JP2012015963A - Terminal device, base station device and radio communication system using them - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make it possible to adequately synthesize signals received at entire antennas to enable an excellent reception property to be obtained, when THP MU-MIMO transmission targeting a virtual reception antenna formed based on a plurality of the reception antennas included by respective terminals is performed.SOLUTION: A terminal device 200 generates propagation path information of a virtual reception antenna, the number of which is less than the number of reception antennas, based on propagation path information observed at a plurality of the reception antennas and feeds it back as CSI to a base station device. The base station device performs THP MU-MIMO transmission based on the CSI. The terminal device 200 obtains a desired signal by MMSE synthesizing between the signals transmitted in this way and the signals received at a plurality of the reception antennas.

Description

  The present invention relates to a terminal device, a base station device, and a radio communication system using them, which transmit a spatially multiplexed signal from a base station having a plurality of transmitting antennas to a terminal device having a plurality of receiving antennas in MIMO transmission or the like About.

  In order to solve the tightness of frequency resources accompanying the increase in data traffic in cellular systems, as a technology to realize high frequency utilization efficiency and high-speed transmission, multiple transmission signals using multiple transmission antennas provided in the base station Active research is being carried out on downlink MIMO (Multiple-Input Multiple-Output) transmission that spatially multiplexes, network MIMO that transmits multiple transmission signals from multiple base stations and transmission, and multi-antenna transmission technology called CoMP. It has been broken. Among these multi-antenna transmission technologies, Multi User-MIMO (MU-MIMO), which transmits spatially multiplexed transmission signals addressed to a plurality of terminals and transmits the signals at the same time, is provided on the base station side even when each terminal has a small number of receiving antennas. It is possible to perform transmission using a transmission antenna or a base station of an adjacent cell effectively, and is attracting attention as a technique for improving frequency utilization efficiency.

  In MU-MIMO transmission, a signal addressed to a plurality of terminals is transmitted using the same resource regardless of whether it is performed within a cell or between cells, so that the reception side of each terminal does not interfere with each other in advance. It is necessary to perform transmission with precoding. Usually, since this precoding is performed based on the channel condition on the terminal side, in the FDD (Frequency Division Duplex) system, the channel is measured on the terminal side, and the measurement result (CSI: Channel State Information). Is fed back to the transmitting side.

  Here, although each terminal is provided with a plurality of receiving antennas, when transmitting transmission signals of one stream (also referred to as a rank) to each terminal, each terminal receives one receiving antenna. If the CSI can be ascertained on the transmission side, transmission with spatial multiplexing can be performed. For this reason, each terminal only needs to feed back the CSI for one receiving antenna to the transmitting side. As a feedback method, one of the plurality of receiving antennas provided in each terminal is determined in advance. A method of feeding back CSI observed by a receiving antenna, a method of feeding back the best channel among the channels observed by a plurality of receiving antennas as CSI, and a virtual one reception by a plurality of receiving antennas There is a method in which an antenna is formed and CSI in the virtual antenna is fed back.

Among these methods, there is a method described in Non-Patent Document 1, for example, as a method of forming one virtual receiving antenna with a plurality of receiving antennas. In Non-Patent Document 1, singular value decomposition (SVD) is performed on a channel matrix observed by a plurality of receiving antennas in a terminal, and a complex conjugate of a left singular vector corresponding to the maximum singular value is obtained. It is described that a result obtained by multiplying a channel matrix by a transposed vector is fed back as CSI. Here, when a propagation path matrix observed by a plurality of receiving antennas of the m-th terminal is H _m and SVD is applied to the propagation path matrix H _m , the following expression (1) is obtained.

Here, Σ _m represents a diagonal matrix having a positive real number component, and U _m and V _m each represents a unitary matrix. At this time, H _m ^′ indicating CSI fed back from the m-th terminal to the transmission side is expressed by the following equation (2).

However, subscripts in parentheses U _m is a row component of the left U _m, the right represents respectively the components in the column of U _m, which means that extract all the notation rows or columns of all . Therefore, in Equation (2), the vector composed of the first column of U _m is complex conjugate transposed and multiplied by the propagation path matrix H _m . The multiplication result is a vector of one row, and is fed back to the transmission side as CSI when a virtual one receiving antenna is formed using a plurality of receiving antennas.

  On the transmitting side where CSI in such a virtual receiving antenna is fed back from each terminal, they are used to perform precoding and transmission so that signals destined for each terminal do not interfere with each other. This precoding method is roughly classified into linear precoding that multiplies a plurality of transmission signals by linear weights and non-linear precoding that sequentially subtracts a known interference signal from the transmission signals and then multiplies linear weights. Although precoding is degraded in characteristics as compared with nonlinear precoding, spatial multiplexing of a plurality of signals can be realized by a very simple process.

  A signal transmitted with linear precoding on the transmission side is received by a plurality of receiving antennas of each terminal, but a desired signal of each terminal is one stream at a time. It is necessary to demodulate after synthesizing the received signals. Such signal synthesis may be performed using the complex conjugate transposed vector of the left singular vector corresponding to the maximum singular value used when forming the virtual receiving antenna, or MMSE (Minimum Mean Square) Error) A configuration may be adopted in which reception weights calculated based on a criterion are used.

Here, assuming that a matrix representing linear precoding performed on the transmission side is P, the MMSE reception weight W _m is expressed as follows. Here, ξ is a ratio of average noise power to signal power, and I represents a unit matrix. By multiplying the reception signal by such a reception weight and synthesizing it, the mean square error between the desired signal and the reception signal can be minimized.

  As described above, when the terminal includes a plurality of receiving antennas, the channel matrix is multiplied by the complex conjugate transposed vector of the left singular vector corresponding to the maximum singular value obtained by performing SVD on the channel matrix. Thus, a virtual antenna (maximum beam) that becomes a very good propagation path can be formed, and spatial multiplexing can be performed with high power efficiency by using CSI in this virtual antenna. Also, since the terminal only needs to feed back CSI for one receiving antenna, the amount of feedback can be reduced.

"Weighted CSI Feedback aided DL CoMPtransmissions," 3GPP TSG-RAN WG1 # 58bis R1-093782, Oct. 2009.

  When linear precoding is performed on the transmission side, the mean square error between the desired signal and the received signal is minimized by synthesizing signals received by a plurality of antennas using the reception weight shown in Equation (3). And good reception characteristics can be obtained. Such synthesis is processing on the receiving side, not only for signals spatially multiplexed by the above-described linear precoding, but also for signals spatially multiplexed by precoding including nonlinear processing (nonlinear precoding). However, it is expected that the reception characteristics of each terminal that receives a spatially multiplexed signal by nonlinear precoding can be improved.

  However, when spatial multiplexing using THP (Tomlinson Harashima Precoding), which is one of the typical non-linear precoding, is performed, signals received by a plurality of receiving antennas of each terminal are expressed by Equation (3). Even if the received weights are combined, it is not possible to perform appropriate combining, and there is a problem that reception characteristics deteriorate. Therefore, in the conventional technique targeted for spatial multiplexing by linear precoding, MU-MIMO in which a spatially multiplexed signal by non-linear precoding is transmitted from a base station and reception signals at a plurality of reception antennas are appropriately combined at a terminal side. The system cannot be realized.

  In view of such circumstances, the present invention forms a virtual receiving antenna from a plurality of receiving antennas of each terminal in a system that simultaneously transmits transmission signals addressed to a plurality of terminals by nonlinear spatial multiplexing. When performing spatial multiplexing for receiving antennas, it is possible to appropriately synthesize signals received by all antennas, and use terminal devices, base station devices, and the like that can obtain good reception characteristics. A wireless communication system is to be provided.

The present invention communicates with a base station apparatus that includes N receiving antennas when N is an integer of 2 or more, and includes a plurality of transmitting antennas and spatially multiplexes signals destined for a plurality of terminal apparatuses using a nonlinear operation. A terminal device,
Based on information about propagation paths observed by the N receiving antennas, information about propagation paths when received by virtual reception antennas less than N is calculated and received by the virtual reception antennas. Information signals that have been spatially multiplexed in the base station apparatus using the information on the propagation path in this case and the information on the propagation paths of other terminal apparatuses are received by the N reception antennas, and the N receptions The information signal received by the antenna is multiplied by a reception weight and combined, and a non-linear operation is performed on the combined information signal to detect a desired information signal.

  The terminal apparatus according to the present invention receives a pilot signal transmitted from the base station apparatus, and calculates the reception weight based on the received pilot signal.

  Further, the terminal apparatus of the present invention receives information related to a coefficient representing interference in spatial multiplexing using the non-linear operation transmitted from the base station apparatus, and receives the reception based on information related to a coefficient representing the interference. The weight is calculated.

  In the terminal apparatus of the present invention, the reception weight is a weight that minimizes a mean square error between the information signal transmitted from the base station and the information signal after the reception weight multiplication. And

  In addition, the terminal device of the present invention obtains information regarding a propagation path when received by the virtual reception antenna by performing singular value decomposition on a matrix representing the propagation path observed by the N reception antennas. The obtained vector or matrix is obtained by multiplying a matrix representing a propagation path observed by the N receiving antennas.

  The terminal apparatus according to the present invention is characterized in that the base station apparatus is notified of information related to a propagation path when received by the virtual receiving antenna.

  The terminal device according to the present invention is characterized in that the non-linear operation is a modulo operation.

In addition, the present invention includes a first terminal device that includes a plurality of transmission antennas and includes N reception antennas when N is an integer equal to or greater than 2. Reception of all terminal devices to be communicated A base station apparatus that performs communication by performing spatial multiplexing on information signals addressed to a plurality of terminal apparatuses, the total of which is greater than the number used for transmission among the transmission antennas,
Information on a propagation path when received by a virtual receiving antenna of less than N, generated based on a propagation path observed in the first terminal apparatus, and information on a propagation path of another terminal apparatus And calculating a coefficient representing transmission weight and interference in spatial multiplexing using a non-linear operation, and spatially multiplexing information signals addressed to a plurality of terminal devices using the transmission weight and the coefficient representing interference. And

  The base station apparatus according to the present invention is characterized in that the transmission weight is multiplied by a pilot signal for channel estimation and transmitted.

  In addition, the base station apparatus of the present invention notifies the terminal apparatus of information related to a coefficient representing the interference.

  The base station apparatus according to the present invention is characterized in that the non-linear operation is a modulo operation.

  Moreover, the present invention is a wireless communication system comprising the terminal device and the base station device.

  By using the present invention, in a system for simultaneously transmitting transmission signals addressed to a plurality of terminals by nonlinear spatial multiplexing, a virtual reception antenna is formed from a plurality of reception antennas of each terminal, and the virtual reception antenna is When performing the target spatial multiplexing, it is possible to appropriately combine signals received by all antennas, and to obtain good reception characteristics.

It is a figure which shows the radio | wireless communications system of MU-MIMO by this invention. It is a block diagram which shows the base station in 1st embodiment. It is a figure which shows the flame | frame of the multiplexed signal produced | generated in 1st embodiment. It is a block diagram which shows the terminal in 1st embodiment. It is a block diagram which shows the terminal in 3rd embodiment. It is a figure which shows the flame | frame of the multiplexed signal produced | generated in 4th embodiment. It is a block diagram which shows the terminal in 4th embodiment.

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

  First, FIG. 1 shows a MU-MIMO system according to the present invention. As shown in FIG. 1, in the MU-MIMO system according to the present invention, a base station 100 having a plurality of transmitting antennas performs nonlinear processing on different signals addressed to terminals 200a, 200b, 200c, and 200d having a plurality of receiving antennas. The transmission is performed using the same resource with spatial multiplexing included. However, the total number of reception antennas of all terminals 200a to 200d to be subjected to spatial multiplexing is larger than the number of transmission antennas included in the base station 100. In the base station 100, virtual transmission is performed from a plurality of reception antennas included in the terminals 200a to 200d. A reception antenna is formed, and the virtual reception antenna is set as a target of nonlinear spatial multiplexing, and transmission processing is performed so that inter-user interference with respect to a signal received by the virtual reception antenna is suppressed. In addition, although the number of terminals is four in FIG. 1, it is not restricted to this. FIG. 1 shows an example of a MU-MIMO system in which signals addressed to a plurality of terminals are transmitted from one transmission source. However, the present invention is not limited to this, and a network in which spatial multiplexing is performed by a plurality of transmission sources. The present invention is also applicable to MU-MIMO. The present invention is applicable not only to base stations and terminals in cellular systems, but also to transmitters and receivers in wireless LAN systems without distinction such as base stations and terminals.

  In the example shown in FIG. 1, each terminal has two reception antennas, an antenna numbered 30-1 and an antenna numbered 30-2. It is assumed that one virtual receiving antenna is formed based on the spatial multiplexing for the virtual receiving antenna. As described above, here, one stream is transmitted from the base station 100 to each terminal. However, the present invention is not limited to this. For example, the terminal may have three or more receiving antennas. In such a case, a configuration may be adopted in which two or more streams are transmitted to one terminal. However, it is assumed that the number of streams transmitted to each terminal is less than the number of reception antennas that each terminal has, and the virtual reception antennas for the number of streams are formed by a plurality of antennas that each terminal has. Further, it is not necessary for all terminals to be spatially multiplexed to have a plurality of antennas, and terminals having only one receiving antenna may be mixed. Furthermore, the number of transmission antennas provided in the base station is not limited to four, and it is sufficient that a plurality of transmission antennas are provided. Also, depending on the system, not all of the plurality of transmission antennas provided in the base station are always used for transmission, and some transmission antennas are selected and used in the base station. May not be sent. In such a case, the present invention targets a situation where the total number of receiving antennas of all terminals subject to spatial multiplexing is larger than the number of transmitting antennas (number of selected transmitting antennas) used for transmission in the base station. It is said.

  In such a MU-MIMO system, in the present invention, a signal subjected to spatial multiplexing using THP based on CSI in a virtual receiving antenna formed from a plurality of receiving antennas is received by each receiving antenna. In the case of reception, a method for appropriately combining these received signals will be shown, and a communication system that can obtain good reception characteristics and is excellent in frequency utilization efficiency will be provided.

(First embodiment)
In the first embodiment of the present invention, first, transmission processing of MU-MIMO (THP MU-MIMO) transmission using THP will be described, and then, when the THP MU-MIMO transmission is performed, Processing for combining signals received by a plurality of antennas included in the terminal will be described.

  First, transmission processing of THP MU-MIMO transmission targeted in the first embodiment will be described using the configuration of the base station. A base station configuration in the present embodiment is shown in FIG. As shown in FIG. 2, the base station 100 in this embodiment includes an upper layer 10, a modulation unit 11, a P / S unit 12, an interference subtraction unit 13, a modulo unit 14, an interference generation unit 15, an S / P unit 16, Transmission weight multipliers 17, 23, signal multiplexer 18, D / A unit 19, radio units 20, 27, transmission antenna unit 21, pilot signal generator 22, transmission weight / interference coefficient calculator 24, receiver 25, A / D section 26 and receiving antenna section 28.

When THP MU-MIMO transmission is performed in the base station 100, first, information about a propagation path (CSI: Channel State Information) fed back from each terminal to be spatially multiplexed is received by the reception antenna unit 28, and the radio unit 27, the received signal is converted to a frequency that can be A / D converted, and the A / D unit 26 converts the analog signal into a digital signal, and then the receiving unit 25 performs demodulation and the like to obtain a propagation path matrix H. However, the CSI fed back from each terminal 200a to 200d is a propagation path between the transmission antenna of the base station 100 and each terminal 200a to 200d. Here, the CSI is formed based on two reception antennas of each terminal. It is assumed that the propagation path is one virtual receiving antenna. As a method for forming this virtual receiving antenna, several methods are conceivable. Here, as an example, the left corresponding to the maximum singular value obtained by applying SVD to the channel matrix observed at the terminal is used. A method of forming one virtual receiving antenna by multiplying a propagation path matrix by a complex conjugate transposed vector of a singular vector shall be used. In such a case, the CSI fed back from each terminal to the base station is expressed by Equation (2), and the CSI fed back from each terminal is summarized in a matrix form at the base station is H ′ = [H ₁ the ^{_{'T H 2' T H 3}} 'T H 4' T] T. Here, H _m ^′ is a 1 × 4 vector representing a propagation path between the virtual antenna of the terminal m and the four antennas of the base station. In addition, any transmission method may be used as an uplink transmission method from each terminal to the base station. In FIG. 2, only one reception antenna is provided, but a plurality of reception antennas are provided. The MU-MIMO transmission may be performed.

  The propagation path matrix obtained in this way is input to the transmission weight / interference coefficient calculation unit 24, and transmission weights and interference coefficients necessary for performing THP MU-MIMO transmission are calculated. These calculation methods will be described later. The interference coefficient is information regarding the interference signal, and an interference signal can be generated based on the interference coefficient. The transmission weight calculated by the transmission weight / interference coefficient calculation unit 24 is input to the transmission weight multiplication units 17 and 23, and the interference coefficient is input to the upper layer 10 and the interference generation unit 15.

  In the upper layer 10, individual data to be transmitted to the terminals 200 a to 200 d is generated and input to the modulation unit 11. Here, since the interference coefficient input from the transmission weight / interference coefficient calculation unit 24 is notified to each terminal, the interference coefficient is converted into digital information and input to the modulation unit 11. As will be described later, since the interference coefficient in the present embodiment is represented by a lower triangular matrix, it is not necessary to notify the terminal of all elements of the lower triangular matrix, and values other than zero, that is, lower triangular elements. Only need to be notified. In addition, since it is necessary to notify each terminal of the coefficient for normalizing the power, the same processing as the interference coefficient is performed. The modulation unit 11 modulates the input digital data and generates a modulated signal such as QPSK or 16QAM. Here, modulation is performed on the four data transmitted to the four terminals, the interference coefficient and the power normalization coefficient that are commonly notified to the four terminals.

  The signal modulated by the modulation unit 11 is input to the P / S unit 12 and the signal multiplexing unit 18. Here, the signal input to the P / S unit 12 is a signal obtained by modulating individual data destined for each terminal that is spatially multiplexed and transmitted, and the signal that is input to the signal multiplexing unit 18 is not spatially multiplexed. This is a signal obtained by modulating an interference coefficient or the like notified to all terminals. The P / S unit 12 to which the data signal addressed to each terminal is input performs parallel-serial conversion. However, in the THP MU-MIMO targeted by the present invention, the interference subtraction unit 13 subtracts the interference signal for each terminal. Since it is performed sequentially, the signals addressed to the four terminals are rearranged so that they are output in order for each modulation signal. Here, it is assumed that modulated signals are output from P / S section 12 in the order of signals addressed to terminals 200a, 200b, 200c, and 200d.

  The interference subtracting unit 13 sequentially performs a process of subtracting the interference signal from the desired signal with respect to the signals output in this way. This is a process of subtracting the interference signal input from the interference generation unit 15 from the input modulation signal, and the interference signal here refers to inter-user interference in THP MU-MIMO transmission. . Specifically, the interference signal is generated based on the modulation signal output before the modulation signal to be subjected to the subtraction process as the desired signal among the signals sequentially output by the P / S unit 12. For example, when a modulated signal addressed to the terminal 200b is processed as a desired signal, an interference signal to be subtracted is generated based on a signal addressed to the terminal 200a processed before the signal addressed to the terminal 200b. Is done. When the modulation signal addressed to the terminal 200c is processed as the desired signal, the interference signal to be subtracted is generated based on the signals addressed to the terminal 200a and the terminal 200b. However, since there is no interference to be subtracted for the signal addressed to the terminal 200a that is first input from the P / S unit 12, the input signal is output as it is for the signal addressed to the terminal 200a.

The signal processed by the interference subtracting unit 13 is input to the modulo unit 14 and subjected to a modulo calculation. The modulo operation is expressed by the following equation, and the output signal f _τ (z) is within the range of [−τ / 2, τ / 2] from the origin, regardless of the value of the input signal z. This is a non-linear calculation that adjusts by adding an appropriate signal to the input signal. However, z is a complex number, j is an imaginary unit, and τ is a real constant determined according to the modulation method. Specifically, when the average power of the modulation symbol is normalized to 1, τ = 2√2 for QPSK, τ = 8 / √10 for 16QAM, and τ = 16 / √42 for 64QAM. Further, floor (x) represents the maximum integer not exceeding x.

  Here, the signal added to the input signal in equation (4) is called a perturbation vector, and the modulo operation is an operation of adding a value (perturbation vector) that is an integral multiple of τ to each of the in-phase component and quadrature component of the input signal. It can also be said that. By using such a modulo operation, it becomes possible to keep the amplitude of the signal within a certain range, so that the signal power that is increased by subtracting the interference signal from the desired signal in the interference subtraction unit 13 is suppressed. A signal obtained by subtracting interference between users in advance can be transmitted while satisfying the prescribed transmission power. However, as described above, since the subtraction of interference is not performed on the signal addressed to the terminal 200a, the amplitude is always within a certain range without adding the perturbation vector, and the modulo calculation is performed. There is no need. Further, a modulo operation may be performed, but the same signal as the input signal is output.

  Such a modulo operation is performed, and a signal output from the modulo unit 14 is input to the interference generation unit 15 and the S / P unit 16. The interference generation unit 15 multiplies the signals sequentially input from the modulo unit 14 by an appropriate value among the interference coefficients input from the transmission weight / interference coefficient calculation unit 24, and subtracts the desired signal from the interference subtraction unit 13. To generate a power interference signal. Here, when the interference coefficient is represented by a matrix, the following lower triangular matrix is obtained, and each element is multiplied by a signal sequentially input from the modulo unit 14 to generate an interference signal.

F _qr shown in the equation (5) represents a coefficient of interference that the signal addressed to the terminal r has on the terminal q, and the subtraction processing in the interference subtraction unit 13 is performed using the modulation signal addressed to the terminal q as the desired signal. In other words, each element in the q-th row of the matrix F is multiplied by a signal from the terminal 200a to the terminal 200-q-1 (a signal output from the modulo unit 14) to generate an interference signal. Here, all the elements in the first row of the matrix F representing the interference coefficient are all zero, which means that there is no interference signal to be subtracted from the signal addressed to the terminal 200a as described above. Show. Also, here, in order to briefly explain the basic process of subtracting the interference signal from the desired signal to suppress inter-user interference, the multiplication result of the interference coefficient and the output signal of the modulo unit 14 is calculated from the desired signal. Although described as subtracting, depending on the calculation method of the matrix F, the sign of each element may be calculated as already including the subtraction process. In such a case, in order to perform appropriate interference suppression processing in the interference subtracting unit 13, the calculated matrix F is inverted in sign and then multiplied with the output signal of the modulo unit 14, or The interference subtraction unit 13 needs to perform addition processing instead of subtraction processing.

  By repeatedly performing the processes such as subtraction of interference signals, suppression of signal power increase by modulo calculation, and generation of interference signals for each signal destined for each terminal, inter-user interference is appropriately suppressed and power efficiency is improved. An excellent transmission signal can be generated.

  The signal in which the inter-user interference is suppressed in this manner is then input to the S / P unit 16, where serial-parallel conversion is performed, and signals destined for each terminal are output in parallel. The transmission weight multiplication unit 17 multiplies the transmission signal by the transmission weight, and the result is input to the signal multiplexing unit 18. However, a signal to be transmitted from the antenna 21-i is input to the signal multiplexing unit 18-i. The signal multiplexing unit 18 multiplexes an individual data signal addressed to each terminal, a pilot signal for propagation path estimation, a signal representing an interference coefficient that is commonly notified to each terminal, and the like. Here, the interference coefficient includes a coefficient used for power normalization. In this embodiment, these signals are multiplexed in time. Specifically, this is processing for generating a frame as shown in FIG.

  Here, the pilot signal for propagation path estimation is a signal known between the transmitter and the receiver, and is used for demodulation of the desired signal on the receiving side. In the present embodiment, as this pilot signal, a pilot signal (signal numbered in FIG. 3) for demodulating a signal representing an interference coefficient that is commonly notified to each terminal, and each spatially multiplexed terminal Two types of pilot signals (signals numbered 1 to 4 in FIG. 3) for demodulating individual data signals addressed thereto are prepared, and multiplexed in the signal multiplexing unit 18 with data signals or the like. However, as shown in FIG. 2, in the present embodiment, the interference coefficient notification signal (information on the interference signal) is transmitted only from the antenna 21-1, and the pilot signal is similarly transmitted from the antenna 21-1. Sent only from Also, four pilot signals for demodulating individual data signals are prepared because the pilot signals transmitted from the transmission antennas are temporally orthogonalized so as not to interfere with each other. The signal is input to the signal multiplexing unit 18 after the transmission weight multiplication unit 23 multiplies the known transmission signal generated by the pilot signal generation unit 22 by the same transmission weight as that multiplied by the data signal.

  The transmission frame as shown in FIG. 3 thus generated in the signal multiplexing unit 18 is converted from a digital signal to an analog signal in the D / A unit 19 and converted into a frequency that can be wirelessly transmitted in the radio unit 20. Are transmitted from the transmission antenna units 21-1 to 21-4, respectively. With such a configuration, when each terminal uses a reception method for combining signals received by a plurality of receiving antennas by notifying each terminal of a common interference coefficient, the signals can be appropriately combined in those terminals. Transmission processing can be performed as possible.

Here, the calculation method of the transmission weight and the interference coefficient in the THP MU-MIMO transmission targeted in this embodiment will be briefly described. However, although there are several types of transmission weight generation methods and the like in THP MU-MIMO transmission, an example in which transmission weights and the like are generated on the basis of MMSE is shown here. First, the transmission weight multiplied by the signal in the transmission weight multipliers 17 and 23 is P, the coefficient for normalizing the transmission power to a specified value is β, the transmission signal vector before the transmission weight multiplication (the output of the S / P unit 16) ) Is v, the covariance matrix of the transmission signal vector v is R _v , and the predetermined total transmission power is E _TX , P, F, and β are calculated by the following equation (6).

  However, y = argminX represents obtaining y that minimizes X, and ε is a value representing an error between the transmission signal and the reception signal expressed by the following equation (7).

Therefore, Expression (6) means that P, F, and β that minimize the mean square error between the transmission signal and the reception signal are obtained. However, when obtaining this solution, there is a constraint condition shown in the lower part of Expression (6), and the first expression of the constraint condition indicates the definition of transmission power. In addition, the second expression of the constraint condition indicates that the matrix F is a lower triangular matrix, and m is 1 to 4 in the present embodiment because it is an example of spatial multiplexing of 4 terminals. Here, S _m = S _{(m, 4-m)} = [I _m , 0 _{m × (4-m)} ], which means that the 1st to m-th elements in the m-th column of the matrix F are zero. Is a matrix. Moreover, e _m denotes the m-th column vector of matrix.

  When P, F, and β that minimize the mean square error between the transmission signal and the reception signal under such constraint conditions are obtained, the respective solutions are expressed as follows. Where ξ represents the average noise power to signal power ratio (the reciprocal of SNR, or can be paraphrased as noise variance when signal power is normalized to 1). Here, the noise power at each terminal is expressed as Although ξ is represented by a scalar as being the same, ξ may be represented by a diagonal matrix when the noise power at each terminal is different.

  Interference between users is generated by multiplying the interference coefficient matrix F as shown in Equation (9) by the transmission signal addressed to each terminal in the interference generation unit 15, and the modulo operation is performed by subtracting the interference between users. By multiplying the signal by the transmission weight P shown in Expression (8) in the transmission weight multipliers 17 and 23, it is possible to realize the MMSE-standard THP MU-MIMO transmission targeted in this embodiment. It becomes. However, as described above, the interference subtraction unit 13 has described that the multiplication result of the interference coefficient and the output signal of the modulo unit 14 is subtracted from the desired signal. However, in the interference coefficient F shown in Equation (9), Since the positive / negative sign including the subtraction process has already been added, in order to perform an appropriate interference suppression process in the interference subtraction unit 13, the positive / negative of the calculated matrix F is inverted, and then the modulo unit 14 It is necessary to perform multiplication with the output signal or to perform addition processing in the interference subtraction unit 13 instead of the subtraction processing.

  The device configuration of each terminal that receives a signal that has been spatially multiplexed in the base station as described above is shown below. The device configuration of terminal 200 in the present embodiment is shown in FIG. As shown in FIG. 4, terminal 200 in the present embodiment includes receiving antenna sections 30-1 and 30-2, radio sections 31-1 and 43, A / D sections 32-1 and 32, and signal separation section 33-. 1, 2, propagation path estimation units 34-1 and 2, propagation path compensation unit 35, MMSE synthesis unit 36, modulo unit 37, demodulation unit 38, upper layer 39, virtual antenna formation unit 40, transmission unit 41, D / A Part 42 and a transmission antenna part 44.

  In this terminal 200, the signal transmitted from the base station 100 shown in FIG. 2 is first received by the receiving antenna units 30-1 and 30-2, and the received signals are converted to frequencies that can be A / D converted by the radio units 31-1 and 31-2. After conversion, the A / D units 32-1 and 32-1 convert the analog signals into digital signals, and the received signals are input to the signal separation units 33-1 and 33-1. In the signal demultiplexing units 33-1 and 33-2, the base station 100 demultiplexes the temporally multiplexed signals as shown in FIG. Specifically, the pilot signal (0 to 4 in FIG. 3) is separated into the propagation path estimation units 34-1 and 3-4, the interference coefficient notification signal is separated into the propagation path compensation unit 35, and the data signal is separated into the MMSE combining unit 36. Enter. However, in the present embodiment, the interference coefficient notification signal that is not spatially multiplexed is acquired by demodulating only the signal received by the receiving antenna unit 30-1, and the signal separation unit 33-2 acquires the pilot signal 0. And the interference coefficient notification signal shall be discarded.

Based on the pilot signals input from the signal separators 33-1 and 33-2, the propagation path estimators 34-1 and 34-2 estimate the propagation path. However, the propagation path estimation unit 34-1 performs propagation path estimation using the pilot signals 0 to 4, and the propagation path estimation unit 34-2 performs propagation path estimation using the pilot signals 1 to 4. At this time, since the pilot signal 0 is not multiplied by the transmission weight, the propagation path estimated using the pilot signal 0 is the transmission antenna unit 21-1 of the base station and the reception antenna unit 30-1 of each terminal. The estimation result is input to the propagation path compensation unit 35. Since pilot signals 1 to 4 are multiplied by transmission weights, the propagation path estimated by propagation path estimation unit 34-n of terminal m is H _mn P, and this estimation result is input to MMSE combining unit 36. Is done. Here, H _mn represents a propagation path estimated by the antenna n of the terminal m.

  The propagation path compensation unit 35 performs propagation path compensation of the interference coefficient notification signal, and the propagation path compensated interference coefficient notification signal is input to the demodulation unit 38. Then, the demodulation unit 38 demodulates the interference coefficient notification signal, and the demodulated interference coefficient notification signal is input to the upper layer 39. In the upper layer 39, the interference coefficient converted into digital information in the base station 100 is reproduced and input to the MMSE combining unit 36.

The MMSE combiner 36 receives the spatially multiplexed data signal from the signal separator 33, the estimated propagation path H _mn P from the propagation path estimator 34, and the interference coefficient from the upper layer 39. Thus, the signals received by the two receiving antennas are combined. Specifically, the received signal vector y _m = [y _m1 received by the two antennas of the terminal m. y _m2 ] ^T is first divided by the power normalization coefficient β used in the base station, and then multiplied by a reception MMSE weight expressed by the following equation to perform synthesis. This reception weight minimizes the mean square error between the signal received at each terminal and the desired signal.

_However, it is ^{_{H m = [H m1 T H}} m2 T] T.

  A signal that is multiplied by such reception weights and combined becomes a vector, and since the m-th signal is a desired signal of terminal m, terminal m inputs the m-th signal to modulo unit 37, After performing the modulo operation represented by the same equation (4) as that performed at the base station, the demodulator 38 demodulates the desired signal. The demodulated signal is output to the upper layer 39.

  Also, in terminal apparatus 200 in the present embodiment, CSI needs to be fed back to base station 100. The fed back CSI is information about the propagation path estimated by propagation path estimation units 34-1 and 34-2, respectively. This is CSI for one receiving antenna that is input to the antenna forming unit 40 and obtained by combining the two propagation path information in the virtual antenna forming unit 40 (vector of 1 row × 4 columns). As described above, as a method for synthesizing the two propagation path information, the complex conjugate of the left singular vector corresponding to the maximum singular value obtained by applying SVD to the propagation path matrix obtained from the two propagation path information. Assume that a method of multiplying a transposed vector by a propagation path matrix is used. By performing such combining, it is only necessary to feed back the combined CSI, that is, the CSI for one receiving antenna, and the base station has a single antenna having a very good propagation path condition. Seems to exist. The virtual antenna forming unit 40 performs SVD on the propagation path matrix obtained from the two propagation path information, and multiplies the propagation path matrix by the complex conjugate transposed vector of the left singular vector corresponding to the maximum singular value obtained there. This multiplication result becomes CSI in the virtual antenna. The combined CSI is input to the transmission unit 41 and transmitted from the transmission antenna unit 44 to the base station via the D / A unit 42 and the radio unit 43. Therefore, as described above, in this embodiment, what is subject to spatial multiplexing in the base station is one virtual antenna formed based on a plurality of antennas possessed by each terminal. . However, in the present embodiment, CSI measurement is performed using pilot signals transmitted from antennas so as to be orthogonal to each other without being multiplied by a transmission weight, and is a frame having a configuration different from that shown in FIG. It shall be measured in

  By configuring the terminal as described above, a virtual antenna can be formed, and CSI in the virtual antenna can be fed back to the base station, and the signal received by the two antennas can be expressed by the equation (11). Therefore, it is possible to synthesize desired signals while reducing inter-user interference, so that reception is better than when demodulating a signal received by one antenna. Characteristics can be obtained.

  Here, in the present embodiment, transmission weights and interference coefficients used in the base station are calculated based on the MMSE standard. However, the present invention is not limited to this, and may be performed using a ZF (Zero Forcing) standard. In addition, calculation using QR decomposition may be performed. However, when QR decomposition is used, it is necessary to divide the received signal by the gain represented by the diagonal component of the triangular matrix obtained by QR decomposition. Therefore, the received pilot signal is multiplied by the vector used to form the virtual antenna. Therefore, it is necessary to estimate this diagonal component. Then, after the reception signal is divided by the gain, the reception signal is synthesized using the reception weight shown in Expression (11).

  Furthermore, the reception weights shown in Expression (11) are described in a matrix format, and the combined signal after reception weight multiplication is a vector, but as described above, the desired signal of each terminal is a combined signal vector. Therefore, it is not always necessary for each terminal to multiply the reception weight in the matrix format shown in Equation (11), and from any row component in the matrix shown in Equation (11). A configured vector may be used as a reception weight. However, in this case, each terminal needs to use a necessary row component in order to appropriately synthesize each desired signal, and in the m-th terminal where the interference signal subtraction process is performed, The terminal m needs to use the m-th line of Equation (11) as a reception weight.

  In this embodiment, an example in which single carrier transmission is used has been described. However, the present invention can also be applied to a system that performs multicarrier transmission. When applied to a multicarrier transmission system, calculation of transmission weights and interference coefficients may be performed for each subcarrier, or may be performed in units in which several subcarriers are grouped. In this embodiment, pilot signals transmitted from each antenna are processed so as to be orthogonal in time. However, in a multicarrier transmission system, pilot signals are arranged on different subcarriers and orthogonalized in the frequency domain. May be transmitted. In such a multi-carrier transmission system, an IFFT (Inverse Fast Fourier Transform) unit, an FFT unit, a GI (Guard Interval) insertion unit, and the like are required for a base station and a terminal.

  In the present embodiment, the interference coefficient is once converted into digital information, and the information is modulated in the same manner as the data signal and notified to the terminal. However, the interference coefficient may be notified as an analog signal. . In such a case, a signal in which each element of the matrix F has an amplitude is transmitted from the base station as an interference coefficient notification signal.

  Furthermore, in this embodiment, the pilot signal for channel estimation used for demodulation of the interference coefficient notification signal is transmitted from only one antenna. However, the configuration is such that the interference coefficient notification signal can be demodulated by all terminals. If there is, it is not limited to this.

  In this embodiment, error correction coding is not performed on the data signal to be transmitted. However, the present invention is not limited to this, and the base station may perform error correction coding and perform decoding at the terminal. Good.

(Second embodiment)
In the first embodiment, the interference subtraction process in the interference subtraction unit 13 of the base station is performed for four terminals in order from the terminal 200a1, but this subtraction process does not necessarily have to be performed in a fixed order. It is good also as a structure which changes a process order appropriately according to. For example, the desired signal addressed to the terminal 200c is first processed, and then the interference signal generated from the desired signal addressed to the terminal 200c is subtracted from the desired signal addressed to the terminal 200a. Further, an interference signal generated from the desired signals addressed to the terminals 200a and 200c is subtracted from the desired signal addressed to the terminal 200d, and finally, from the desired signals addressed to the terminal 200b, the desired signals addressed to all other terminals. This means that processing such as subtracting the generated interference signal is performed. By appropriately determining the order of such processing according to the propagation path, it becomes possible to improve power efficiency and obtain good reception characteristics as compared with the case where processing is performed in a predetermined order. be able to. Such a processing order is also called ordering, and in the present embodiment, a case will be described in which ordering is adaptively changed.

  First, a base station configuration when performing adaptive ordering as a target in this embodiment will be described. The base station in the case of performing adaptive ordering targeted in the present embodiment can be realized with the same configuration as the base station 100 shown in FIG. 2, but the ordering described above is performed for each terminal. In the transmission weight / interference coefficient calculation unit 24 that has received the CSI from, it is performed in units of frames as shown in FIG. There are several methods for ordering, such as a method called BLAST, in which the order of processing is determined according to the SNR of each terminal. In the present invention, the ordering method is not limited, and a multicarrier transmission system is used. When applying to the above, ordering may be performed for each subcarrier. The transmission weight / interference coefficient calculation unit 24 calculates transmission weights and interference coefficients in consideration of ordering. However, when processing in a predetermined ordering order is performed as in the first embodiment, interference is calculated. While the matrix F representing the coefficients was calculated as a lower triangular matrix, the matrix F when adaptive ordering is performed is represented as follows. However, the ordering order shows an example in which the terminal 200c, the terminal 200a, the terminal 200d, and the terminal 200b are used as described above.

F _qr shown in Expression (12) represents the coefficient of interference that the signal addressed to the terminal r has on the terminal q, as in the first embodiment.

  The matrix F representing the interference coefficient calculated in this way is input to the upper layer 10 and the interference generation unit 15, but the matrix F is not necessarily a simple lower triangular matrix as shown in the first embodiment. In the upper layer 10, it is necessary to generate an interference coefficient notification signal so that each terminal can correctly reconstruct the matrix F. In the first embodiment, only elements other than zero of the matrix F are sent to the terminal as interference coefficient notification signals, but can be realized by sending all elements including zero.

However, when all elements including zero are transmitted as an interference coefficient notification signal, the amount of information increases and transmission efficiency decreases. Therefore, the interference coefficient notification signal includes not only elements other than zero of the matrix F. It can be efficiently realized by including information indicating how many non-zero elements are included in which row of the matrix. This is because, when the matrix F as shown in the equation (12) is calculated, in addition to each non-zero element [f ₁₃ f ₂₃ f ₂₁ f ₂₄ f ₄₃ f ₄₁ ], [1 3 0 2 ], The number of elements other than zero in each line may be notified to each terminal. That is, this is equivalent to notifying each terminal of the ordering order. Each terminal that has received such an interference coefficient notification signal can correctly reconstruct a matrix F representing the interference coefficient as shown in Expression (12) and use it for calculating the reception weight as shown in Expression (11). .

  In the upper layer 10 of the base station, rearrangement of individual data to be transmitted to each terminal is also performed in the ordering order. As described in the first embodiment, when THP MU-MIMO transmission is performed, it is necessary to sequentially perform a process of subtracting an interference signal from a desired signal, and this order matches the ordering order. Therefore, the higher layer 10 that receives the matrix F from the transmission weight / interference coefficient calculation unit 24 determines the ordering order, and performs a process of rearranging data so that appropriate interference subtraction is performed. When the matrix F is expressed by Expression (12), the rearrangement is performed so that the interference signals are subtracted in the order of the terminal 200c, the terminal 200a, the terminal 200d, and the terminal 200b. However, this rearrangement process can also be performed by notifying the P / S unit 12 of information relating to ordering.

  As described above, by generating an interference coefficient notification signal in consideration of ordering and performing appropriate subtraction of interference, the base station can be configured even when the ordering order is adaptively changed. The other blocks, for example, the transmission weight multiplying units 17 and 23 and the signal multiplexing unit 18 may have the same configuration as that shown in the first embodiment.

Next, the device configuration of the terminal in the present embodiment will be described. The terminal in the present embodiment can also be realized with the same configuration (FIG. 4) as in the first embodiment, similarly to the configuration of the base station. However, as described above, since the matrix F is not necessarily a lower triangular matrix, in order to appropriately combine the signals received by the spatial multiplexing target antenna and the auxiliary antenna using the reception weight shown in Equation (11). Therefore, it is necessary to appropriately reconstruct the matrix F based on the interference coefficient notification signal notified from the base station in consideration of the order of ordering. In this embodiment, it is assumed that this reconfiguration is performed in the upper layer 39 in FIG. Further, in Equation (11), the matrix F is subtracted from the unit matrix I _{4 × 4.} However, when adaptive ordering in the present embodiment is performed, this unit matrix is also rearranged in accordance with the ordering. Need to do. Specifically, the rearrangement is performed so that each element of 1 in the unit matrix is to the right of each element other than zero in the matrix F. For example, when the matrix F is represented by Expression (12), I _{4 × 4} used in Expression (11) is represented as follows.

  Such rearrangement is performed in the upper layer 39 similarly to the reconstruction of the matrix F, and is input to the MMSE synthesis unit 36 together with the reconstructed matrix F. However, the rearrangement of the unit matrix may be performed on the rows or the columns. Then, the MMSE combining unit 36 generates a reception weight represented by Equation (10) and multiplies the signals received by the two antennas, respectively, thereby combining the desired signal components while reducing the inter-user interference. Is possible. Other blocks may be configured to perform the same processing as that shown in the first embodiment. By adopting such a terminal configuration, even when adaptive ordering is performed, It becomes possible to synthesize signals received by antennas, and it is possible to obtain better reception characteristics as compared with the case of demodulating signals received by one antenna.

(Third embodiment)
In the embodiments so far, one virtual antenna is formed based on a plurality of receiving antennas of each terminal, and spatial multiplexing is performed in a system that performs THP MU-MIMO transmission in a base station using CSI in the virtual antenna. In this example, received signals are received by a plurality of receiving antennas, and the received signals are combined by MMSE combining. In this embodiment, unlike these examples, an example will be described in which a terminal receives signals using a plurality of virtual antennas, and MMSE-combines the signals received by these virtual antennas. Here, the base station apparatus in the present embodiment may have the same configuration as the base station apparatus in the first or second embodiment.

FIG. 5 shows the device configuration of the terminal in such a case. As illustrated in FIG. 5, the terminal device according to the present embodiment can be realized with a configuration in which a virtual antenna receiving unit 50 is added to the terminal device illustrated in FIG. 4. The virtual antenna receiving unit 50 multiplies the signals received by the two receiving antennas by weights, and generates a signal as if it was received by the two virtual antennas. Specifically, when the virtual antenna forming unit 40 calculates the CSI to be fed back, SVD is performed on the channel matrix to obtain a singular vector, but the obtained singular vector is notified to the virtual antenna receiving unit 50. The received signal received by the two antennas is multiplied by a singular vector. However, in the virtual antenna forming unit 40, the CSI is generated by multiplying the channel matrix by the complex conjugate transposed vector of only the left singular vector corresponding to the maximum singular value. However, in the virtual antenna receiving unit 50, the maximum singular value is generated. Using the singular vector corresponding to the minimum singular value, the received signal is multiplied by the complex conjugate transpose matrix of the matrix composed of these two vectors. This is the received signal vector received by the two antennas of the terminal m as y _m = [y _m1 When y _m2 ] ^T , the received signal vector y _m ^′ = [y _m1 ^′ y _m2 ^′ ] ^T received by the virtual antenna (which is the output of the virtual antenna receiving unit 50) is expressed by the following equation: Is shown.

However, U _m is a unitary matrix obtained by SVD of the propagation path matrix H _m as in Equation (1), and Equation (14) is a matrix composed of the first column to the second column of U _m. This means that y _m is multiplied by the complex conjugate transpose matrix. By performing such an operation, signals are received by two virtual antennas, a virtual antenna that obtains a gain corresponding to the maximum singular value and a virtual antenna that obtains a gain corresponding to the minimum singular value. The antenna receiving unit 50 outputs two signals received by these two virtual antennas.

  Then, the two signals output in this way are combined by the MMSE combining unit 36 using the reception weight represented by the equation (11). By performing such processing, it is possible to receive a signal spatially multiplexed with respect to one virtual antenna of each terminal by using a plurality of virtual antennas and to combine the received signals.

  Normally, the singular vector corresponding to the largest singular value and the singular vector corresponding to the smallest singular value are orthogonal, so spatial multiplexing for the virtual antenna formed using the singular vector corresponding to the largest singular value is performed. When performing, a signal is not received by the virtual antenna formed using the singular vector corresponding to the minimum singular value. However, when a precoding method that does not completely cancel inter-user interference is used, or when inter-user interference occurs due to the influence of time fluctuation of the propagation path, it is formed using a singular vector corresponding to the minimum singular value. Any signal is also received by the virtual antenna, and by performing the synthesis according to the present invention, the signals received by the two virtual antennas can be synthesized and good reception characteristics can be obtained.

(Fourth embodiment)
In the above embodiment, the matrix F representing the interference coefficient is notified from the base station apparatus to each terminal apparatus. However, the present invention is not limited to this, and the components necessary for the synthesis of the desired signal according to the present invention are included in the matrix F. It is good also as a structure which estimates in each terminal device. This is because, since the lower triangular component of the matrix F is equal to the lower triangular component of HP, the pilot signal multiplied by the matrix P representing the transmission weight is transmitted from the base station apparatus and passed through the propagation path H. This is possible by receiving a pilot signal equivalently passing through the propagation path at each terminal device and performing propagation path estimation using the received pilot signal.

  Thus, when estimating the matrix F representing the interference coefficient in each terminal, it is not necessary to notify the interference coefficient from the base station to each terminal. Therefore, the interference coefficient notification signal and the pilot signal necessary for demodulating the interference coefficient notification signal (the pilot signal with 0 added in FIG. 3) are deleted from the frame configuration shown in FIG. Transmission can be performed using such a frame configuration. However, pilot signals 1 to 4 shown in FIG. 6 are pilot signals for demodulating individual data signals. Similarly to the configuration shown in FIG. The transmission weight multiplier 23 multiplies the same transmission weight as that multiplied by. The pilot signal is orthogonalized so as not to interfere with each other on the terminal side. In other words, the pilot signal multiplied by the transmission weight and transmitted from the base station is orthogonalized in time so that it can be received on the terminal side as transmitted without interfering with each transmission antenna of the base station. ing. Here, pilot signal 1 is transmitted from antenna 1 of the base station, pilot signal 2 is transmitted from antenna 2 of the base station, pilot signal 3 is transmitted from antenna 3 of the base station, and pilot signal 4 is transmitted from antenna 4 of the base station. It is assumed that the data is transmitted so that it can be received by each terminal.

  As shown in FIG. 6, the orthogonalized pilot signal is received and the HP, that is, the matrix F is estimated at each terminal based on the received pilot signal. However, not all the lower triangular components can be estimated. Absent. For example, when the ordering order is the order of the terminals 1, 2, 3, and 4, in the terminal 1, it is possible to estimate from the pilot signal 1 that can be estimated from the pilot signal 1 and the pilot signal 2 in the HP Is the component in the first row and second column of the HP, the component that can be estimated from the pilot signal 3 is the component in the first row and third column of the HP, and the component that can be estimated from the pilot signal 4 is the component in the first row and fourth column of the HP It becomes. That is, only the first line of the HP can be estimated in the terminal 1. Similarly, the terminal 2 can estimate the second line of HP, the terminal 3 can estimate the third line of HP, and the terminal 4 can estimate the first line of the fourth line of HP. It becomes only.

Therefore, in such a case, it is impossible to calculate a weight as shown in Expression (11). However, as described above, since each terminal needs only one row of the matrix represented by Expression (11) to synthesize a desired signal, the right side of Expression (11) I _4x4 −F may be replaced with a vector using only one row of the matrix F that can be estimated by each terminal device, and the weight may be calculated. In this case, the _{I 4x4} -F first line of the terminal 1, the second line of the terminal 2, _{I 4x4} -F, third line of the terminal 3, _{I 4x4} -F, four lines of _{I 4x4} -F the terminal 4 Since the reception weight capable of combining the desired signals can be calculated by using each eye, the necessary reception weight can be calculated by estimating each row of the HP using the pilot signal described above. However, since the first row of the matrix F is all zero, it is not necessary for the terminal 1 to estimate the interference coefficient. When using the weight of Equation (11), the synthesized signal is a vector, but a weight obtained by replacing I _4x4 −F with a vector using only one row of the matrix F that can be estimated by each terminal apparatus is used. In this case, the synthesized signal becomes one value, and this obtained value becomes the desired signal.

  Here, in the present invention, each terminal feeds back CSI in a virtual antenna, and is intended for a system that performs spatial multiplexing in a base station based on CSI in each virtual antenna fed back from each terminal. In order to estimate the necessary components of the matrix F at each terminal, it is necessary to receive the pilot signal transmitted from the base station with the virtual antenna. FIG. 7 shows the apparatus configuration of the terminal that receives the pilot signal with the virtual antenna and estimates the row component of the matrix F necessary for the MMSE combining shown in Equation (11).

  As shown in FIG. 7, the terminal device according to the present embodiment has a configuration in which the configuration for demodulating the interference coefficient notification signal is deleted from the configuration shown in FIG. 4 and a new interference coefficient estimation unit 60 is provided. . The interference coefficient estimator 60 includes a channel estimation value estimated from the pilot signals 1 to 4 and a vector used when forming the virtual antenna (the left singularity corresponding to the maximum singular value in the first embodiment). Vector) and the vector used to form the virtual antenna is multiplied from the estimated channel matrix from the left side to calculate the estimated channel value when the pilot signal is received by the virtual antenna. Is done. As described above, the CSI fed back to the base station is a propagation path when it is received by the virtual antenna. Since the base station performs spatial multiplexing using the CSI, the pilot signal must be received by the virtual antenna. For example, the equivalent propagation path HP including the linear transmission weight, that is, each row of the matrix F representing the interference coefficient cannot be estimated. Therefore, the interference coefficient estimation unit 60 multiplies the propagation path estimated value by the vector used when forming the virtual antenna, estimates an equivalent propagation path when the pilot signal is received by the virtual antenna, Of each row of the matrix F representing the interference coefficient, a necessary row component is estimated at each terminal.

  Then, any row component of the matrix F representing the estimated interference coefficient is notified to the MMSE combining unit 36 and used for combining signals received by the two antennas. The desired signal obtained by MMSE synthesis in the MMSE synthesis unit 36 is modulo-modulated and demodulated in the demodulation unit 38 in the same manner as in the previous embodiments. With such a configuration, it is not necessary to notify each terminal device of the matrix F representing the interference coefficient from the base station device, and the transmission efficiency can be improved. Note that the base station apparatus in this case has a configuration in which the portion for outputting the interference coefficient to the higher layer 10 is deleted from the transmission weight / interference coefficient calculation unit 24 of FIG. Further, as described above, the pilot signal that is not multiplied by the transmission weight (pilot signal 0) is not transmitted. Furthermore, in the present embodiment, an example related to the ordering order similar to that in the first embodiment has been described. However, the present invention can also be applied to a configuration in which the ordering order is adaptively changed as in the second embodiment. In such a case, each terminal needs to estimate the row component of the matrix F required for each MMSE synthesis in accordance with the ordering order.

  In the above embodiment, as a method of forming a virtual antenna in each terminal, SVD is applied to the propagation path matrix observed by a plurality of receiving antennas, and the complex conjugate transposed vector of the left singular vector corresponding to the maximum singular value is obtained. Although the method of multiplying the received signal is targeted, the present invention is not limited to this, and the received signal is processed as if it were received by fewer than N receiving antennas by performing processing such as multiplying the received signal by N receiving antennas with a weight. The present invention is applicable as long as the method is obtained. For example, a method of simply adding signals received by a plurality of receiving antennas and processing them as if they were received by one virtual receiving antenna, or one virtual receiving by adaptively switching the receiving antennas There is a method of forming an antenna, and by performing signal synthesis according to the present invention, when a virtual antenna is formed using any method, a plurality of signals spatially multiplexed for the virtual antenna are received. A desired signal can be obtained by receiving with an antenna and appropriately combining the received signals.

  Moreover, in the above embodiment, although the structure which feeds back the propagation path information in each terminal device from each terminal device to a base station apparatus was shown, in a TDD (Time Division Duplex) system, an uplink and a downlink are shown. Since duality is established, it is not always necessary to provide feedback, and spatial multiplexing may be performed based on propagation path information observed by the base station apparatus.

  Further, in the case where MMSE combining of signals received by a plurality of antennas at each terminal has been shown in the above embodiment, a signal received by one antenna is demodulated or received by a virtual antenna. Compared to the case of processing, a good reception characteristic can be obtained, but when the terminal is located in the vicinity of the base station and a very high SNR is obtained, the characteristic difference is reduced, and there is an advantage of combining MMSE. May be smaller. Even in a situation where the terminal is stationary, it is possible to obtain a certain level of good reception characteristics by performing demodulation such as demodulating only the signal received by any one antenna or receiving by a virtual antenna. it can. In addition, if the MMSE combining method according to the present invention is applied with a small error, it is necessary to notify each terminal of the interference coefficient calculated by the base station, and the transmission efficiency is higher than when not reporting such information. Since there is a problem that it is lowered, it is possible to realize a more efficient system construction by applying the present invention only in an appropriate situation instead of always performing signal synthesis according to the present invention. Conceivable. Therefore, a configuration may be adopted in which application / non-application of the MMSE combining according to the present invention is switched according to the received SNR at the terminal, the moving speed of the terminal, and the like.

  Further, the program that operates in the mobile station apparatus and the base station apparatus related to the present invention is a program (program that causes a computer to function) that controls the CPU and the like so as to realize the functions of the above-described embodiments related to the present invention. Information handled by these devices is temporarily stored in the RAM at the time of processing, then stored in various ROMs and HDDs, read out by the CPU, and corrected and written as necessary. As a recording medium for storing the program, a semiconductor medium (for example, ROM, nonvolatile memory card, etc.), an optical recording medium (for example, DVD, MO, MD, CD, BD, etc.), a magnetic recording medium (for example, magnetic tape, Any of a flexible disk etc. may be sufficient. In addition, by executing the loaded program, not only the functions of the above-described embodiment are realized, but also based on the instructions of the program, the processing is performed in cooperation with the operating system or other application programs. The functions of the invention may be realized.

  In the case of distribution in the market, the program can be stored and distributed in a portable recording medium, or transferred to a server computer connected via a network such as the Internet. In this case, the storage device of the server computer is also included in the present invention. Moreover, you may implement | achieve part or all of the mobile station apparatus and base station apparatus in embodiment mentioned above as LSI which is typically an integrated circuit. Each functional block of the mobile station apparatus and the base station apparatus may be individually made into a processor, or a part or all of them may be integrated into a processor. Further, the method of circuit integration is not limited to LSI, and may be realized by a dedicated circuit or a general-purpose processor. In addition, when an integrated circuit technology that replaces LSI appears due to progress in semiconductor technology, an integrated circuit based on the technology can also be used.

  The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and the design and the like within the scope not departing from the gist of the present invention are also claimed. Included in the range.

DESCRIPTION OF SYMBOLS 10 Upper layer 11 Modulation part 12 P / S part 13 Interference subtraction part 14 Modulo part 15 Interference generation part 16 S / P part 17, 23 Transmission weight multiplication part 18 Signal multiplexing part 19 D / A part 20, 27 Radio | wireless part 21 Transmission Antenna unit 22 Pilot signal generation unit 23 Transmission weight multiplication unit 24 Transmission weight / interference coefficient calculation unit 25 Reception unit 26 A / D unit 28 Reception antenna unit 30 Reception antenna unit 30-1, 2 Reception antenna 31 Radio unit 32 A / D Unit 33 signal separation unit 34 propagation path estimation unit 35 propagation path compensation unit 36 MMSE synthesis unit 37 modulo unit 38 demodulation unit 39 upper layer 40 virtual antenna formation unit 41 transmission unit 42 D / A unit 43 radio unit 44 transmission antenna unit 50 virtual Antenna receiving unit 60 Interference coefficient estimating unit 100 Base station 200 Terminal

Claims (12)

A terminal apparatus that communicates with a base station apparatus that includes N receiving antennas when N is an integer equal to or greater than 2, and includes a plurality of transmitting antennas and spatially multiplexes signals addressed to a plurality of terminal apparatuses using a nonlinear operation. There,
Based on information about propagation paths observed by the N receiving antennas, information about propagation paths when received by virtual reception antennas less than N is calculated and received by the virtual reception antennas. Information signals subjected to spatial multiplexing in the base station apparatus using information on propagation paths in the case and information on propagation paths of other terminal apparatuses are received by the N receiving antennas,
Information signals received by the N receiving antennas are multiplied by a reception weight and combined,
A terminal device, wherein a non-linear operation is performed on the combined information signal to detect a desired information signal.   The terminal apparatus according to claim 1, wherein the terminal apparatus receives a pilot signal transmitted from the base station apparatus, and calculates the reception weight based on the received pilot signal.   The information regarding the coefficient showing the interference in the spatial multiplexing using the said nonlinear calculation transmitted from the said base station apparatus is received, The said receiving weight is calculated based on the information regarding the coefficient showing the said interference. Item 3. The terminal device according to Item 2.   The reception weight is a weight that minimizes an average square error between the information signal transmitted from the base station and the information signal multiplied by the reception weight. The terminal device described in 1.   The information about the propagation path when received by the virtual receiving antenna is a vector or matrix obtained by singular value decomposition of a matrix representing the propagation path observed by the N receiving antennas. The terminal device according to claim 1, wherein the terminal device is obtained by multiplying a matrix representing a propagation path observed by a receiving antenna.   The terminal device according to claim 1, wherein the base station device is notified of information regarding a propagation path when received by the virtual receiving antenna.   The terminal device according to claim 1, wherein the nonlinear calculation is a modulo calculation. A first terminal device including a plurality of transmission antennas and N reception antennas when N is an integer equal to or greater than 2 is included, and the total of reception antennas of all terminal devices to be communicated is the transmission A base station device that performs communication by performing spatial multiplexing on information signals addressed to a plurality of terminal devices that are larger than the number of antennas used for transmission,
Information on a propagation path when received by a virtual receiving antenna of less than N, generated based on a propagation path observed in the first terminal apparatus, and information on a propagation path of another terminal apparatus And calculating a coefficient representing transmission weight and interference in spatial multiplexing using a non-linear operation, and spatially multiplexing information signals addressed to a plurality of terminal devices using the transmission weight and the coefficient representing interference. Base station apparatus.   The base station apparatus according to claim 8, wherein the transmission weight is multiplied by a pilot signal for channel estimation and transmitted.   The base station apparatus according to claim 8 or 9, wherein information on a coefficient representing the interference is notified to the terminal apparatus.   The base station apparatus according to claim 8, wherein the non-linear operation is a modulo operation. A terminal device according to any one of claims 1 to 7;
A base station apparatus according to any one of claims 8 to 11,
A wireless communication system comprising:

Images