JP2008252301A - Radio communication device and transmission method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a radio communication device which compensates IQ imbalance without impairing transmission efficiency. <P>SOLUTION: The device estimates a propagation path response at each subcarrier from a known signal for propagation path estimation included in an orthogonal demodulated signal obtained by carrying out the orthogonal demodulation of a received OFDM (Orthogonal Frequency Division Multiplexing) signal, carries out the Fourier transformation of the orthogonal demodulation signal, and obtains the received signal at each subcarrier. The device demodulates the signal transmitted by: the first and the second subcarriers using a first subcarrier whose transmission path estimation precision is relatively low among a plurality of subcarriers; the received signal of the second subcarrier which is arranged in a symmetric position of the first subcarrier making a center frequency in a frequency band of the OFDM signal as an axis; and the transmission path response of a third subcarrier whose transmission path estimation precision is relatively high in the plurality of subcarriers, and which is arranged near the first subcarrier, and the fourth subcarrier which is arranged in a symmetric position of the third subcarrier making the center frequency as an axis. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a wireless communication apparatus.

  If the frequency bandwidth used for communication is expanded for the purpose of speeding up wireless communication, the propagation delay time difference of the multipath propagation path cannot be ignored. In an environment where signals having different propagation delay times arrive, waveform distortion due to intersymbol interference is a major factor that degrades communication quality. In such an environment, the Orthogonal Frequency Division Multiplexing (OFDM) transmission system is known as a system that can compensate for waveform distortion caused by intersymbol interference even when signals having different propagation delay times are received. It has been.

  Since signals are complex signals in the OFDM transmission system, it is necessary to use a quadrature modulator in the transmitter and a quadrature demodulator in the receiver. At this time, if a difference occurs in the amplitude of the in-phase component and the quadrature component in the quadrature modulator and the quadrature demodulator, or if a phase error occurs in the 90 ° phase shifter, the frequency axis with respect to the center frequency among the plurality of subcarriers of the OFDM signal Two subcarriers that are symmetrically above each other (among those two subcarriers, subcarriers in the higher frequency band than the center frequency are subcarriers in the upper sideband, and subcarriers in the lower frequency band than the center frequency are Signals “sometimes referred to as“ lower sideband subcarriers ”) interfere with each other, greatly limiting transmission performance.

Under such circumstances, a technique has been proposed in which the amount of interference between upper and lower subcarriers is estimated, and signals are determined by maximum likelihood estimation or spatial filtering using both signals received by the upper and lower subcarriers.
A. Tarighat, R. Bagheri, and AHSayed, “Compensation Schemes and Performance Analysis of IQ Imbalances in OFDM Receivers,” IEEE Trans. Signal Processing, vol.8, no.8, pp.3257-3268, Aug.2005. Hiroyuki Kamada, Kei Sakaguchi, Junmichi Araki, "Time IQ Imbalance Compensation Method in MIMO-OFDM," IEICE Technical Report, A / P2005-94, pp.55-60, Oct.2005.

  If the method of Non-Patent Document 1 is used, distortion caused by the quadrature modulator / demodulator can be compensated with high accuracy. However, it is necessary to perform propagation path responses of signals that interfere with each other between upper and lower subcarriers that are symmetrical with respect to the center frequency. Therefore, a signal twice as long as a known signal for propagation path estimation required in a normal OFDM transmission system is required, and there is a problem that transmission efficiency deteriorates.

  On the other hand, in the method of Non-Patent Document 2, the propagation path response of a signal that interferes between upper and lower subcarriers symmetrical with respect to the center frequency is realized by estimating the impulse response in the time domain.

  Since Non-Patent Document 2 requires only a known signal for channel estimation having the same length as that of conventional OFDM transmission, there is no deterioration in transmission efficiency. However, in this method, the accuracy of propagation path estimation depends on the signal sequence of known signals for propagation path estimation. Therefore, depending on the signal series, the propagation path estimation accuracy deteriorates, and if such a propagation path estimation result is used for correction of waveform distortion, the transmission characteristics are greatly degraded. Also, depending on the radio communication system, it is possible to take measures to improve throughput and communication quality on the transmission side by feeding back the propagation path estimation result to the transmission side. Transmission characteristics will be greatly degraded.

  As described above, the conventional wireless communication apparatus has a problem that the transmission performance is largely limited by the accuracy of the amplitude and phase of the quadrature modulator / demodulator.

  In addition, the conventional wireless communication device requires a known signal for propagation path estimation longer than usual when compensating for waveform distortion caused by imperfection of the quadrature modulator / demodulator, which degrades transmission efficiency. There was a problem.

  In addition, when trying to estimate the channel response using a short known channel estimation signal, the channel estimation accuracy depends on the sequence of the known channel estimation signal. Will be obtained. If such a propagation path estimation result is used for correction of waveform distortion or feedback to the transmission side, transmission characteristics may be greatly degraded.

  In view of the above problems, the present invention is based on the incompleteness of the quadrature modulator / demodulator, and therefore, between two subcarriers (upper and lower sideband subbands) that are symmetrical with each other on the frequency axis with the center frequency as an axis. It is an object of the present invention to provide a wireless communication apparatus capable of compensating for interference occurring between carriers) without degrading transmission efficiency and improving transmission characteristics.

  The wireless communication device includes: (a) a reception unit that receives an OFDM (Orthogonal Frequency Division Multiplexing) signal including a plurality of subcarriers; and (b) an orthogonal demodulation unit that orthogonally demodulates the OFDM signal to obtain an orthogonal demodulated signal; (C) propagation path estimation means for estimating a propagation path response for each subcarrier from a known signal for propagation path estimation included in the orthogonal demodulated signal; and (d) Fourier transform the orthogonal demodulated signal for each subcarrier. (E) a first subcarrier having a relatively low propagation path estimation accuracy among the plurality of subcarriers and a center frequency of the frequency band of the OFDM signal as an axis. The received signal with the second subcarrier arranged at a position symmetrical to one subcarrier and the first sub with relatively high propagation path estimation accuracy among the plurality of subcarriers Using the third subcarrier arranged near the carrier and the propagation path response of the fourth subcarrier arranged at a position symmetrical to the third subcarrier with the center frequency as an axis, Demodulating means for demodulating signals transmitted on the first and second subcarriers.

  According to the present invention, the interference generated between two subcarriers located symmetrically on the frequency axis with the center frequency as an axis due to the incompleteness of the orthogonal modulator / demodulator is compensated without degrading the transmission efficiency.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

(First embodiment)
FIG. 1 shows an example of the configuration of a transmission apparatus according to the first embodiment. In FIG. 1, a transmission signal is generated by applying a channel coding to an information signal to be transmitted by an encoder 181 as shown in FIG. Since redundancy is added to the signal by performing channel coding in this way, the transmission speed of the information signal is deteriorated, but the communication quality is improved by error correction, resulting in high throughput characteristics as a result. There is expected.

  The encoding method used by the encoder 181 is any encoding method such as a Reed-Solomon code, a convolutional code, a turbo code, or LDPC (Low Density Parity Check Codes) as long as the wireless receiver is known. May be. Also, a plurality of encoding methods may be implemented and the encoding method may be changed for each frame to be transmitted. Note that, depending on the encoding method used in the encoder 181, the correlation between adjacent codes is high, and therefore it is preferable to rearrange signals using an interleaver 191 as shown in FIG. 3. At this time, the interleaver 191 may perform rearrangement according to any rule as long as the wireless reception device is a known rearrangement method.

  The transmission signal generated as described above is modulated for each subcarrier in the modulator 101. Modulation methods used in the modulator 101 include phase modulation such as BPSK, QPSK, and 8PSK, and quadrature amplitude modulation such as 16QAM, 64QAM, and 256QAM. However, in the present embodiment, the modulation method in the modulation unit 101 is not limited to the above-described method. The modulation unit 101 may use any of the modulation schemes described above as long as the modulation scheme is known by the wireless reception device, or may use a modulation scheme other than these.

  In general, in OFDM transmission, information is not transmitted on all subcarriers, but is received on some subcarriers in order to correct local frequency shift, phase shift, and propagation path fluctuation between the transmitter and receiver. The device sends a known signal. Here, the subcarrier is called a pilot subcarrier, and the subcarrier that transmits information is called a data subcarrier.

  Pilot generation section 111 generates a predetermined sequence of signals for transmission on pilot subcarriers.

  The inverse Fourier transform unit 121 converts the data subcarrier signal and the pilot subcarrier signal into a time domain signal. Here, the inverse Fourier transform in the inverse Fourier transform unit 121 may use IDFT (Inverse Digital Fourier Transform) or IFFT (Inverse Fast Fourier Transform). Any method may be used as long as the discrete inverse Fourier transform is applied to the transmission symbol generated for each subcarrier. Further, even if the signal is cyclically shifted in the time domain and output by shifting, the shift amount may be the same within one frame.

  Each of the in-phase component (I component or I channel signal) signal and the quadrature component (Q component or Q channel signal) signal obtained by performing the inverse Fourier transform in the inverse Fourier transform unit 121 is GI (Guard Interval, A guard interval) input unit 131 inputs a signal called a guard interval or a cyclic prefix for each OFDM symbol. This is a technique commonly used in OFDM transmission to maintain signal periodicity and avoid intersymbol interference in the frequency domain when multipath signals with different propagation delay times arrive. Description is omitted.

  In addition to the signal information described above, in wireless communication, generally known signals for synchronization and propagation path estimation are transmitted. As an example of a communication system that transmits such a known signal, a frame format in IEEE802.11a will be described with reference to FIG. In FIG. 13, the SP (short preamble) 411 at the head of the frame is a known signal for synchronization, and the subsequent LP (long preamble) 421 is a known signal for propagation path estimation. When transmitting such a known signal, a switch that connects either the GI adding unit 131 or the known signal generating unit 121 to the radio unit 151 and connecting the known signal generating unit 121 and the radio unit 151 to each other. The wireless unit 151 transmits the known signal generated by the known signal generation unit 121.

  The transmission of the known signal has been described above by taking the IEEE 802.11a frame format as an example. However, the signal that can be received by the wireless communication apparatus according to the present embodiment is not limited to IEEE 802.11a.

  The I component signal and the Q component signal including the information signal and the known signal are input to the radio unit 151.

  FIG. 15 shows a configuration example of the radio unit 151. A digital-to-analog converter (DAC) 151a, a low-pass filter (LPF) 151b, a quadrature modulator 151c, a frequency converter 151d, and an amplifier 151e are included. Including.

  The DAC 151a converts each of the input I component signal and Q component signal into analog signals, removes unnecessary high frequency components from the I component and Q component analog signals obtained by the LPF 151b, and then performs quadrature modulation. A quadrature modulation is performed by the device 151c to generate a quadrature modulation signal. Further, the quadrature modulation signal is converted into a radio frequency signal by the frequency converter 151 d and then transmitted through the amplifier 151 e and the transmission antenna 161.

  As the transmission antenna 161, any antenna may be used as long as it can transmit a signal in a desired frequency band.

  FIG. 11 shows a configuration example of the receiving apparatus according to the first embodiment that receives a signal transmitted from the transmitting apparatus of FIG.

  11 includes a receiving antenna 261, a radio unit 251, a GI removing unit 201, a propagation path estimating unit 221, a Fourier transform unit 211, a received signal combining unit 231, a phase correcting unit 241, and a demodulating unit 251.

  The receiving antenna 261 may be any antenna as long as it can receive a signal in a desired frequency band.

  FIG. 16 shows a configuration example of the wireless unit 251. The radio unit 251 includes an amplifier 251a, a frequency converter 251b, a quadrature demodulator 251c, a low-pass filter (LPF) 251d, and an analog-to-digital converter (ADC) 251e.

  The radio frequency signal received by the receiving antenna 261 is input to the frequency converter 251b via the amplifier 251a, converted into an intermediate frequency signal, and then orthogonally demodulated by the orthogonal demodulator 251c to generate a baseband signal ( I component and Q component analog signals). The obtained analog signals of I component and Q component are converted into digital signals of I component and Q component by ADC 251e after unnecessary high frequency components are removed by LPF 251d.

  The wireless unit 251 outputs digital signals of I component and Q component.

  In FIG. 16, the received radio frequency signal is once converted into an intermediate frequency signal and then converted into a complex baseband signal. However, the present invention is not limited to this. A radio frequency signal may be directly converted into a baseband signal without using an intermediate frequency as long as it can be converted into a baseband signal in a complex representation. In this case, the frequency converter 251b is not necessary.

  In FIG. 16, analog quadrature demodulation is performed. However, the present invention is not limited to this, and quadrature demodulation may be performed after first being converted into a digital signal by ADC.

  As described above, the I component signal and the Q component signal output from the radio unit 251 are respectively input to the GI removal unit 201. As described in the explanation of the transmitting apparatus, a guard interval is generally added in OFDM transmission. GI removal section 201 in FIG. 11 removes a signal having a guard interval length added at the time of transmission from each of the input I component signal and Q component signal.

  The I component signal and Q component signal from which the guard interval has been removed are input to the Fourier transform unit 211. In OFDM transmission, a signal is transmitted using a plurality of subcarriers, and a signal transmitted on each subcarrier can be extracted by applying a discrete Fourier transform. The Fourier transform unit 211 in FIG. 11 applies a discrete Fourier transform for each signal of 1 OFDM symbol length to the I component signal and the Q component signal from which the guard interval is removed. Here, the discrete Fourier transform may be calculated using DFT (Digital Fourier Transform) or may be calculated using FFT (Fast Fourier Transform). Any method may be used as long as discrete Fourier transform can be performed for each OFDM symbol.

  Next, a signal output from the Fourier transform unit 211 will be described.

  In the signal transmitted from the transmission apparatus of FIG. 1, the analog signal of the I component (I channel) and the Q component (Q channel) is quadrature modulated by the quadrature modulator 151c of the radio unit 151 and then the radio frequency of the signal is transmitted by the frequency converter 151e. Although it is converted into a signal and transmitted, when the quadrature modulator 151c is configured by an analog circuit, it is generally difficult to keep the gains of the I channel and Q channel the same. Further, when an individual difference occurs between the I-channel digital-analog converter and the Q-channel digital-analog converter, it is equivalent to adding different gains to the I-channel and the Q-channel. Furthermore, it is difficult to accurately generate a phase difference of 90 degrees when generating an I-channel signal and a Q-channel signal in an orthogonal modulator. As a result, in the output of the quadrature modulator 151c, the I channel and the Q channel have different amplitudes, and the phase difference also deviates from 90 degrees.

The above phenomenon can be modeled as shown in FIG. In FIG. 17, u _i (t) and u _q (t) represent baseband signals transmitted on the I channel and the Q channel, respectively.

Multipliers 531 and 532 multiply these baseband signals by a local frequency signal generated from transmitter 511 and a local frequency signal obtained by shifting the phase of the local frequency signal by 90 degrees.

In FIG. 17, as described above, the gain difference between the I channel signal and the Q channel signal including the individual difference of the DAC is represented by the gains G _i ^(t) and G _q that are different for the I channel signal and the Q channel signal, respectively. Expressed by multiplying ^(t) . In FIG. 17, the phenomenon in which the phase difference between the I channel signal and the Q channel signal described above does not become 90 degrees is expressed by multiplying the Q channel by a sine wave whose phase is shifted by θ ^(t) . .

As a result, the output of the quadrature modulator 151c is expressed by the following equation (1).

However, f represents a local frequency. When the above signals are expressed in a matrix in an equivalent low-frequency system, they can be expressed as the following equation (2).

Here, m _i (t) and m _q (t) represent I-channel and Q-channel transmission signals, respectively, and an equivalent low-band complex transmission signal m (t) = m _i ( t) + jm _q (t) (where j ² = −1)
Can be represented by the following formulas (3), (4), (5), (6).

In this way, by using the complex conjugate u ^* (t) of the transmission signal u (t), the signal can be expressed in a complex manner even in an environment where the orthogonality is lost due to the influence of IQ imbalance. As can be seen from equation (3), the transmission signal is not only distorted by α ^(t) due to the influence of IQ imbalance, but also the complex conjugate signal u ^* (t) is radiated unnecessarily. . As will be described later, this complex conjugate signal becomes interference, which limits the performance of OFDM transmission (for example, equations (19) to (21)).

  Next, IQ imbalance in the quadrature demodulator 251c in the receiving apparatus will be considered. FIG. 18 shows the IQ imbalance generated by the quadrature demodulator 251c modeled by imperfection of the quadrature demodulator 251c.

  In the quadrature demodulator 251c, two sine waves (the local frequency signal generated from the transmitter 512 and the local frequency signal obtained by shifting the phase of the local frequency signal by 90 degrees) are different in phase by 90 degrees from the input received signal. By multiplying each and applying LPF 251d, I channel signal and Q channel signal are obtained.

  However, the quadrature demodulator 251c, like the quadrature modulator 151c, is difficult to accurately generate a phase difference of 90 degrees, and the I channel and the Q channel depend on the gain of the filter and the individual difference of the analog-digital converter. Generally, the gains of these are also different.

In FIG. 18, the band signal is
y _i (t) cos (2πft) −y _q (t) sin (2πft)
In other words, the I channel signal r _i (t) and the Q channel signal r _q (t) output from the quadrature demodulator 251c can be expressed by the following equation (7).

Therefore, an equivalent low-frequency received signal in consideration of the influence of IQ imbalance r (t) = r _i (t) + jr _q (t)
Can be expressed by the following formulas (8), (9), (10), and (11).

As described above, in the quadrature demodulator, as in the quadrature modulator, an equivalent low-frequency signal in an environment where the orthogonality is lost can be expressed in a complex manner. Further, in the receiver as well as the transmitter, not only the distortion of α ^(r) expressed by the equation (10) occurs in the received signal due to the influence of IQ imbalance but also the complex conjugate as expressed by the equation (8). It can be seen that the signal y ^* (t) is added.

Next, the influence of the multipath propagation path between the transmission device and the reception device will be described.
When the number of paths of the propagation path is L, the delay time of each path is τ ₁ , and the complex amplitude of each path is h ₁ , the received signal y (t) can be expressed by the following equation (12).

When the influence of the distortion of the transmission apparatus represented by Expression (3) is added to Expression (12), the following Expression (13) is obtained.

Further, the influence of IQ imbalance in the multipath propagation environment is expressed by the following expressions (14), (15), and (16) by adding the influence of IQ imbalance of the receiving apparatus represented by Expression (8). Can do.

From the equation (14), similarly to the case where the distortion of the quadrature modulator or the quadrature demodulator is shown, even if the influence of the propagation path response is taken into consideration, the influence of the IQ imbalance is the transmission signals u (t) and u ^* (t ).

  Finally, the influence of IQ imbalance when performing OFDM transmission will be described.

In the case of OFDM transmission, the received signal represented by Equation (14) is Fourier transformed and converted into a frequency domain signal, and then demodulated. The relationship of the following formulas (17) and (18) is established between the time shift rule of Fourier transform and the Fourier transform of the complex conjugate signal.

Therefore, the signal x ^(k) received by the k-th subcarrier from the equations (14), (17), and (18) can be expressed as the following equations (19), (20), and (21). .

Here, h ^(k) and h ^(−k) represent the propagation path responses of the k-th and −k-th subcarriers not including the influence of IQ imbalance, respectively. s ^(k) and s ^(−k) represent modulated signals transmitted on the k-th and −k-th subcarriers, respectively.

  As described above, when IQ imbalance occurs in OFDM transmission, the signal output from the Fourier transform unit in the receiving apparatus outputs a signal obtained by interfering with the signals of subcarriers at symmetrical positions with the center frequency as an axis. I understand. FIG. 19 shows this phenomenon.

  The reception signal of the -kth subcarrier interferes with the reception signal of the kth subcarrier, and the signal of the kth subcarrier interferes similarly in the -kth subcarrier. Therefore, since the received signal of the kth subcarrier and the received signal of the -kth subcarrier include the signals transmitted on the kth and -kth subcarriers, the two subcarriers By simultaneously estimating signals transmitted on two subcarriers using the received signal, it is possible to perform demodulation even in an environment where distortion due to IQ imbalance occurs.

As shown in the following equation (22), a complex conjugate of the received signal of the −kth subcarrier is obtained, and a received signal vector obtained by combining the received signals of the two subcarriers is considered.

Here, n ^(k) and n ^(−k) represent vectors obtained by combining noise vectors in the kth subcarrier and the −kth subcarrier.

  Equation (22) is equivalent to a MIMO (Multiple Input Multiple Output) signal having a spatial multiplexing number of “2” and a receiving antenna number of “2”. If the propagation path response can be estimated, reception is performed with two subcarriers. It is possible to simultaneously estimate signals transmitted on two subcarriers using the obtained signal. At this time, the same algorithm as that used in conventional MIMO can be applied as the reception algorithm, and the influence of IQ imbalance can be compensated by estimating the signal in this way.

  However, the known signal for propagation path estimation is generally designed on the assumption that the signals of the kth subcarrier and the −kth subcarrier interfere, as in the IEEE802.11a frame format shown in FIG. Not. Therefore, it is impossible to distinguish between the channel response of a signal that interferes with another subcarrier and the channel response of a signal transmitted on the subcarrier, and signals of two subcarriers cannot be estimated simultaneously.

Here, a time domain signal before Fourier transform is performed by the Fourier transform unit 211 shown in Expression (14) will be considered. Here, there are L signals (1 = 0 to L−1) to be estimated by the channel response including the influence of IQ imbalance, and the number of guard interval samples is the OFDM transmission including pilot subcarriers. If the number of subcarriers to be used is less than half of the number of subcarriers to be used and the number of paths L of the multipath propagation path is smaller than the number of samples in the guard interval, the propagation path estimation known signal (propagation path estimation known signal 421) and its complex impulse response h _l of conjugate signal ^'and it is possible to estimate the h ^_l''.

The propagation path estimation unit 221 accumulates the received signal of the section in which the propagation path estimation known signal 421 is transmitted for one OFDM symbol (one OFDM symbol includes N samples r (0) to r (N−1)). The reception vector is obtained as shown in the following equation (23).

Further, the number of guard interval samples is set to N _g , and the signal u (t) in the section in which the propagation path estimation known signal is transmitted is set to L _p (t), thereby using Equation (14). (23) can be expressed as the following equation (24).

Thus, by using a generalized inverse matrix of the matrix A as shown in the following equation (28) can be estimated as impulse response h _l ^{'h _l''} a (l = 0~L-1).

By Fourier transforming h _l ^'and h ^_l'' obtained as described above, it is possible to estimate the channel response h α ^_(k) and h β ^_(k) of each subcarrier.

That is, by performing a Fourier transform on h _l ^′ , a propagation path response h _α ^{(k) of} a signal transmitted on the kth subcarrier and received on the kth subcarrier is obtained as shown in FIG. . Further, by ^performing Fourier transform on h _{l ″} , as shown in FIG. 19, the propagation path response h _β ^(k) of the signal transmitted on the −k th subcarrier and received on the k th subcarrier is obtained. can get.

Here, the Fourier transform may apply the discrete Fourier transform according to the equations (30) and (31), and fills in “0” until the number of samples of the OFDM symbol is equal, and applies the fast Fourier transform. It doesn't matter. If it is possible to obtain the resulting h _l ^'and h _l channel response of each subcarrier ^from'' h ^{α _(k)} and h β ^_(k) may be used any method.

  As described above, the method using the least square method as the channel response estimation method of the channel estimation unit 221 has been described. However, the channel estimation unit in the present invention is not limited to the least square method. The noise power may be estimated and the mean square error minimum method may be used, or any method can be used as long as the time domain signal can be used to estimate the frequency domain propagation response. I do not care.

  On the other hand, the local frequencies generated by the transmitter 511 (FIG. 17) used in the quadrature modulator 151c of the transmission device and the transmitter 512 (FIG. 18) used in the quadrature demodulator 251c of the reception device are exactly the same frequency. It is difficult to set, and generally there is a frequency offset between the transmission device and the reception device. Further, different phase noises are generated in the radio unit 151 of the transmitting end device and the radio unit 251 of the receiving device. Due to the influence of these frequency offset and phase noise, the received signal is subjected to a different phase error for each symbol.

Thus, before demodulation, the phase correction unit 241 corrects the phase error using the pilot subcarrier signal for each symbol. The received signal of Equation (22) that has received the phase error ψ can be expressed by the following Equation (32).

Since the signal transmitted by the pilot subcarrier is a known signal, a replica signal can be created as shown in the following equation (37) using the estimated propagation path response.

The inner product of the replica signal of Expression (37) and the received signal represented by Expression (32) is obtained as shown in the following Expression (39).

  Assuming that the propagation path estimation unit 221 correctly estimates the propagation path estimation, the part enclosed in parentheses in the first term on the right side of the equation (39) is in a Hermitian form, and thus becomes a real number. If the noise component of the second term on the right side is negligible, Ψ can be estimated by obtaining the declination of equation (39).

  As described above, the phase error can be estimated using only one pilot subcarrier. However, in order to reduce the influence of noise, the estimation may be performed using a plurality of pilot subcarriers. In this case, the phase error can be estimated by obtaining the sum angle of the equation (39) obtained for each pilot subcarrier.

The phase correction unit 241 corrects the phase error by multiplying all the data subcarriers by the inverse characteristic e ^−jΨ of the phase error estimated as described above, as shown in the following equation (40).

  The method for estimating and correcting the phase error for each OFDM symbol has been described above, but the estimation is performed by recursively adding the phase error estimated in the past or the inner product of the replica signal and the received signal using the forgetting factor. You can go.

  As described above, the phase correction unit 241 corrects the phase error of the signal received by each subcarrier, for example, as shown in Expression (40), and then performs demodulation on the signal whose phase error is corrected. Performs demodulation.

For example, when a demodulation method based on ZF (zero forcing criterion) (hereinafter referred to as a ZF method) is used as a demodulation method, the propagation path of each subcarrier obtained by the propagation path estimation unit 221 as shown in the following expression (41) The signal transmitted on the kth subcarrier by multiplying the phase-corrected received signal x _k ^′ by the inverse matrix of the matrix H of the response (Equation (30) and Equation (31)) and the −kth subcarrier The signal transmitted with can be separated.

  Although the ZF method has been described as an example of the demodulation method in the demodulator 251, the demodulation may be performed using any other method. Demodulation may be performed based on the MMSE standard (minimum mean square error), or demodulation may be performed based on the maximum likelihood estimation method. Any method may be used as long as the signal transmitted on the kth subcarrier and the signal transmitted on the -kth subcarrier can be separated.

  By estimating the channel response using the time domain signal as described above, it is possible to perform demodulation even in a situation where interference between subcarriers occurs due to the influence of IQ imbalance.

  However, the method of performing propagation path estimation using a signal in the time domain has a problem that the estimation accuracy depends on the sequence of transmitted propagation path known signals. When the ratio between the maximum eigenvalue and the minimum eigenvalue of the matrix A shown in Expression (25) (hereinafter referred to as the condition number) is small, the estimation accuracy does not deteriorate, but the condition number changes depending on the signal sequence, When the condition number is large, the characteristics are significantly degraded due to noise enhancement or the like.

  Further, such a deterioration in propagation path estimation accuracy does not occur equally in each subcarrier, but has a feature that appears prominently at locations where discontinuities occur in the signal. In general, in OFDM transmission, in order to prevent interference with an adjacent channel, several subcarriers at both ends of the band are often not used for transmission. Therefore, discontinuities occur in the subcarriers at both ends, and the channel response estimation accuracy of the subcarrier may be greatly degraded as compared to other subcarriers.

  An example in which the propagation path estimation accuracy of each subcarrier is obtained is shown in FIG. In FIG. 20, the horizontal axis represents the subcarrier number, and the vertical axis represents the mean square value of the estimation error of the channel response of each subcarrier normalized by the average power of the channel response. Thus, it can be seen that the estimation errors of the subcarriers at both ends are significantly larger than those of the other subcarriers. When a known signal for channel estimation having such channel estimation accuracy characteristics is used, the signal received by the subcarrier has a low channel estimation accuracy, so that the received power is high and the demodulation accuracy is generally high. In this case, the demodulation accuracy is lowered, the demodulation accuracy is not sufficiently obtained, and the transmission characteristics are deteriorated.

Therefore, the demodulator 251 does not use the propagation path estimation results of subcarriers (for example, the i-th and -i-th subcarriers) whose propagation path estimation accuracy deteriorates, but instead has a high propagation estimation precision. A channel estimation result of a subcarrier having a path correlation is selected. Then, as shown in Expression (41), the inverse matrixes of the matrix H of the channel response (Expression (30) and Expression (31)) of the selected subcarrier are phase-corrected i-th and −i-th. The signal transmitted on the i-th subcarrier and the signal transmitted on the -i-th subcarrier are obtained by multiplying the received signal x _i ^′ of the sub-carrier of.

Although it is difficult to measure the propagation path estimation error during communication, since the known signal for propagation path estimation is a predetermined sequence, it is easy to estimate the accuracy of the propagation path estimation before configuring the device. . As an example, in the case where a known signal for channel estimation that can obtain a channel estimation error as shown in FIG. 20 is used, the channel estimation accuracy of the −28th and 28th subcarriers at both ends of the band is extremely high. In addition, since the propagation path estimation accuracy of the −27th and 27th subcarriers is also poor, the propagation path estimation result of the subcarrier is not used. As shown in the following equations (42) and (43), the channel estimation values of the −28th and −27th subcarriers are, for example, those of the −26th subcarrier having high channel estimation accuracy and high channel correlation. Select the propagation path estimation result. Similarly, the channel estimation results of the 26th subcarrier having high channel estimation accuracy and channel correlation are selected as the channel estimation values of the 28th and 27th subcarriers. Further, as shown in the following equation (44), the propagation path estimation results of other subcarriers are used as they are.

  As a result, since the influence of the propagation path estimation result obtained with subcarriers with low propagation path estimation accuracy on demodulation is reduced, it is possible to mitigate degradation in reception performance. In addition, subcarriers with poor propagation path estimation accuracy that can be known in advance from known signals for propagation path estimation can be reduced because it is not necessary to perform propagation path estimation.

  As described above, when IQ imbalance distortion occurs in OFDM transmission, two subcarriers at symmetrical positions with the center frequency as an axis interfere with each other. Although the influence of IQ imbalance can be corrected by estimating the propagation path response including this interference, generally, the preamble is not designed so that the interference component can be estimated. This channel response can be estimated by estimating the impulse response using the time domain signal, but the estimation accuracy of the subcarriers at both ends of the frequency band is improved due to the presence of a sequence of known signals for channel estimation and the presence of guard bands. There is a feature that lowers. Therefore, instead of using the channel estimation results of subcarriers with degraded channel estimation accuracy, instead of using the channel estimation results of subcarriers with high channel estimation accuracy and channel correlation, Performance degradation can be alleviated.

  In addition, since the propagation path estimation accuracy for each subcarrier is inevitably determined when the known signal for propagation path estimation and the propagation path estimation method are determined, the estimation result of the subcarrier with low estimation accuracy is obtained by the propagation path estimation accuracy obtained in advance. Instead of using, the estimation result of subcarriers with high estimation accuracy and propagation path correlation is used as an alternative to reduce the influence on demodulation.

(Second Embodiment)
A second embodiment of the present invention will be described.

  When performing OFDM transmission, the influence of IQ imbalance cannot be ignored, and when subcarriers at symmetrical positions with respect to the center frequency interfere, the propagation path including the interference signal components using the time domain signal For subcarriers whose response is estimated and whose channel estimation accuracy is poor, the channel estimation results of other subcarriers with high channel estimation accuracy and channel correlation, and the signals received by two subcarriers that interfere with each other The point of demodulating the signal transmitted on the two subcarriers is the same as in the first embodiment. The second embodiment is different from the first embodiment in that the transmitting device transmits signals by performing spatial multiplexing using a plurality of antennas, and the receiving device performs reception using a plurality of receiving antennas. is there.

  FIG. 4 is a configuration example of the transmission apparatus according to the second embodiment, and shows a case where there are two antennas. In FIG. 4, the same parts as those in FIG.

  In FIG. 4, a transmission signal is generated by applying channel coding to an information signal to be transmitted. As an example of a method for generating the transmission signal 1 and the transmission signal 2 at this time, as illustrated in FIG. 5, the information signal transmitted by the encoder 181 is encoded and then converted into a parallel signal by the serial conversion unit 171. There is a method of generating a plurality of transmission signals. Here, the serial-parallel conversion may be performed bit by bit, or may be performed in units of bits corresponding to the modulation multilevel number applied by the modulators 101 and 102. Also, conversion may be performed for each number of bits included in the OFDM symbol or in other units. The number of bits is a predetermined number, and conversion may be performed in any unit as long as the wireless reception device is known.

  Further, as described in the first embodiment, there is an encoding scheme in which the correlation between adjacent codewords is high depending on the encoding scheme applied by the encoder 181, and such a signal is transmitted to the same subcarrier, or If assigned to adjacent subcarriers, continuous errors may occur depending on the propagation path response of the subcarriers. In order to prevent such an error, signals may be rearranged using interleavers 191 and 192 as shown in FIG. At this time, the interleavers 191 and 192 may rearrange according to any rule as long as the wireless communication devices are in a known order, or may be rearranged according to the same rule between 191 and 192, or different rules. You can sort by.

  In addition, as shown in FIGS. 7 and 8, the information signal is first subjected to serial-parallel conversion by the serial-parallel converter 171 and then a plurality of encoders (for example, two encoders 181 and 182 corresponding to two antennas) are added. May be used to encode the signal. At this time, the encoders 181 and 182 may perform encoding using the same encoding method, or different encoding methods may be used. Since transmission signal 1 and transmission signal 2 are received by the wireless communication device through different propagation paths by performing spatial multiplexing, they are encoded by each encoder as shown in FIG. 9 and FIG. The signals from the encoders may be mixed and output as transmission signals 1 and 2 using the signal switching unit 172 so that the signals are included in the respective spatially multiplexed signals.

  Returning to the description of FIG. 4, the transmission signal 1 and the transmission signal 2 generated in this way are modulated for each subcarrier by the modulator 101 and the modulator 102, respectively. Since the modulation scheme applied by the modulators 101 and 102 is the same as that of the first embodiment, detailed description thereof is omitted. The modulators 101 and 102 may use the same modulation method or different modulation methods.

  The pilot generation unit 112 generates a pilot subcarrier signal as described in the first embodiment. At this time, the same pilot signal may be transmitted by the spatially multiplexed signal 1 and the spatially multiplexed signal 2 or different signals may be transmitted. Further, the pilot signal may be transmitted using only one spatially multiplexed signal. Any transmission means may be used as long as the wireless communication apparatus transmits a known signal.

  The inverse Fourier transform units 121 and 122, the GI addition units 131 and 132, the radio units 151 and 152, and the antennas 161 and 162 are the same as those in the first embodiment, and detailed description thereof is omitted. Note that antennas 161 and 162 may be antennas having the same performance, or antennas having different performances. Any transmitting antenna may be used as long as a signal having a desired frequency can be transmitted.

  The known signal generation unit 142 generates a known signal necessary for performing MIMO-OFDM transmission with the receiving apparatus. An example of a frame format in MIMO-OFDM transmission is shown in FIG.

  The frame in FIG. 14 has a frame configuration for coexisting with IEEE802.11a shown in FIG. 13. L-STF 411, L-LTF 421, and L-SIG 431 in FIG. 14 are SP411 and LP421 shown in FIG. The same signal as SIG431 is transmitted. The HT-SIG 451 includes information necessary for performing MIMO-OFDM demodulation, and includes a modulation scheme, a coding scheme, a coding rate, a spatial multiplexing number, a signal length, and the like. The L-STF 412, L-LTF 422, L-SIG 432, and HT-SIG 452 may not be transmitted, and the same signals as the L-STF 411, L-LTF 421, L-SIG 431, and HT-SIG 451 are transmitted. Alternatively, a cyclically shifted signal may be transmitted.

  In addition, HT-LTFs 471 to 473 are known signals for channel estimation for estimating the channel response of MIMO-OFDM, and are designed so that the channel response of each signal spatially multiplexed for each subcarrier can be estimated. Is done. Here, as an example of the header signal, the format of FIG. 15 has been described. However, the frame format in the present invention is not limited to the format of FIG. 15, and the format of the header signal is also limited to the format of FIG. is not. In addition, the above signals are given as examples of information necessary for demodulation. However, if the receiving apparatus can demodulate, it is not always necessary to transmit all these signals as header signals, and other signals are used as header signals. It may be included and transmitted.

  Next, FIG. 12 shows a configuration example of a receiving apparatus according to the second embodiment. 12 includes a plurality (for example, two) of antennas 261 and 262, a plurality of (for example, two) radio units 251 and 252 connected to the antennas 261 and 262, and a radio unit. 151 and 152 GI (guard interval) removal units 201 and 202 connected to the GI removal units 201 and 202, respectively, and a plurality of (for example, two) Fouriers connected to the GI removal units 201 and 202, respectively. Converters 211 and 212, propagation path estimation unit 222 connected to GI removal units 201 and 202, reception signal combining unit 232 connected to Fourier transformers 211 and 212, and propagation path estimation unit 222 MIMO demodulation preprocessing unit 311, received signal combining unit 232, phase correction unit 242 connected to propagation path estimation unit 222, phase A MIMO demodulation unit 301 connected to the correction unit 242 and the MIMO demodulation preprocessing unit 311 is included.

  Here, the antennas 161 and 162, the radio units 151 and 152, the GI removal units 201 and 202, and the Fourier transform units 211 and 212 are the same as those in the first embodiment, and detailed description thereof is omitted.

  Next, as in the first embodiment, the Fourier transform unit 211 in the receiving device when IQ imbalance occurs due to imperfections of the quadrature modulator 151c and the quadrature demodulator 251c in the radio unit of the transmitting device and the receiving device. And 212 outputs.

  First, a signal received by the antenna 261 is considered. In the case of the OFDM transmission shown in the first embodiment, it has been shown that the output of the equation (19) is obtained. However, in the case of the MIMO-OFDM transmission, only a signal transmitted by spatial multiplexing is added. . Here, each spatially multiplexed signal is received after passing through a different quadrature modulator and propagation path, so the distortion value and propagation path response value of the quadrature modulator differ, but the position is symmetrical about the center frequency. The characteristics of interfering signals between subcarriers in the same are the same. In addition, since only the spatially multiplexed signals are added to the signal received by the receiving antenna 262, there is no significant difference in properties, except that the propagation path response and the distortion value of the orthogonal demodulator are different. .

Therefore, the received signal of the OFDM transmission of Expression (19) can be expanded to MIMO-OFDM and expressed as the following Expression (47).

In Equation (47), x ^(k) represents a reception vector having the reception signal of the kth subcarrier output from each of the Fourier transform units 211 and 212 as an element. H _α ^(k) is a propagation path response of the kth subcarrier whose i-th row and jth column component is the jth spatially multiplexed signal output from the ith Fourier transform unit. Represents a propagation path matrix. H _β ^(k) is the propagation path of the k th subcarrier from which the i th j column component is output from the i th Fourier transform unit as the j th spatially multiplexed signal with the −k th sub carrier. Represents a response path matrix. s ^(k) represents a transmission vector having each signal spatially multiplexed on the kth subcarrier as an element, and s ^(-k) is a transmission having each signal spatially multiplexed on the -kth subcarrier as an element. Represents a vector.

The complex conjugate of the received vector of the k-th subcarrier expressed in the same manner as Expression (47) is obtained, and the received vector combined with the received vector of the k-th subcarrier can be expressed as the following Expression (48). it can.

  Expression (48) is obtained by extending each element in Expression (22) in the first embodiment to a vector or a matrix, and only the number of elements is increased. Therefore, if a channel response of a spatially multiplexed signal and a signal from a subcarrier that causes interference can be obtained, distortion due to IQ imbalance can be compensated for in MIMO-OFDM transmission as in the first embodiment. it can.

  Consider propagation path estimation in MIMO-OFDM transmission applied by the propagation path estimation unit 222. Assume that a propagation path estimation known signal as shown in FIG. 14 is transmitted. Here, the propagation path estimation preambles (HT-LTF) 471 to 474 transmit the same series of signals, and only the propagation path estimation preamble 473 transmits a signal in which the sign of the propagation path estimation preamble 471 is inverted. . When such a propagation channel estimation preamble (known signal for propagation channel estimation) is transmitted, and when the influence of propagation channel fluctuation or frequency offset is small, the received signal in the section in which the propagation channel estimation preamble 471 is transmitted When the sum of the received signals transmitted by the propagation path estimation preamble 473 is obtained, the propagation path estimation preamble 471 and the propagation path estimation preamble 473 are offset because they are inverted, and the second spatially multiplexed It can be seen that only these signals remain in the received signal. On the other hand, when the difference between the received signals in the section is calculated, it can be seen that the signals of the spatially multiplexed signal 1 are added in phase, and the spatially multiplexed signal 2 is canceled out. Thus, when an orthogonal sequence is transmitted as a known signal for propagation path estimation, it is possible to easily separate the intermultiplexed signals.

  Therefore, each of the signals output from the GI removal units 201 and 202 obtains the sum and difference of the signals in the section in which the propagation path estimation known signal is received, so that each OFDM signal to which a single known signal is transmitted is obtained. Can be extracted. As a result, the propagation of the j-th spatially multiplexed signal output from the i-th GI removal unit in the procedure of equations (23), (28), (30), and (31) is performed in exactly the same manner as in the first embodiment. Since the path response can be obtained including the subcarrier component that causes interference, the propagation path response of each subcarrier can be estimated from the signal in the time domain in the case of MIMO-OFDM transmission as in the case of OFDM transmission.

The phase correction unit 242 extends the equations (33) to (36) in the case of OFDM transmission described in the first embodiment to MIMO-OFDM transmission as shown in the following equations (49) to (51). As in the first embodiment, the phase error can be estimated according to the equations (37) and (39).

Similarly to the first embodiment, the phase correction unit 242 multiplies all the data subcarriers by the inverse characteristic e ^−jΨ of the phase error estimated as described above, as shown in the equation (40). Thus, the phase error is corrected.

  The MIMO demodulator 301 demodulates the signal whose phase error has been corrected by the phase corrector 242. Demodulation is performed only by increasing the number of elements of the reception vector, channel matrix, and transmission vector as compared with the demodulation in the demodulation unit 251 in the first embodiment, and is demodulated by the same method as shown in Expression (41). Detailed description will be omitted.

  Further, the propagation path estimation in the propagation path estimation unit 222 estimates each element of the MIMO propagation path one by one, causing the same problem as in the OFDM transmission described in the first embodiment, and the same phenomenon Is obtained. Therefore, similarly to the method described in the first embodiment, the propagation path estimation accuracy can be obtained in advance. As shown in the equations (42) and (43), the subcarrier whose propagation path estimation accuracy is deteriorated. Demodulates signals transmitted on two subcarriers using signals received on two subcarriers that interfere with each other using channel estimation results of subcarriers with high channel estimation accuracy and channel correlation To do.

  In the second embodiment, the case where the number of spatial multiplexing is “2” and the number of reception antennas is “2” as MIMO-OFDM transmission has been described as an example. It is not limited to numbers. Only the number of elements in the vectors and matrixes in equations (49) to (51) increases, and basically reception can be performed in the same manner.

  As described above, according to the second embodiment, IQ imbalance occurs due to imperfections of the quadrature modulator and the quadrature demodulator, and the signals of the subcarriers in the upper sideband and the lower sideband are mutually connected. When MIMO-OFDM transmission is performed in an environment that interferes with the signal, the channel response including interference components is estimated using the time domain signal, and both subcarriers are transmitted using both received signals of each subcarrier. When the received signal is demodulated, it is possible to mitigate the degradation of reception performance caused by the known signal symbol sequence for channel estimation.

(Third embodiment)
A third embodiment of the present invention will be described.

  In this embodiment, the influence of IQ imbalance cannot be ignored in OFDM transmission or MIMO-OFDM transmission, and when subcarriers at symmetrical positions with respect to the center frequency interfere with each other, the time domain signal is used for the interference signal. The channel response is estimated including the component, and for subcarriers with poor channel estimation accuracy, the channel estimation results of subcarriers with high channel estimation accuracy and channel correlation, and two subcarriers that interfere with each other The point of demodulating the signal transmitted on the two subcarriers using the signal received in step 2 is the same as in the first to second embodiments. The third embodiment is different from the first to second embodiments in that the weighting factor is multiplied so that not only the subcarriers near the both ends of the band but also the likelihood of the subcarriers around the center subcarrier is reduced. It is a point to do.

  In the first embodiment, when estimating a channel response including an interference signal using a signal in the time domain, the estimation accuracy of the channel response is degraded depending on a sequence of known signals for channel estimation, and A phenomenon in which the estimation accuracy near the discontinuous subcarriers deteriorates has been described. When the condition number of the matrix shown in Expression (25) is small, such degradation of the propagation path estimation accuracy does not occur so much, but as the condition number increases, a plurality of subcarriers in the vicinity of both ends are not used. Degradation in subcarrier propagation path estimation accuracy appears significantly.

  In OFDM transmission, the signal of subcarrier number “0” has a greater influence of unnecessary signals than other subcarriers due to the leakage of the local frequency in the radio section and the influence of unnecessary signals of DC components, and is generally used for transmission. Absent. Therefore, as the condition number of Expression (25) increases, the channel estimation accuracy of subcarriers near the center frequency also deteriorates. An example of the estimation accuracy of propagation path estimation when using such a signal sequence is shown in FIG.

  In FIG. 21, as in FIG. 20, the horizontal axis indicates the subcarrier number, and the vertical axis indicates the mean square value of errors in channel estimation normalized by the average power of the channel. Note that the number of conditions in the equation (25) when using the propagation path estimation known signal in FIG. 20 is about “30”, whereas the equation (25) using the propagation path estimation known signal in FIG. The condition number of 25) is about “200”, which is about seven times larger.

  As a result, the degradation of the propagation path estimation accuracy near the subcarriers at both ends of the band is the same as in FIG. 20, but the degradation of the propagation path estimation precision of the subcarriers near the center frequency is also greater. I understand. When such a known signal for channel estimation is used, the channel estimation values of subcarriers near the center frequency as well as the subcarriers near the both ends of the band are subchannels with high channel estimation accuracy and channel correlation. By using the carrier propagation path estimation result, it is possible to mitigate the deterioration of transmission characteristics.

  For example, when a known signal for channel estimation that can obtain a channel estimation error as shown in FIG. 21 is used, the channel estimation accuracy of the −1st and first subcarriers near the center frequency is extremely high. In addition, since the propagation path estimation accuracy of the second and second subcarriers is also poor, the propagation path estimation result of the subcarrier is not used. As shown in the following equations (53) and (54), the channel estimation values of the −1st and −2nd subcarriers are, for example, those of the −3rd subcarrier having high channel estimation accuracy and a channel correlation. Similarly, the propagation path estimation result of the third subcarrier having a high propagation path estimation accuracy and a propagation path correlation is used as the propagation path estimation value of the first and second subcarriers.

  In addition, since the channel estimation accuracy of the −58th and 58th subcarriers at both ends of the band is extremely poor, and the channel estimation accuracy of the −57th and 57th subcarriers is also poor, the channel estimation of the subcarrier is concerned. The result is not used. The channel estimation values of the −58th and −57th subcarriers are, for example, the channel estimation results of the −56th subcarrier having high channel estimation accuracy and channel correlation, and similarly, the 58th and 57th subcarriers. The channel estimation result of the 56th subcarrier having high channel estimation accuracy and channel correlation is used as the channel estimation value of the subcarrier.

Further, the propagation path estimation results of subcarriers near the center frequency and subcarriers other than both ends of the band are used as they are.

  As described above, according to the present embodiment, not only the subcarriers near the both ends of the band but also the propagation channel estimation values of the subcarriers near the central subcarrier have high propagation path estimation accuracy and have subcarrier correlation. By using the propagation path estimation result, it is possible to mitigate deterioration of transmission characteristics.

(Fourth embodiment)
A fourth embodiment of the present invention will be described.

  In the MIMO-OFDM system, the influence of IQ imbalance cannot be ignored, and when subcarriers at symmetrical positions around the center frequency interfere with each other, the propagation path response including the interference signal component using the time domain signal As in the second embodiment, attention is paid to the problem that the channel estimation result deteriorates because the channel estimation accuracy is low for a specific subcarrier. The fourth embodiment differs from the second embodiment in that the wireless communication apparatus determines a weight vector for each subcarrier based on the propagation path estimation result, and transmits using a directional beam generated by the weight vector. It is a point to do.

  FIG. 22 shows a configuration example of a transmission unit of the wireless communication apparatus according to the fourth embodiment. In FIG. 22, the same parts as those in FIG. That is, in FIG. 22, a weight matrix generation unit 601 and a weight matrix multiplication unit 602 are added. Although omitted in FIG. 22, the wireless communication apparatus according to the fourth embodiment includes a receiving unit having the same configuration as that of FIG. In this case, the antennas of the reception unit and the transmission unit may be shared.

  The weight matrix multiplication unit 602 is provided in the preceding stage of the inverse Fourier transform units 121 and 122, and performs processing of multiplying the signals output from the modulators 101 and 102 by the weight matrix supplied from the weight matrix generation unit 601.

  Hereinafter, the case where the number of transmission antennas and the number of reception antennas are two will be described as an example with respect to the weight matrix generation unit 601 and the weight matrix multiplication unit 602.

The weight matrix generation unit 601 generates a weight based on a propagation path response estimation result between the transmission device and the reception device. The propagation path response estimation result is obtained by using, for example, a propagation path response matrix H ^(k) for each subcarrier obtained by the propagation path estimation unit 222 of another wireless communication apparatus having the configuration shown in FIG. It may be done. Since the other wireless communication apparatus includes a transmission unit having the same configuration as that of FIG. 4, the propagation path response matrix for each subcarrier obtained is transmitted through this transmission unit. The radio communication apparatus in FIG. 23 may receive the propagation path response matrix H ^(k) for each subcarrier and use it in the weight matrix generation unit 601.

  Each of modulators 101 and 102 generates a signal modulated for each subcarrier (modulated signal for each subcarrier), and outputs the signal to weight matrix multiplication section 602.

  Weight matrix multiplication section 602 receives the modulation signal for each subcarrier output from modulators 101 and 102 and multiplies the weight obtained by weight matrix generation section 601 for each modulation signal.

The weight matrix generated by the weight matrix generation unit 601 is known to be optimized by performing singular value decomposition (hereinafter referred to as SVD) on the propagation path response matrix. The carrier channel response matrix H ^(k) can be expressed by the following equation (101) by performing singular value decomposition.

In equation (101), the superscript H represents a complex conjugate transpose, and diag [] represents a diagonal matrix. Further, v ₁ ^(k) and v ₂ ^(k) are orthogonal vectors that satisfy the following equation (102), and u ₁ ^(k) and u ₂ ^(k) are orthogonal that satisfy the following equation (103). Is a vector.

However, δ _ij is a Kronecker delta that can be expressed by the following equation (104).

The weight matrix generation unit 601 calculates the weight matrix V ^(k) corresponding to the kth subcarrier from the channel response matrix H ^{(k) of} the ^kth subcarrier and the equations (101) to (104). To do.

Weight matrix multiplication section 602 multiplies weight matrix V ^(k) for each subcarrier for each subcarrier signal. When the kth subcarrier signal output from the modulator 101 is s ₁ ^(k) and the kth subcarrier signal output from the modulator 102 is s ₂ ^(k) , the weight matrix multiplying unit 602 The output k-th subcarrier signal can be expressed by the following equation (105).

  As a method for extracting the signal transmitted from the received signal by the MIMO demodulator on the receiving side, the ZF method for multiplying the generalized inverse matrix of the propagation path estimation result, the MMSE method for multiplying the weight matrix that minimizes the mean square value, and the replica signal May be used, or a method of multiplying the weight matrix U obtained by singular value decomposition of the propagation path estimation result H may be used.

  In the fourth embodiment, the receiving means is not limited to a specific method, and any method other than the above or the above may be used. As described above, a method of performing transmission by multiplying a transmission signal by a weight matrix V obtained by singular value decomposition of a propagation path estimation result to form a directional beam is known as a theoretically optimal transmission method.

As described in the second and third embodiments, when the propagation path response estimation method including the interference signal using the time domain signal is used, the sequence dependency of the known signal for propagation path estimation and the communication are not performed. The presence of the carrier deteriorates the propagation path estimation accuracy of a specific subcarrier. Therefore, in the case of a transmission method in which the transmission matrix is multiplied by the weight matrix V ^(k) obtained by singular value decomposition of the above-described propagation path estimation result, the inaccurate propagation path estimation result is directly subjected to singular value decomposition If it is used as the weight matrix V, an optimum directional beam cannot be formed, and transmission characteristics deteriorate due to inter-stream interference.

  Therefore, in the fourth embodiment, instead of using the channel estimation result of the subcarrier whose channel estimation accuracy deteriorates, as an alternative, the channel estimation result of the subcarrier having high channel estimation accuracy and channel correlation is used. Use. Although it is difficult to measure the propagation path estimation error during communication, since the known signal for propagation path estimation is a predetermined sequence, it is easy to estimate the accuracy of the propagation path estimation before configuring the device. . When a propagation path estimation known signal that can obtain a propagation path estimation error as shown in FIG. 20 is used, the propagation path estimation accuracy of the −28th and 28th subcarriers at both ends of the band is extremely poor. Next, the channel estimation accuracy of the −27th and 27th subcarriers is also poor.

Therefore, the weight matrix generation unit 601 does not use the propagation path estimation result of the subcarrier with poor propagation path estimation accuracy, and the weight matrix corresponding to the −28th and −27th subcarriers is, for example, propagation path estimation. The channel response matrix of the −26th subcarrier having high accuracy and channel correlation is obtained by singular value decomposition. That is, the weight matrices V ⁽⁻²⁸⁾ and V ⁽⁻²⁷⁾ corresponding to the −28th and −27th subcarriers are obtained by singular value decomposition of the propagation path response matrix H ⁽⁻²⁶⁾ .

Similarly, the weight matrices corresponding to the 28th and 27th subcarriers are obtained by singular value decomposition of the propagation response matrix of the 26th subcarrier having high propagation path estimation accuracy and a propagation path correlation. That is, the weight matrices V ⁽²⁸⁾ and V ⁽²⁷⁾ corresponding to the 28th and 27th subcarriers are obtained by singular value decomposition of the channel response matrix H ⁽²⁶⁾ .

For subcarriers with good propagation path estimation accuracy other than the above, each propagation path estimation result H ^(k) is subjected to singular value decomposition to calculate a weight matrix V ^(k) .

  As a result, the compensation effect for the deterioration of the propagation path estimation accuracy can be easily performed without increasing the apparatus scale, and stream interference can be reduced or eliminated.

Also, as in the example described above, for example, the weight matrices V ⁽²⁸⁾ and V ⁽²⁷⁾ used for the 27th and 28th subcarriers are exactly the same as the weight matrix V ⁽²⁶⁾ used for the ^26th , For the 27th and 28th subcarriers with low propagation path estimation accuracy, the SVD calculation can be omitted, and the calculation amount and power consumption can be reduced.

  As described in the third embodiment, when the number of conditions in Expression (25) increases, for example, as shown in FIG. 21, the propagation path estimation accuracy of subcarriers near the center frequency also deteriorates. . Therefore, in such a case, the subcarrier confidence matrix near the center frequency is used not only when calculating the weight matrix of the subcarrier near the both ends of the band but also when calculating the weight matrix of the subcarrier near the center frequency. In addition, the deterioration of the transmission characteristics can be alleviated by calculating the weight matrix in the same manner as described above using the propagation path matrix of the subcarrier having high propagation path estimation accuracy and the propagation path correlation.

That is, in FIG. 21, when V ⁽⁻¹⁾ and V ⁽⁻²⁾ are obtained as weight matrices of the −1st and −2nd subcarriers with poor propagation path estimation accuracy, ^The weight matrices V ^(-1) and V ^(-2) are calculated from the equations (101) to (104) using the propagation path matrix H ^(-3) of the third subcarrier for ^(k). To do. Even when the weight matrices V ⁽¹⁾ and V ^{(2 )} of the first and second subcarriers are obtained, the propagation path matrix H ^{(3 of the} third subcarrier is added to H ^(k) of Equation (101). ⁾ , Weight matrices V ⁽¹⁾ and V ⁽²⁾ are calculated from equations (101) to (104). Alternatively, the calculation of the weight matrices of the −1st and −2nd subcarriers with poor propagation path estimation accuracy is omitted, and the weight matrices obtained for the −3rd subcarrier are the −1st and −2nd You may use as a weight matrix of a subcarrier.

In FIG. 21, when obtaining the weight matrices V ⁽⁻⁵⁸⁾ and V ⁽⁻⁵⁷⁾ of the −58th and −57th subcarriers with poor propagation path estimation accuracy, H ^{(k )} , The weight matrices V ⁽⁻⁵⁸⁾ and V ⁽⁻⁵⁷⁾ are calculated from the equations (101) to (104) using the propagation path matrix H ⁽⁻⁵⁶⁾ of the ^−56th subcarrier. Further, when obtaining the weight matrices V ⁽⁵⁸⁾ and V ⁽⁵⁷⁾ of the 58th and 57th subcarriers, the propagation path matrix H of the 56th subcarrier is expressed by H ^(k) in equation (101). ^{Using (56)} , weight matrices V ⁽⁵⁸⁾ and V ⁽⁵⁷⁾ are calculated from equations (101) to (104). Alternatively, the calculation of the weight matrix of the −58th and −57th subcarriers with poor propagation path estimation accuracy is omitted, and the weight matrix obtained for the −56th subcarrier is −58th and −57th. You may use as a weight matrix of a subcarrier.

  Note that the other wireless communication device of the communication partner may transmit the channel response matrix for all subcarriers, but as described above, the channel response matrix of the subcarrier with low channel estimation accuracy is Since it is not used, there is no need to bother with feedback. Therefore, only the channel response matrix for subcarriers with high channel estimation accuracy among all subcarriers may be transmitted. As a result, the amount of feedback from the receiving side can be reduced, and an improvement in throughput can be expected.

Also, as shown in FIG. 23, the other wireless communication apparatus of the communication partner includes a weight matrix generation unit 701 having the same function as the weight matrix generation unit 601 in FIG. As described above, a weight matrix for each subcarrier may be generated using the channel response matrix H ^(k) for each carrier. Although not shown in FIG. 23, the other wireless communication device of the communication partner includes a transmission unit having the same configuration as that of FIG. Then, the obtained weight matrix for each subcarrier is transmitted through this transmitter. As described above, weight matrix multiplication section 602 does not use a weight matrix obtained by singular value decomposition of a subcarrier propagation path response matrix with poor transmission path estimation accuracy. There is no need. Therefore, we do not transmit the weight matrix obtained by singular value decomposition of the channel response matrix of subcarriers with low channel estimation accuracy, but perform singular value decomposition of the channel response matrix of subcarriers with low channel estimation accuracy. Only the obtained weight matrix may be transmitted. As a result, the amount of feedback can be reduced and throughput can be improved.

  As described above, when the propagation path information is known on the transmission side, a quadrature beam is formed by multiplying the transmission signal with the weight obtained by singular value decomposition of the propagation path response, and optimal transmission with MIMO is possible. A certain beam forming transmission can be performed. However, depending on the known signal for channel estimation and the channel estimation method, there is a problem that the channel estimation accuracy of a specific subcarrier becomes extremely low, and the weight obtained by singular value decomposition of such channel estimation results If used, a quadrature beam is not formed, resulting in stream interference and degradation of transmission characteristics. Therefore, based on the known signal for channel estimation and the channel estimation accuracy determined by the channel estimation method, the channel estimation results of subcarriers with low estimation accuracy are not used, the channel estimation accuracy is high, and channel correlation The propagation path estimation result of a subcarrier having a high is used. As a result, inter-stream interference can be reduced or eliminated, and transmission characteristics can be improved.

  The method of the present invention described in the embodiment of the present invention is a program that can be executed by a computer, such as a magnetic disk (flexible disk, hard disk, etc.), an optical disk (CD-ROM, DVD, etc.), a semiconductor memory, etc. It can be stored in a medium and distributed.

  Note that the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. In addition, various inventions can be formed by appropriately combining a plurality of components disclosed in the embodiment. 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.

The figure which showed the structural example of the principal part of the transmission part of the radio | wireless communication apparatus which transmits the signal which the radio | wireless communication apparatus which concerns on 1st and 3rd embodiment receives. The figure which showed the structural example for encoding the information signal to transmit and producing | generating a transmission signal. The figure which showed the structural example for encoding the information signal to transmit and producing | generating a transmission signal. The figure which showed the structural example of the principal part of the transmission part of the radio | wireless communication apparatus which transmits the signal which the radio | wireless communication apparatus which concerns on 1st, 3rd and 4th embodiment receives. The figure which showed the structural example for encoding the information signal to transmit and producing | generating a transmission signal. The figure which showed the structural example for encoding the information signal to transmit and producing | generating a transmission signal. The figure which showed the structural example for encoding the information signal to transmit and producing | generating a transmission signal. The figure which showed the structural example for encoding the information signal to transmit and producing | generating a transmission signal. The figure which showed the structural example for encoding the information signal to transmit and producing | generating a transmission signal. The figure which showed the structural example for encoding the information signal to transmit and producing | generating a transmission signal. The figure which showed the structural example of the principal part of the receiving part of the radio | wireless communication apparatus concerning 1st and 3rd embodiment. The figure which showed the structural example of the principal part of the receiving part of the radio | wireless communication apparatus concerning 2nd and 3rd Embodiment. The figure which shows the frame format of IEEE802.11a. The figure which shows the frame format in MIMO-OFDM transmission. The figure which showed the structural example of the radio | wireless part in the radio | wireless communication apparatus of a transmission side. The figure which showed the structural example of the radio | wireless part in the receiving side radio | wireless communication apparatus. The figure which showed the structural example of the orthogonal modulator. The figure which showed the structural example of the orthogonal demodulator. The figure for demonstrating the phenomenon which arises when IQ imbalance exists in OFDM transmission. The figure which shows an example of the estimation error (propagation accuracy) of propagation path estimation for every subcarrier. The figure which shows the other example of the estimation error (channel estimation precision) of the channel estimation for every subcarrier. The figure which showed the structural example of the principal part of the transmission part of the radio | wireless communication apparatus which concerns on 4th Embodiment. The figure which showed the structural example of the principal part of the receiving part of the radio | wireless communication apparatus which concerns on 4th Embodiment.

Explanation of symbols

  251: Radio unit, 201: GI removing unit, 211: Fourier transform unit, 221 ... Propagation path estimation unit, 231 ... Received signal combining unit, 241 ... Phase correction unit, 251 ... Demodulation unit.

Claims (12)

A receiving means for receiving an OFDM (Orthogonal Frequency Division Multiplexing) signal including a plurality of subcarriers;
Orthogonal demodulation means for orthogonally demodulating the OFDM signal and obtaining an orthogonal demodulated signal;
A propagation path estimation means for estimating a propagation path response for each subcarrier from a known signal for propagation path estimation included in the orthogonal demodulated signal;
Fourier transform means for Fourier transforming the orthogonal demodulated signal to obtain a received signal for each subcarrier;
The first subcarrier having a relatively low propagation path estimation accuracy among the plurality of subcarriers and the center frequency of the frequency band of the OFDM signal are arranged at positions symmetrical to the first subcarrier. The received signal with the second subcarrier, and the third subcarrier arranged near the first subcarrier having the relatively high propagation path estimation accuracy among the plurality of subcarriers and the center Using the propagation path response with the fourth subcarrier arranged at a position symmetrical to the third subcarrier about the frequency as an axis, the signals transmitted on the first and second subcarriers are Demodulation means for demodulating;
A wireless communication apparatus comprising: Receiving means for receiving a plurality of spatially multiplexed OFDM (Orthogonal Frequency Division Multiplexing) signals each including a plurality of subcarriers with a plurality of antennas;
Orthogonal demodulation means for orthogonally demodulating the plurality of OFDM signals to obtain a plurality of orthogonal demodulation signals;
A propagation path estimation means for estimating a propagation path response for each subcarrier from a known signal for propagation path estimation included in each of the plurality of orthogonal demodulation signals;
Fourier transform means for Fourier transforming each of the plurality of quadrature demodulated signals to obtain a received signal for each subcarrier from each quadrature demodulated signal;
Among the plurality of subcarriers obtained from each orthogonal modulation signal, a first subcarrier having a relatively low propagation path estimation accuracy, and the first subcarrier with the center frequency of the frequency band of the OFDM signal as an axis The received signal with the second subcarrier arranged at a symmetric position and the first subcarrier in the vicinity of the first subcarrier having a relatively high propagation path estimation accuracy among the plurality of subcarriers Using the channel response of the third subcarrier and the fourth subcarrier arranged at a position symmetrical to the third subcarrier with the center frequency as an axis, the first and second Demodulation means for demodulating a signal transmitted on a subcarrier;
A wireless communication apparatus comprising:   The radio communication apparatus according to claim 1 or 2, wherein the first and second subcarriers are arranged near both ends of the frequency band.   The radio communication apparatus according to claim 1 or 2, wherein the first and second subcarriers are arranged in the vicinity of the center frequency. The propagation path estimation means includes
3. The propagation path response for each subcarrier is estimated from the known signals by separating a known signal for propagation path estimation included in each orthogonal demodulation signal from the plurality of orthogonal demodulation signals. Wireless communication device. The propagation path estimation means includes
(A) Fourier transform of an impulse response of a known signal for channel estimation in the OFDM signal to obtain a channel response of a signal transmitted by each subcarrier and received by the subcarrier; (b) The impulse response of the complex conjugate signal of the known signal is Fourier transformed to be transmitted on each subcarrier and received on another subcarrier arranged at a position symmetrical to the subcarrier with the center frequency as an axis. 3. The wireless communication apparatus according to claim 1, wherein a propagation path response of the signal is obtained. A wireless communication device that transmits and receives a plurality of OFDM (Orthogonal Frequency Division Multiplexing) signals each including a plurality of subcarriers,
Means for obtaining a channel response matrix for each subcarrier;
For subcarriers having a relatively low propagation path estimation accuracy among the plurality of subcarriers, other subcarriers in the vicinity of the subcarrier having a relatively high propagation path estimation accuracy among the plurality of subcarriers. A weight matrix is calculated by performing singular value decomposition on the propagation path response matrix of a carrier, and for the subcarriers with relatively high propagation path estimation accuracy among the plurality of subcarriers, A calculation means for calculating a weight matrix by performing singular value decomposition on a propagation path response matrix and obtaining a plurality of weight matrices corresponding to each of the plurality of subcarriers;
Generating means for generating a plurality of modulated signals for each subcarrier;
Multiplying means for multiplying the modulation signal for each subcarrier by the weight matrix corresponding to the subcarrier;
An inverse Fourier transform means for performing an inverse Fourier transform on a modulation signal for each subcarrier multiplied by the weight matrix;
A wireless communication device. A wireless communication device that transmits and receives a plurality of OFDM (Orthogonal Frequency Division Multiplexing) signals each including a plurality of subcarriers,
Means for obtaining a channel response matrix for each subcarrier;
Corresponding to each subcarrier having a relatively high channel estimation accuracy by performing singular value decomposition on the channel response matrix of each subcarrier having a relatively high channel estimation accuracy among the plurality of subcarriers. A calculation means for obtaining a weight matrix;
Generating means for generating a plurality of modulated signals for each subcarrier;
For the modulated signal of a subcarrier having a relatively low propagation path estimation accuracy among the plurality of subcarriers, in the vicinity of the subcarrier having a relatively high propagation path estimation accuracy among the plurality of subcarriers Multiplying the weight matrix corresponding to another subcarrier, the modulation signal of the subcarrier having a relatively high propagation path estimation accuracy among the plurality of subcarriers corresponds to the subcarrier. Multiplication means for multiplying the weight matrix;
An inverse Fourier transform means for performing an inverse Fourier transform on the modulation signal for each subcarrier multiplied by the weight matrix;
A wireless communication device.   The radio communication apparatus according to claim 9, wherein the acquisition unit acquires the channel matrix of a subcarrier having a relatively high channel estimation accuracy among the plurality of subcarriers. A receiving step of receiving an OFDM (Orthogonal Frequency Division Multiplexing) signal including a plurality of subcarriers;
An orthogonal demodulation step of orthogonally demodulating the OFDM signal to obtain an orthogonal demodulated signal;
A propagation path estimation step for estimating a propagation path response for each subcarrier from a known signal for propagation path estimation included in the orthogonal demodulated signal;
A Fourier transform step of Fourier transforming the orthogonal demodulated signal to obtain a received signal for each subcarrier;
The first subcarrier having a relatively low propagation path estimation accuracy among the plurality of subcarriers and the center frequency of the frequency band of the OFDM signal are arranged at positions symmetrical to the first subcarrier. The received signal with the second subcarrier, and the third subcarrier arranged near the first subcarrier having the relatively high propagation path estimation accuracy among the plurality of subcarriers and the center Using the propagation path response with the fourth subcarrier arranged at a position symmetrical to the third subcarrier about the frequency as an axis, the signals transmitted on the first and second subcarriers are A demodulation step for demodulating;
Including receiving method. A transmission method for transmitting a plurality of OFDM (Orthogonal Frequency Division Multiplexing) signals each including a plurality of subcarriers,
An acquisition step of acquiring a channel response matrix for each subcarrier;
For subcarriers having a relatively low propagation path estimation accuracy among the plurality of subcarriers, other subcarriers in the vicinity of the subcarrier having a relatively high propagation path estimation accuracy among the plurality of subcarriers. A weight matrix is calculated by performing singular value decomposition on the propagation path response matrix of a carrier, and for the subcarriers with relatively high propagation path estimation accuracy among the plurality of subcarriers, Calculating a weight matrix by singular value decomposition of the channel response matrix, and calculating a plurality of weight matrices corresponding to each of the plurality of subcarriers;
Generating a plurality of modulated signals for each subcarrier; and
A multiplication step of multiplying the modulation signal for each subcarrier by the weight matrix corresponding to the subcarrier;
An inverse Fourier transform step of performing an inverse Fourier transform on the modulation signal for each subcarrier multiplied by the weight matrix;
Including sending method. A transmission method for transmitting a plurality of OFDM (Orthogonal Frequency Division Multiplexing) signals each including a plurality of subcarriers,
An acquisition step of acquiring a channel response matrix for each subcarrier;
Corresponding to each subcarrier having a relatively high channel estimation accuracy by performing singular value decomposition on the channel response matrix of each subcarrier having a relatively high channel estimation accuracy among the plurality of subcarriers. A calculation step for obtaining a weight matrix;
Generating a plurality of modulated signals for each subcarrier; and
For the modulated signal of a subcarrier having a relatively low propagation path estimation accuracy among the plurality of subcarriers, in the vicinity of the subcarrier having a relatively high propagation path estimation accuracy among the plurality of subcarriers Multiplying the weight matrix corresponding to another subcarrier, the modulation signal of the subcarrier having a relatively high propagation path estimation accuracy among the plurality of subcarriers corresponds to the subcarrier. A multiplication step for multiplying the weight matrix;
An inverse Fourier transform step of performing an inverse Fourier transform on the modulation signal for each subcarrier multiplied by the weight matrix;
Including sending method.

Images