JP5797177B2 - Wireless communication method, wireless transmission device, wireless reception device, and wireless communication system - Google Patents

Description

  The present invention relates to a wireless communication method, a wireless transmission device, a wireless reception device, and a wireless communication system, and, for example, relates to wireless communication in which signal transmission is performed simultaneously from a plurality of antennas.

  As a background art in this technical field, a MIMO (Multiple Input, Multiple Output) system is widely known. The MIMO scheme is a technique for transmitting signals from a plurality of antennas and separating the transmission signals by receiving signals from the plurality of antennas. Another background art in this technical field is BICM-ID (Bit Interleaved Coded Modulation-Iterative Decoding / Detecting). The BICM-ID scheme is an iterative process in which, on the receiving side, the result of decoding the demodulated signal is returned to the demodulating side and demodulated again to output a result with higher reliability than the previous result. This is a technology for performing reception by (Iterative Decoding / Detecting).

  For example, Non-Patent Document 1 is a technique in which encoding is performed using an LDPC (Low Density Parity Check) code and transmission is performed from a plurality of antennas, and on the receiving side, a MIMO demodulation process is incorporated in the repetition process of Sum-Product decoding of the LDPC code. Is disclosed. In Non-Patent Document 1, code design is performed using Extrinsic Information Transfer (EXIT) analysis (see Non-Patent Document 2) that analyzes how the reliability of information propagates in iterative processing. It has been shown that characteristics close to the Shannon limit (theoretical limit) can be realized.

S. ten Brink, G. Kramer and A. Ashikhmin, “Design of Low-Density Parity-Check Codes for Modulation and Detection,” IEEE Transactions on Communications, Vol. 52, No. 4, pp.670-678, April 2004 . S. ten Brink, “Convergence Behavior of Iteratively Decoded Parallel Concatenated Codes,” IEEE Transactions on Communications, Vol. 49, No. 10, pp.1727-1737, October 2001. Kenichi Higuchi et al., “Multi-antenna Wireless Transmission Technology Part 3: Signal Separation Technology in MIMO Multiplexing”, NTT DoCoMo Technical Journal Vol. 14 No.1, pp. 66-75, April, 2006. Masaru Matsumoto et al., “Basics of Turbo Equalization and Information Theoretical Consideration”, IEICE Transactions B, Vol. J90-B No. 1, pp. 1-16, January 2007. T. Yano and T. Matsumoto, “Arithmetic Extended-Mapping for BICM-ID with Repetition Codes”, Proceedings 2009 International ITG Workshop on Smart Antennas (WSA 2009), pp. 304-311, February 2009.

  In the conventional technique as described above, as described in Non-Patent Document 1, how well the EXIT curves in two processes (for example, demodulation process and decoding process) connected via an interleaver are matched. Determines communication performance. That is, when the two curves intersect, the result of the iterative process converges to a point where the reliability of the information is not sufficiently improved, and the information cannot be correctly decoded. On the other hand, when there is a large gap between the two curves, the decoding is completed with a small number of iterations, but the communication rate is low. Therefore, in order to design a BICM-ID communication method capable of performing communication at the highest possible communication rate, the EXIT curves in the two processes connected via the interleaver are designed to be as close as possible without crossing each other. It is desirable.

  However, in the BICM-ID system using LDPC shown in Non-Patent Document 1, since the MIMO demodulation process is incorporated in the iterative process, the EXIT is caused by a change in the propagation path existing between the transmission side and the reception side. The curve changes. Therefore, the EXIT curves are not matched with each other due to the fluctuation of the propagation path, and there is a problem that the reception signal cannot be accurately decoded on the reception side. Therefore, it is desired that the signal can be accurately decoded on the receiving side even if the propagation path varies.

  The present invention has been made in view of such a situation, and provides a technique for enabling a signal to be accurately decoded on the receiving side even when there is a change in propagation path.

  In order to solve the above problems, in the present invention, transmission data is encoded and modulated in encoding units, and transmitted from m (m is an integer of 2 or more) antennas. At this time, beam forming processing is executed by multiplying each of modulation data generated from transmission data of one coding unit by a plurality of different beam forming matrices. Then, the modulated data subjected to the beam forming process is transmitted from m antennas. Here, a plurality of beamforming weight vectors constituting a plurality of different beamforming matrices take a number of different values exceeding m.

  According to the present invention, it is possible to realize BICM-ID communication that can restore a transmission signal without error even when the propagation path fluctuates.

  Further features related to the present invention will become apparent from the description of the present specification and the accompanying drawings. The embodiments of the present invention can be achieved and realized by elements and combinations of various elements and the following detailed description and appended claims.

  It should be understood that the description herein is merely exemplary and is not intended to limit the scope of the claims or the application of the invention in any way.

It is a figure which shows the example of the propagation environment to which this invention is applied. It is a figure which shows the example of the change of the propagation path matrix of the propagation environment to which this invention is applied. It is a figure which shows the EXIT curve of a MIMO demodulator. It is a figure which shows the example of the EXIT chart of a receiver. It is a figure which shows the other example of the EXIT chart of a receiver. It is a figure which shows the further another example of the EXIT chart of a receiver. It is a figure which shows the EXIT curve of a MIMO demodulator at the time of applying this invention. It is a figure which shows the example of the EXIT chart of the receiver at the time of applying this invention. It is a figure which shows schematic structure of the radio | wireless communications system by this invention. It is a figure which shows the schematic structural example of the transmitter by the 1st Embodiment of this invention. It is a figure which shows the schematic structural example of the receiver by the 1st Embodiment of this invention. It is a figure which shows the structural example of the encoder of the transmitter by 1st Embodiment. It is a figure which shows operation | movement of an encoder. It is a figure which shows operation | movement of the modulator of the transmitter by 1st Embodiment. It is a figure which shows the internal structural example of the beam forming process part (transmission side beam forming process part) in a transmitter. It is a figure which shows the structural example of a beam forming matrix production | generation part. It is a figure which shows the example of the table in a beam forming matrix production | generation part. It is a figure which shows the other structural example of a beam forming matrix production | generation part. It is a figure which shows the example of the other table in a beam forming matrix production | generation part. It is a figure which shows the example of the other table in a beam forming matrix production | generation part. It is a figure which shows the example of the other table in a beam forming matrix production | generation part. It is a figure which shows the example of the other table in a beam forming matrix production | generation part. It is a figure which shows the structural example of the beam forming process part (reception side beam forming process part) in a receiver. It is a figure which shows the schematic structural example of the transmitter by the 2nd Embodiment of this invention. It is a figure which shows the schematic structural example of the receiver by the 2nd Embodiment of this invention. It is a figure which shows the structural example of the variable node decoder of the receiver by 2nd Embodiment. It is a figure which shows the structural example of the check node decoder of the receiver by 2nd Embodiment. It is a flowchart for demonstrating the example of the transmission side resending process performed in each embodiment of this invention. It is a flowchart for demonstrating the example of the receiving side resending process performed in each embodiment of this invention.

  The present invention relates to random beamforming BICM-ID wireless communication.

  Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the accompanying drawings, functionally identical elements may be denoted by the same numbers. The attached drawings show specific embodiments and implementation examples based on the principle of the present invention, but these are for understanding the present invention and are not intended to limit the present invention. Not used.

  This embodiment has been described in sufficient detail for those skilled in the art to practice the present invention, but other implementations and configurations are possible without departing from the scope and spirit of the technical idea of the present invention. It is necessary to understand that the configuration and structure can be changed and various elements can be replaced. Therefore, the following description should not be interpreted as being limited to this.

(1) Outline and principle of the present invention In the present invention, beam forming using various beam forming matrices is performed within one codeword (coding unit: frame) on the transmitting side (coding unit (frame)). A different beamforming matrix is applied to each modulation symbol in the signal). As a result, the EXIT curve on the MIMO demodulation processing side can be averaged, and the BICM-ID can be designed according to the averaged EXIT curve. According to the present invention, since the change in the propagation path matrix that changes the EXIT curve is generated in advance by transmission beam forming, the average EXIT of the effective propagation path even when there is a change in the actual propagation path. The curve does not change, and even when the propagation path fluctuates, it is possible to prevent the consistency between the EXIT curves.

  Hereinafter, the principle and outline of the present invention will be described using specific examples.

  FIG. 1 is a diagram schematically illustrating a MIMO communication path having a strong propagation path correlation. In the propagation path as shown in FIG. 1, when transmitting from two antennas (two points on the left side in FIG. 1) and receiving with two antennas (two points on the right side in FIG. 1), It is assumed that there is a reflector at the distance d on the left side of the transmission antenna. Here, it is assumed that the transmission antennas are separated by λ / 4 (λ is a wavelength) as shown in the figure. The transmission signal is reflected and received with an interference wave having a phase difference of π (4d / λ + 1) on one side and a phase difference of π (4d / λ) on the other side. In addition, since the two receiving antennas are arranged so as to be perpendicular to the traveling direction of the radio wave, both received signals are the same (that is, the propagation path is degenerated). (Receiver side correlation is strong)). At this time, if the transmission signals from the two transmission antennas are t1 and t2, and the reception signals of the two reception antennas are r1 and r2, the propagation path can be expressed as a matrix with 2 rows and 2 columns.

  FIG. 2 is a diagram illustrating changes in the propagation path matrix when the transmission side moves and d changes. In FIG. 2, the attenuation coefficient and the complex phase term common to all the elements of the matrix are omitted for simplicity. As shown in FIG. 2, in some cases, only t1 or t2 is received (a state in which only one information is lost and the other information is completely lost), and in other cases, a signal in which t1 and t2 are mixed is received. Has been received. In general, even in the case where a “mixed” signal is received in MIMO communication, the transmission signals t1 and t2 can be restored using signals having different “mixing methods” received by a plurality of receiving antennas. In the above example, since the reception signals of the two reception antennas are exactly the same, the transmission signals t1 and t2 cannot be separated by normal MIMO demodulation processing. However, a BICM-ID that performs repetitive processing incorporating MIMO demodulation processing can correctly restore the transmission signal. In the BICM-ID, MIMO demodulation processing can be performed using information from the decoding processing. If complete information on the other signal, for example t2, is obtained from the decoding process, t2 can be removed from the signal in which t1 and t2 are mixed, and t1 can be extracted.

FIG. 3 is a diagram illustrating how the EXIT curve of the MIMO demodulation process varies. In FIG. 3, the horizontal axis indicates information I _{A, M} obtained from the decoding process (when 0 is no information and 1 is completely decoded), and the vertical axis is information I _{E, M} (output by the MIMO demodulation process). 0 indicates no information and 1 indicates complete decoding). While the transmission signals t1 and t2 are not mixed, in the propagation path where only one of the signals is received and the other is lost (the value on the vertical axis has almost no change: the EXIT curve has no inclination), it is obtained from the decoding process. Therefore, the MIMO demodulation process outputs certain information regardless of the information IA _{and M} obtained from the decoding process. However, in this case, since half of the information is lost, the obtained information amount _{IE, M} is less than 0.5.

On the other hand, in a propagation path in which both are mixed, only information amount of 0.4 or less is output at I _{A, M} = 0 where there is no information obtained from the decoding process, but the information obtained from the decoding process is completely I _{When A, M} = 1, an information amount exceeding 0.7 can be output. Therefore, as shown in FIG. 4, the BICM-ID combined with the convolutional code with the constraint length K = 4 and the coding rate R = 1/3 can restore the transmission signal with almost no error. In FIG. 4, the solid line shows the EXIT curve of the MIMO demodulation process in the propagation path where t1 and t2 shown in FIG. 3 are mixed with equal gain, and the broken line shows the constraint length K = 4 and the coding rate R = 1. The EXIT curve of the decoding process of the / 3 convolutional code is shown.

The EXIT curve of the decoding process of the convolutional code is opposite to the EXIT curve of the MIMO demodulation process, the vertical axis is the information amount I _{A, D} passed from the MIMO demodulation process, and the horizontal axis is the information amount output by the decoding process of the convolutional code _{IE and D} (see FIG. 4). First, when starting the BICM-ID process, the decoding process of the convolutional code cannot output any information (IA _{, M} = 0). Therefore, the MIMO demodulation process outputs an information amount of about _{IE, M} = 0.36. When this information is input to the convolutional code decoding process as I _{A, D} , an information amount of about −IE _{, D} = 0.52 is output as a decoding result as a result of the convolutional code decoding process. When this output is given to the MIMO demodulation process as I _{A, M} and the MIMO demodulation process is performed again, an information amount of about _{IE, M} = 0.52 is output. If this is input again to the convolutional code decoding process as I _{A, D} , the convolutional code decoding process outputs information of about _{IE, D} = 0.90. Thereafter, when the processing is repeated in the same manner, the decoding result -IE _{, D} of the decoding processing of the convolutional code becomes almost 1 as shown by the stepped dotted line in FIG. 4, and there is no error in the decoding result.

Next, considering the case where the transmission side in FIG. 1 moves a little and becomes a propagation path in which t1 and t2 are hardly mixed, the inclination of the EXIT curve of the MIMO demodulation process becomes small as shown in FIG. As a result, in the combination with the convolutional code having the constraint length K = 4 and the coding rate R = 1/3, the information amount _{IE, M} output by the MIMO demodulation process is about 0.5, and the convolutional code is decoded. The amount of information _{IE and D} output by the process converges to a point of about 0.85, and no error-free decoding can be performed no matter how many times it is repeated.

  In this regard, for example, if a convolutional code having a constraint length K = 4 and a coding rate R = 1/5 is used, decoding can be performed correctly in any propagation path as shown in FIG. The data rate is reduced to 3/5 times as compared with the above example, which is not desirable.

  Therefore, in the present invention, as described above, various transmission beamforming is performed within one codeword. For example, for a propagation path in which a propagation path matrix is expressed as in Expression 1 (the transmission side does not recognize the propagation path) and a transmission signal from two antennas is mixed and received as in Expression 2. Using the beamforming matrix B1 of Equation 3, beamforming is performed on the transmission signal.

  Then, as shown in Equation 4, the effective channel matrix represented by the product of the beamforming matrix and the channel matrix remains to be received by mixing t1 and t2.

  On the other hand, when beam forming is performed on the transmission signal using the beam forming matrix B2 of Equation 5, only t1 is an effective propagation channel matrix represented by the product of the beam forming matrix and the propagation channel matrix as shown in Equation 6. It is received and t2 is not received.

  Therefore, when beamforming is performed by mixing B1 of Equation 3 and B2 of Equation 5 in one codeword, only one of t1 and t2 is received and symbols received with t1 and t2 mixed with equal gain. The EXIT curve becomes an average curve of both.

  Similarly, let us consider a propagation path in which a propagation path matrix is expressed as in Expression 7, and only a transmission signal t1 from two antennas is received as in Expression 8, and t2 is not received.

  Transmission beamforming is performed on such a propagation path using the beamforming matrix B1 of Expression 3. In this case, the effective propagation path matrix is represented by the product of the beamforming matrix and the propagation path matrix, but as can be seen from Equation 9, only t1 is received and t2 is not received.

  On the other hand, when beam forming is performed on the transmission signal using the beam forming matrix B2 of Expression 5, an effective propagation path matrix represented by a product of the beam forming matrix and the propagation path matrix is represented by Expression 10, and t1, t2 is received with equal gain.

  Therefore, also in this case, when beam forming is performed by mixing B1 of Equation 3 and B2 of Equation 5 in one codeword, symbols received by mixing t1, t2 with equal gain, t1, Symbols in which only one of t2 is received are mixed, and the EXIT curve is also an average curve of both.

  As described above, by mixing a plurality of types of beamforming in one codeword, various “mixing methods (mixed propagation paths)” can be realized in one codeword. Therefore, even when the propagation path changes, the average EXIT curve can be prevented from changing. The average EXIT curve obtained in this way is, for example, the thick line in FIG. Since this does not change even when the propagation path changes, as shown in FIG. 8, for example, a convolutional code with a constraint length K = 9 and a coding rate R = 2/5 (the EXIT curve is a broken line in FIG. 25). ) And the BICM-ID combined, it is possible to always perform error-free decoding even when the transmission side (transmitter) in FIG. 1 moves.

  In the example of FIG. 1, for the sake of simplicity, the explanation has been made using the real coefficient propagation path of two antennas. In general, when a propagation path matrix is represented by H, it can be represented by a product of three matrices as shown in Equation 11 using unitary matrices U and V and a diagonal matrix Σ using singular value decomposition.

Here, V ^H represents Hermitian conjugate of V. Therefore, when the beamforming matrix B coincides with V (Equation 12), an effective channel matrix can be expressed as Equation 14 from the property of the unitary matrix (Equation 13).

  When the propagation path is degenerated as described above (when the signal correlation is strong on the receiving side), only the first transmission signal is received because only the first element of the diagonal matrix Σ has a value. It becomes a propagation path.

  On the other hand, when the beam forming matrix B is expressed by Equation 15 for m transmission antennas, a reception signal in which all transmission signals are mixed at an equal level is obtained.

  In other words, the various beamforming matrices used in the present invention are V values in the singular value decomposition shown in Equation 11 and various approximate values obtained by transformation shown in Equation 15 with respect to various propagation path matrices H. It is necessary to include.

  In addition, for the complex coefficient channel matrix H, V is also a complex coefficient, so the beam forming matrix must also be a matrix having various complex coefficients. For example, in order to receive only t1 with respect to the propagation channel matrix H3 having a complex coefficient as shown in Expression 16, a beamforming matrix as shown in Expression 17 is received with t1 and t2 mixed with equal gain. In order to achieve this, a beam forming matrix as shown in Equation 18 is required.

  Here, a more generalized propagation path matrix can be expressed as Equation 19.

In Equation 19, V ^H is described as a column vector, but it is actually an m × m matrix having m row vectors. B is also described as a row vector, but is actually an m × m matrix having m column vectors.

If the propagation path is degenerated (if the receiving side correlation signal is strong), the equation 19, since the lambda >> epsilon _k, indicating the value only the values of the portion surrounded by a dotted line is greater in Equation 19 It will be. That is, only that component contributes to reception, and if only one of them (eg, λν ′ ₁ b ₂ ) has a dominant value, the corresponding transmission The signal is received without being mixed with other components. On the other hand, when there is no protruding value of λ or ε _k (all of similar values), all transmission signals are mixed and received.

  As described above, the beamforming matrix must be a unitary matrix (or a constant multiple thereof) so that the channel capacity is not impaired by beamforming. Therefore, the EXIT curve of the MIMO demodulation process can be averaged by performing beamforming on a signal in one codeword using a unitary matrix having various complex coefficients, and the propagation path is A BICM-ID communication system that can be demodulated without error even when it fluctuates can be configured.

  Note that a beamforming matrix close to the matrix shown in Expression 12 for any propagation path exists in various beamforming matrices means that any beamforming weight vector is used for any complex unit vector. This means that the magnitude (absolute value) of the inner product with (column vector of matrix B) is close to 1. If the two coincide completely, the inner product is 1, and the inner product of another beamforming weight vector included in the beamforming matrix including the beamforming weight vector and an arbitrary complex unit vector is 0. Therefore, only the transmission signal beamformed with the weight vector having the inner product of 1 is received, and the transmission signal beamformed with the weight vector having the inner product of 0 is lost. When the inner product is not completely 1, the transmission signal beamformed by another weight vector is received in a state of being slightly leaked and mixed, but it seems that one transmission signal is received with a dominant strength. If there is, it does not have to completely coincide with the beamforming matrix shown in Expression 12.

  Therefore, when a plurality of beamforming matrices used in the present invention are designed so that any beamforming matrix includes a beamforming weight vector in which the absolute value of the inner product with an arbitrary complex unit vector is larger than a predetermined value. Good. The predetermined value is ideally close to 1, but should be at least 1 / √2 or more. This is because by setting the ratio to 1 / √2 or more, the ratio of the power occupied by one transmission signal in the reception signal can be set to 1/2 or more.

  Further, the beamforming of the present invention is essentially different from a system in which signals are arranged and transmitted on orthogonal channels typified by eigenmode MIMO transmission. When the number of transmission antennas is m, the beamforming weight vector used in eigenmode transmission is a vector corresponding to at most m orthogonal channels. In the beam forming process of the present invention, since various beam forming matrices are applied as described above, the number of beam forming weight vectors exceeding the number m of antennas is used. This is because at most m beamforming weight vectors cannot include an inner product with an arbitrary complex unit vector larger than a predetermined value (for example, 1 / √2). For example, in the case of the beamforming matrices B1 and B2 used in the calculations of the above formulas 1 to 10, the beamforming weight vectors (column vectors of the matrix B) are (1, 0), (0, 1), (√2 / 2, √2 / 2), (−√2 / 2, √2 / 2), and the number thereof is four. That is, in this case, since the number of antennas m is 2, the number of beamforming weight vectors exceeds the number of antennas.

(2) 1st Embodiment 1st Embodiment demonstrates the example of the BICM-ID communication method which can restore | restore a transmission signal correctly, even when a propagation path changes.

<Configuration of communication system>
FIG. 9 is a diagram showing a schematic configuration of the communication system according to the present embodiment. The transmission data is transmitted from a plurality of antennas 3 (two in FIG. 9) by the transmitter 1, and a signal received by the antenna 3 of the receiver is demodulated by the receiver 2 and outputs reception data.

  FIG. 10 is a diagram illustrating an internal configuration example of the transmitter 1. The transmitter 1 includes an encoder 10, an interleaver 11, serial / parallel converters 12 and 14, a modulator 13, and a beamforming processing unit (transmission-side beamforming processing unit) 15. The transmission signal subjected to the forming process is output from the antenna 3.

  The transmission data is encoded by the encoder 10. The codeword of the encoding result is output with the bit order changed by the interleaver 11. The output of the interleaver 11 is converted by the serial / parallel converter 12 so that the number of bits modulated into one modulation symbol is output simultaneously. The output of the serial / parallel converter 12 is modulated by the modulator 13. The modulated signal is converted by the serial / parallel converter 14 so that the number of modulation symbols transmitted from a plurality of antennas is simultaneously output. The output of the serial / parallel converter 14 is beam-formed by the transmission-side beam forming processing unit 15 and supplied to the antenna 3.

  FIG. 11 is a diagram illustrating an internal configuration example of the receiver 2. The receiver 2 includes a propagation path estimation processing unit 16, a beamforming processing unit (receiving side beamforming processing unit) 17, a MIMO demodulator 18, and a parallel / serial converter 19, an interleaver 20, and a decoder 21. And an interleaver 11 and a serial / parallel converter 22.

  The propagation path estimation processing unit 16 estimates the propagation path matrix between the transmission and reception antennas from the signal received by the antenna. Although not specifically described in the description of the transmitter 1, a method in which the receiver 2 periodically transmits a known reference signal for estimating the propagation path and estimates the propagation path from the received reference signal is widely known. ing. The reception-side beamforming processing unit 17 calculates an effective channel matrix from the estimated channel matrix using the same beamforming matrix used in the beamforming process performed by the transmitter 1. The MIMO demodulator 18 outputs a demodulation result in an LLR format using an effective channel matrix, a received signal, and an a priori LLR (Log Likelihood Ratio) supplied from the decoder 21. LLRs for a plurality of bits included in a plurality of simultaneously received symbols are converted by a parallel / serial converter 19 and supplied to a deinterleaver 20. The deinterleaver 20 returns the bits whose order has been changed by the interleaver of the transmitter 1 to the original order and outputs it. The LLR returned to the original order is decoded by the decoder 21 and a posteriori LLR is calculated. The a posteriori LLR is converted to an extrinsic LLR by subtracting the LLR input to the decoder 21 and supplied to the interleaver 11. The interleaver 11 changes the bit order of the external (Extrinsic) LLR again and outputs the result. The external (Extrinsic) LLR whose order has been changed is converted so that bits included in symbols simultaneously processed by the serial / parallel converter 22 are output simultaneously, and the MIMO demodulator 18 is converted into an a priori LLR. To be supplied. After a sufficient number of iterations, the decoder 21 outputs decoded data (Rx Data).

<Configuration of encoder>
FIG. 12 is a diagram illustrating a circuit configuration example of the encoder 10. As the encoder 10, for example, a convolutional encoder can be used as shown in FIG.

  The convolutional encoder shown in FIG. 12 is an encoder having a coding rate R = 2/5 and a constraint length K = 9 showing the EXIT curve shown in FIG. However, in FIG. 12, a convolution operation unit with a coding rate R = 1/3 is used, but a part of the code as shown in FIG. 13 is obtained by the puncture and serial / parallel converter 100 arranged at the output. Is a code with a coding rate R = 2/5.

  When a convolutional encoder as shown in FIG. 12 is used for the encoder 10, the decoder 21 can perform a decoding process using a BCJR (Bahl, Cocke, Jelinek and Raviv) algorithm, and calculate an a posteriori LLR. Is widely known.

<Modulator and demodulator>
FIG. 14 is a diagram illustrating QPSK (Quadrature Phase Shift Keying) modulation. As the modulator 13 in the transmitter 1 described above, a QPSK modulator that performs modulation shown in FIG. 14 can be used.

  Further, the MIMO demodulator 18 of the receiver 2 preferably calculates the LLR by MLD (Maximum Likelihood Detection) processing as shown in Non-Patent Document 3, but other methods are shown in Non-Patent Document 4. The LLR may be calculated by SC-MMSE (Soft Cancellation followed by Minimum Mean Squared Error filter) processing.

<Configuration of Transmitter Beamforming Processing Unit>
FIG. 15 is a diagram illustrating a configuration example of the transmission-side beamforming processing unit 15.

  The transmission side beamforming processing unit 15 includes a pseudo-random number generator 101, a beamforming matrix generation unit 102, and a matrix × vector calculator 103.

  The pseudo random number generator 101 generates a pseudo random number including a plurality of bits and supplies the pseudo random number to the beam forming table 102. The pseudo random number generator on the transmission side and the reception side must be the same. This is because if they are the same, the same sequence is generated and the same beamforming matrix is generated.

  As will be described later, the beamforming generation unit 102 holds, for example, a plurality of tables, and generates and outputs a beamforming matrix that is a complex matrix according to a pseudo random number.

  A complex vector composed of a plurality of modulation symbols supplied from the modulator 13 is multiplied by the beamforming matrix by the matrix × vector computing unit 103, and a beamforming process is executed and output.

  Note that the pseudo-random number is generated from one codeword and outputs a different value while processing a plurality of modulation symbols, so that beam forming is performed using different matrices.

<Beam forming table>
FIG. 16 is a diagram illustrating an internal configuration of the beamforming matrix generation unit 102.

  The beamforming matrix generation unit 102 includes a plurality of beamforming tables 105-1 to 105-5 and a plurality of matrix calculators 104. The beamforming table can be realized as one large table that selects and outputs the beamforming matrix according to the input random number, but here the beamforming matrix is calculated by combining small tables. It is configured to do.

  As will be described later, the same table is used for each of the tables 1_105-1 and 4_105-4, and the tables 2_105-2 and 5_105-5.

  Further, as shown in FIG. 16, regarding the output from each table, among the pseudo random number sequences b14 to b0, the values from b1 to b0 from table 1, the values from b5 to b3 from table 2, A matrix is output based on the values b8 to b6 from 3, and the values b14 to b12 from Table 5, and these are multiplied and output as a beamforming matrix.

<Table configuration example>
FIG. 17 is a diagram illustrating a configuration example of the tables 1 to 5-105-1 to 5. In the first embodiment, Table 1 and Table 4, and Table 2 and Table 5 are common to each other, and the output matrix is made different by making a part of the input pseudo-random number sequence different. ing. Table 3 is used to generate a plurality of 2-by-2 unitary matrices of real coefficients, and the matrix generated from Table 1 has a complex phase applied to the vector in the first row of the matrix generated from Table 3. Giving. Similarly, the matrix generated from Table 2 gives a complex phase to the vector in the second row. Tables 4 and 5 give complex phases to the vectors in the first and second columns, respectively.

  Although the case where the number of transmission antennas is two has been described here, the configuration of FIG. 18 is adopted instead of the configuration of FIG. 16 in the case of three transmission antennas, and FIG. 19A is used instead of the table of FIG. Through Tables 1 to 7_105-1 to 105-7 of D to D, a random beamforming matrix can be obtained similarly. The same applies when there are four or more transmission antennas.

<Receiving side beamforming processing unit>
FIG. 20 is a diagram illustrating a configuration example of the reception-side beamforming processing unit 17.

  The reception side beamforming processing unit 17 also includes a pseudo random number generator 101, a beamforming matrix generation unit 102, and a matrix × matrix multiplier 104. The pseudorandom number generator 101 and the beamforming table 102 are the same as those of the transmission side beamforming processing unit 15.

  The generated beamforming matrix (B) is multiplied by the propagation path matrix (H: estimated by the propagation path estimation processing unit 16) input by the matrix × matrix multiplier 104 to obtain an effective propagation path matrix (HB ) Is output. Note that the pseudo-random number generator 101 generates the same random number as that on the transmission side.

  In the first embodiment described above, the convolutional code is used as the code, and the case where the modulation is QPSK modulation is described as an example. However, as shown in Non-Patent Document 5, as the code (with single parity check), as the repeated code and the modulation It is also preferred to use Arithmetic Extended Mapping.

(3) Second Embodiment The second embodiment of the present invention relates to a configuration example using an LDPC (Low Density Parity Check) code as a code. The MIMO_BICM-ID system using the LDPC code is described in Non-Patent Document 1, but the present invention combines this with random beam forming. The table used for generating the beamforming matrix is the same as that in the first embodiment (FIGS. 17 and 19A to D).

<Configuration example of transmitter>
FIG. 21 is a diagram illustrating a configuration example of the transmitter 1 according to the second embodiment.

  The transmitter 1 includes an LDPC encoder 23, serial / parallel converters 12 and 14, a modulator 13, and a beamforming processing unit (transmission-side beamforming processing unit) 15.

  The transmission data is encoded by the LDPC encoder 23 and output. The encoded data is converted by the serial / parallel converter 12 and output. Since the subsequent configuration and operation are the same as those in the first embodiment (FIG. 1), the description thereof is omitted. However, unlike the first embodiment, the interleaver 11 is omitted, but is unnecessary because the LDPC encoder 23 substantially includes the interleaver.

<Example of receiver configuration>
FIG. 22 is a diagram illustrating a configuration example of the receiver 2 according to the second embodiment.

  The receiver 2 includes a propagation path estimation processing unit 16, a beamforming processing unit (receiving-side beamforming processing unit) 17, a MIMO demodulator 18, and a variable node decoding unit (VND) 24, and an interleaver 20. A node decoder (CND) 25 and an interleaver 11 are included. The operations of the propagation path estimation processing unit 16, the reception side beamforming processing unit 17, and the MIMO demodulator 18 are the same as those in the first embodiment. As the operation of the BICM-ID, as in Non-Patent Document 1, a combination of variable node processing and MIMO demodulation processing in the Sum-Product algorithm, which is a decoding algorithm for LDPC codes, and a check node decoder (Check Node Decoder) processing is connected through an interleaver and deinterleaver to perform repeated processing. Specifically, an LLR (Log Likelihood Ratio) output from the MIMO demodulator is processed in a variable node decoder 24, and an external LLR is output. The LLR output from the variable node decoder 24 is changed in bit order by the deinterleaver 20 and supplied to the check node decoder 25. The LLR processed by the check node decoder 25 is again passed to the variable node decoder 24 via the interleaver 11. The variable node decoder 24 calculates and outputs an a priori LLR supplied to the MIMO demodulator 18 using the LLR passed from the interleaver 11. Here, the method of changing the bit order in the deinterleaver 20 and the interleaver 11 is determined by the bit arrangement of the parity check matrix of the LDPC code.

<Configuration of variable node decoder>
FIG. 23 is a diagram illustrating an internal configuration example of a variable node decoder 24 used in the second embodiment. In FIG. 23, the variable node order is 4, but in the case of other orders, the number of switches 110 and the number of inputs / outputs of the parallel / serial converter 110 and serial / parallel converter 112 are almost the same. It is.

  When the variable node decoder 24 calculates the external LLR supplied to the deinterleaver 20 from the external LLR output from the MIMO demodulator 18, the switch 108 is on the upper side and the switches 110-1 to 110-4. Are used by switching one of the bits for calculating an extrinsic LLR to the upper side and the other to the lower side.

  The adder 109 supplies the external (extrinsic) LLR supplied from the MIMO demodulator 18 via the parallel-serial converter 106 and the LLR supplied from the interleaver 11 for bits other than the bit for calculating the external (extrinsic) LLR. (Pre-LLR) is added. At this time, by sequentially switching one of the switches 110-1 to 110-4 to the upper side and the other to the lower side and performing the same calculation, a plurality of extrinsic LLRs to be supplied to the deinterleaver 20 Is calculated and supplied to the deinterleaver 20 via the parallel / serial converter 111. When calculating an a priori LLR supplied from the a priori LLR supplied from the interleaver 11 to the MIMO demodulator 18, the switch 108 is on the lower side and the switches 110-1 to 110-4 are all on the lower side. Switch to the side. In this case, the adder 109 adds all the a priori LLRs supplied from the deinterleaver 20 via the check node decoder 25 and the serial / parallel converter 112, and passes through the serial / parallel converter 107. This is supplied to the MIMO demodulator 18.

  When decoding data (Rx data) is output after a sufficient number of repetitions, the switch 108 can be switched to the upper side and the switches 110-1 to 110-4 can all be switched to the lower side. The adder 109 adds and outputs all the external LLR supplied from the MIMO demodulator 18 and the prior LLR supplied from the interleaver.

<Check node decoder>
FIG. 24 is a diagram illustrating a configuration example of a check node decoder 25 used in the second embodiment.

  The check node decoder 25 includes a serial / parallel converter 113, switches 115-1 to 115-4, an LLR box sum calculator 116, and a parallel / serial converter 114. In FIG. 24, the case of the check node order 4 is shown, but in the case of other orders, only the number of switches 115 and the number of inputs / outputs of the parallel / serial converter 114 and serial / parallel converter 113 are changed. It is almost the same.

  In FIG. 24, the switches 114-1 to 114-4 are used by switching one of the bits for calculating an extrinsic LLR to the lower side and the other to the upper side.

  The LLR box sum calculator 116 calculates a prior LLR supplied from the deinterleaver 20 via the serial / parallel converter 113 for bits other than the bit for calculating the external LLR. The calculation of the LLR box sum calculator 116 is expressed by Expression 20.

  Further, the functions in Expression 19 are expressed by Expression 21 and Expression 22, respectively. This is the same as the check node process of the Sum-Product decoding process.

  Then, the same calculation is performed by sequentially switching one of the switches 115-1 to 115-4 to the lower side and the other to the upper side. Thereby, a plurality of external LLRs to be supplied to the interleaver 11 are calculated and supplied to the interleaver 11 via the parallel / serial converter 114.

(4) Modification (i) In the first and second embodiments, the effective propagation path matrix is changed by random beamforming processing that switches the beamforming matrix a plurality of times within one codeword (frame). This provides an BICM-ID communication system that averages the EXIT curve of the MIMO demodulation processing in the codeword (frame) and can restore the transmission signal without error even when the propagation path fluctuates.

  However, when the energy of the entire received code word fluctuates (there is a gain corresponding to the number of antennas provided and this gain fluctuates itself), the EXIT curve of the MIMO demodulation process fluctuates vertically in the vertical axis direction. Resulting in. In this case, for example, transmission power control such as that used in a CDMA (Code Division Multiple Access) method may be used in combination. That is, when the EXIT curve decreases due to energy fluctuation, control is performed to increase the power. Through the transmission power control, the energy of the entire received codeword becomes constant, and it is possible to prevent the EXIT curve of the MIMO demodulation process from fluctuating together with random beamforming.

(Ii) It is also preferable to use H-ARQ (Hybrid Auto Repeat reQuest) instead of transmission power control. In this case, FIG. 25 shows a flowchart of the transmission process, and FIG. 26 shows a flowchart of the reception process.

<Transmission processing: FIG. 25>
The transmitter 1 encodes transmission data for one frame (S2501).

  Next, the transmitter 1 divides the codeword obtained by the encoding process into N partial codes (S2502).

  Here, the transmitter 1 first sets the variable k = 1 (k = 1,..., N) (S2503), and starts transmitting the kth partial code (S2504).

  Then, the transmitter 1 determines whether an ACK (when decoding is successful on the receiving side) has been received from the receiver 2 (S2505). When the ACK is received, the data transmission process for one frame ends. When ACK is not received (when NACK is received), the process proceeds to S2506.

  In S2506, the transmitter 1 determines whether k = N. If k = N, the process proceeds to S2507, and the transmitter 1 again executes transmission of the first partial code in the frame with k = 1 (S2504). On the other hand, if k = N is not the case, the process proceeds to S2508, and the transmitter 1 sets k = k + 1 and executes transmission of the next partial code in the frame (S2504).

<Reception processing: FIG. 26>
The receiver 2 receives data for one frame in units of partial codes. At this time, first, a variable k = 1 (k = 1,..., N) is set (S2601), and the kth part The code is received (S2602).

  Next, the receiver 2 performs a decoding process using the partial code received so far (S2603).

  Then, the receiver 2 determines whether or not the decoding is successful for the frame (S2604). When decoding is successful (even if a partial code for one frame has not been received), the process proceeds to S2605, and the receiver 2 transmits an ACK to the transmitter 1 (S2605). finish. On the other hand, if the data for one frame cannot be decoded, the process proceeds to S2606, and the receiver 2 transmits a NACK to the transmitter 1.

  Subsequently, the receiver 2 determines whether k = N (S2607). When k = N, the process proceeds to S2608, and the receiver 2 executes the reception process from the first partial code again in the frame with k = 1 (S2602). On the other hand, if k = N is not the case, the process proceeds to S2609, and the receiver 2 sets k = k + 1 and executes the reception process of the next partial code in the frame (S2602).

  As described above, in H-ARQ, the divided codeword is gradually transmitted, and the codeword is repeatedly transmitted until energy necessary for decoding is received on the receiving side. If reception is not successful even after transmission of one codeword is completed, transmission of the same codeword is continued again. In this case, the transmission side beamforming processing unit 15 and the reception side beamforming processing unit 17 are transmitted using a larger number of effective channel matrices than when the transmission side beamforming processing unit 15 and the reception side beamforming processing unit 17 are operated so as to have a beamforming matrix different from the first time. Is preferable.

(5) Summary (i) In the present invention, the MIMO communication method is used, and the transmitting side performs different beam forming processing on each bit of the modulated data in one frame and transmits the data. On the receiving side, the estimated propagation path and the beam forming matrix used at the time of transmission are multiplied by the same matrix and used for MIMO demodulation. By doing so, the EXIT curve in one frame can be averaged during decoding and demodulation, and the signal can be restored without error on the receiving side even when the propagation path varies. .

In the present invention, a plurality of different beamforming matrices include a plurality of beamforming weight vectors constituting the same. This weight vector corresponds to a column vector of the beamforming matrix, and is set to take different values exceeding the number m of antennas. By doing so, it is possible to guarantee that a certain transmission signal is received with a dominant strength on the receiving side, and it is possible to perform restoration on the receiving side without error. For example, when a beamforming matrix having a complex coefficient is used, at least one of a plurality of different beamforming matrices has an absolute value of an inner product of the arbitrary complex unit vector with the complex unit vector larger than 1 / √2. It is preferable to include a beamforming weight vector. More specifically, it means that the absolute value of λν ₁ ′ b ₂ in Equation 19 above is greater than 1 / √2.

  In the first embodiment, an interleave process is performed on encoded data for one frame, and the encoded data for one frame after the interleave process is modulated. Then, the beam forming process is executed by multiplying each modulation symbol of the modulation data for one frame after the interleaving process by a different beam forming matrix. Encoding is a process that associates codeword bits, but by interleaving and modulating, the correlation between adjacent bits in one frame can be ignored, and LLR computation is performed in the demodulation and decoding processes. Can be regarded as an operation for an independent event. Even if the bit data in one frame after interleaving includes an error, information can be obtained from the relevant bit data in one frame before interleaving, so that demodulation is performed via the interleaver. By repeatedly executing the decoding process, information is gradually collected, and the transmission data restoration process on the receiving side can be accurately executed.

  On the other hand, in the second embodiment, encoded data is generated by performing LDPC encoding on transmission data of an encoding unit (for one frame). In this case, unlike the first embodiment, the interleaving process is not executed on the transmission side.

(Ii) The transmission process and the reception process according to the above-described embodiment of the present invention can be realized by storing a program corresponding to each process in a memory and executing the program by the processor. .

  In this case, a storage medium in which the program code is recorded is provided to the system or apparatus, and the computer (or CPU or MPU) of the system or apparatus reads the program code stored in the storage medium. In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiments, and the program code itself and the storage medium storing the program code constitute the present invention. As a storage medium for supplying such program code, for example, a flexible disk, CD-ROM, DVD-ROM, hard disk, optical disk, magneto-optical disk, CD-R, magnetic tape, nonvolatile memory card, ROM Etc. are used.

  Also, based on the instruction of the program code, an OS (operating system) running on the computer performs part or all of the actual processing, and the functions of the above-described embodiments are realized by the processing. May be. Further, after the program code read from the storage medium is written in the memory on the computer, the computer CPU or the like performs part or all of the actual processing based on the instruction of the program code. Thus, the functions of the above-described embodiments may be realized.

  Further, by distributing the program code of the software that realizes the functions of the embodiment via a network, it is stored in a storage means such as a hard disk or memory of a system or apparatus, or a storage medium such as a CD-RW or CD-R And the computer (or CPU or MPU) of the system or apparatus may read and execute the program code stored in the storage means or the storage medium when used.

  Finally, it should be understood that the processes and techniques described herein are not inherently related to any particular apparatus, and can be implemented by any suitable combination of components. In addition, various types of devices for general purpose can be used in accordance with the teachings described herein. It may prove useful to build a dedicated device to perform the method steps described herein. Various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiments. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, constituent elements over different embodiments may be appropriately combined.

  Although the present invention has been described with reference to specific examples, these are in all respects illustrative rather than restrictive. Those skilled in the art will appreciate that there are numerous combinations of hardware, software, and firmware that are suitable for implementing the present invention. For example, the described software can be implemented in a wide range of programs or script languages such as assembler, C / C ++, perl, shell, PHP, Java (registered trademark).

  Furthermore, in the above-described embodiment, control lines and information lines are those that are considered necessary for explanation, and not all control lines and information lines on the product are necessarily shown. All the components may be connected to each other.

  In addition, other implementations of the invention will be apparent to those skilled in the art from consideration of the specification and embodiments of the invention disclosed herein. Various aspects and / or components of the described embodiments can be used singly or in any combination in a computerized storage system capable of managing data. The specification and specific examples are merely exemplary, and the scope and spirit of the invention are indicated in the following claims.

DESCRIPTION OF SYMBOLS 1 ... Transmitter 2 ... Receiver 3 ... Antenna 10 ... Encoder 11 ... Interleaver 12, 14, 22, 107, 112, 113 ... Serial to parallel converter 13 .. Modulator 15... Transmission side beamforming processing unit 16... Propagation path estimation processing unit 17... Reception side beamforming processing unit 18... MIMO demodulator 19, 106, 111, 114. Parallel-serial converter 20 ... deinterleaver 21 ... decoder 23 ... LDPC encoder 24 ... variable (variable) node decoder 25 ... check node decoder 100 ... puncture and Serial / parallel converter 101... Pseudorandom number generator 102... Beam forming matrix table 103... Matrix × vector multiplier 104. X matrix multipliers 105-1 to 105-7 ... matrix table 108, 110-1 to 110-4, 115-1 to 115-4 ... changeover switch 109 ... adder 116 ... LLR box Sum calculator

Claims (12)

A wireless communication method for encoding and modulating transmission data and transmitting from m (m is an integer of 2 or more) antennas,
An encoding step of encoding the transmission data for each encoding unit to generate encoded data;
A modulation step of modulating the encoded data to generate modulation data;
A transmission-side beamforming process execution step for executing beamforming process by multiplying each of the modulated data generated from the transmission data of one coding unit by a plurality of different beamforming matrices;
Transmitting the beamformed modulation data from the m antennas, and
The wireless communication method characterized in that a plurality of beamforming weight vectors constituting the plurality of different beamforming matrices take a number of different values exceeding m. In claim 1,
When a beamforming matrix having a complex coefficient is used, at least one of the plurality of different beamforming matrices is a beam whose absolute value of the inner product of the arbitrary complex unit vector and the complex unit vector is greater than 1 / √2. A wireless communication method comprising a forming weight vector. In claim 1,
And an interleaving process execution step for performing an interleaving process on the encoded data,
In the modulation step, the encoded data after the interleaving process is modulated to generate modulated data,
In the beam forming process execution step, each of the modulated data generated after the interleaving process is multiplied by the plurality of different beam forming matrices. In claim 1,
In the encoding step, the encoded data is generated by performing LDPC encoding on transmission data of the encoding unit. The claim 1, further comprising:
A receiving step of receiving the modulated data transmitted by the transmitting step by the m antennas;
A propagation path estimation processing step for estimating a propagation path from the received modulation data and outputting a propagation path matrix;
An effective propagation path matrix generation step of generating an effective propagation path matrix by multiplying the propagation path matrix by the same beamforming matrix as the beam forming matrix used in the transmission side beamforming process execution step;
A demodulation step of generating demodulated data by performing MIMO demodulation on the received modulated data using the effective channel matrix and a decoding result by a decoder;
Decoding the demodulated data and generating received data, and
A radio communication method, wherein a plurality of beamforming weight vectors constituting a beamforming matrix used on a receiving side take a number of different values exceeding m. A wireless transmission device that encodes and modulates transmission data and transmits m (m is an integer of 2 or more) antennas,
An encoding processor that encodes the transmission data for each encoding unit to generate encoded data;
A modulation processing unit that modulates the encoded data to generate modulation data;
A transmission-side beamforming processing unit that performs beamforming processing by multiplying each of the modulation data generated from the transmission data of one coding unit by a plurality of different beamforming matrices;
A transmitter that transmits the beam-formed modulated data from the m antennas,
A plurality of beamforming weight vectors constituting the plurality of different beamforming matrices take a number of different values exceeding m. In claim 6,
When a beamforming matrix having a complex coefficient is used, at least one of the plurality of different beamforming matrices is a beam whose absolute value of the inner product of the arbitrary complex unit vector and the complex unit vector is greater than 1 / √2. A wireless transmission device including a forming weight vector. In claim 6,
And an interleaving unit that performs an interleaving process on the encoded data.
The modulation unit modulates the encoded data after the interleaving process to generate modulation data,
The radio transmission apparatus, wherein the beamforming processing unit multiplies each of the modulated data generated after the interleaving process by the plurality of different beamforming matrices. In claim 6,
The radio transmission apparatus, wherein the encoding processing unit generates the encoded data by performing LDPC encoding on the transmission data of the encoding unit. A wireless receiver that receives data transmitted from the wireless transmitter according to claim 6,
A receiving unit that receives the modulation data transmitted by the wireless transmission device by the m antennas;
A propagation path estimation processing unit that estimates a propagation path from the received modulation data and outputs a propagation path matrix;
An effective channel matrix generation unit that multiplies the channel matrix by the same beamforming matrix as the beamforming matrix used by the transmission side beamforming processing unit, and generates an effective channel matrix;
A MIMO demodulator that generates demodulated data by demodulating the received modulated data using the effective channel matrix and a decoding result by a decoder;
A decoding unit that decodes the demodulated data and generates received data,
A radio receiving apparatus characterized in that a plurality of beamforming weight vectors constituting a beamforming matrix used on the receiving side take a number of different values exceeding m. In claim 10,
When a beamforming matrix having a complex coefficient is used, at least one of the plurality of different beamforming matrices is a beam whose absolute value of the inner product of the arbitrary complex unit vector and the complex unit vector is greater than 1 / √2. A wireless receiver comprising a forming weight vector. A wireless communication system having a wireless transmission device and a wireless reception device,
The wireless transmission device
m (m is an integer of 2 or more) antennas;
An encoding processing unit that encodes transmission data for each encoding unit to generate encoded data;
A modulation processing unit that modulates the encoded data to generate modulation data;
A transmission-side beamforming processing unit that performs beamforming processing by multiplying each of the modulation data generated from the transmission data of one coding unit by a plurality of different beamforming matrices;
A transmitter that transmits the beam-formed modulated data from the m antennas,
The wireless receiver is
a receiving unit that receives modulation data transmitted by the wireless transmission device using m (m is an integer of 2 or more) antennas;
A propagation path estimation processing unit that estimates a propagation path from the received modulation data and outputs a propagation path matrix;
An effective channel matrix generation unit that multiplies the channel matrix by the same beamforming matrix as the beamforming matrix used by the transmission side beamforming processing unit, and generates an effective channel matrix;
A MIMO demodulator that generates demodulated data by demodulating the received modulated data using the effective channel matrix and a decoding result by a decoder;
A decoding unit that decodes the demodulated data and generates received data,
A wireless communication system, wherein a plurality of beamforming weight vectors constituting a beamforming matrix used in both the wireless transmission device and the wireless reception device take a number of different values exceeding m.

Images