KR101176432B1 - System, method, and apparatus for wireless communication - Google Patents

Abstract

The disclosed invention efficiently realizes SVD-MIMO communication with a small number of singular value decomposition (SVD) processing of a high channel matrix of computational load. The receiver obtains the channel matrix H using the reference signal from the transmitter, obtains the transmission weight V of the downlink, and obtains the reception weight U ^H by the SVD of the channel matrix H. The receiver then transmits a reference signal weighted to U * to the transmitter, with the conjugate matrix U * of U being the uplink transmission weight. The transmitter receives the U * weighted reference signal and separates the received reference signal into a transmission weight V and a diagonal matrix D of the downlink based on the attributes of the unitary matrix.

Channel matrix, outlier decomposition, unitary matrix, transmit weight, receive weight, diagonal matrix, reference signal, uplink, downlink

Description

Wireless communication system, wireless communication device and wireless communication method {SYSTEM, METHOD, AND APPARATUS FOR WIRELESS COMMUNICATION}

The present invention includes the subject matter of Japanese Patent Application No. 2004-140485 filed with the Japan Patent Office on May 10, 2004, the entire contents of which are incorporated herein by reference.

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a wireless communication system, a wireless communication device, a wireless communication method, and a computer program for realizing broadband wireless transmission between a plurality of wireless stations, such as a wireless local area network (LAN). A wireless communication system in which transmission capacity is increased by pairing a transmitter and a receiver having a plurality of antennas and performing communication (Multi Input Multi Output (MIMO)) in which a plurality of logical channels are formed using spatial multiplication. A communication device, a method of wireless communication, and a computer program.

More specifically, the present invention relates to a wireless communication system, a wireless communication apparatus for performing a closed-loop MIMO transmission using singular value decomposition (SVD) of a channel matrix having transmission / reception antenna pairs as elements. It relates to a wireless communication method and a computer program, and more particularly, to a wireless communication system, a wireless communication device, and a wireless communication method for realizing SVD-MIMO communication with fewer times of singular value decomposition processing of a high channel matrix of computational load, and Relates to a computer program.

By computer networking represented by LAN, it is possible to efficiently share information resources and device resources. Here, a wireless LAN is attracting attention as a system for releasing a user from LAN wiring by a conventional wired system. According to the wireless LAN, most of the wired cable can be omitted in a work space such as an office and a communication terminal such as a personal computer (PC) can be moved relatively easily.

In recent years, the demand is remarkably increasing with the high speed and low price of a wireless LAN system. In particular, introduction of a personal area network (PAN) has been considered in order to establish a small-scale wireless network and perform information communication between a plurality of electronic devices existing in the human body. For example, different radio communication systems and radio communication apparatuses are defined by using a frequency band for which the supervisory authority does not need a license such as the 2.4 GHz band or the 5 GHz band.

Examples of standard standards for wireless networks include the Institute of Electrical and Electronics Engineers (IEEE) 802.11 (see, for example, Non Patent Literature 1), and HiperLAN / 2 (for example, Non Patent Literature 2 or Non Patent Literature). 3), IEEE 302.15.3, Bluetooth communication, and the like. As for the IEEE 802.11 standard, there are extension standards such as IEEE 802.11a (for example, refer to Non-Patent Document 4), 802.11b, and 802.11g due to a difference in a wireless communication method, a frequency band to be used, and the like.

The IEEE 802.11a standard supports a modulation scheme that achieves a maximum communication speed of 54 Mbps. However, there is a demand for a wireless standard capable of realizing a new high bit rate as a communication speed. For example, IEEE 802.11n is devising the next generation wireless LAN standard aiming at the development of a high speed wireless LAN technology exceeding 100MBPS in effective throughout.

MIMO communication is attracting attention as one of the technologies for realizing the high speed of wireless communication. This includes a plurality of antenna elements on both the transmitter side and the receiver side, and realizes a spatial multiplex transmission path (hereinafter referred to as a " MIMO channel ") to increase transmission capacity and improve communication speed. It is a technique to achieve. Since MIMO communication uses spatial multiplexing, it effectively uses frequency bands.

In the MIMO communication system, a transmitter distributes and transmits transmission data to a plurality of antennas, transmits the data using a plurality of virtual MIMO channels, and a receiver obtains received data by signal processing from signals received by the plurality of antennas. It is a communication method using the characteristics of and is different from a simple transmit and receive adaptive antenna array.

4 conceptually illustrates a MIMO communication system. As shown in the figure, each of the transceivers is equipped with a plurality of antennas. On the transmitting side, a plurality of transmission data are space / time encoded, multiplexed, distributed to M antennas, and transmitted to a plurality of MIMO channels. On the receiving side, the receiving signals received by the N antennas via the channel are space / time decoded. Receive data can be obtained. The channel model in this case is composed of the propagation environment (transfer function) around the transmitter, the structure of the channel space (transmission function), and the propagation environment (transfer function) around the receiver. When the signals transmitted from each antenna are multiplexed, cross talk occurs, but each signal multiplexed by the signal processing on the receiving side can be accurately extracted without cross talk.

Various methods have been proposed as a configuration method of MIMO transmission, but how to exchange channel information between transmission and reception according to the arrangement of antennas is a big problem for implementation.

In order to exchange channel information, it is easy to transmit existing information (preamble information) only from the transmitting side to the receiving side. In this case, the transmitter and the receiver perform spatial multiplexing independently of each other. It is called a transmission method. As a further development of this method, there is also a closed loop MIMO transmission method in which the preamble information is fed back from the reception side to the transmission side to create an ideal spatial orthogonal channel between transmission and reception.

As an open-loop MIMO transmission method, for example, V-BLAST (Vertical Bell Laboratories Layered Space Time) method is mentioned (for example, refer patent document 1). In the transmitting side, a signal is simply multiplexed for each antenna without being given an antenna weight coefficient matrix. In other words, the feedback procedure for obtaining the antenna weight coefficient matrix is omitted at all. Before transmitting the multiplexed signal, the transmitter inserts a training signal for performing channel estimation on the receiver side, for example, by time division for each antenna. In contrast, the receiver estimates the channel using the training signal in the channel estimating unit and calculates the channel information matrix H corresponding to each antenna pair. By appropriately combining zero-focusing and cancellation, the SN ratio is improved by utilizing the antenna degree of freedom caused by the cancellation, thereby increasing the accuracy of decoding.

As an ideal form of closed-loop MIMO transmission, an SVD-MIMO method using singular value decomposition (SVD) of a propagation path function is known (for example, non-patent literature). 5).

5 conceptually illustrates an SVD-MIMO transmission system. In SVD-MIMO transmission, a UDV ^H is obtained by singular value decomposition of a numerical matrix, that is, a channel information matrix H, using channel information corresponding to each antenna pair as an element, a V is given as an antenna weight coefficient matrix on the transmitting side, and received. U ^H is given as an antenna weight coefficient matrix on the side. Thus, each MIMO channel is represented as a diagonal matrix D having the square root of each singular value λ _{i in} the diagonal element, and can transmit multiplexed signals without crosstalk at all. In this case, both of the transmitter side and the receiver side can realize a plurality of logically independent transmission paths that are spatial division, that is, spatial orthogonal multiplexing.

According to the SVD-MIMO transmission method, theoretically, the maximum communication capacity can be achieved. For example, if the transceiver has two antennas, the maximum transmission capacity can be obtained twice.

Here, the mechanism of the SVD-MIMO transmission method will be described in detail. When the number of antennas of the transmitter is M, the transmission signal x is represented by a vector of M × 1. When the number of antennas of the receiver is N, the received signal y is represented by a vector of N × 1. In this case, the channel characteristic is expressed as a numeric matrix of N x M, that is, the channel matrix H. The element h _ij of the channel matrix H is a transfer function from the j th transmit antenna to the i th receive antenna. The received signal vector y is expressed by multiplying the transmission signal vector by the channel information matrix and adding the noise vector n as shown in the following equation (1).

[Equation 1]

As described above, the singular value decomposition of the channel information matrix H yields the following equation (2).

&Quot; (2) "

Here, the antenna weight coefficient matrix V on the transmitting side and the antenna weight matrix U on the receiving side are unitary matrices that satisfy the following expressions (3) and (4), respectively.

&Quot; (3) "

&Quot; (4) "

Specifically, the normalized eigenvectors of HH ^H are arranged in the antenna weight matrix U ^H of the receiving side, and the antenna weight matrix V of the transmitting side is arranged in the normalized eigenvector of H ^H H. In addition, D is a diagonal matrix having the square root of the singular value λ _i of H ^H H HH ^H or the diagonal elements. The magnitude | size is a smaller number among the transmission antenna number M and the reception antenna number N, and is a square matrix of the magnitude of min (M, N), and becomes a diagonal matrix.

[Equation 5]

In the above description, the singular value decomposition in the real number has been described, but there is a point to pay attention to the singular value decomposition in the case of extending to the imaginary number. U and V are matrices composed of eigenvectors. However, even when the eigenvector is manipulated such that the norm is 1, i.e., normalization, the eigenvectors are not singular, and there are numerous eigenvectors with different phases. Depending on the phase relationship between U and V, the above formula (2) may not hold. That is, although U and V are correct, respectively, they rotate arbitrarily by phase. In order to completely match the phase, V is usually obtained as an eigenvector of H ^H H. And U is multiplied by V from the right side to both sides of said Formula (2), and is calculated | required as follows.

&Quot; (6) "

When the transmitting side weights and transmits using the antenna weighting coefficient matrix V, and the receiving side weights and receives the antenna weighting coefficient matrix U ^H , since U and V are unitary matrices, U is N × min (M, N). And V is the following formula, which is Mxmin (M, N).

[Equation 7]

Here, the reception signal y and the transmission signal x are not vectors determined by the number of transmission antennas and reception antennas, but are (min (M, N) × 1) vectors.

Since D is a diagonal matrix, each transmission signal can be received without crosstalk. Since the amplitude of each independent MIMO channel is proportional to the square root of the singular value lambda, the magnitude of power of each MIMO channel is proportional to lambda.

Since the noise component n is also a eigenvector whose law is normalized to 1, U ^H n does not change its noise power. As the size, U ^H n becomes a (min (M, N)) vector and is the same size as y and x.

As described above, in the SVD-MIMO transmission, a plurality of logically independent MIMO channels can be obtained at the same frequency and at the same time and without crosstalk. In other words, it is possible to transmit a plurality of data by wireless communication using the same frequency at the same time, thereby improving the transmission speed.

In general, the number of MIMO channels obtained in the SVD-MIMO communication system corresponds to the smaller one of the number of transmission antennas M and the number of reception antennas N min [M, n]. The antenna weight coefficient matrix V on the transmission side is composed of transmission vectors vi for MIMO channels (V = [v ₁ , v ₂ , ..., v _{min [M, N]} ]). In addition, the number of elements of each transmission vector vi is the number of transmission antennas M. FIG.

In general, in the closed loop type MIMO scheme represented by SVD-MIMO, the transmitter side may calculate the optimal antenna weight coefficient for the antenna in consideration of the information of the propagation path. Further, by optimizing the coding rate and modulation scheme for the bit stream of each transmitting antenna, more ideal information transmission can be realized.

On the other hand, the introduction of the closed loop MIMO scheme as a real system also requires a problem that a large frequency of feedback from the receiver side to the transmitter side is required when the channel variation is large due to the movement of the transceiver. In the SVD-MIMO communication system, it is not easy to perform singular value decomposition in real time, and a setup procedure for transmitting the derived V or U ^H to the counterpart in advance is required.

Considering the amount of information in the antenna coefficient matrix V of the transmitting side, taking an example of IEEE 802.11a, orthogonal frequency division multiplexing (OFDM) communication scheme, which is one of LAN systems to which SVD-MIMO transmission is applied. Try it. If the number of transmission / reception antenna elements is three, the antenna coefficient matrix V on the transmitting side becomes a 3 × 3 matrix, and the number of elements is nine. If it is expressed by real number and complex number of 10-bit precision per element, and matrix V for 52 carriers is required, it is 9,360 bits (= 9 (number of elements of matrix) x 2 (complex real part, imaginary part) x) 10 (bits) x 52 (number of OFDM subcarriers) must be fed back from the receiver to the transmitter.

Here, a description will be given of the fact that it must be considered when configuring the actual SVD-MIMO transmission / reception system.

In the basic type of the SVD-MIMO transmission system, the receiver performs singular value decomposition of the acquired channel matrix H to obtain a weight vector U ^H for reception and a transmission weight vector V for use in the transmitter, and feeds this V back to the transmitter. . The transmitter uses this V as the transmission weight.

However, since the amount of information in the transmission weighting matrix V fed back to the transmitter is large, when the V information is extracted and sent, the orthogonal state between the MIMO channels is broken due to an error with the V information. It occurs.

Therefore, normally, after feeding back the transmission weight matrix V obtained at the receiver side to the transmitter side, the transmitter weights the reference signal using the matrix V and transmits it to the transmitter, and the receiver side acquires the channel matrix again. If the channel matrix is H, the receiver can obtain a channel matrix called HV from the reference signal transmitted by weighting V.

On the receiver side, the inverse of this HV is obtained and used as a weight for reception. Since H = UDV ^H , HV and its inverse become as follows.

[Equation 8]

It is only necessary to multiply the stream of each separated MIMO channel by an integer obtained from each diagonal element λ _i of the diagonal matrix D after using U ^H as a normal SVD-MIMO for the weight for reception.

On the transmission side, the matrix V is used as the transmission weight, and the receiver side uses the inverse matrix of the HV and the weight for reception is the same as that of the normal SVD-MIMO. none. Therefore, such a configuration can be actually employed.

[Patent Document 1] Japanese Patent Application Laid-Open No. 10-84324

[Non-Patent Document 1 International Standard ISO / IEC 8802-11: 1999 (E) ANSI / IEEE Std 802.11, 1999 Edition, Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications

[Non-Patent Document 2] ETSI Standard ETSI TS 101 761-1V 1.3.1 Broadband Radio Access Networks (BRAN); HIPERLAN Type 2; Data Link Control (DLC) Layer; Part 1: Basic Data Transport Functions

[Non-Patent Document 3] ETSI TS 101 761-2V 1.3.1 Broadband Radio Access Networks (BRAN); HIPERLAN Type 2; Data Link Control (DLC) Layer; Part 2: Radio Link Control (RLC) sublayer

[Non-Patent Document 4] Supplement to IEEE Standard for Information technology-Telecommunications and information exchange between systems-Local and metropolitan area networks-Specific requirements-Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) specifications: High- speed Physical Layer in the 5GHZ Band

[Non-Patent Document 5] http://radio3.ee.uec.ac.jp/MIMO(IEICE＿TS).pdf (as of October 24, 2003)

In order to perform SVS-MIMO communication, acquisition of a channel matrix or the like is required. On the other hand, in the conventional wireless communication system, while the collision is avoided by the CSMA / CA system, for example, in order to solve the concealed terminal problem, the acquisition of the transmission right according to the so-called RTS / CTS step is performed. Is done. Therefore, the acquisition of the channel matrix can be realized by using each packet such as RTS, CTS, DATA, ACK, etc. in the same control procedure as described below (see Fig. 6). However, before starting the RTS / CTS sequence, it is assumed that the transmitter has acquired the transmission weight V in advance.

(Step 1)

The transmitter sends an RTS packet to the receiver. The reference signal is added to the RTS packet.

(Step 2)

The receiver acquires the channel matrix H from the reference signal of the received RTS packet.

(Step 3)

The receiver determines from the acquired channel matrix H how many independent spatial channels can be used in which modulation scheme.

When receiving an RTS, there is a case where a receiver wants to determine a modulation scheme. For example, this is a case where a transmission operation of a neighboring station is to be stopped until the end of a packet with a network allocation vector (NAV) added to the CTS. In the case of setting the Short NAV, it is necessary to determine the modulation method and bit rate in the channel in order to calculate the time required for data transmission. In order to determine which modulation scheme to send, the receiver side needs to singularly decompose the channel matrix H and know each MIMO channel state, i.

(Step 4)

Return the CTS from the receiver side to the transmitter side. A reference signal for channel matrix estimation is added to the CTS.

(Step 5)

The transmitter acquires the reverse channel matrix H from the CTS reference signal sent from the receiver.

Then, when a calibration is performed to compensate for the difference in characteristics of the analog circuits belonging to each antenna of the transmitter and the analog circuits belonging to each antenna of the receiver, the forward and reverse transfer functions become equal. A calibration method for the difference in the characteristics of analog circuits in a transceiver is described in, for example, the Japanese Examined Patent Application Publication (JP-B) 2003-426294 specification already assigned to the present applicant.

(Step 6)

The transmitter performs singular value decomposition of the acquired reverse direction H to determine the forward transmission weight V. FIG. Of course, the forward transmission weight V obtained by singular value decomposition at the receiver side may be directly transmitted to the transmitter, but the amount of information is too large. Thus, the receiver sends a reference signal with a smaller amount of data in this way, and the transmitter acquires V as above.

(Step 7)

The transmitter transmits a data packet in response to receiving the CTS signal from the receiver. The V-weighted reference signal is added to the head of this data packet, and immediately after that, user data (paid load) is transmitted.

(Step 8)

The receiver acquires the channel matrix HV from the reference signal weighted by V, and receives the user data using the inverse matrix (see equation (8)) as a weight for reception.

In case of performing SVD-MIMO communication according to the above communication step, when performing one-time data transmission by the RTS / CTS sequence, the calculation of the singular value decomposition in total of step 3 to step 6 and the one in step 8 Since the inverse of a matrix must be calculated, there is a problem that the computational load on the transceiver becomes excessive. In particular, the calculation of singular value decomposition uses a large number of multipliers and the like, so that the power consumption at the time of execution is also large. Therefore, there is a demand to reduce the calculation amount of singular value decomposition even once.

This invention is made | formed in view of the above-mentioned technical subject, The main objective is the outstanding wireless communication which can expand transmission capacity by performing MIMO communication which formed the several logical channel using spatial multiplexing. A system, a wireless communication device and a wireless communication method, and a computer program are provided.

Another object of the present invention is to provide an excellent wireless communication system, a wireless communication device, and a wireless communication method, and a computer, which can efficiently perform closed-loop MIMO transmission by using singular value decomposition of a channel matrix having a transmit / receive antenna pair as an element. To provide a program.

Another object of the present invention is to provide an excellent wireless communication system, a wireless communication device and a wireless communication method, and a computer program which can efficiently realize SVD-MIMO communication with a small number of singular value decomposition processing of a high channel matrix of computational load. To provide.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problem, and a first aspect thereof is a wireless communication system for performing spatial multiplexing by determining transmission weights and reception weights using a channel matrix having transmission / reception antenna pairs as elements. When performing downlink data transmission from the communicator to the second communicator, the second communicator transmits a weighted reference signal with a transmission weight in the uplink direction, and the first communicator is based on the received reference signal. A radio communication system characterized by separating the transmission weight component in the downlink direction and the component in the diagonal matrix and performing data transmission using the obtained transmission weight component.

However, the term "system" as used herein refers to a logical collection of a plurality of devices (or functional modules for realizing a specific function), and it does not specifically ask whether each device or function module is in a single housing.

In the wireless communication system according to the present invention, for example, by using the MIMO communication method, it is possible to expand the transmission capacity by using a plurality of spatially multiple transmission paths, that is, the MIMO channel, and improve the communication speed. In this case, the transmitter and the receiver are each provided with a plurality of antennas, the transmitter distributes the transmission data to a plurality of streams and transmits a weighted stream from each transmit antenna, and the receiver weights the streams to each receive antenna. Received stream

Moreover, in the wireless communication system which concerns on this invention, the closed loop type MIMO communication system represented by SVD-MIMO transmission can be employ | adopted. In this case, the transmitter obtains an optimal transmission antenna weighting coefficient according to the feedback information from the receiver.

In a normal wireless communication system, while performing access control in accordance with CSMA / CA, the communication station acquires a transmission right in the RTS / CTS order. In this case, in order to perform SVD-MIMO communication, the reference signal for channel matrix acquisition is added to each packet of RTS, CTS, and DATA.

However, when performing SVD-MIMO communication according to such a communication sequence, each time the transmitter and the receiver receive each other's packets, the transmitter and the receiver calculate singular value decomposition or inverse matrix for obtaining a weight matrix based on the added reference signal. You must perform arithmetic processing. In particular, the calculation of singular value decomposition uses a large number of multipliers and the like, so that the power consumption at the time of execution is also large.

Thus, in the present invention, when the communicator receives the reference signal weighted with the transmission weight, the communicator uses the unitary matrix attribute as appropriate to separate the received signal into a reverse transmission weight component and a diagonal matrix component. did. As a result, the overall singular value decomposition can be reduced as a whole of the system, and unnecessary calculation amount can be reduced.

For example, the second communicator as a receiver in the SVD-MIMO transmission system obtains the channel matrix H using the reference signal added to the RTS packet from the first communicator as the transmitter, and further singularly resolves it. The transmit weight V and the receive weight U ^H of the downlink from the transmitter to the receiver can be obtained. The receiver then uses the conjugate matrix U * of U as the transmission weight of the uplink, and sends a reference signal weighted with U * to the CTS packet spatially multiplexed with U *.

On the other hand, when the transmitter receives a reference signal weighted by U *, the transmitter can separate the received signal into a reverse weight, that is, a weighting component V for transmission in the downlink and a diagonal matrix D. Specifically, upon receiving a reference signal weighted with U *, the channel matrix of the uplink is represented as a transpose matrix H ^T with respect to H (= UDV ^H ) of the channel matrix of the downlink, using H ^T U *. = V * DU ^T = Get V * D. The transmission weight V of the downlink is separated from V * D based on the attributes of the unitary matrix.

Therefore, the singular value decomposition for one time is reduced on the transmitter side, so that unnecessary calculation amount can be reduced.

Further, a second aspect of the present invention is a process for performing spatial multiplexing by determining transmission weights and reception weights by using a matrix UDV ^H obtained by singular value decomposition of a channel matrix H having transmission / reception antenna pairs as elements. A computer program coded computer readable to execute on a computer system, the method comprising: acquiring a downlink channel matrix H in accordance with a reference signal received from the communicating partner when performing downlink data communication from the communicating partner; A singular value decomposition step of performing singular value decomposition of the obtained channel matrix H into UDV ^H , and forwarded from the communication partner in a forward direction using a reception weight U ^H obtained by singular value decomposition from the obtained channel matrix H; A reception step of weighting and receiving user data and decomposing the singular values of the obtained channel matrix H Based on the results, the computer program comprising: a transmission step of using a conjugate matrix U * for U in the reverse direction of transmission weight to the communication partner, and transmits to the weighted reference signal.

In addition, a third aspect of the present invention is a process for performing spatial multiplexing by determining transmission weights and reception weights using a matrix UDV ^H obtained by singular value decomposition of a channel matrix H having transmission / reception antenna pairs as elements. A computer program coded computer-readable to execute on a computer system, wherein when performing downlink data communication with a communication partner, the reference signal weighted by U * as a transmission weight of U as a conjugate weight matrix U of the uplink is used. A receiving step of receiving and obtaining H ^T U * = V * DU ^T = V * D, a separating step of separating the downlink transmission weight V from V * D based on the attributes of the unitary matrix, and V And a transmitting step of performing a weighted transmission of the downlink using the computer program.

A computer program according to each of the second and third aspects of the present invention defines a computer program that is computer readable coded to realize predetermined processing on a computer system. In other words, by installing the computer programs according to the second and third aspects of the present invention in a computer system, a cooperative action is exerted on the computer system, and the communication devices which perform data transmission in the downlink respectively operate. By activating a plurality of such communication apparatuses and establishing a wireless network, the same effects as those of the wireless communication system according to the first aspect of the present invention can be obtained.

According to the present invention, there is provided an excellent wireless communication system, a wireless communication device and a wireless communication method, and a computer program, which can efficiently perform closed-loop MIMO transmission using singular value decomposition of a channel matrix having a transmit / receive antenna pair as an element. Can provide.

Further, according to the present invention, an excellent radio communication system, radio communication apparatus and radio communication method, and computer program capable of efficiently realizing SVD-MIMO communication with a small number of singular value decomposition processing of a high channel matrix of a computational load in a small number of times. Can be provided.

According to the present invention, when the communicator receives a reference signal weighted with a transmission weight from a communication partner, the communicator uses the unitary matrix attribute appropriately, thereby transmitting the received signal in a reverse transmission weight component and a diagonal matrix component. By separating by, it is possible to obtain the transmission weight of the local station without calculating the singular value decomposition.

Other objects, features and advantages of the present invention will become apparent in accordance with the following detailed description of the embodiments of the present invention or the accompanying drawings.

1 is a diagram showing the configuration of a radio communication apparatus according to an embodiment of the present invention.

2 is a diagram schematically showing a functional configuration of the channel characteristic acquisition and transmission weight separator 37.

3 is a diagram illustrating an operation for performing MIMO communication in the order of RTS / CTS.

4 is a diagram conceptually illustrating a MIMO communication system.

5 is a diagram conceptually illustrating an SVD-MIMO transmission system.

6 is a diagram illustrating an operation for performing bidirectional MIMO communication in the order of RTS / CTS.

EMBODIMENT OF THE INVENTION Hereinafter, the Example of this invention is described in detail, referring drawings.

The present invention relates to a MIMO communication system in which a transmitter having a plurality of antennas and a receiver having a plurality of antennas are paired and spatially multiplexed to communicate.

In a normal wireless communication system, while performing access control in accordance with CSMA / CA, the communication station acquires a transmission right in the RTS / CTS order. In this case, in order to perform SVD-MIMO communication, the reference signal for channel matrix acquisition is added to each packet of RTS, CTS, and DATA.

However, when performing SVD-MIMO communication according to such a communication sequence, each time the transmitter and the receiver receive each other's packets, the transmitter and the receiver calculate singular value decomposition or inverse matrix for obtaining a weight matrix based on the added reference signal. You must perform arithmetic processing. In particular, the calculation of singular value decomposition uses a large number of multipliers and the like, so that the power consumption at the time of execution is also large.

Thus, in the present invention, when the communicator receives a reference signal weighted with transmission weights, the communicator divides the received signal into transmission weight components and diagonal matrix components in the opposite direction by appropriately using the unitary matrix attribute. I made it. As a result, the overall singular value decomposition can be reduced as a whole of the system, and unnecessary calculation amount can be reduced.

1 shows a configuration of a radio communication apparatus according to an embodiment of the present invention.

The radio communication apparatus shown in FIG. 1 is provided with two transmission / reception antennas 11a and 11b, and can perform data transmission by SVD-MIMO system. That is, at the time of transmission, a transmission antenna weighting factor is assigned to each transmission signal to be multiplexed, space / time encoded, distributed to two antennas 11a and 11b, and transmitted to the channel. On the receiving side, two antennas (via a channel) Receive antenna weight coefficients are given to the multiplexed signals received by 11a and 11b) and space / time decoded to obtain received data. However, the gist of the present invention is not limited to two antennas, but may be three or more.

Each transmission / reception antenna 11a and 11b is connected in parallel via a switch 12a and 12b, respectively, and transmits a signal over a predetermined frequency channel to another radio communication device line, or Collect signals from other wireless communication devices. However, the switches 12a and 12b exclusively connect the transmission / reception antennas 11a and 11b with one of the transmission system or the reception system, and it is assumed that the transmission and reception cannot be performed in parallel.

Each transmission system includes a modulation coding unit 21, a transmission weight multiplier 22, an IFFT 23, a preamble / reference granting unit 24, a D / A converter 25, and a transmission analog processing unit 26. Equipped with.

The modulation encoding unit 21 encodes the transmission data sent from the upper layer of the communication protocol into an error correction code, and simultaneously maps the transmission signal in the signal space by a predetermined modulation scheme such as BPSK, QPSK, or 16 QAM. At this point, the existing data series may be inserted into the modulation symbol series as pilot symbols according to the pilot symbol insertion pattern and timing. A pilot signal composed of an existing pattern is inserted for each subcarrier or at intervals of several subcarriers.

The transmission weight multiplier 22 multiplies the encoded transmission signal by the transmission weight matrix V to obtain a plurality of MIMO channels by spatial multiplexing. The transmission weight matrix V is constructed based on the feedback information sent from the communication partner, and is set in the transmission weight multiplier 22.

The IFFT 23 converts the modulated serial format signal into parallel data for several parallel carriers according to the number and timing of parallel carriers, and then inverse Fourier transforms for the FFT size according to a predetermined FFT size and timing. Is done. Here, in order to remove the inter-symbol interference, a guard interval period may be provided between every two OFDM symbols. The time width of the guard interval is determined by the situation of the propagation path, i.e., the maximum delay time of the delay wave which affects the demodulation. The parallel data is converted into a serial signal, which is converted into a transmission signal on the time axis while maintaining mutual orthogonality of carriers on the frequency axis. The preamble / reference providing unit 24 adds a preamble signal and a reference signal to the head of transmission signals such as RTS, CTS, and DATA packets.

When the radio communication apparatus operates as the transmitting side of the downlink transmission, the preamble / reference granting unit 24 multiplies V by the V as a transmission weight as a reference signal sent to the receiving side on the downlink.

In addition, when the radio communication apparatus operates as the receiving side of the downlink transmission, the preamble / reference granting unit 24 multiplies the reference signal sent to the transmitting side on the uplink by U *, and multiplies the conjugate matrix U * of U. Used as the transmission weight of the uplink. In this case, the channel characteristic acquisition and transmission weight separator 37 acquires the channel matrix H using the reference signal sent from the downlink transmission side, and also singularly decomposes it, thereby downlinking from the transmitter to the receiver. The transmission weight V and the reception weight U ^H are obtained, and U * is given to the preamble / reference granting unit 24 as the transmission weight for the reference signal. The details of the operation of the channel characteristic acquisition and transmission weight separation section 37 will be described later.

This transmission signal is converted into an analog baseband signal by the D / A converter 25 and upconverted to the RF frequency band by the transmission analog processing unit 26, and then each MIMO channel from the antenna 11 is converted. Is sent out.

On the other hand, each receiving system includes a reception analog processing unit 31, an A / D converter 32, a synchronization acquisition unit 33, an FFT 34, a reception weight multiplier 35, and a demodulation decoding unit 36. And a channel characteristic acquisition and transmission weight separation unit 37.

The signal received from the antenna 11 is down-converted from the RF frequency band to the baseband signal by the receiving analog processing unit 31 and converted into a digital signal by the A / D converter 32.

Subsequently, the received signal as serial data is converted into parallel data according to the synchronization timing detected by the synchronization acquisition unit 33, and arranged (in this case, one OFDM symbol including up to a guard interval is collected). The FFT 34 performs Fourier transform on a signal corresponding to the effective symbol length, and extracts a signal of each subcarrier to convert a signal on the time axis into a signal on the frequency axis.

In the channel characteristic acquisition and transmission weight separator 37, first, a channel matrix H is obtained using a reference signal weighted for each signal multiplexed by a communication partner. The channel matrix H can be singularly resolved to obtain the transmission weighting matrix V, the reception weighting matrix U ^H and the diagonal matrix D. When the reference signal is sent from the communication counterpart at predetermined intervals, the channel characteristic acquisition and transmission weight separator 37 updates each time with a new channel matrix H, and decomposes this outlier.

The reception weight matrix U ^H obtained by singular value decomposition of the channel matrix is set in the reception weight multiplier 35 of the own apparatus, and the transmission weight matrix V is fed back to the communication counterpart. However, as the number of credit weighting matrix U is not ^H, it is also possible to use D ^_ U ^H as the inverse matrix of HV (see the above expression (8)). In addition, the transmission weight matrix V obtained is given to the transmission weight multiplier 22.

In addition, when the radio communication apparatus operates as the receiving side of the downlink transmission, the channel characteristic acquisition and transmission weight separator 37 transmits the conjugate matrix U * of U obtained by singular value decomposition of the channel matrix H for the reference signal. The weight is given to the preamble / reference granting unit 24 as a weight.

In addition, the channel characteristic acquisition and transmission weight separator 37 can obtain the transmission weight V without performing singular value decomposition. That is, when the wireless communication apparatus operates as a transmitting side of the downlink transmission, when the uplink receives a reference signal weighted to U * from the receiving side, the uplink channel matrix is H (=) of the downlink channel matrix. Using what is represented as the transpose matrix H ^T for UDV ^H ), H ^T U * = V * DU ^T = V * D is obtained. The transmission weight V of the downlink is separated from V * D based on the attributes of the unitary matrix. The transmission weight matrix V thus obtained is given to the transmission weight multiplier 22.

The details of the operation of the channel characteristic acquisition and the transmission weight separation unit 37 are yielded later.

The reception weight multiplier 35 spatially separates the spatial multiplexed reception signal by multiplying the reception weight matrix U ^H or D ^_ U ^H obtained by singular value decomposition of the channel matrix H with the reception signal.

The weighted received signal is further demodulated and decoded by the demodulation decoding unit 36 to demap the received signal in the signal space by a predetermined method such as BPSK, QPSK, 16 QAM, and at the same time, correct and decode the received signal to be received data. Passed to upper layer of protocol.

Here, in the case of performing the access operation according to the CSMA / CA communication method according to the RTS / CTS step, the wireless communication device operating as a transceiver adds a reference signal for channel matrix acquisition to each packet of RTS, CTS, and DATA, A transmission / reception weight component is obtained using the reference signal. In general, the weight matrix can be obtained by singular value decomposition of the channel matrix. However, since the calculation of singular value decomposition uses a large number of multipliers and the like, there is a problem that the power consumption at the time of execution is large. In the present embodiment, the channel characteristic acquisition and transmission weight separation unit 37 can acquire the weight component without performing singular value decomposition.

In FIG. 2, the function structure of the channel characteristic acquisition and transmission weight separation part 37 is shown typically. As shown, the channel characteristic acquisition and transmission weight separation unit 37 is constituted by a channel matrix acquisition unit, an outlier decomposition unit, and a transmission weight separation unit.

The channel matrix acquisition unit obtains the channel matrix H by using a reference signal weighted discretely for each signal transmitted by the communication partner. The singular value decomposition unit performs singular value decomposition of the obtained channel matrix H to obtain a transmission weighting matrix V, a reception weighting matrix U ^H, and a diagonal matrix D. The obtained reception weight matrix U ^H is set in the reception weight multiplier 35 of the own apparatus, and the transmission weight matrix V is fed back to the communication counterpart. However, as the number of credit weighting matrix U is not ^H, it is also possible to use D ^_ U ^H as the inverse matrix of HV (see the above expression (8)). Moreover, the conjugate matrix U * of U obtained by singular value decomposition of the channel matrix H is given to the preamble / reference granting unit 24 as the transmission weight for the reference signal.

When the reference signal is transmitted from the communication counterpart at predetermined intervals, the channel characteristic acquisition and transmission weight separation unit 37 updates the new channel matrix H each time, and singular value decomposition is performed.

When the transmission weight separation unit receives the U * weighted reference signal on the uplink from the receiving side, it indicates that the uplink channel matrix is represented as a transpose matrix H ^T with respect to H (= UDV ^H ) of the downlink channel matrix. To obtain H ^T U * = V * DU ^T = V * D. The transmission weight V of the downlink is separated from V * D based on the attributes of the unitary matrix. The transmission weight matrix V thus obtained is given to the transmission weight multiplier 22.

Here, the arithmetic processing for separating the transmission weight V from the reference signal weighted by U * received in the uplink from the reception side in the transmission weight separation unit will be described. In the following description, however, the downlink from the transmitter to the receiver is in the forward direction, and the uplink from the receiver to the transmitter is in the reverse direction.

In this case, the forward channel matrix is expressed by the following equation. However, the number of antennas of the transmitter is n and the number of antennas of the receiver is m. In addition, the transfer function from the j-th antenna of the transmitter to the i-th antenna of the receiver is represented by h _ij .

[Equation 9]

In contrast, the reverse channel matrix is expressed as follows.

&Quot; (10) "

In other words, the reverse channel matrix is expressed as the transpose matrix H ^T of the forward channel matrix H.

Using the reference signal from the transmitter, the forward channel matrix H is obtained at the receiver side, and the singular value decomposition is as follows.

&Quot; (11) "

Therefore, the channel matrix in the reverse direction from the receiver to the transmitter becomes as follows.

[Equation 12]

The receiver transmits to the reference signal transmitted in the uplink to the transmitter by using the conjugate matrix U * of the matrix U obtained by singular value decomposition of the forward channel matrix H as the transmission weight as the transmission weight.

In this case, the transmitter receives the reference signal over U * in the reverse channel matrix H ^T. Therefore, the transmitter can obtain the following equation.

&Quot; (13) "

If only V can be separated from V * D, then outlier decomposition is unnecessary on the transmitter side. The method of separating V and D from V * D will be described later.

V is a unitary matrix and is composed of singular value vectors for the number of MIMO channels (= min [n, m]). Each eigenvector u is normalized in Law 1.

&Quot; (14) "

Therefore, the integrated UD = V * D obtained at the receiver side is obtained by multiplying the singular value by the elements of the eigenvectors in each column, as shown in the following equation.

&Quot; (15) "

Here, the law of the elements of the matrix is calculated for each column of V * D. Since the magnitude of the law for each column consists of the diagonal elements of the matrix D, V * can be separated.

From the transmitter side, V is transmitted as a reference signal and user data is transmitted immediately after. On the receiver side, HV is obtained from the reference signal multiplied by V. User data is received as the inverse of the HV as a weight for reception (see Expression (8)).

By using the method described above, by performing singular value decomposition once and inverse matrix operation once, the same result as that of two singular value decomposition and one retrograde column operation can be obtained.

An operation step for performing MIMO communication such as RTS, CTS, and DATA in the wireless communication system according to the present embodiment will be described below with reference to FIG. 3. However, the process for acquiring the transmission weight matrix V at the transmitter side is assumed to be performed in advance.

(Step 1)

From the transmitter, send a reference signal to the receiver. At the receiver, the channel matrix H can be obtained.

(Step 2)

At the receiver, outlier decomposition of the channel matrix H (see equation (10)) is obtained, and the matrix U is obtained.

(Step 3)

From the receiver to the transmitter, transmit a reference signal weighted to U *.

(Step 4)

In the transmitter, V * DU ^T U * = V * D is obtained from the reference signal weighted by U * (see equation (13)).

(Step 5)

These are separated by calculating the law of each column of the acquired V * D. As a result, D can be obtained and V can also be obtained without performing singular value decomposition.

(Step 6)

From the transmitter side, the V-weighted reference signal is transmitted and immediately after that, user data is transmitted.

(Step 7)

At the receiver side, the channel matrix HV = UD is obtained by receiving the reference signal weighted to V from the transmitter side.

(Step 8)

The receiver calculates the inverse of the channel matrix UD and maintains this as a weight for reception.

(Step 9)

The user weight transmitted from the transmitter is decoded using the reception weight obtained in step (8).

In the above, this invention was demonstrated in detail, referring a specific Example. However, it will be apparent to those skilled in the art that modifications and substitutions of the above embodiments can be made without departing from the spirit of the invention.

The present invention can be applied to various wireless communication systems that perform data transmission by spatial multiplexing. That is, the gist of the present invention and its application are not necessarily limited to the spatial division, that is, the spatial orthogonal multiplex transmission scheme, such as the SVD-MIMO scheme. The present invention can be preferably applied to other similar communication systems that transmit and receive weighted bills according to the channel matrix H obtained from the propagation path situation.

In short, the present invention has been disclosed in the form of illustration, and the description of the present specification should not be limitedly interpreted. In order to determine the gist of the present invention, reference should be made to the appended claims.

Claims (11)

A wireless communication system for performing spatial multiplexing by determining transmission weights and reception weights by using a channel matrix including transmission / reception antenna pairs as elements. In the case of performing data transfer from a first communication device to a second communication device, The second communicator transmits a reference signal weighted with a transmission weight in a direction from the second communicator to the first communicator, The first communicator divides the received reference signal into a transmission weight and a diagonal matrix in a direction from the first communicator to the second communicator, and performs data transmission using the transmission weight obtained by the separation. Wireless communication system. The method of claim 1, The wireless communication system performs spatial multiplexing by determining transmission weights and reception weights by using a matrix UDV ^H obtained by singular value decomposition of a channel matrix H having transmission and reception antenna pairs as elements. The second communicator obtains a channel matrix H using the reference signal from the first communicator, and transmits a weight V from the first communicator as a transmitter to the second communicator as a receiver by outlier decomposition of the channel matrix H. And obtain a reception weight U ^H , and transmit the reference signal weighted by the conjugate matrix U * of U in the direction from the second communicator to the first communicator as the transmission weight in the direction from the second communicator to the first communicator, When the first communicator receives the reference signal weighted with U *, the first communicator uses the attributes of a unitary matrix to transmit the received signal V in the direction from the first communicator to the second communicator, and the diagonal matrix D. Wireless communication system characterized in that separated by. 3. The method of claim 2, The first communicator obtains H ^T U * = V * DU ^T = V * D by receiving a reference signal weighted U * in the direction from the second communicator to the first communicator, and the channel matrix in the direction Is a transpose matrix H ^T for the channel matrix H (= UDV ^H ) in the direction from the first communicator to the second communicator, wherein the first communicator is in the direction from the first communicator to the second communicator based on the attributes of the unitary matrix. And the transmission weight V is separated from V * D. A radio communication apparatus for performing spatial multiplexing by determining transmission weights and reception weights by using a matrix UDV ^H obtained by singular value decomposition of a channel matrix H having transmission / reception antenna pairs as elements, When performing data communication from the communication partner in the direction from the first communication device to the second communication device, A channel matrix acquiring unit for acquiring a channel matrix H in a direction from a first communicator to a second communicator in accordance with a reference signal received from the communication partner; An outlier decomposition unit that performs outlier decomposition of the obtained channel matrix with respect to UDV ^H ; A receiving unit for receiving the user data transmitted in the forward direction from the communication counterpart and weighting the user data using the reception weight U ^H obtained from the obtained channel matrix H through the singular value decomposition; And A transmitter that weights a reference signal using a transmission weight of the reverse direction to the communication counterpart by using the obtained conjugate matrix U * based on the result of the singular value decomposition of the obtained channel matrix H Wireless communication device comprising a. A wireless communication apparatus for performing spatial multiplexing by determining transmission weights and reception weights by using a channel matrix including transmission / reception antenna pairs as elements, A receiver which receives a reference signal weighted with a transmission weight; A separator for separating the received reference signal into a transmission weight V and a diagonal matrix D in a reverse direction; And Transmitter for transmitting data weighted by the transmission weight obtained by the separator in the reverse direction of the direction in which the reference signal is received Wireless communication device comprising a. The method of claim 5, When performing data communication in the direction from the first communication device to the second communication device to the communication partner, The receiving unit receives a reference signal weighted by the conjugate matrix U * of U as a transmission weight in the direction from the second communicator to the first communicator, The separator separates the received signal into a transmission weight V and a diagonal matrix D in the direction from the first communicator to the second communicator, And the transmitting unit transmits data weighted by V in the direction from the first communication device to the second communication device. The method of claim 6, The channel matrix in the direction from the second communicator to the first communicator is a transpose matrix H ^T of H (= UDV ^H ) of the channel matrix in the direction from the first communicator to the second communicator, The receiving unit obtains H ^T U * = V * DU ^T = V * D by receiving a reference signal weighted with U * in the direction from the second communicator to the first communicator, And the separator separates the transmission weight V in the direction from the first communicator to the second communicator from V * D based on the attributes of the unitary matrix. A wireless communication method for performing spatial multiplexing by determining transmission weights and reception weights by using a matrix UDV ^H obtained by singular value decomposition of a channel matrix H having transmission / reception antenna pairs as elements, In the case of performing data transfer from the communication partner in the direction from the first communicator to the second communicator, Acquiring a channel matrix H in a direction from a first communicator to a second communicator based on the reference signal received from the communication counterpart; A singular value decomposition step of performing singular value decomposition on the obtained channel matrix H into UDV ^H ; Receiving user data transmitted in the forward direction from the communication counterpart and weighting using the received weight U ^H obtained by singular value decomposition from the obtained channel matrix H; And Based on the obtained singular value decomposition of the channel matrix H, weighting a reference signal using a transmission weight of the reverse direction to the communication counterpart by using the conjugate matrix U * of U and transmitting the reference signal Wireless communication method comprising a. A wireless communication method for performing spatial multiplexing by determining transmission weights and reception weights by using a matrix UDV ^H obtained by singular value decomposition of a channel matrix H having transmission / reception antenna pairs as elements, In the case of performing data transfer from the first communicator to the second communicator to the communication partner, Receiving a weighted reference signal with the conjugate matrix U * of U as the transmission weight in the direction from the second communicator to the first communicator, and obtaining H ^T U * = V * DU ^T = V * D; Separating the transmission weight V in the direction from the first communicator to the second communicator from V * D based on the attributes of the unitary matrix; And A transmission step of weighting and transmitting data using a transmission weight V in the direction from the first communicator to the second communicator Wireless communication method comprising a. delete delete

Images