JP2009071493A - Wireless communication system, wireless communication device, and wireless communication method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To appropriately perform space multiple transmission by singular value decomposition by excluding the influence of the temporal change of a channel matrix, and a carrier frequency error and a sampling frequency error between transmission and reception. <P>SOLUTION: A transmitter estimates the channel matrix through the use of a training sequence included in the header of a CTS packet which is received from a receiver, continues to update the channel matrix by channel tracking in a subsequent payload part, calculates a transmission weight matrix, based on the latest channel matrix estimation value, and performs the space multiple transmission of the data packet. The deterioration of a communication characteristic on a receiver side is prevented, which is caused by the carrier frequency error and the sampling frequency error on a transmission side. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a radio communication system using spatial multiplexing based on a channel matrix having a channel transfer function as an element for each combination of transmission and reception antennas between radio communication apparatuses each having a plurality of antennas, as well as a radio communication apparatus and radio communication More particularly, the present invention relates to a wireless communication system that performs closed-loop spatial multiplexing transmission by sharing channel information between wireless communication apparatuses, a wireless communication apparatus, and a wireless communication method.

  More specifically, the present invention relates to a radio communication system, a radio communication apparatus, and a radio communication method for performing spatial multiplexing (beam forming) of a plurality of transmission streams using a transmission weight matrix obtained by singular value decomposition of a channel matrix. In particular, the present invention relates to a radio communication system, a radio communication apparatus, and a radio communication method for performing spatial multiplexing transmission by singular value decomposition while eliminating the influence of time variation of a channel matrix and carrier frequency error between transmission and reception and sampling frequency error. .

  A wireless network is attracting attention as a system free from wiring in the conventional wired communication system. As a standard regarding a wireless network, there can be mentioned IEEE (The Institute of Electrical and Electronics Engineers) 802.11 and IEEE 802.15.

  For example, in IEEE802.11a / g, an OFDM (Orthogonal Frequency Division Multiplexing) modulation scheme, which is one of multicarrier schemes, is adopted as a standard for wireless LANs. In the OFDM modulation method, transmission data is distributed and transmitted to a plurality of carriers having mutually orthogonal frequencies, so that the bandwidth of each carrier is narrow, the frequency utilization efficiency is very high, and frequency selective fading interference Strong.

  Further, the IEEE 802.11a / g standard supports a modulation scheme that achieves a communication speed of 54 Mbps at the maximum, but a next-generation wireless LAN standard capable of realizing a higher bit rate is required.

  MIMO (Multi-Input Multi-Output) communication has attracted attention as one of the technologies for realizing high-speed wireless communication. This is a communication system that includes a plurality of antenna elements on both the transmitter side and the receiver side, and realizes a spatially multiplexed stream. On the transmission side, a plurality of transmission data are multiplexed by applying space / time codes, distributed to a plurality of transmission antennas, and transmitted to a channel. On the other hand, on the receiving side, received data received by a plurality of receiving antennas via a channel can be spatially / temporally decoded to obtain received data without crosstalk between streams.

  A MIMO communication system is basically composed of wireless communication devices each having a plurality of antennas. From the viewpoint of packet transmission, the source of the packet is an initiator and the reception destination is positioned as a receiver, but from the viewpoint of beam forming, the side that transmits a packet by spatially multiplexing a plurality of streams is positioned as a beamformer. The side that receives the spatially multiplexed packet and separates the packet is positioned in Beamformee. In the following, communication from Beamformer to Beamforme is positioned as a “forward” or “downlink” link, and communication from Beamformee to Beamformer is positioned as a “reverse” or “uplink” link.

  According to the MIMO communication system, it is possible to increase the transmission capacity according to the number of antennas without increasing the frequency band, thereby achieving an improvement in communication speed. Also, since spatial multiplexing is used, the frequency utilization efficiency is good. MIMO is a communication method using channel characteristics and is different from a simple transmission / reception adaptive array.

  In MIMO communication, in order to spatially separate original stream signals from spatially multiplexed received signals, a channel matrix H is obtained by some method, and a plurality of streams are spatially multiplexed from the channel matrix H by a predetermined algorithm.・ It is necessary to perform spatial separation processing.

  First, a method for obtaining the channel matrix H will be described. The channel matrix H is generally used to estimate the channel transfer function based on the difference between the actually received signal and the known sequence by transmitting and receiving a known training sequence between wireless communication devices that are communication partners, The channel transfer functions for each path for the combination of transmitting and receiving antennas are arranged in a matrix format. N training sequences are transmitted from one wireless communication device having N transmission antennas, and the N training sequences received for each reception antenna are transmitted from the other wireless communication device having M reception antennas. Can be used to obtain an M × N (row × column) channel matrix H.

  FIG. 6 shows an example (Mixed format) of a packet format defined by IEEE802.11n adopting the OFDM_MIMO communication system. As shown in the figure, in the PHY layer header, an HT-LTF (High Throughput-Long consisting of a known training sequence used for channel estimation for each input signal spatially modulated (mapped) on the receiving side of the packet. Training Field) is included. On the transmission side of the packet, HT-LTF is transmitted in a time division manner from each transmission antenna. Therefore, the number of HT-LTF fields corresponding to the number of spatial streams is added (for example, 2 HT-LTFs are added in a 2 × 2 MIMO communication system). On the receiving side of the packet, an antenna reception weight matrix can be obtained by calculating an inverse matrix of the acquired channel matrix, and this weight matrix is multiplied by a reception vector having a spatial multiplexed signal received by each reception antenna as an element. Thus, the spatially multiplexed stream can be spatially separated.

  Note that the leading portion of the PHY layer header includes a legacy preamble portion compatible with IEEE802.11a / g, such as L-STF (Legacy Short Training Field) and L-LTF (Legacy Long Training Field). ing. Prior to HT-LTF used for channel matrix estimation, these legacy preamble parts can be used to find a packet, acquire synchronization, and measure a carrier frequency error and a sampling frequency error between transmission and reception. (In the IEEE 802.11n specification, for the mixed format packet shown in FIG. 6, the carrier frequency error and the sampling frequency error are estimated from the L-STF, L-LTF, HT-STF, and HT-LTF. In addition, for a green field format packet (not shown), a carrier frequency error and a sampling frequency error can be estimated from HT-GF-STF and HT-LTF.)

  Next, a method for realizing spatial multiplexing of a plurality of streams will be described. The MIMO communication system is roughly classified into an open loop type that does not share channel matrix information between packet transmission and reception and a closed loop type that shares channel matrix information between packet transmission and reception.

  In the open-loop type MIMO communication system, the receiving side of a packet acquires a channel matrix H, and multiplies a reception vector whose element is a reception signal for each reception antenna by a reception weight matrix calculated from the channel matrix. The multiplexed transmission stream is spatially separated. As a technique for obtaining a reception weight matrix from the channel matrix H, for example, Zero Force (zeroization standard) and MMSE (Minimum Mean Square Error) are known. In the open loop type, the procedure for feedback of channel information from the receiver to the transmitter is omitted, so that the MIMO communication system can be configured relatively simply.

  On the other hand, in a closed-loop type MIMO communication system, channel information is shared between transmission and reception, and on the packet transmission side, transmission antenna weights obtained from the channel matrix are multiplied to perform appropriate beamforming toward the receiver. An ideal spatial orthogonal channel can be created.

  In a closed-loop type MIMO communication system, there are two types of procedures of “Implicit feedback” and “Explicit feedback” as a procedure for feeding back information on a channel matrix between transmission and reception of packets.

  In the impact feedback, a wireless communication device (that is, a beamformer) that performs spatial multiplexing transmission of a packet estimates an upstream channel matrix using a training sequence included in a packet transmitted from a communication partner (that is, a beamforme), and both Under the assumption that the direction channel characteristics are reciprocal, a downlink channel matrix is calculated, and beam forming, that is, spatial multiplexing transmission of a packet, is performed using a transmission weight matrix obtained from the channel matrix. In order for the channel characteristics to be reversible, calibration of the RF circuit in the communication system is necessary.

  In explicit feedback, when Beamformee estimates a downlink channel matrix using a training sequence sent from Beamformer, and returns a packet including the channel matrix as data to Beamformer, Beamformer receives from the received channel matrix. Beam weighting is performed by calculating a transmission weight matrix. Alternatively, the beamformer further calculates a transmission weight matrix for the beamformer to perform beam forming from the estimated channel matrix, and returns a packet including the transmission weight matrix as data (not the channel matrix) to the beamformer. In explicit feedback, the weight matrix is calculated based on the estimated channel matrix in the forward direction, so that it is not necessary to assume the reversibility of the channel.

As an ideal form of closed-loop type MIMO transmission, an SVD-MIMO system using singular value decomposition (SVD) of a channel matrix H is known. In SVD-MIMO transmission, UDV ^H is obtained by singular value decomposition of the channel matrix H. On the beamformer side, V (= U ^* (where the superscript * means a complex conjugate transpose of the matrix)) is used as a transmission antenna weight matrix, and a transmission vector whose element is a transmission signal for each transmission antenna Is multiplied by the weight matrix V to transmit a beam-formed packet toward Beamformee. On the Beamformee side, typically, (UD) ⁻¹ obtained by singular value decomposition of the channel matrix is given as a reception antenna weight matrix, and this is multiplied by the reception vector.

Of the matrices U, D, and V obtained by singular value decomposition, D is a diagonal matrix having the square root of each singular value λ _i corresponding to the quality of each spatial stream as a diagonal element (subscript i is i The second spatial stream). By arranging the singular values λ _{i in the} descending order of the values in the diagonal elements of the diagonal matrix D, and assigning power distribution and modulation scheme according to the communication quality represented by the magnitude of the singular values to each stream Realize a plurality of logically independent transmission paths that are spatially orthogonally multiplexed, and the beamforme side can extract the original signal sequences without any crosstalk, and theoretically achieve the best performance (for example, , See Patent Document 1).

  Consider a method for acquiring channel information in each of the beamformer and beamformee in a closed-loop type MIMO communication system.

  The downlink link channel matrix from Beamformer to Beamforme is set to H, and the reverse uplink channel matrix is set to H ′, and each is subjected to singular value decomposition as follows.

  Here, assuming that the reversibility of the propagation path, that is, the transfer function of the downlink and uplink is equivalent, the downlink channel matrix H is equal to the uplink channel matrix H ′, and the following equation (3) It is expressed as

  Substituting equations (1) and (2) into equation (3) gives the following.

  That is, the transmitting antenna weight matrix V used on the beamformer side is as follows.

  On the Beamformer side, the signal sequence of the packet is distributed to a plurality of transmission streams. A transmission vector having a transmission signal for each transmission stream as an element is set to X, and this is multiplied by a transmission antenna weight matrix V to be spatially multiplexed, and a transmission signal for each transmission antenna branch is expressed by the following equation (6). A transmission vector Y as an element is obtained.

  In the beamformee, a spatially multiplexed data packet arrives via a propagation path represented by a channel matrix H. That is, the beamformee receives the packet in a state where the channel matrix H is multiplied by the spatially multiplexed transmission vector Y as shown in the following equation (7).

According to the IEEE802.11n packet format employing the OFDM_MIMO communication scheme, the beamformee can estimate the channel matrix using the training sequence HT-LTF (described above) included in the PHY layer header of the packet. As can be seen from the above equation (7), the channel matrix estimated from the HT-LTF of the packet weighted by Beamformer with the weighting matrix V is UD. Therefore, Beamformee obtains a reception weight matrix (UD) ⁻¹ corresponding to this UD and multiplies the spatially multiplexed reception vector HY (= UDX), thereby spatially separating the original stream X. it can.

  In wireless networking such as IEEE 802.11, multiple users on the same communication channel perform random access while avoiding contention, so that CSMA (Carrier Sense Multiple Access) / CA (Collision Aviation): collision Avoidance) method is used. Also, as a methodology for solving the hidden terminal problem at that time, RTS (Request to Send) / CTS (Clear to Send) communication procedure is used. When using an RTS / CTS procedure in a closed loop MIMO communication system, for example, Beamformee transmits a CTS packet including a training sequence, and Beamformer is a channel obtained based on the training sequence included in the CTS packet. A transmission weight matrix for spatial multiplexing is calculated from the matrix, and a data packet including a training sequence in which a weight is given to each transmission stream to be multiplexed may be transmitted (for example, refer to Patent Document 2). thing).

  FIG. 7 schematically shows a communication sequence in accordance with the RTS / CTS procedure in a closed loop type MIMO communication system. However, the channel matrix is assumed to be reversible between the uplink and the downlink.

  The beamformer as a transmission request source transmits an RTS packet to the beamformee as a transmission destination. Beamformee estimates the downlink channel matrix H using the HT-LTF of the received RTS packet.

In addition, the beamformer estimates the uplink channel matrix H ′ using the HT-LTF of the CTS packet returned from the beamformee. Then, Beamformer calculates the transmission weight matrix U ′ ^* (see the above equation (5)) based on the assumption of reversibility (see the above equation (4)), and the payload of the data packet A transmission vector X in which a portion is distributed to a plurality of transmission streams is multiplied by a transmission weight matrix U ′ ^* and spatially multiplexed to obtain a transmission vector Y having a transmission signal for each transmission antenna / branch as an element (the above formula ( 6)) and transmit.

Since the data packet arrives at Beamformee through the propagation path represented by the channel matrix H, the reception vector of the data payload portion is HY (= UDX) (see the above equation (7)). Beamforme estimates the UD as a channel matrix in the downlink direction from the HT-LTF of the data packet weighted and transmitted by the beamformer with the weight matrix V. Then, Beamformee obtains a reception weight matrix (UD) ⁻¹ corresponding to this UD, and multiplies the reception vector HY (= UDX) of the data payload portion of the data packet to multiplex the transmission stream. Are spatially separated into the original stream X.

  In the communication sequence example shown in FIG. 7, both Beamformer and Beamformee use the beam forming (spatial multiplexing and spatial separation processing) as it is from the channel matrix estimated from the training sequence HT-LTF included in the header of the packet received from the other party. ) Are used respectively.

  However, it is considered that the training sequence is generally short in time, and it cannot be expected that the channel matrix H is accurately estimated only by that portion. Further, the carrier frequency error and the sampling frequency error between transmission and reception are measured using another training sequence L-LTF transmitted prior to the MIMO training sequence HT-LTF (described above), and the same applies to these. However, due to the short time of the training sequence, it is difficult to measure with sufficient accuracy, and this also causes an estimation error of the channel matrix H. Since the weight matrix calculated from the channel matrix with insufficient accuracy still contains an error, there is a problem that the degree of improvement of the communication characteristics is lowered even if beam forming is performed.

Further, if it can be assumed that the channel matrix H does not substantially vary, the channel matrix H may be obtained only once during the entire RTS / CTS procedure. Fluctuates over time. The beamformer estimates the channel matrix H _CTS using the HT-LTF _CTS of the CTS packet received from the beamforme, and multiplies the transmission weight matrix U _CTS ^* calculated based on the channel matrix H _CTS ^* to space the payload portion of the data packet. Multiplex transmission is performed, but the channel matrix at the time when Beamformee receives HT-LTF _DATA of the data packet varies from the channel matrix H _CTS at the time of CTS packet transmission. For this reason, the channel matrix U _DATA D _DATA estimated by Beamformee from HT-LTF _DATA is not accurate for spatially separating a transmission stream spatially multiplexed by the transmission weight matrix U _CTS ^* .

  When Beamformee performs reception processing of the payload portion of the data packet, if the UD estimated from the HT-LTF of the data packet does not correspond to the channel matrix at that time, the spatially multiplexed payload portion of the data packet is Since orthogonalization (space separation) cannot be performed, an increase in packet error rate (PER), an increase in required SNR, and a deterioration in communication quality are caused.

JP-A-2005-323217, paragraphs 0018 to 0043 JP 2005-142715 A

  An object of the present invention is to provide an excellent wireless communication system, a wireless communication apparatus, and a wireless communication method capable of suitably performing closed-loop spatial multiplexing transmission by sharing channel information between wireless communication apparatuses. It is in.

  A further object of the present invention is to provide an excellent radio communication system capable of suitably performing spatial multiplexing (beam forming) of a plurality of transmission streams using a transmission weight matrix obtained by singular value decomposition of a channel matrix, and A wireless communication device and a wireless communication method are provided.

  A further object of the present invention is to provide an excellent radio communication system capable of suitably performing spatial multiplexing transmission by singular value decomposition by eliminating the influence of time variation of a channel matrix and carrier frequency error and sampling frequency error between transmission and reception. And providing a wireless communication device and a wireless communication method.

The present invention has been made in consideration of the above-mentioned problems. Packets are wirelessly transmitted using a plurality of antennas, and a transmission weight matrix is obtained from a channel matrix estimated from received packets, and the packets are weighted and transmitted. A wireless communication system comprising: a first wireless communication device that performs a weighting reception of a packet by obtaining a reception weight matrix from a channel matrix estimated from the received packet, wherein the packet is channel matrix estimated Training series and payload for
The first wireless communication apparatus estimates a channel matrix using a training matrix for channel matrix estimation included in a packet received from the second wireless communication apparatus, and a training sequence for channel matrix estimation of the received packet An error component of a channel matrix caused by at least one of a carrier frequency error and a sampling frequency error with the second wireless communication apparatus is obtained using a section other than the above, and transmitted using a channel matrix estimation result obtained by correcting the error component A weight matrix is generated, and a packet to be transmitted to the second wireless communication apparatus is weighted and transmitted with the transmission weight.
The second wireless communication apparatus receives a weighted transmission of the packet transmitted by the first wireless communication apparatus using a reception weight matrix obtained from a channel matrix estimated from a channel matrix estimation training sequence included in the packet. To start the
This is a wireless communication system.

  However, “system” here refers to a logical collection of a plurality of devices (or functional modules that realize specific functions), and each device or functional module is in a single housing. It does not matter whether or not.

  MIMO communication is known as a wireless communication technology that realizes high-speed and large-capacity transmission. Based on a channel matrix with the transfer function for each combination of transmission and reception antennas as the element, multiple streams are spatially multiplexed from the transmission side. At the same time, the original stream signal can be spatially separated from the reception signal of each antenna on the reception side. In particular, the closed-loop MIMO communication system can create an ideal spatial orthogonal channel by performing appropriate beam shaping toward the receiver by multiplying the transmission antenna weight based on the channel matrix on the transmitter side as well. .

  In a MIMO communication system represented by IEEE802.11n, a channel matrix can be acquired using a training sequence HT-LTF included in a PHY layer header at the beginning of a packet. In the SVD-MIMO communication system, for example, when performing communication operation according to the RTS / CTS procedure, the beamformer estimates the channel matrix using the training sequence included in the header of the CTS packet from the beamforme, and singular value decomposition thereof. The data packet is spatially multiplexed with the transmission weight matrix obtained as described above, and one beamforme uses the training sequence included in the header of the data packet to estimate the channel matrix, and the reception weight matrix obtained based on this The data payload portion can be spatially separated with

  However, since the training sequence is short in time, the channel matrix cannot be estimated with sufficient accuracy, or the measurement accuracy of the carrier frequency error and sampling frequency error between transmission and reception is not sufficient, resulting in a channel matrix estimation error. There is.

  Further, since the channel matrix fluctuates with time, there is a problem that the weight matrix obtained from the channel matrix estimated at different times of Beamformer and Beamformee does not correspond. That is, the time when the beamformer used to estimate the channel matrix used to spatially multiplex the payload portion of the data packet and the time when the beamformee used to spatially separate the payload portion of the data packet The channel matrix in which the beamformee is estimated from the training sequence included in the header of the data packet is not an accurate one for spatially separating the data payload portion spatially multiplexed on the beamformer side, and a packet error This increases the rate, increases the required SNR, and degrades the communication quality.

  On the other hand, in the wireless communication system according to the present invention, the first wireless communication apparatus operating as a beamformer uses a training sequence for channel matrix estimation included in a packet received from the second wireless communication apparatus operating as a beamforme. In addition to estimating the channel matrix, an error component of the channel matrix caused by at least one of a carrier frequency error and a sampling frequency error with the second wireless communication apparatus is obtained using a section other than the training sequence, A transmission weight matrix is generated using the channel matrix estimation result in which the error component is corrected, and subsequently, a packet to be transmitted to the second wireless communication apparatus is weighted and transmitted with the transmission weight.

  For example, when packet exchange is performed according to the RTS / CTS procedure, the first wireless communication device as the transmission request source uses the training sequence included in the header of the CTS packet from the second wireless communication device that is the transmission destination. In addition to estimating the channel matrix, the carrier frequency error and the sampling frequency error are obtained with high accuracy using the subsequent payload portion, and the channel matrix estimated value is continuously updated using them. Therefore, the first wireless communication apparatus calculates a transmission weight matrix from the latest channel matrix estimation value as a beamformer, and multiplies such a transmission weight matrix by a transmission vector to perform spatial multiplexing transmission of data packets. It is possible to prevent deterioration of communication characteristics due to a carrier frequency error or a sampling frequency error between transmission and reception on the second wireless communication apparatus, that is, the beamforme side.

When the first wireless communication device performs the communication procedure of transmitting the second packet addressed to the second wireless communication device following the reception of the first packet from the second wireless communication device, The wireless communication apparatus performs singular value decomposition on the uplink channel matrix H ′ (= U′D′V ′ ^H ) estimated from the received first packet, and transmits the first packet following the first packet. A transmission weight matrix V for weighting 2 packets can be obtained. Further, the second wireless communication apparatus receives the second packet weighted and transmitted by the first wireless communication apparatus via a channel composed of the downlink channel matrix H, and is estimated from the received second packet. Singular value decomposition of the downstream channel matrix H (= UDV ^H ) to obtain a reception weight matrix (UD) ⁻¹ for weighted reception of the second packet.

  The first wireless communication apparatus uses a payload portion after the training sequence for channel matrix estimation of the first packet, and a channel matrix caused by at least one of a carrier frequency error and a sampling frequency error with the second wireless communication apparatus The second packet is weighted and transmitted using a transmission weight matrix obtained from a channel matrix estimation result obtained by correcting the error component. The first wireless communication apparatus obtains a transmission weight matrix using the channel matrix estimation result at a position closer to the second packet and performs weighted transmission, whereby the second wireless communication apparatus receives the second packet. In this case, it is possible to prevent the reception characteristics from being deteriorated due to the influence of channel fluctuation.

  At this time, the first wireless communication apparatus estimates the channel matrix until reaching the end of the payload portion after the training sequence for channel matrix estimation of the first packet, so that the packet is transmitted based on the fresher channel matrix. Weighted transmission can be performed, and a reception weight matrix corresponding to the transmission weight matrix can be obtained from the second packet on the second wireless communication apparatus side.

  Alternatively, in the first wireless communication apparatus, when the arithmetic circuit is not sufficiently fast, if the channel matrix is continuously updated until the end of the received first packet, the channel matrix is subjected to singular value decomposition to obtain a transmission weight matrix. The process will not be in time for the transmission of the second packet. Therefore, in consideration of the frame interval from the end of reception of the first packet to the start of transmission of the second packet, transmission of the second packet in the payload portion after the training sequence for channel matrix estimation of the first packet. The channel matrix may be updated until the time when the transmission weight matrix calculation is in time.

  The first wireless communication device can update the channel matrix by channel tracking in the payload portion after the training sequence for channel matrix estimation of the packet received from the second wireless communication device. For channel tracking, a tracking algorithm such as LMS can be used. That is, when the payload portion of the received packet such as CTS arrives, the received signal is decoded, and then re-encoded to feed back this signal sequence, and the channel matrix H is followed by the LMS tracking algorithm based on this feedback information.

  Alternatively, even if the first wireless communication device is not equipped with a channel tracking mechanism, the first wireless communication device is inserted at the phase point of the payload portion after the channel matrix estimation training sequence of the packet received from the second wireless communication device. An error component of the channel matrix (other than the channel transfer function) caused by at least one of a carrier frequency error and a sampling frequency error with the second wireless communication device is obtained using the pilot signal, and the error component is calculated. A transmission weight matrix can be generated using the corrected channel matrix estimation result.

  As described above, the first wireless communication apparatus updates the channel matrix by channel tracking in the payload portion after the training sequence of the packet received from the second wireless communication apparatus. However, when at least a part of the signal series transmitted from each antenna of the second wireless communication apparatus is an extended training series that does not include a payload portion after the training series for channel matrix estimation, the corresponding channel The matrix component cannot be channel-tracked in the section after the training sequence.

  In such a case, the first wireless communication apparatus, with respect to the channel matrix component corresponding to the extended training sequence that was first obtained based on the extended training sequence, only the error component other than the temporal variation of the channel matrix. to correct. Further, the channel matrix component corresponding to the normal channel matrix estimation training sequence followed by the payload portion is caused by at least one of the carrier frequency error and the sampling frequency error with the second wireless communication apparatus, which is obtained by channel tracking. An error component due to temporal variation of the channel matrix is corrected. Then, a transmission weight matrix may be generated from a channel matrix obtained by combining these corrected channel matrix components.

  Here, if the temporal variation of the channel matrix is large, the matrix component that cannot be updated by the channel / track and the matrix component that can be tracked by the channel / track are so far apart that the channel matrix has sufficient estimation accuracy. Can no longer be obtained. Therefore, in such a case, the reception on the second wireless communication apparatus side is obtained by performing the singular value decomposition of the channel matrix by focusing only on the matrix components that can be tracked by the channel / track, and obtaining the transmission weighting matrix. It is possible to minimize the deterioration of characteristics.

  Advantageous Effects of Invention According to the present invention, an excellent wireless communication system, a wireless communication apparatus, and a wireless communication that can suitably perform spatial multiplexing transmission of a plurality of transmission streams using a transmission weight matrix obtained by singular value decomposition of a channel matrix. A communication method can be provided.

  In addition, according to the present invention, excellent wireless communication that can suitably perform spatial multiplexing transmission by singular value decomposition by eliminating the influence of temporal variation of the channel matrix and carrier frequency error between transmission and reception and sampling frequency error. A system, a wireless communication apparatus, and a wireless communication method can be provided.

  According to the present invention, the beamformer estimates a carrier frequency error and a sampling frequency error using portions other than various training sequences in the received packet, and performs spatial multiplexing using a transmission weight matrix calculated from a channel matrix obtained by correcting these errors. As a result, it is possible to prevent degradation of reception characteristics due to the error on the beamforme side.

In addition, according to the present invention, the beamformer is fresher (that is, with higher estimation accuracy) updated by following the channel track than the channel matrix estimated from the training sequence HT-LTF of the received packet (CTS packet). A transmission weighting matrix of a packet (data packet) to be transmitted immediately after the singular value decomposition of the channel matrix is calculated. That is, since the transmission weight matrix V closer to the transmission weight matrix V corresponding to the time when the beamformer transmits the packet (data packet) can be used, that is, the V ^H V in the following equation (8) is converted into a unit matrix. You can get closer.

  As a result, the Beamformee side can obtain a more accurate channel matrix UD from the training sequence HT-LTF of the data packet received after transmitting the CTS packet, and can improve the separation accuracy of the multiplexed signal in the data payload portion. It is possible to increase the communication quality by reducing the packet error rate.

  Further, according to the present invention, when a beamformer receives a packet including an extended training sequence that does not include a payload portion and cannot be followed by a channel track, the beamformer corresponds to a training sequence that can be updated by a channel track including the payload portion. The channel matrix component is made to be in a fresh state by correcting the error between the transmitter and the receiver and the temporal variation of the channel transfer function, and for the matrix component corresponding to the extended training sequence that cannot be updated by the channel track, the carrier frequency between the transmitter and receiver Error and sampling frequency error are corrected, and a channel matrix combining these matrix components is decomposed into singular values and used as a weighting matrix for subsequent transmission packets for spatial multiplexing of packets to be transmitted immediately thereafter. Therefore, on the Beamformee side, the channel matrix UD can be more accurately derived from the data payload portion of the transmission packet to improve the separation quality of the multiplexed signal and improve the communication quality.

  In addition, according to the present invention, the channel matrix component that cannot be updated by the channel track is greatly deviated from the channel matrix component that cannot be updated by the channel track and the channel element that can be followed by the channel track, and the estimated channel transmission If the function estimation accuracy cannot be obtained, the channel matrix focused only on the matrix components that can be tracked by the channel and track is decomposed into singular values, and this is used as the weighting vector for the subsequent transmission packets. Degradation of characteristics can be minimized.

  Further, according to the present invention, even if the temporal variation of the channel is completely negligible, the beamformer uses not the channel matrix estimated by only the temporally short training signal but the other data portion. The transmission weight matrix of the next transmission packet is obtained from the channel matrix updated by the channel track. Since the channel matrix is updated by using a signal of a longer time, the channel estimation accuracy whose upper limit is determined by the SNR of the training signal portion, compared with the case where the channel matrix is estimated only from the training sequence, It is possible to improve beyond the upper limit, and it is possible to prevent deterioration of beam forming characteristics due to poor estimation accuracy.

  Other objects, features, and advantages of the present invention will become apparent from more detailed description based on embodiments of the present invention described later and the accompanying drawings.

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

  The radio communication system according to the present invention includes a transmitter and a receiver each having a plurality of antennas, and performs MIMO communication in a closed loop type. From the viewpoint of packet transmission, the wireless communication device that is the source of the data packet is an initiator, and the wireless communication device that is the destination of the data packet is positioned as the receiver, but from the viewpoint of beam forming, multiple streams are spatially multiplexed. The wireless communication device that transmits the packet in this manner is Beamformer, and the wireless communication device that receives the spatially multiplexed packet is Beamformee. In the following description, communication from the beamformer to the beamforme is regarded as a “forward” or “downlink” link, and communication from the beamformee to the beamformer is regarded as a “reverse” or “uplink” link.

  In the radio communication system according to the present invention, as a procedure for performing feedback regarding the channel matrix, for example, spatial multiplex transmission is performed in accordance with the implicit feedback procedure. In the impact feedback, the beamformer separates the spatial stream training from the beamformee, assembles an estimated channel matrix in the reverse direction from the separated training sequence, and uses the transmission weight matrix for beam formation obtained based on the channel matrix. Then, communication is performed by forming a transmission packet into a beam. However, it is assumed that the channel matrix is reversible.

  FIGS. 1 and 2 show configurations of a transmission system and a reception system of a wireless communication apparatus that can operate mainly as an impact feedback beamformer in the wireless communication system according to the present invention, respectively. The number of antennas shared by the transmission / reception system is N, and this N is, for example, a maximum of four if conforming to the IEEE specifications. However, in each figure, only two antenna branches are drawn to avoid the confusion of the drawings. . In addition, although description is abbreviate | omitted below, the transmission system and reception system of the radio | wireless communication apparatus which operate | move as Beamforme can also be comprised as FIG.1 and FIG.2, respectively.

  The transmission data supplied from the data generator 100 is scrambled by the scrambler 102. Next, the encoder 104 performs error correction coding. The encoded serial data sequence is input to the serial / parallel (S / P) conversion unit 106 and distributed to N transmission streams.

  In each transmission stream, the transmission signal is punctured by the puncture 108 in accordance with the data rate given for each stream, and is interleaved by the interleaver 110 so that the same interleaving is not performed among a plurality of streams. The mapper 112 applies any one of BPSK, QPSK, 16QAM, and 64QAM, maps the bit sequence for each transmission stream to the IQ signal space, and becomes a complex baseband signal.

  The selector 111 inserts a training sequence at an appropriate timing into the transmission signal for each interleaved spatial stream, and supplies the training sequence to the mapper 112. In the IEEE802.11n specification, training sequences such as L-STF, L-LTF, HT-STF, and HT-LTF are defined for mixed format packets. Also, training sequences such as HT-GF-STF and HT-LTF are defined for the green field format packet (described above).

  When performing beam forming, that is, spatial multiplexing of a plurality of transmission streams, the transmission weight matrix V is calculated by singular value decomposition of the channel matrix H by the beam forming transmission weight matrix calculation unit 114a in the spatial multiplexing unit 114. Then, the transmission weight matrix multiplication unit 114b multiplies the transmission vector X having each transmission stream as an element by this transmission weight matrix V, thereby generating a transmission stream for each transmission antenna branch (Equations (1) to (1)). See 6)).

  The channel matrix H can be estimated using, for example, the training sequence HT-LTF included in the received packet from the communication partner (Beamforme) in the space separation unit 216 of the reception system. In this embodiment, the channel matrix H is continuously updated by channel tracking even in the payload portion of the packet following the training sequence, and the transmission weight matrix multiplication unit 114b transmits the transmission weight matrix V calculated from the latest channel matrix estimation value. The spatial multiplexing transmission can be performed using the above, but the details will be described later.

  The fast Fourier inverse transform unit (IFFT) 116 converts the subcarriers arranged in the frequency domain into a time axis signal, and the guard insertion unit 118 adds a guard interval. The band is limited by the digital filter 120, converted to an analog signal by the DA converter (DAC) 122, up-converted to an appropriate frequency band by the RF unit 124, and then transmitted from each transmission antenna to the propagation path. Is sent out. The Implicit feedback is based on the premise that the channel characteristics are reversible, and the RF unit 124 is calibrated.

  On the other hand, the data received from the communication partner (Beamforme) through the channel is subjected to analog processing in the RF unit 228 in each receiving antenna branch, converted into a digital signal by the AD converter (ADC) 226, and then to the digital filter 224. Entered. The Implicit feedback is based on the premise that the channel characteristics are reversible, and the RF unit 228 is calibrated.

  Subsequently, after processing such as packet discovery, timing detection, correction of carrier frequency error and sampling frequency error is performed in the synchronization circuit 222, the guard interval added to the head of the data transmission interval is detected by the guard removal unit 220. Remove. Then, the fast axis transform unit (FFT) 218 turns the time axis signal into a frequency axis signal.

  In the space separation unit 216, space separation processing is performed on the spatially multiplexed received signal. Specifically, the estimated channel matrix H is assembled from the training sequence HT-LTF included in the PHY header of the packet received from the communication partner (Beamformee), and the antenna reception weight matrix is calculated based on the channel matrix H. In the present embodiment, the space separation unit 216 re-encodes the signal sequence decoded by the decoder 204 (described later) in the payload portion of the packet following the training sequence, and performs decision feedback (Decision Feedback). The channel matrix H is continuously updated by channel tracking. Therefore, transmission weight matrix multiplication section 114b can perform spatial multiplexing transmission using transmission weight matrix V calculated from the latest channel matrix estimation value. Details of the internal configuration of the space separation unit 216 and the channel tracking operation will be described later.

  The channel equalization circuit 214 further performs residual frequency offset correction, channel tracking, and the like on the signal sequence for each stream. The demapper 212 demaps the received signal in the IQ signal space, the deinterleaver 210 deinterleaves, and the depuncture 208 depunctures at a predetermined data rate.

  The parallel / serial (P / S) conversion unit 206 performs parallel / serial conversion on a plurality of received streams and combines them into one stream. This data synthesizing process is an operation reverse to the data distribution performed on the transmission side. Then, after error correction decoding by the decoder 204, the descrambler 202 descrambles, and the data acquisition unit 200 acquires the received data.

  FIG. 3 shows a configuration inside the space separation unit 216.

  The channel matrix estimation unit 309 separates the training sequence HT-LTF included in the PHY header of the received packet, assembles an uplink estimated channel matrix H, and stores this in the channel matrix holding unit 308.

  On the other hand, in the payload portion after the training sequence of the received packet, the received signal is subjected to error correction decoding by the decoder 204, then re-encoded by the encoder 301, and further distributed to a plurality of streams by the serial / parallel converter 302. The stream is punctured by the puncture 303, interleaved by the interleaver 304, and mapped to the signal space by the mapper 305. Then, the decision feedback unit 306 makes a decision feedback (Decision Feedback) to the channel matrix tracking unit 307 for the signal sequence including a plurality of streams.

  The channel matrix tracking unit 307 inputs the reception signal sequence of each reception antenna branch before being spatially separated by multiplying the reception weight matrix, and the signal sequence decoded by the decoder 204 and re-encoded, and receives the LMS (Least) Using a tracking algorithm such as (Mean Square), the channel matrix H is tracked over the payload portion of the received packet. Specifically, the channel matrix tracking unit 307 obtains an error component caused by a carrier frequency error or a sampling frequency error between transmission and reception, and obtains a channel matrix estimation result obtained by correcting these error components. Since the channel matrix H fluctuates in time and causes a carrier frequency error and a sampling frequency error, an error component resulting from the carrier frequency error and the sampling frequency error measured from the training sequence L-LTF is represented by the training sequence HT-LTF. In addition to removing from the estimated channel matrix, error components due to carrier frequency error and sampling frequency error are obtained in the payload portion after the training sequence HT-LTF, and updated to a channel matrix in which these are corrected.

  Alternatively, when the channel matrix tracking unit 307 is not equipped with a channel / track mechanism, the channel matrix tracking unit 307 uses at least one of the carrier frequency error and the sampling frequency error using the pilot signal inserted at the phase point of the data payload portion. An error component of the resulting channel matrix is obtained, and a transmission weight matrix can be generated using a channel matrix estimation result obtained by correcting the error component.

  In this way, the channel matrix that the channel matrix tracking unit 307 updates every moment is stored in the channel matrix holding unit 308.

  For example, when the CTS packet is received in the RTS / CTS procedure, the channel matrix estimation unit 309 separates the training sequence HT-LTF included in the PHY header to obtain the channel matrix, but the channel matrix tracking unit 307 determines the packet time. It is possible to obtain a channel matrix followed by the channel track.

  For the channel matrix component estimated from the Extension HT-LTF that is included in the received packet but cannot be updated by the channel track, the channel matrix tracking unit 307 performs error components (carrier frequency between transmission and reception) other than the temporal variation of the channel matrix. A value obtained by correcting only an error or a sampling frequency error is used. For the channel matrix component corresponding to HT-LTF that is included in the received packet and can be updated by the channel track, a fresh value updated by the channel track is used. A channel matrix obtained by combining these channel matrix components is stored in the channel matrix holding unit 308.

  Reception weight matrix calculation section 310 calculates an antenna reception weight matrix based on channel matrix H stored in channel matrix holding section 308. When the wireless communication apparatus operates as Beamformee and the received packet is beamformed, the estimated channel matrix is equal to UD when the singular value decomposition is performed (see Equation (7)). An antenna reception weight matrix is calculated. Reception weight matrix multiplication section 311 performs spatial decoding of the spatially multiplexed signal by performing matrix multiplication of the reception vector having each reception stream as an element and the antenna reception weight matrix, and obtains an independent signal sequence for each stream. .

  Further, the estimated channel matrix H stored in the channel matrix holding unit 308 is passed to the transmission system beam forming transmission weight matrix calculation unit 114a as a downlink channel matrix (it is assumed that the channel matrix is reversible). . As described above, the transmission weight matrix calculation unit 114a calculates the transmission weight matrix V by performing singular value decomposition on the channel matrix H, and the transmission weight matrix multiplication unit 114b applies this transmission vector X to each transmission stream as an element. The transmission weight matrix V is multiplied, thereby generating a transmission stream for each transmission antenna branch.

  The space separation unit 216 continues to update the channel matrix by channel tracking even in the payload portion after the header, and the channel matrix holding unit 308 stores the latest channel matrix estimation value. It should be fully understood that the transmission weight matrix can be calculated from the estimated channel matrix values and the transmission vector can be multiplied by a more accurate transmission weight matrix to perform spatial multiplexing transmission of data packets.

For example, when transmitting a data packet in the RTS / CTS procedure, the transmission weight matrix calculation unit 114a uses the CTS packet time instead of the channel matrix obtained from the training sequence HT-LTF included in the PHY header of the immediately preceding CTS packet. The latest channel matrix obtained by following the channel and track by the singular value is subjected to singular value decomposition to obtain U ′ ^* (= V) as a transmission weight. Multiply U ' ^* .

  In addition, the channel matrix tracking unit 307, regarding the channel matrix component estimated from the Extension HT-LTF that is included in the received packet but cannot be updated by the channel / track, is an error component other than the temporal variation of the channel matrix (between transmission and reception). The channel matrix component corresponding to the HT-LTF that is included in the received packet and can be updated by the channel track is updated by the channel track. A channel matrix in which these channel matrix components are combined is stored in the channel matrix holding unit 308 using a value with a high freshness. The transmission weight matrix calculation unit 114a performs singular value decomposition on such a channel matrix, and uses the result as a weight matrix for subsequent transmission packets.

  Here, the operation of the transmission system shown in FIG. 1 and the reception system shown in FIGS. 2 to 3 when the wireless communication apparatus transmits a data packet through the RTS / CTS procedure as a beamformer in the impact feedback will be described with reference to FIG. Will be described with reference to FIG.

  The beamformer transmits an RTS packet addressed to the beamformee as a communication partner from the transmission system, and in response to this, the beamformee returns a CTS packet.

  In the beamformer reception system, first, the channel matrix H ′ in the uplink direction is estimated using the HT-LTF of the received CTS packet. However, the channel matrix is further tracked by the CTS packet time, and the channel matrix is continuously updated. .

In the Beamformer transmission system, transmission is performed based on the assumption of reversibility (see the above equation (4)) from the latest upstream channel matrix H ′ obtained by channel-tracking for the time of CTS packets. A weight matrix U ′ ^* (see the above equation (5)) is calculated. Then, the transmission vector X in which the payload portion of the data packet is distributed to a plurality of transmission streams is multiplied by the transmission weight matrix U ′ ^* to be spatially multiplexed, and the transmission vector Y having the transmission signal for each transmission antenna branch as an element (See equation (6) above) and transmit.

  Since this data packet arrives via the propagation path represented by the channel matrix H, the reception vector of the data payload portion is HY (= UDX) (see the above equation (7)).

The channel matrix H ′ used by Beamformer to calculate the transmission weight matrix U ′ ^* is fresh for Beamforme estimated at a time closer to the time when the training sequence HT-LTF of the data packet is received. Therefore, on the Beamforme side, a more accurate channel matrix UD can be estimated from the training sequence HT-LTF included in the header of the data packet. In other words, the beamformee obtains a reception weight matrix (UD) ⁻¹ more accurately corresponding to the transmission weight matrix U ′ ^* used for spatial multiplexing of the payload portion of the data packet on the beamformer side, and performs spatial separation. it can. This can prevent an increase in packet error rate (PER), an increase in required SNR, and a deterioration in communication quality.

  Finally, the operation method in the channel matrix tracking unit 307 in the space separation unit 216 will be summarized.

(1) When correcting only error components other than the channel transfer function Referring to the pilot signal of the received signal, the carrier frequency error and the sampling frequency error between the beamformer and the beamformee are detected (in the OFDM transmission method, etc. A pilot signal serving as a reference for performing amplitude and phase fluctuation compensation in the complex baseband after quasi-synchronous detection is inserted, for example, at the phase point of the payload portion). A transmission weighting matrix is calculated from a channel matrix in which these errors are corrected. This method is also effective in a wireless communication system not equipped with a channel track mechanism.

(2) When the arithmetic circuit is sufficiently fast The beamformer performs channel tracking to the end of the received packet (for example, CTS packet), and obtains the final channel matrix H of the received packet. Then, U ′ ^* is calculated as a transmission weight matrix. At the time of transmission of the subsequent data packet, the transmission weight matrix U ′ ^* is multiplied and transmitted. Ideal operation can be realized.

(3) When the operation of channel track and singular value decomposition is not in time in the time interval between the received packet and the data packet In IEEE 802.11, it is short for transmission of high priority frames such as RTS / CTS frames and ACK frames. A frame interval (SIFS: Short Inter-Frame Space) is used. For this reason, it is assumed that the calculation of the channel track and the singular value decomposition is not in time before the beamformer receives the CTS packet and starts transmission of the data packet.

In such a case, the beamformer reversely calculates the time required for the calculation of the singular value decomposition so that the calculation of the transmission weight matrix U ′ ^* is in time when the data packet is transmitted, and the timing preceding the next transmission packet is approximately that much. The transmission weight matrix is calculated using the channel matrix that has been channel-tracked and estimated.

(4) When there is an Extension HT-LTF and the time variation of the channel matrix H is small IEEE 802.11n stipulates that the HT-LTF is composed of two parts. The first part consists of 1 to 4 HT-LTFs, which are necessary for the spatial separation of the data payload part and are also referred to as DATA HT-LTFs (herein simply referred to as HT-LTFs). Sometimes referred to as DATA HT-LTF). Also, Extension HT-LTF is defined as a training sequence for investigating the number of extra spatial dimensions in the MIMO channel. The second part is optional, consists of 0 to 4 HT-LTFs, and is used to investigate the extra spatial dimension of the MIMO channel that is not used for spatial multiplexing transmission of the data payload part, Also called Extra HT-LTF.

  Unlike the DATA HT-LTF, the MIMO channel that transmits the Extension HT-LTF does not include a payload portion thereafter, and therefore cannot be channel-tracked. Therefore, when the beamformer channel-tracks the channel matrix H as a result of channel tracking, there is little time variation of the channel matrix H, and only the channel matrix component corresponding to the HT-LTF that can be updated by the channel track is updated. Perform singular value decomposition. In addition, for matrix components corresponding to Extension HT-LTF that cannot be updated by the channel / track, error components (carrier frequency error and sampling frequency error between transmission and reception) other than temporal fluctuation of the channel matrix are set in the first Extension HT-LTF. Correct for this. Then, a singular value decomposition is performed on the channel matrix obtained by combining these corrected channel matrix components, and the result is used to obtain a transmission weighting matrix for subsequent transmission packets.

For example, a packet (CTS packet) transmitted from Beamformee consists of three MIMO channels. Of these, DATA HT-LTF is transmitted on two MIMO channels # 1 to # 2, but one MIMO channel When Extension HT-LTF is transmitted (see FIG. 5), channel tracking is possible since the payload continues after DATA HT-LTF, but the field of the payload is extended after Extension HT-LTF. Since it is not accompanied, channel tracking is impossible. When the transfer function from the beamformee's i-th transmit antenna to the beamformer's j-th receive antenna is set to h _ji , the channel matrix is expressed by the following equation (9) (where the number of beamformee transmit antennas is 3 And the number of Beamformer receiving antennas is two).

For the transfer functions h ₁₁ , h ₁₂ , h ₂₁ , and h ₂₁ obtained from the HT-LTF of the streams # 1 to # 2, a channel track is possible, but the transfer obtained from the HT-LTF of the stream # 3. Channel tracking is not possible for the functions h ₁₃ and h ₂₃ . Therefore, at the end of the fed back packet (CTS packet), the channel matrix is updated as shown in the following equation (10). As shown in the equation, the matrix elements in the first and second columns of the channel matrix are updated, but the channel matrix element in the third column is not updated.

  Further, when an error component other than the temporal variation of the channel matrix (carrier frequency error between transmission and reception, sampling frequency error, etc.) is Δ and corrected together, the channel matrix at the end of the fed back packet (CTS packet) Is as shown in the following equation (11).

  Therefore, the beamformer in this case may use the result obtained by singular value decomposition of the channel matrix shown in the above equation (11) as the transmission weight matrix of the subsequent transmission packet (data packet).

(5) When there is an Extension HT-LTF and the temporal variation of the channel matrix H is large If the temporal variation of the channel matrix H is large as a result of channel tracking, the channel corresponding to the Extension HT-LTF Matrix components are discarded, and singular value decomposition processing is performed using only matrix elements corresponding to channel-trackable HT-LTFs.

(6) Method for Confirming Temporal Variation of Channel Matrix Beamformer receives received packets (CTS packets) to the end, and estimates a carrier frequency error and a sampling frequency error between the transmitter and the receiver. Then, with this correction value, this error component is removed from the HT-LTF near the head of the packet, and the channel matrix is estimated. As a result of receiving up to the end of the packet, the error is removed from the channel transfer function in which both the time variation of the channel matrix and the error between the transmitter and receiver are corrected, and the HT-LTF near the head of the packet. A correlation value with the estimated channel transfer function is obtained, and the magnitude of temporal variation of the channel matrix is determined using this correlation value.

  The present invention has been described in detail above with reference to specific embodiments. However, it is obvious that those skilled in the art can make modifications and substitutions of the embodiment without departing from the gist of the present invention.

  In the present specification, for the sake of simplicity of explanation, only the embodiment in which the transmitting terminal performs “direct mapping” in which a stream is directly assigned to an antenna branch has been described. However, “spatial expansion”, stream, and antenna branch are described. Even when a conversion method that does not correspond to the one-to-one is adopted, the present invention can be similarly applied.

  In the present specification, the description has been made mainly on the embodiment applied to IEEE802.11n which is an extension standard of IEEE802.11. However, the gist of the present invention is not limited to this. For example, Mobile WiMax (Worldwide Interoperability for Microwave) based on IEEE802.16e, IEEE802.20 which is a high-speed wireless communication standard for mobiles, and a high-speed wireless PAN (Personal Area) using the 60 GHz (millimeter wave) band IEEE 802.15.3c, Wireless HD that enables transmission of uncompressed HD (High Definition) video using 60 GHz (millimeter-wave) band wireless transmission, MIMO communication such as 4th generation (4G) mobile phones, etc. The present invention can be similarly applied to various wireless communication systems employing the method.

  In short, the present invention has been disclosed in the form of exemplification, and the description of the present specification should not be interpreted in a limited manner. In order to determine the gist of the present invention, the claims should be taken into consideration.

FIG. 1 is a diagram illustrating a configuration of a transmission system of a wireless communication apparatus that can mainly operate as a beamformer of an implicit feedback. FIG. 2 is a diagram illustrating a configuration of a reception system of a wireless communication apparatus that can operate mainly as a beamformer of an implicit feedback. FIG. 3 is a diagram illustrating a configuration in the space separation unit 216. FIG. 4 is a diagram illustrating a communication sequence when a wireless communication apparatus transmits a data packet through an RTS / CTS procedure as a beamformer in an impact feedback. FIG. 5 is a diagram for explaining a channel matrix update method when Extension HT-LTF is included. FIG. 6 is a diagram showing an example (mixed format) of a packet format defined by IEEE802.11n adopting the OFDM_MIMO communication system. FIG. 7 is a diagram schematically showing a communication sequence in accordance with the RTS / CTS procedure in a closed loop type MIMO communication system. FIG. 8 is a diagram showing how the channel matrix fluctuates on the assumption that the RTS / CTS procedure is executed.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 ... Data generator 102 ... Scrambler 104 ... Encoder 106 ... Serial-parallel conversion part 108 ... Puncture 110 ... Interleaver 111 ... Selector 112 ... Mapper 114 ... Spatial multiplexing part 114a ... Transmission weight matrix calculation part 114b for beam generation ... Transmission weight matrix calculation unit 116... Fast Fourier inverse transform unit (IFFT)
118 ... Guard insertion part 120 ... Digital filter 122 ... DA converter (DAC)
DESCRIPTION OF SYMBOLS 124 ... RF part 200 ... Data acquisition part 202 ... Descrambler 204 ... Decoder 206 ... Parallel / serial conversion part 208 ... Depuncture 210 ... Deinterleaver 212 ... Demapper 214 ... Channel equalization circuit 216 ... Spatial separation part 218 ... Fast Fourier transform Department (FFT)
DESCRIPTION OF SYMBOLS 220 ... Guard removal part 222 ... Synchronous circuit 224 ... Digital filter 226 ... AD converter (ADC)
228 ... RF unit 301 ... encoder 302 ... serial / parallel conversion unit 303 ... puncture 304 ... interleaver 305 ... mapper 306 ... decision feedback unit 307 ... channel matrix tracking unit 308 ... channel matrix holding unit 309 ... channel matrix estimation unit 310 ... Reception weight matrix calculation unit 311... Reception weight matrix multiplication unit

Claims (29)

A first wireless communication device that wirelessly transmits a packet using a plurality of antennas, obtains a transmission weight matrix from a channel matrix estimated from the received packet, and weights and transmits the packet, and is estimated from the received packet A wireless communication system comprising a second wireless communication device that obtains a reception weight matrix from a channel matrix and weights and receives a packet, wherein the packet includes a training sequence for channel matrix estimation and a payload,
The first wireless communication apparatus estimates a channel matrix using a training matrix for channel matrix estimation included in a packet received from the second wireless communication apparatus, and a training sequence for channel matrix estimation of the received packet An error component of a channel matrix caused by at least one of a carrier frequency error and a sampling frequency error with the second wireless communication apparatus is obtained using a section other than the above, and transmitted using a channel matrix estimation result obtained by correcting the error component A weight matrix is generated, and a packet to be transmitted to the second wireless communication apparatus is weighted and transmitted by the transmission weight matrix thereafter,
The second wireless communication apparatus receives a weighted transmission of the packet transmitted by the first wireless communication apparatus using a reception weight matrix obtained from a channel matrix estimated from a channel matrix estimation training sequence included in the packet. To start the
A wireless communication system. The channel between the first and second wireless communication devices is assumed to be reversible in the channel matrix, and the first wireless communication device receives the first packet from the second wireless communication device. When transmitting the second packet to the second wireless communication device,
The first wireless communication apparatus performs singular value decomposition on an uplink channel matrix H ′ (= U′D′V ′ ^H ) estimated from the received first packet, and follows the first packet. Obtaining a transmission weight matrix V for weighting the second packet to be transmitted;
The second wireless communication apparatus receives the second packet weighted and transmitted by the first wireless communication apparatus via a channel including a downlink channel matrix H, and is estimated from the received second packet. Singular value decomposition of the downlink channel matrix H (= UDV ^H ) to obtain a reception weight matrix (UD) ⁻¹ for weighted reception of the second packet,
The wireless communication system according to claim 1. When the first wireless communication device transmits the second packet to the second wireless communication device following the reception of the first packet from the second wireless communication device,
The first wireless communication apparatus uses a payload portion after the training sequence for channel matrix estimation of the first packet to generate a channel caused by at least one of a carrier frequency error and a sampling frequency error with the second wireless communication apparatus An error component of the matrix is obtained, and the second packet is weighted and transmitted using a transmission weight matrix obtained from a channel matrix estimation result obtained by correcting the error component.
The wireless communication system according to claim 1. The first wireless communication apparatus is caused by at least one of a carrier frequency error and a sampling frequency error with the second wireless communication apparatus until reaching the end of the payload portion after the training sequence for channel matrix estimation of the first packet. An error component of a channel matrix to be transmitted, and weighted transmission of the second packet using a transmission weight matrix obtained from a final channel matrix estimation result obtained by correcting the error component at the end.
The wireless communication system according to claim 3. The first wireless communication device considers a frame interval from the end of reception of the first packet to the start of transmission of the second packet, and includes a payload portion after the training sequence for channel matrix estimation of the first packet. An error component of the channel matrix resulting from at least one of a carrier frequency error and a sampling frequency error with the second wireless communication apparatus is obtained until a time point at which transmission weight matrix calculation is in time for transmission of the second packet. The second packet is weighted and transmitted using a transmission weight matrix obtained from a final channel matrix estimation result obtained by correcting the error component at the time point.
The wireless communication system according to claim 3. The first wireless communication device performs carrier frequency error with the second wireless communication device by channel tracking in a payload portion after a training sequence for channel matrix estimation of a packet received from the second wireless communication device. Alternatively, an error component of a channel matrix caused by at least one of sampling frequency errors is obtained, and a transmission weight matrix is generated using a channel matrix estimation result obtained by correcting the error component.
The wireless communication system according to claim 1. The first radio communication device uses the pilot signal inserted at the phase point of the payload portion after the training sequence for channel matrix estimation of the packet received from the second radio communication device, and uses the second radio An error component of a channel matrix resulting from at least one of a carrier frequency error or a sampling frequency error with a communication device is obtained, and a transmission weight matrix is generated using a channel matrix estimation result obtained by correcting the error component.
The wireless communication system according to claim 1. When at least a part of the signal sequence transmitted from each antenna of the second wireless communication device is an extended training sequence that does not include a payload portion after the training sequence for channel matrix estimation,
The first wireless communication apparatus obtains a channel matrix component corresponding to a normal channel matrix estimation training sequence followed by a payload portion by carrier tracking error or sampling with the second wireless communication apparatus, which is obtained by channel tracking. Corrects error components due to temporal variations in channel matrix caused by at least one of frequency errors, and corrects error components other than temporal variations in channel transfer function for channel matrix components corresponding to extended training sequences. A transmission weight matrix is generated from the channel matrix obtained by combining the channel matrix components after
The wireless communication system according to claim 6. When the first wireless communication apparatus corrects an error component by channel tracking for a channel matrix component corresponding to a normal channel matrix estimation training sequence followed by a payload portion, the first wireless communication device estimates the channel matrix of the first packet. The error component of the channel matrix resulting from at least one of the carrier frequency error and the sampling frequency error with the second wireless communication apparatus is obtained until reaching the end of the payload portion after the training sequence for use, and the error component is corrected at the end Weighted transmission of the second packet using a transmission weight matrix obtained from the final channel matrix estimation result;
The wireless communication system according to claim 8. When the first wireless communication device corrects an error component by channel tracking for a channel matrix component corresponding to a normal channel matrix estimation training sequence followed by a payload portion, the first wireless communication device starts receiving the first packet. Considering the frame interval until the start of transmission of the second packet, the transmission weight matrix can be calculated in time for the transmission of the second packet in the payload portion after the channel matrix estimation training sequence of the first packet. From the final channel matrix estimation result obtained by obtaining an error component of the channel matrix due to at least one of the carrier frequency error and the sampling frequency error with the second wireless communication device until the time and correcting the error component at the time A second packet is weighted and transmitted using a transmission weight matrix obtained;
The wireless communication system according to claim 8. The first wireless communication apparatus, for the channel matrix component corresponding to the extended training sequence, has a temporal relationship of a channel transfer function caused by at least one of a carrier frequency error and a sampling frequency error with the second wireless communication apparatus. Correct error components other than fluctuations,
The wireless communication system according to claim 8. When the variation of the channel matrix is large, the first wireless communication apparatus generates a transmission weight matrix from the channel matrix obtained by removing the channel matrix component estimated from the extended training sequence.
The wireless communication system according to claim 8. The first wireless communication apparatus, for a channel matrix component corresponding to a normal channel matrix estimation training sequence followed by a payload portion, until the end of the payload portion after the channel matrix estimation training sequence of the first packet. Generate a transmission weight matrix from the channel matrix updated by channel tracking.
The wireless communication system according to claim 12. For the channel matrix component corresponding to a normal channel matrix estimation training sequence followed by a payload portion, the first wireless communication apparatus sets a frame interval from the end of reception of the first packet to the start of transmission of the second packet. In consideration, the channel matrix updated by channel tracking until the time when the calculation of the transmission weight matrix is in time for the transmission of the second packet in the payload portion after the training sequence for channel matrix estimation of the first packet Generate a transmission weight matrix from
The wireless communication system according to claim 12. A wireless communication device that includes a plurality of antennas and wirelessly transmits a packet to and from a communication partner, the packet including a training sequence for channel matrix estimation and a payload,
A packet receiving means for receiving a packet from the communication partner;
Channel matrix estimation means for estimating a channel matrix using a training sequence for channel matrix estimation included in a packet received from the communication partner;
An error component of a channel matrix caused by at least one of a carrier frequency error or a sampling frequency error with the communication partner is obtained using a section other than the training sequence for channel matrix estimation of the packet from the communication partner, and the error component is corrected Channel matrix tracking means for estimating the channel matrix;
Transmission weight matrix generation means for generating a transmission weight matrix using a channel matrix estimation result by the channel matrix tracking means;
Packet transmission means for weighting and transmitting packets to be transmitted to the communication partner thereafter using the transmission weight matrix;
A wireless communication apparatus comprising: The channel between the communication partners is assumed to be reversible in the channel matrix, and when the second packet is transmitted to the communication partner following the reception of the first packet from the communication partner,
The transmission weight matrix generation means performs singular value decomposition on an upstream channel matrix H ′ (= U′D′V ′ ^H ) estimated from the first packet by the channel matrix tracking means to obtain a first packet. Subsequently, a transmission weight matrix V for weighting the second packet to be transmitted is obtained.
The wireless communication apparatus according to claim 15. When transmitting the second packet to the communication partner following the reception of the first packet from the communication partner,
The channel matrix tracking means obtains an error component of the channel matrix caused by at least one of a carrier frequency error and a sampling frequency error with the communication partner using a payload portion after the training sequence for channel matrix estimation of the first packet. Estimating a channel matrix with the error component corrected,
The wireless communication apparatus according to claim 15. The channel matrix tracking means includes a channel caused by at least one of a carrier frequency error and a sampling frequency error with the second wireless communication apparatus until reaching the end of the payload portion after the training sequence for channel matrix estimation of the first packet. Obtaining an error component of the matrix and estimating a final channel matrix with the error component corrected at the end,
The wireless communication apparatus according to claim 17. The channel matrix tracking means considers the frame interval from the end of reception of the first packet to the start of transmission of the second packet, and includes the first portion of the payload portion after the training sequence for channel matrix estimation of the first packet. Until a time point at which transmission weight matrix calculation is in time for transmission of the second packet, an error component of the channel matrix resulting from at least one of a carrier frequency error and a sampling frequency error with the communication partner is obtained, and the error component at that time point Estimate the final channel matrix corrected for,
The wireless communication apparatus according to claim 17. The channel matrix tracking means includes a channel caused by at least one of a carrier frequency error or a sampling frequency error with the communication partner by channel tracking in a payload portion after a channel matrix estimation training sequence of a packet received from the communication partner. Obtaining an error component of the matrix and estimating a channel matrix in which the error component is corrected;
The wireless communication apparatus according to claim 15. The channel matrix tracking means uses the pilot signal inserted at the phase point of the payload portion after the training sequence for channel matrix estimation of the packet received from the communication partner, or the carrier frequency error or sampling frequency with the communication partner Obtaining an error component of the channel matrix caused by at least one of the errors, and estimating a channel matrix in which the error component is corrected;
The wireless communication apparatus according to claim 15. When at least a part of the signal sequence transmitted from each antenna of the communication partner is an extended training sequence that does not include a payload portion after the training sequence for channel matrix estimation,
For the channel matrix component corresponding to the normal channel matrix estimation training sequence followed by the payload portion, the channel matrix tracking means determines at least one of the carrier frequency error and the sampling frequency error with the communication partner obtained by channel tracking. Correct the error component due to the temporal variation of the channel matrix caused by it, correct the error component other than the temporal variation of the channel transfer function for the channel matrix component corresponding to the extended training sequence, and correct these channel matrix components To estimate the channel matrix,
The wireless communication apparatus according to claim 20. When the channel matrix component corresponding to a normal channel matrix estimation training sequence followed by a payload portion is corrected for an error component by channel tracking, the channel matrix tracking means performs channel matrix estimation training for the first packet. An error component of a channel matrix resulting from at least one of a carrier frequency error or a sampling frequency error with the communication partner until the end of the payload portion after the series is obtained, and a final channel matrix in which the error component is corrected at the end is obtained. presume,
The wireless communication apparatus according to claim 22. When the channel matrix component corresponding to the normal channel matrix estimation training sequence followed by the payload portion is corrected for error components by channel tracking, the channel matrix tracking means performs the second from the end of reception of the first packet. In consideration of the frame interval until the start of packet transmission of the first packet, in the payload portion after the channel matrix estimation training sequence of the first packet, at the time when the transmission weight matrix is calculated by the time of transmission of the second packet An error component of a channel matrix resulting from at least one of a carrier frequency error or a sampling frequency error with the communication partner is obtained, and a final channel matrix in which the error component is corrected is estimated at the time point.
The wireless communication apparatus according to claim 22. For the channel matrix component corresponding to the extended training sequence, the channel matrix tracking means other than temporal variation of the channel transfer function due to at least one of a carrier frequency error and a sampling frequency error with the second wireless communication apparatus To correct the error component of
The wireless communication apparatus according to claim 22. When the variation of the channel matrix is large, the first channel matrix tracking means estimates a channel matrix obtained by removing a channel matrix component estimated from the extended training sequence.
The wireless communication apparatus according to claim 22. For the channel matrix component corresponding to the normal channel matrix estimation training sequence followed by the payload portion, the first channel matrix tracking means performs channel processing until reaching the end of the payload portion after the channel matrix estimation training sequence of the first packet.・ Estimate the channel matrix updated by tracking,
The wireless communication apparatus according to claim 27. The channel matrix tracking means considers the frame interval from the end of reception of the first packet to the start of transmission of the second packet for a channel matrix component corresponding to a normal channel matrix estimation training sequence followed by a payload portion. Then, the channel matrix updated by channel tracking until the time when the calculation of the transmission weight matrix is in time until the transmission of the second packet in the payload portion after the training sequence for channel matrix estimation of the first packet is obtained. presume,
The wireless communication apparatus according to claim 27. A wireless communication method comprising a plurality of antennas and wirelessly transmitting a packet with a communication partner, wherein the packet includes a channel matrix estimation training sequence and a payload,
A packet receiving step of receiving a packet from the communication partner;
A channel matrix estimation step of estimating a channel matrix using a training sequence for channel matrix estimation included in a packet received from the communication partner;
An error component of the channel matrix caused by at least one of a carrier frequency error or a sampling frequency error with the communication partner is obtained using a section other than the training sequence for channel matrix estimation of the packet from the communication partner, and the error component is corrected A channel matrix tracking step for estimating the channel matrix;
A transmission weight matrix generation step of generating a transmission weight matrix using a channel matrix estimation result by the channel matrix tracking means;
A packet transmission step of weighting and transmitting a packet to be transmitted to the communication partner thereafter with the transmission weight matrix;
A wireless communication method comprising:

Images