corresponding to the transfer function matrix information. Performing a transfer function calculation process for calculating function matrix information;
A channel information feedback method comprising: The first wireless communication apparatus and the second wireless communication apparatus each adopt an orthogonal frequency division multiplexing modulation scheme using K (K> 1: K is an integer) subcarriers,
The first wireless communication apparatus has N (N> 1: N is an integer) transmission / reception antennas,
The second wireless communication apparatus has M (M> 1: M is an integer) transmission / reception antennas,
In the step of performing the first orthogonal pilot signal provision processing or the second orthogonal pilot signal provision processing of the second wireless communication device,
Said plurality of said first quadrature pilot signals applied to each of the wireless control packet transmitted from the transmitting and receiving antenna or the second orthogonal pilot signals, when decomposed into components for each of the sub-carrier, the Each signal transmitted from each of the transmission / reception antennas is set so as not to include at least one signal component of the subcarrier,
In the step of performing the channel estimation process,
Of the K subcarriers, the jth (1 ≦ j ≦ M: j is an integer) th of the second wireless communication apparatus of the kth (1 ≦ k ≦ K: k is an integer) th subcarrier. When acquiring transfer function matrix information between the transmission / reception antenna and the i-th (1 ≦ i ≦ N: i is an integer) -th transmission / reception antenna of the first wireless communication device, the second wireless communication device When the component of the kth subcarrier is included in the first orthogonal pilot signal or the second orthogonal pilot signal transmitted from the jth transmission / reception antenna, the first orthogonal pilot signal , or the second orthogonal pilot signal or the signal summation of these linear, by calculating a cross-correlation between the signal of the known pattern, the second of the j-th reception antenna of the wireless communication device, And the first Perform direct transfer function acquisition process to directly obtain the transfer function matrix information between the i-th antenna line communication apparatus,
Contains components of said first quadrature pilot signal or said k-th subcarrier in the second quadrature pilot signals within, sent from the second of the j-th reception antenna of the wireless communication device If not, the k-th of the j th receiving antenna and the first wireless of the obtained in the direct transfer function acquisition process at least one subcarrier adjacent to the subcarrier second wireless communication device and characterized in that the i-th transfer function interpolation relates matrix information or extrapolation complementary to a transfer function complementary and acquires the transfer function matrix information of the subcarriers by calculation, between the transmitting and receiving antennas of the communication device The channel information feedback method according to claim 1 or 2. The first wireless communication device includes a wireless unit for each of the transmission / reception antennas,
The radio unit has α _h as a complex physical quantity indicating an amplification factor and a phase rotation amount of a transmission signal in the radio unit connected to the h- th antenna of the own device, and a received signal in the h- th radio unit. When the complex physical quantity indicating the amplification factor and the amount of phase rotation is β _h ,
In the first wireless communication device,
Performing calibration coefficient acquisition processing for each of the transmission / reception antennas to acquire α _h / β _h which is a ratio of the complex physical quantity of the device at a predetermined period;
In the predetermined cycle, the amplification ratio and phase rotation amount of the transmission signal in the radio unit, or reception in the radio unit so that the ratio of the complex physical quantity takes the same value across all the transmission / reception antennas of its own device Performing a calibration process for adjusting a signal amplification factor and a phase rotation amount for each of the transmission and reception antennas,
In the step of performing the transfer function correction process of the first wireless communication device,
Communication from the first wireless communication device to the second wireless communication device as a matrix obtained by transposing a transfer function matrix between the transmitting and receiving antennas for communication from the second wireless communication device to the first wireless communication device. The transfer function matrix information of the said transmission / reception antenna with respect to is given. The channel information feedback method as described in any one of Claim 1 to 3 characterized by the above-mentioned. Comprises a first and second wireless communication apparatus having a plurality of transmitting and receiving antenna, a radio communication system for transmitting a plurality of signal sequences in the same frequency channel by between said plurality of transmitting and receiving antennas by spatially multiplexing Because
The second wireless communication device is
And quadrature pilot signal applying means for applying to the head region of the radio control packet used for each of the radio control transmitted each of the plurality of orthogonal known pattern signals from the plurality of transmitting antennas as Cartesian pilot signal,
Signal virtual orthogonalization means for performing the same processing as the orthogonal processing of the orthogonal known pattern signal to the information area following the head area of the radio control packet;
Signal transmitting means for transmitting the radio control packet without spatial multiplexing using the plurality of transmission antennas;
Equipped with a,
The first wireless communication device is
Signal receiving means for receiving the radio control packet using the plurality of receiving antennas;
Orthogonal pilot signal extraction means for extracting a signal in a region to which the orthogonal pilot signal is added from the received radio control packet;
Channel estimation means for generating transfer function matrix information between the transmitting and receiving antennas for communication in the direction from the second wireless communication device based on the relationship between the extracted signal and the known pattern signal;
Based on the generated transfer function matrix information, the transfer function matrix information between the transmitting and receiving antennas for communication in the direction of the second wireless communication device is calculated from the own device that is in the opposite direction to the communication direction corresponding to the transfer function matrix information. Transfer function matrix calculating means for
A wireless communication system comprising: Comprises a first and second wireless communication apparatus having a plurality of transmitting and receiving antenna, a radio communication system for transmitting a plurality of signal sequences in the same frequency channel by between said plurality of transmitting and receiving antennas by spatially multiplexing Because
The second wireless communication device is
1st orthogonal pilot signal provision which assign | provides each of several orthogonal known pattern signal to the head area | region of the radio | wireless control packet used for each radio | wireless control transmitted from the said several transmission antenna as 1st orthogonal pilot signal Means,
Signal virtual orthogonalization means for performing the same processing as the orthogonal processing of the orthogonal known pattern signal to the information area following the head area of the radio control packet;
Any one of the first orthogonal pilot signals or each of a plurality of orthogonal known pattern signals obtained by multiplying the first orthogonal pilot signal by a predetermined coefficient is used as the second orthogonal pilot signal. Second orthogonal pilot signal assigning means for assigning to each of the radio control packets transmitted from the transmitting antennas;
Signal transmitting means for transmitting the radio control packet without spatial multiplexing using the plurality of transmission antennas;
Equipped with a,
The first wireless communication device is
Signal receiving means for receiving the radio control packet using the plurality of receiving antennas;
Orthogonal pilot signal extraction means for extracting a signal in a region to which the first orthogonal pilot signal and the second orthogonal pilot signal are added from the received radio control packet;
Channel estimation means for generating transfer function matrix information between the transmitting and receiving antennas for communication in the direction of the own apparatus from the second wireless communication apparatus based on a relationship between the signal obtained by linearly adding the extracted signals and the known pattern signal; ,
Based on the generated transfer function matrix information, the transfer function matrix information between the transmitting and receiving antennas for communication in the direction of the second wireless communication device is calculated from the own device that is in the opposite direction to the communication direction corresponding to the transfer function matrix information. Transfer function calculating means for
A wireless communication system comprising:

Images