JP2006295629A - Radio communication system, radio communication apparatus and radio communication method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To accurately remove a frequency error before estimating a channel in a receiver for performing MIMO (Multi Input Multi Output)_OFDM (Orthogonal Frequency Division Multiplexing) communication. <P>SOLUTION: On the receiver side, a frequency error is estimated by phase comparison within a preamble for synchronization acquisition and frequency error estimation, the frequency error is estimated by comparing the phase of the preamble with that of a preamble for channel matrix estimation obtained from one of transmitting antennas and more accurate frequency error correction can be performed by using both the estimated results. For instance, the phase of an antenna receiving weight vector found out from a channel matrix is rotated to correct a remaining frequency error. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a radio communication system for performing MIMO (Multi Input Multi Output) communication using a plurality of logical channels formed by a pair of a transmitter having a plurality of antennas and a receiver having a plurality of antennas. The present invention relates to a radio communication apparatus and a radio communication method, and in particular, a radio communication system that performs MIMO communication using an OFDM (Orthogonal Frequency Division Multiplexing) communication system that multiplex-transmits a plurality of subcarriers orthogonal to each other, and radio communication The present invention relates to an apparatus and a wireless communication method.

  More particularly, the present invention relates to a radio communication system, a radio communication apparatus, and a radio communication method for accurately obtaining a frequency error between transmitters / receivers performing MIMO_OFDM communication, and in particular, to calculate an antenna reception weight vector on the receiver side. In parallel, the present invention relates to a wireless communication system, a wireless communication apparatus, and a wireless communication method that improve frequency error estimation accuracy.

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

  For example, in IEEE802.11a / g, an OFDM (Orthogonal Frequency Division Multiplexing) modulation scheme, which is one of multicarrier schemes, is adopted as a standard for wireless LANs. In the OFDM modulation scheme, the frequency of each carrier is set so that the subcarriers are orthogonal to each other within a symbol interval. That subcarriers are orthogonal to each other means that the peak point of the spectrum of an arbitrary subcarrier always coincides with the zero point of the spectrum of another subcarrier. According to the OFDM modulation scheme, transmission data is distributed and transmitted to a plurality of carriers having different frequencies, so that the bandwidth of each carrier is narrow, the frequency utilization efficiency is very high, and it is resistant to frequency selective fading interference.

  An OFDM transmitter assigns multiple data to be output by serial / parallel conversion of information sent serially for each symbol period slower than the information transmission rate, and modulates amplitude and phase for each subcarrier. And by performing inverse FFT on the plurality of subcarriers, the subcarriers are converted to time axis signals and transmitted while maintaining the orthogonality of each subcarrier on the frequency axis. The OFDM receiver performs the reverse operation, that is, performs FFT to convert the time-axis signal to the frequency-axis signal, demodulates each subcarrier in accordance with the modulation method, and performs parallel / serial conversion. The information sent with the original serial signal is reproduced.

  Further, although the IEEE 802.11a standard supports a modulation scheme that achieves a communication speed of 54 Mbps at the maximum, a wireless standard capable of realizing a higher bit rate is required.

  MIMO (Multi-Input Multi-Output) communication is attracting attention as one of the technologies for realizing high-speed wireless communication. This is a communication system that includes a plurality of antenna elements on both the transmitter side and the receiver side, and realizes a spatially multiplexed transmission path (hereinafter also referred to as “MIMO channel”). Transmitter distributes transmission data to a plurality of antennas at a transmitter. On the other hand, each signal can be extracted without crosstalk by performing signal processing on spatial signals received by a plurality of antennas in the receiver (see, for example, Patent Document 1). For example, in IEEE 802.11a / n, an OFDM_MIMO scheme using OFDM for primary modulation is employed.

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

  FIG. 4 conceptually shows the MIMO communication system. The MIMO transmitter is provided with two antennas, that is, a transmission antenna 1 and a transmission antenna 2, and one MIMO receiver is also provided with two reception antennas 1 and 2. Here, the propagation path of transmission antenna 1 and reception antenna 1 is propagation path a, the propagation path of transmission antenna 2 and reception antenna 1 is propagation path b, the propagation path of transmission antenna 1 and reception antenna 2 is propagation path c, and transmission antenna 2 and the propagation path of the receiving antenna 2 are defined as a propagation path d. Then, the transmitter assigns the transmission data sequence X1 to the transmission antenna 1 and the transmission data sequence X2 to the transmission antenna 2, and the receiver receives the reception data sequence Y1 at the reception antenna 1, and It is assumed that the reception data series Y2 has been received. The propagation path condition in this case can be expressed as the following equation (1).

If the channel matrix H at this time is defined as in the following equation (2), an inverse matrix H ⁻¹ of the channel matrix H can be obtained as the antenna reception weight matrix W as in the following equation (3).

Therefore, by multiplying the reception signal sequences Y1 and Y2 by the inverse matrix H ⁻¹ of the channel matrix H as in the following equation (4), the reception signal sequences X1 and X2 are obtained as in the following equation (5).

  Although FIG. 4 shows the case where there are two transmission / reception antennas, if the number of antennas is two or more, a MIMO communication system can be constructed in the same manner. On the transmission side, a plurality of transmission data are multiplexed by applying space / time codes, distributed to M transmission antennas, and transmitted to the MIMO channel. On the other hand, on the receiving side, received data can be obtained by space / time decoding received signals received by N receiving antennas via a MIMO channel. Ideally, only the smaller number of transmission / reception antennas (MIN [M, N]) MIMO streams are formed.

  As described above, in order to spatially separate each stream signal x from the spatially multiplexed received signal y, the MIMO receiver obtains the channel matrix H by some method and receives the reception weight from the channel matrix H by a predetermined algorithm. It is necessary to obtain the matrix W.

  The channel matrix H expressed by the above equation (2) is generally transmitted and received by transmitting and receiving known sequences on the transmission side and the reception side. In other words, a, b, c, d) are arranged in a matrix format. When the number of transmitting antennas is N and the number of receiving antennas is M, the channel matrix is an M × N (row × column) matrix. Therefore, M × N known sequences, that is, training signals are transmitted from the transmitter, and the receiver can obtain the channel matrix H using the training signals.

  Here, if a plurality of training signals are simultaneously transmitted without countermeasures, the receiving side cannot determine which antenna has been transmitted. Therefore, the training signal for each transmitting antenna is time-divisionally (ie, time-divided) from the transmitter side. The time division method of acquiring the channel matrix H based on the training signal received by each receiving antenna is applied on the receiver side (see FIG. 5).

  As a relatively simple algorithm for obtaining the reception weight matrix W from the channel matrix H, Zero Force (zeroization standard), MMSE (Minimum Mean Square Error), and the like are known. Zero Force is a logic-based method that completely eliminates crosstalk, while MMSE is a logic-based method that maximizes the ratio of signal power and square error (sum of crosstalk power and noise power). It is.

  Here, there is a problem of frequency offset in the OFDM communication device. This is due to a subtle error in the frequency of the oscillator mounted on each transceiver. For example, an oscillator having an accuracy of about 20 ppm is used in a wireless LAN, and an error of the oscillator is observed as a phenomenon of phase rotation of a received signal in a digital part on the receiver side.

  In the case of OFDM communication, the frequency domain OFDM signal is a set of Sync functions. Another Sync function peak comes to the section of each Sync function. For this reason, in the MIMO_OFDM communication system, interference must be caused between subcarriers unless the frequency error is first accurately removed before channel estimation (see the dotted line in FIG. 6).

  As a general countermeasure against the frequency offset, a preamble including a known pattern is added to the head of the packet from the transmitter side. The receiver uses this preamble to acquire synchronization and observe the frequency offset with the transmitter, and the frequency offset is corrected by reversely rotating the data phase in response to the frequency shift.

  In the MIMO_OFDM communication system, channel estimation is performed in the preamble after removing the frequency error. In FIG. 7, obtaining the direction (phase) and magnitude (amplitude) of the arrow over all subcarriers corresponds to channel estimation. The frequency error that cannot be removed by the initial frequency correction is the phase rotation of the entire subcarrier of the OFDM symbol following the preamble.

  Accurate frequency error estimation is particularly important in broadband wireless communications. Therefore, in IEEE802.11a or the like, the frequency error of the received signal is corrected by estimating the frequency error using the preamble, and the frequency error remaining in the data symbol is also tracked. However, if the initial frequency error estimation is not accurate, tracking cannot follow the residual frequency error, so that an error floor where the packet error does not disappear even if the SN ratio is increased is likely to occur.

JP 2002-44051 A

  An object of the present invention is to provide an excellent radio communication system, radio communication apparatus, and radio communication method capable of more accurately performing initial frequency error estimation using a preamble of a packet on the receiver side when performing MIMO_OFDM communication. Is to provide.

  A further object of the present invention is to provide an excellent radio communication system, radio communication apparatus, and radio communication method capable of accurately removing a frequency error when performing channel estimation in a receiver that performs MIMO_OFDM communication. is there.

  A further object of the present invention is to improve the accuracy of the initial frequency error estimation using the preamble and perform more reliable channel estimation in a receiver that performs MIMO_OFDM communication, thereby achieving effective spatial separation in the data portion. An object of the present invention is to provide an excellent wireless communication system, wireless communication apparatus, and wireless communication method that can be performed.

The present invention has been made in consideration of the above problems, and MIMO communication using a plurality of logical channels formed by a pair of a transmitter having a plurality of antennas and a receiver having a plurality of antennas. A wireless communication system for performing
The transmitter adds a first preamble for frequency error estimation to a packet to be transmitted and a second preamble whose phase can be compared with the first preamble,
The receiver performs a phase comparison within the first preamble of the received packet to estimate a frequency error, and performs a phase comparison between the first preamble and the second preamble to estimate a frequency error.
This is a wireless communication system.

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

  The present invention relates to a MIMO communication system using a plurality of logical channels formed by a pair of a transmitter having a plurality of antennas and a receiver having a plurality of antennas. According to the MIMO communication system, it is possible to increase the transmission capacity according to the number of antennas without increasing the frequency band, thereby achieving an improvement in communication speed. In the present invention, the OFDM modulation scheme is applied in order to improve the frequency utilization efficiency and solve the problem of delay distortion in a multipath environment.

  Here, in the wireless communication system, there is a problem of frequency error due to a slight error in the frequency of the local oscillator mounted in each of the transmitter and the receiver. In MIMO communication, interference must occur between subcarriers unless the frequency error is first accurately removed before channel estimation. The frequency error of the received signal can be corrected by estimating the frequency error using the preamble, and the remaining frequency error can also be tracked using the data symbol. It becomes impossible to follow.

  In MIMO communication, in order to create a channel matrix on the receiver side, preambles corresponding to the number of communication channels (streams) are added to the packet (see the above and FIG. 5). Thus, it is possible to perform frequency error estimation with these preambles, and it is considered that the estimation accuracy can be increased.

  However, if a channel matrix is created after estimating the frequency error of these additional preambles and correcting them, and then an antenna reception weight matrix calculation process such as inverse matrix calculation (or singular value decomposition) is performed, MIMO synthesis starts. It takes too much time to do so, and it interferes with communication.

  In addition, even if the frequency error is estimated using only the original preamble, the residual frequency error remains, and the time interval between the preamble and the data symbol for frequency error estimation is increased by the amount of the added preamble, resulting in a phase difference due to the residual frequency error. Becomes larger in proportion to the time interval, which causes a problem in communication quality.

  Also, in MIMO communication, it is expected that the space between the preamble and data (payload) will be wide due to the stream, and the antenna reception weight vector will not match the data symbol due to the residual frequency error.

  Therefore, in the MIMO_OFDM communication system according to the present invention, the receiver side performs phase comparison between corresponding symbols in one preamble to estimate and correct the frequency error, and then a plurality of phases that can be further compared. The estimation accuracy is improved by estimating the frequency error by the same phase comparison between the two preambles.

  On the transmitter side, a preamble for synchronization acquisition and frequency error estimation and a plurality of preambles (training signals) for estimating a channel matrix are added to the header portion of the transmission packet. Since the channel matrix is obtained by arranging the propagation paths of the transmission and reception antenna combinations in a matrix format, the preamble from each transmission antenna is transmitted in time division for channel matrix estimation. In the present invention, any one of the preambles for channel matrix estimation is made phase-comparable with the preamble for frequency error estimation.

  On the receiver side, the frequency error is estimated and corrected by phase comparison within the preamble for frequency error estimation, and then the phase comparison is performed between the preamble for frequency error estimation and one for channel matrix estimation and the preamble. To estimate the frequency error. Since the phase comparison between the preambles is longer than the phase comparison within the preamble, more accurate estimation and correction of the frequency error can be obtained.

  Here, in order to perform phase comparison between preambles, both must be in a phase-comparable state. If these preambles are combined with the same antenna transmission weight vector, phase comparison is possible. Therefore, what number preamble for channel estimation is used for phase comparison may be determined in advance between transmission and reception.

  In multi-carrier transmission schemes such as OFDM, the frequency error can be estimated by comparing the phases of corresponding sample points within and between preambles. For example, in the training field constituting the preamble, when signals are repeated in the first half and the second half, the phase comparison is performed at the n-th sample point in the first half and the second half in the preamble. Further, between preambles, phase comparison is performed at the m-th sample point from each head.

  As a basic operation, the receiver corrects the frequency error of the received signal based on the phase comparison in the preamble for frequency error estimation, but if an error occurs in the calculation of the frequency offset due to noise or other effects, etc. Frequency error may remain. Therefore, in parallel to estimating the channel matrix using each preamble for channel matrix estimation, the receiver compares the phase within at least one preamble for channel matrix estimation or between the first and second preambles. By calculating the residual frequency error by phase comparison and correcting this, the influence of the frequency error can be eliminated.

  In such a case, the receiver can correct the residual frequency error by rotating the phase of the antenna reception weight matrix obtained from the channel matrix. In this case, the received signal can be spatially separated into each stream signal using the antenna reception weight matrix from which the influence of the residual frequency error is removed.

  Alternatively, the receiver may correct the residual frequency error by rotating the phase of the channel matrix of the MIMO-combined communication channel. In this case, MIMO combining is performed using the antenna reception weight matrix including the residual frequency error component, but the received signal can be decoded with the influence of the residual frequency error as channel estimation.

  Alternatively, the receiver may correct the residual frequency error by rotating the phase of the received data of the MIMO-combined communication channel. In this case, after MIMO combining with the antenna reception weight matrix including the residual frequency error component, the phase of the received signal is rotated by the residual frequency error and returned to the correct signal point in the phase space to influence the residual frequency error. After removing, decryption is performed.

  Further, the receiver may correct the residual frequency error by adjusting the correction phase amount for each stream in accordance with the interval between each preamble for channel matrix estimation and data transmitted in a time division manner.

  When correcting the residual frequency error, the receiver can measure the frequency error in parallel with estimating the channel matrix. Therefore, the channel error is estimated after the frequency error of the additional preamble is estimated and corrected, and the processing time is shorter than when processing such as inverse matrix calculation or singular value decomposition is performed. Never come. In addition, since the correction phase amount is adjusted for each reception stream according to the interval between the preamble and the data, the influence of the residual frequency error can be more accurately removed, and the communication quality is improved.

  Advantageous Effects of Invention According to the present invention, when performing MIMO_OFDM communication, an excellent radio communication system, radio communication apparatus, and radio communication method capable of more accurately performing initial frequency error estimation using a packet preamble on the receiver side. Can be provided.

  Furthermore, according to the present invention, it is possible to provide an excellent radio communication system, radio communication apparatus, and radio communication method capable of accurately removing a frequency error when performing channel estimation in a receiver that performs MIMO_OFDM communication. Can do.

  In addition, according to the present invention, in a receiver that performs MIMO_OFDM communication, it is possible to perform frequency error estimation with high accuracy in parallel with calculation of an antenna reception weight vector using a preamble part. A communication system, a wireless communication apparatus, and a wireless communication method can be provided. That is, the receiver can perform effective spatial separation in the data portion by improving the accuracy of the initial frequency error estimation using the preamble and performing more reliable channel estimation.

  According to the present invention, since the estimation accuracy of the frequency error can be increased, the effective bit rate can be improved particularly in the high bit rate mode of the MIMO communication.

  In addition, according to the present invention, as the number of antenna elements increases, the influence of frequency error estimation deviation due to the increase in the interval between the preamble and data (payload) transmitted in a time division manner is reduced, and higher-order MIMO is reduced. Multiple effects can be reliably obtained even in communication.

  Further, according to the present invention, the frequency correction process can be performed reliably, so that the MIMO communication apparatus can use a cheaper oscillator.

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

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

  The present invention relates to MIMO communication in which a transmitter and a receiver each having a plurality of antennas are paired to transmit a spatially multiplexed signal. In the MIMO communication system, transmission data is distributed and transmitted to a plurality of antennas in a transmitter, and each signal is extracted without crosstalk by performing signal processing on spatial signals received by the plurality of antennas in a receiver. According to the MIMO communication system, it is possible to increase the transmission capacity according to the number of antennas without increasing the frequency band, thereby achieving an improvement in communication speed. Also, since spatial multiplexing is used, the frequency utilization efficiency is good.

  The communication system according to the present invention is a MIMO_OFDM communication system using an OFDM modulation scheme. The OFDM modulation scheme is a multicarrier transmission scheme in which the frequency of each subcarrier is set so that the subcarriers are orthogonal to each other within a symbol interval. That subcarriers are orthogonal to each other means that the peak point of the spectrum of an arbitrary subcarrier always coincides with the zero point of the spectrum of another subcarrier. According to the OFDM modulation scheme, the frequency utilization efficiency is very high and it is strong against frequency selective fading interference.

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

  Each transmission / reception antenna 11a and 11b is connected in parallel with a transmission system and a reception system via switches 12a and 12b, respectively, and wirelessly transmits signals to other wireless communication devices on a predetermined frequency channel, or Collects signals sent from other wireless communication devices. However, the switches 12a and 12b connect the transmission / reception antennas 11a and 11b exclusively to one of the transmission system and the reception system, and cannot perform transmission and reception in parallel.

  Each transmission system includes an encoding unit 21, an IFFT 22, a guard applying unit 23, a preamble / reference adding unit 24, a D / A converter 25 for each antenna, and an analog processing unit 26 for transmission.

  The encoding unit 21 encodes transmission data transmitted from an upper layer of the communication protocol with an error correction code, and transmits the transmission signal in a signal space by a predetermined modulation method such as BPSK, QPSK, 16QAM, 64QAM, 256QAM. To map. Furthermore, a plurality of MIMO channels are obtained by spatial multiplexing by multiplying the encoded transmission signal by a predetermined transmission weight matrix. At this point, a known data sequence may be inserted as a pilot symbol into the modulation symbol sequence according to the pilot symbol insertion pattern and timing. A pilot signal having a known pattern is inserted for each subcarrier or at intervals of several subcarriers.

  In IFFT 22, the modulated serial signal is converted into parallel data corresponding to the number of parallel carriers in accordance with the number of parallel carriers and the timing, and then subjected to inverse Fourier transform for the FFT size according to a predetermined FFT size and timing. Do.

  The guard providing unit 23 provides guard interval sections before and after one OFDM symbol in order to remove intersymbol interference. The time width of the guard interval is determined by the state of the propagation path, that is, the maximum delay time of the delayed wave that affects the demodulation. Then, the signal is converted into a serial signal, converted into a signal on the time axis while maintaining the orthogonality of each carrier on the frequency axis, and used as a transmission signal.

  The preamble / reference adding unit 24 adds a preamble signal or a reference signal to the head of a transmission signal such as an RTS, CTS, or DATA packet.

  A transmission signal for each antenna is converted into an analog baseband signal by each D / A converter 25 and further up-converted to an RF frequency band by each transmission analog processing unit 26, and then transmitted from each antenna 11. Transmitted to the MIMO channel.

  On the other hand, each reception system includes a reception analog processing unit 31 and an A / D converter 32 for each antenna, a synchronization acquisition unit 33, a frequency offset compensation unit 34, an FFT 35, a space separation unit 36, and a phase rotation compensation. A section 37 and a decoder 38 are included.

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

  The digital baseband signal of each antenna system is collected by converting the received signal as serial data into parallel data in accordance with the synchronization timing detected by the synchronization acquisition unit 33 (including up to the guard interval here). Signals for one OFDM symbol are collected).

  The frequency offset compensator 34 estimates a frequency error included in the received signal, and performs initial frequency correction on each digital baseband signal based on the estimated value.

  A preamble for synchronization acquisition and frequency error estimation and a plurality of preambles (training signals) for estimating a channel matrix are added to the header portion of the packet. In the present embodiment, the frequency offset compensator 34 performs initial frequency error estimation and correction by phase comparison in the preamble for synchronization acquisition and frequency error estimation, and also performs phase comparison in each preamble for channel matrix estimation, In addition, by performing phase comparison between a plurality of preambles capable of phase comparison, a more accurate estimation of the residual frequency error is performed in parallel with the estimation of the channel matrix. Details of the residual frequency error estimation and correction processing will be described later.

  The FFT 35 converts the signal for the effective symbol length by Fourier transform into a signal on the time axis into a signal on the frequency axis, and decomposes the received signal into subcarrier signals.

  The spatial separation unit 36 generates a channel matrix H for each subcarrier based on the FFT output of the preamble part for channel matrix estimation added to the packet header, and calculates an antenna reception weight matrix from the channel matrix. . Then, by multiplying the reception signal from each antenna by the antenna reception weight matrix, the FFT output of the data portion of the packet is MIMO-combined for each subcarrier and separated into a plurality of independent MIMO streams.

The space separation unit 36 can calculate the inverse matrix H ⁻¹ as an antenna reception weight matrix from the channel matrix H estimated using the preamble based on an algorithm such as Zero Force or MMSE. Alternatively, the channel matrix H may be subjected to singular value decomposition to obtain a reception weight matrix U ^H as the antenna weight matrix W.

  In the present embodiment, the space separation unit 36 performs a correction process for the residual frequency error obtained by the frequency offset compensation unit 34 in parallel with the estimation of the channel matrix and the MIMO synthesis. For example, the residual frequency error is corrected by rotating the phase with respect to the antenna reception weight matrix calculated from the channel matrix. Alternatively, after MIMO combining with the antenna reception weight matrix including the residual frequency error component, the phase of the received signal is rotated by the residual frequency error and returned to the correct signal point in the phase space to influence the residual frequency error. Remove. Details of the residual frequency error estimation and correction processing will be described later.

  The phase rotation compensation unit 37 corrects the phase rotation amount by the residual frequency offset and the phase noise for each MIMO channel subjected to MIMO synthesis. Since the residual frequency offset and phase noise are the same between the MIMO channels, the observations of the phase rotation amount for each MIMO channel obtained using each pilot subcarrier are integrated, that is, averaged, and phase rotation more resistant to noise. The quantity can be obtained. Then, phase compensation is performed for each MIMO channel using the obtained phase rotation amount.

  The phase rotation compensation unit 37 uses, for example, a phase difference indicating a frequency offset or timing offset as a vector on the IQ plane, obtains an average value using the magnitude of this vector as a weight, and performs phase correction based on the frequency offset or timing offset. The amount can be estimated. A method for removing a residual frequency error included after MIMO synthesis is described in, for example, Japanese Patent Application No. 2004-378944 already assigned to the present applicant.

  After the phase rotation correction, the decoding unit 38 demodulates from the modulation point on the phase space to the original value.

  As described above, according to the MIMO communication method, it is possible to increase the transmission capacity according to the number of antennas without increasing the frequency band, thereby achieving an improvement in communication speed. On the other hand, accurate frequency error estimation is particularly important in broadband wireless communication.

  For example, as specified in IEEE802.11a, the frequency error of the received signal can be corrected by estimating the frequency error with the preamble, and the remaining frequency error can be tracked with the data symbol. If it is not accurate, tracking cannot follow the residual frequency error.

  Therefore, in the MIMO_OFDM communication system according to the present embodiment, the receiver side performs phase comparison between corresponding symbols in one preamble to estimate a frequency error, and between a plurality of preambles capable of phase comparison. By estimating the frequency error by the same phase comparison, the estimation accuracy is improved.

  On the transmitter side, a preamble for synchronization acquisition and frequency error estimation and a plurality of preambles (training signals) for estimating a channel matrix are added to the header portion of the transmission packet. In MIMO communication, a transmitter transmits a preamble from each transmission antenna in time division for channel matrix estimation. In the receiver, the channel matrix can be obtained by estimating the propagation path of the path corresponding to the transmission / reception antenna combination from the preamble and arranging these estimated values in a matrix form. In the present embodiment, any one of the preambles for channel matrix estimation is made phase-comparable with the preamble for frequency error estimation.

  In such a case, the receiver side estimates the frequency error by phase comparison in the preamble for synchronization acquisition and frequency error estimation, and also includes one preamble for synchronization acquisition and frequency error estimation, one for channel matrix estimation, and a preamble. More accurate frequency error estimation and correction can be obtained by phase comparison between the two and estimating the frequency error. That is, the receiver compares the phase within at least one preamble for channel matrix estimation or between the first and second preambles in parallel with estimating the channel matrix using each preamble for channel matrix estimation. By calculating the residual frequency error by phase comparison and correcting this, the influence of the frequency error can be removed from the received stream after spatial separation.

  Hereinafter, a detailed description will be given of a processing mechanism for correcting the residual frequency error in parallel with the estimation of the channel matrix.

  FIG. 2 schematically shows a packet format used in the MIMO communication system according to the present embodiment.

  At the head of the packet is an L-STF for packet discovery. In L-STF, the signal is repeated 10 times, and the receiver can find the packet by taking autocorrelation.

  Reference numeral 100 is a preamble called L-LTF (Legacy-Long Training Field) following L-STF. The frequency offset compensator 34 on the receiver side uses the LTF for channel estimation and simultaneously estimates the frequency error by phase comparison. The LTF consists of two OFDM symbols, and the signal is repeated in the first half and the second half. Since one OFDM symbol has 64 sample points, as shown by reference numeral 110, a phase comparison is performed between the n-th sample point in the first half and the n-th sample point in the second half (where n is 0 to 63). (Integer), the average of the 64 phase comparison results is taken to estimate the frequency error within this preamble. Measurement accuracy can be improved by averaging the phase comparison results at a plurality of sample points (hereinafter the same).

  In conventional IEEE802.11a, SIG (Signal Field) and data symbols (payload) are followed, but in MIMO communication, as shown in the figure, after SIG, a channel matrix consisting of channel estimation between transmitting and receiving antennas is obtained. A preamble (101 to 104) called HT-LTF (High Throughput LTF) is added, and data symbols (105 to 108) obtained by spatially multiplexing each transmission stream follow.

  As shown in the figure, the transmitter transmits time-division additional preambles (101 to 104) for channel estimation from each antenna (each additional preamble is switched for each transmission antenna or each combination of antenna transmission weight vectors). Sent).

The spatial separation unit 36 on the receiver side can obtain the channel matrix H by estimating the propagation path of the path corresponding to the combination of transmission and reception antennas from each additional preamble and arranging these estimated values in a matrix format. Further, an inverse matrix H ⁻¹ of the channel matrix H estimated as the antenna reception weight matrix is obtained. In subsequent data symbols, the received signal from each antenna is multiplied by its inverse matrix H ⁻¹ to perform MIMO synthesis to generate a plurality of independent channels (reception streams).

  Here, if the additional preamble HT-LTF (101 to 104) has a length of about two OFDM symbols as in the L-LTF and the signal repeats in the first half and the second half, the reference number 111 ˜114, phase comparison is performed between the n-th sample points in the first half and the n-th sample point in the second half in each preamble (where n is an integer from 0 to 63), and the average of the phase comparison results of 64 points is calculated. By taking this, the frequency error in the preamble can be estimated as in the case of L-LTF.

  Further, if any of the additional preambles HT-LTF (101 to 104) and the first preamble L-LTF 100 are transmitted by the same antenna (or the same antenna transmission weight vector), the phase comparison is also performed between these two preambles. Is possible. In the example shown in FIG. 2, the transmitter transmits the additional preamble 101 with the same antenna weight vector as that of the first preamble 100 and is capable of phase comparison. Therefore, as indicated by reference numeral 115, the transmitter is added to the first preamble 100. By performing phase comparison between corresponding sample points between the preambles 101 and taking the average of 128 phase comparison results, frequency error estimation between the preambles can be performed.

  In the phase comparison between preambles indicated by reference numeral 115, the comparison interval is wider than that of the non-preamble phase comparison indicated by reference numeral 110, so that the frequency error can be measured more accurately.

  Here, after all frequency error measurements indicated by reference numerals 110 to 115 in FIG. 2 are performed, a channel matrix is estimated from the additional preamble HT-LTF (101 to 104), and an antenna reception weight matrix such as inverse matrix calculation is performed. If this is calculated, time is wasted and communication is hindered.

  Therefore, in this embodiment, the receiver measures and corrects the initial frequency error, and then measures the residual frequency error in parallel with the estimation of the channel matrix. The frequency offset compensator 34 first performs frequency error measurement and correction within the first preamble L-LTF (100) indicated by reference numeral 110. Thereafter, the frequency offset compensator 34 measures the residual frequency error using the additional preamble, as indicated by reference numerals 111 to 115, and in parallel with this, passes the additional preamble to the spatial separation unit 36, The antenna reception weight vector is calculated by estimating the channel matrix and calculating the inverse matrix thereof.

  For example, the spatial separation unit 36 can correct the residual frequency error by rotating the phase of the antenna reception weight matrix obtained from the channel matrix. In this case, the received signal of each antenna can be spatially separated into each stream signal by performing MIMO synthesis using the antenna reception weight matrix from which the influence of the residual frequency error is removed.

  Alternatively, the spatial separation unit 36 may correct the residual frequency error by rotating the phase of the channel matrix of the MIMO-combined communication channel. In this case, MIMO combining is performed using the antenna reception weight matrix including the residual frequency error component, but the received signal can be decoded with the influence of the residual frequency error as channel estimation.

  Alternatively, the residual frequency error may be corrected by rotating the phase of the received data of the communication channel subjected to MIMO synthesis. In this case, after MIMO combining with the antenna reception weight matrix including the residual frequency error component, the phase of the received signal is rotated by the residual frequency error and returned to the correct signal point in the phase space to influence the residual frequency error. To be decoded.

  Thus, on the receiver side, the frequency offset compensator 34 performs the frequency error measurement using the additional preamble HT-LTF (101 to 104) in parallel with the estimation of the channel matrix in the spatial separator 36. Can be done. Therefore, the channel error is estimated after the frequency error of the additional preamble is estimated and corrected, and the processing time is shorter than when processing such as inverse matrix calculation or singular value decomposition is performed. Never come. In addition, since the correction phase amount is adjusted for each reception stream according to the interval between the preamble and the data, the influence of the residual frequency error can be more accurately removed, and the communication quality is improved.

Here, the channel matrix H and its inverse matrix H ⁻¹ as the antenna reception weight matrix are set as shown in the following equation (5). However, a, b, c, and d correspond to symbols of the additional preamble HT-LTF (101 to 104), and subscripts 0 to 3 are serial numbers of the reception branches.

  If the phase rotation amount per additional preamble is ω and the length of each additional preamble is T, the channel matrix (hereinafter also referred to as “tilt H”) when the residual frequency error f (= 2π / ωT) remains is It can be expressed as equation (6).

  The channel matrix in which the residual frequency error remains, that is, each column of tilt H is arranged in the order of time of channel estimation at each receiving antenna obtained from the additional preamble. That is, in channel estimation of the second, third, and fourth columns obtained from the second, third, and fourth additional preambles, each element is rotated from the first column by ω, 2ω, and 3ω, respectively. In the following equation (7), each row of the inverse matrix of the tilt H is obtained. Each row of the inverse matrix of the tilt H becomes an antenna reception weight vector, but as can be seen from the equation, the phases ω, 2ω, and 3ω caused by the residual frequency error are reversed for each business.

In other words, as shown in the following equation (8), the inverse matrix of the tilt H has a form in which the phase of the inverse matrix H ^{−1 of the} channel matrix that is not affected by the residual frequency error is rotated in units of rows.

While calculating the inverse matrix of the channel matrix in the spatial separation unit 36, the frequency offset compensation unit 34 uses the additional preamble HT-LTF (101 to 104) as indicated by reference numerals 111 to 115. The residual frequency error ω per additional preamble is measured and passed to the space separation unit 36. Then, the spatial separation unit 36 estimates the tilt H including the residual frequency error as the channel matrix, but corrects the inverse matrix later to provide an accurate inverse matrix H ⁻¹ that does not include the residual frequency error. Is calculated.

  In the above equation (6), the phase at the leading additional preamble 101 is used as a reference, and the phase of the leading data symbol 105 is rotated by 4ω. Therefore, in the spatial separation unit 36, the antenna reception weight vector in the first row of the inverse matrix of tilt H is 4ω, the antenna reception weight vector in the second row is 4ω−ω = 3ω, and the antenna reception weight vector in the third row is 4ω. -2ω = 2ω The antenna reception weight vector in the fourth row is subjected to phase correction of 4ω-3ω = ω, so that the influence of the residual frequency error can be removed from each antenna reception weight vector.

  With such phase correction, each antenna reception weight vector becomes an antenna reception weight vector suitable for the OFDM symbol position. The antenna reception weight vector in the second and lower rows contains a component that returns ω to 3ω, so that the phase of the data symbol can be matched with the time point of the leading additional preamble by correcting 3ω to ω. .

For example, if the transmission signal from the MIMO transmitter is x, the reception signal at the MIMO receiver is y, and the channel matrix is H, y = Hx. The frequency offset compensator 34 measures the phase rotation amount ω per additional preamble HT-LTF and gives it to the space separator 36. In the residual frequency error correction method described above, the spatial separation unit 36 obtains its inverse matrix from the tilt H that is a channel matrix that still includes the residual frequency error, and then, as shown in the following equation (9), An accurate inverse matrix H ⁻¹ is calculated by multiplying each row vector of the inverse matrix by a matrix that gives a phase rotation that corrects the residual frequency error for the additional preamble.

  As another method, there is a method in which the phase of the channel estimation value of the MIMO composite channel is corrected instead of correcting the phase of the antenna reception weight vector in the space separation unit 36. In this case, when the spatial separation unit 36 obtains the inverse matrix as an antenna reception weight matrix from the tilt H that is a channel matrix that still contains the residual frequency error (see equation (8)), this is received signal. Multiply y directly to obtain tilt x as a stream signal containing residual frequency error (see equation (10)).

  Then, the tilt x is correctly decoded while referring to the channel estimation shown in the following equation (11).

  Further, as another method for removing the influence of the residual frequency error, there is a method of correcting the data symbol for each corresponding stream. In this case, when the spatial separation unit 36 obtains an inverse matrix as an antenna reception weight matrix from the tilt H that is a channel matrix that still contains the residual frequency error, the spatial separation unit 36 directly multiplies the received signal y by this to obtain the residual frequency error. Tilt x is obtained as a stream signal including. Then, as shown in the following equation (12), the correct transmission signal x is decoded by multiplying the tilt x by a matrix that gives a phase rotation that corrects the residual frequency error for the additional preamble.

In a normal MIMO receiver, by multiplying the inverse matrix H ⁻¹ of the channel matrix H to become a unit matrix, the original transmission signal x is decoded by directly multiplying the reception signal by the inverse matrix H ^−1. . On the other hand, when the estimated channel matrix is affected by the residual frequency error, even if the inverse matrix is multiplied, it does not become a unit matrix. The second method corresponds to performing no phase correction of the combined channel estimation value by adding a phase rotation corresponding to the interval between the preamble position and the data symbol position in each received stream. Further, the third method corresponds to performing correction by giving the data symbol a reverse rotation of the phase according to the interval between the preamble position and the data symbol position in each received stream.

  FIG. 3 illustrates in detail the frequency offset compensation unit 34 and the space separation unit 36 in the reception system of the communication apparatus illustrated in FIG. 1. In the figure, four reception systems are included, but the gist of the present invention is not limited to the number.

  The frequency offset compensation unit 34 includes reception buffers 200 and oscillators 201 for each reception system, a synchronization detection unit 205 that detects synchronization using a packet preamble, and a frequency error estimation unit 206. In this embodiment, the frequency error estimation unit 206 performs initial frequency error estimation using the channel estimation preamble L-LTF and frequency error correction 212 for the oscillators 201. The residual frequency error is estimated using the preamble HT-LTF, and the phase rotation amount ω per additional preamble is measured and output to the space separation unit 36.

The space separation unit 36 includes an antenna reception weight matrix generation unit 207 and a MIMO channel synthesis unit 208. The antenna reception weight matrix generation unit 207 estimates the channel matrix H using the additional preamble HT-LTF (101 to 104) transmitted in a time division manner for each transmission stream, and further uses the inverse matrix H ⁻¹ as the antenna reception weight matrix. Calculate as MIMO channel combining section 208 multiplies the received signal by the antenna reception weight matrix to perform MIMO channel combining and spatially separates each received stream.

  In the present embodiment, the antenna reception weight matrix generation unit 207 corrects the residual frequency error by rotating the phase of the antenna reception weight matrix obtained from the channel matrix based on the phase rotation amount ω per additional preamble. Alternatively, the MIMO channel synthesis unit 208 corrects the residual frequency error by rotating the phase of the channel matrix of the MIMO-combined communication channel, or the residual frequency error by rotating the phase of the received data of the MIMO-communicated communication channel. Is to be corrected.

  Each received stream after the spatial separation is equalized and the residual frequency is removed by the equalization and phase timing tracking unit 203... And demodulated to the original value from the modulation point on the phase space by the decoding unit 204. Is done.

  Hereinafter, the processing procedure for correcting the residual frequency error in the frequency offset compensator 34 and the spatial separator 36 will be described in detail.

  The frequency error estimation unit 206 first estimates the frequency error using the preamble L-LTF (100), and sends the estimated value 212 to each of the oscillators 201, 201 ′, 201 ″, 201 ′ ″. Frequency correction of received signals after LTF (100) is performed.

  The additional preambles HT-LTF (101 to 104) transmitted in a time-sharing manner for each transmission stream are Fourier-transformed by the respective FFTs 202, 202 ′, 202 ″, 202 ′ ″, and the frequency domain preambles 211, 211 ′, 211 ″, 211 ′ ″. The antenna reception weight matrix generation unit 207 estimates a channel matrix from these additional preambles 211 and further calculates an inverse matrix thereof as an antenna reception weight matrix. The channel matrix estimated here is a tilt H (see Equation (6)) affected by the residual frequency error, and an inverse matrix of the tilt H (see Equation (7)) is obtained. become.

  On the other hand, the frequency error estimation unit 206 measures the residual frequency error by using at least one of the additional preambles HT-LTF (101 to 104) and performing phase comparison in any one of 111 to 114. In some cases, phase comparison is performed using the section 115 between the additional preamble HT-LTF (101 in the example shown in FIG. 2) that is phase-comparable with the preamble L-LTF and the L-LTF. To measure the residual frequency error. Then, the final residual frequency error 213 is estimated by, for example, averaging these measurement results, and sent to the space separation unit 36. For example, the phase rotation amount ω per additional preamble is output as the residual frequency error 213.

  The space separation unit 36 removes the influence of the residual frequency error based on the acquired residual frequency error 213. That is, the antenna reception weight matrix generation unit 207 applies the residual frequency error 213 for each row to the obtained inverse matrix (see Equation (6)), and corrects the phase of the antenna reception weight vector for each stream. The result 214 is sent to the MOMO channel combining unit 208 (see equation (9)). MIMO channel combining section 208 performs MIMO combining by multiplying the data symbol received from each antenna by the antenna reception weight vector, and generates received data of each independent stream.

  As another method for removing the influence of the residual frequency error in the spatial separation unit 36, the phase of the channel estimation value of the MIMO synthesis channel may be corrected instead of correcting the phase of the antenna reception weight vector. . In this case, the antenna reception weight matrix generation unit 207 obtains the inverse matrix as an antenna reception weight matrix from the tilt H that is a channel matrix that still contains the residual frequency error (see Equation (10)). Is directly multiplied by the received signal y to obtain a tilt x as a stream signal including a residual frequency error (see equation (10)). Then, the MIMO channel combining unit 208 correctly decodes the tilt x while referring to the channel estimation shown in the above equation (11).

  Further, as still another method for removing the influence of the residual frequency error in the spatial separation unit 36, the data symbol may be corrected for each corresponding stream. In this case, when the antenna reception weight matrix generation unit 207 obtains an inverse matrix as an antenna reception weight matrix from the tilt H that is a channel matrix that still includes the residual frequency error, the antenna reception weight matrix generation unit 207 directly multiplies the reception signal y by Tilt x is obtained as a stream signal including a residual frequency error (see equation (10)). Then, as shown in the above equation (12), the MIMO channel synthesizing unit 208 multiplies the tilt x by a matrix that gives a phase rotation that corrects the residual frequency error for the additional preamble, thereby obtaining an accurate transmission signal x. Is decrypted.

  The present invention has been described in detail above with reference to specific embodiments. However, it is obvious that those skilled in the art can make modifications and substitutions of the embodiment without departing from the gist of the present invention. That is, the present invention has been disclosed in the form of exemplification, and the contents described in the present specification should not be interpreted in a limited manner. In order to determine the gist of the present invention, the claims should be taken into consideration.

FIG. 1 is a diagram illustrating a configuration of a MIMO_OFDM communication apparatus according to an embodiment of the present invention. FIG. 2 is a diagram schematically showing a packet format used in the MIMO communication system. FIG. 3 is a diagram showing in detail the frequency offset compensation unit 34 and the space separation unit 36 in the reception system of the communication apparatus shown in FIG. FIG. 4 is a diagram conceptually showing the MIMO communication system. FIG. 5 is a diagram for explaining a training signal transmission method. FIG. 6 is a diagram illustrating a state in which inter-subcarrier interference occurs due to a frequency error in the MIMO_OFDM communication system. FIG. 7 is a diagram for explaining a method for performing channel estimation in the MIMO_OFDM communication system.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 ... Antenna 12 ... Switch 21 ... Encoding part 22 ... IFFT
DESCRIPTION OF SYMBOLS 23 ... Guard provision part 24 ... Preamble / reference provision part 25 ... D / A converter 26 ... Analog processing part for transmission 31 ... Analog processing part for reception 32 ... A / D converter 33 ... Synchronization acquisition part 34 ... Frequency offset compensation 35: FFT
36: Spatial separation unit 37 ... Phase rotation amount compensation unit 38 ... Decoding unit 200 ... Reception buffer 201 ... Oscillator 202 ... FFT
203 ... Equalization and phase timing tracking unit 204 ... Decoding unit 205 ... Synchronization detection unit 206 ... Frequency error estimation unit 207 ... Antenna reception weight matrix generation unit 208 ... MIMO channel synthesis unit

Claims (12)

A wireless communication system for performing MIMO (Multi Input Multi Output) communication using a plurality of logical channels formed by a pair of a transmitter having a plurality of antennas and a receiver having a plurality of antennas,
The transmitter adds a first preamble and a second preamble that can be phase-compared with the first preamble to a packet to be transmitted,
The receiver performs a phase comparison within the first preamble of the received packet to estimate a frequency error, and performs a phase comparison between the first preamble and the second preamble to estimate a frequency error.
A wireless communication system. The transmitter combines the first and second preambles with the same antenna transmission weight vector to enable phase comparison.
The wireless communication system according to claim 1. The transmitter transmits a preamble for frequency error estimation to a packet to be transmitted and a preamble for estimating a channel matrix on the receiver side in a time division manner from each antenna or each antenna combination. The channel matrix estimation preamble transmitted from the antenna is a second preamble that can be phase-compared with the frequency error estimation preamble.
The wireless communication system according to claim 1. The receiver corrects the frequency error of the received signal based on the phase comparison in the frequency error estimation preamble, and then in parallel with estimating the channel matrix using each channel matrix estimation preamble, Determining and correcting residual frequency error by phase comparison within at least one preamble for matrix estimation or by phase comparison between first and second preambles;
The wireless communication system according to claim 3. A wireless communication apparatus that includes a plurality of antennas and performs MIMO communication using a plurality of logical channels generated in pairs with a transmitter having a plurality of antennas, wherein the transmitter transmits a first packet to a packet to be transmitted. And a second preamble whose phase can be compared with the first preamble,
A frequency error correction unit that estimates and corrects a frequency error using the preamble;
Estimating a channel matrix of a spatially multiplexed channel using the preamble and obtaining an antenna reception weight matrix from the estimated channel matrix; and
A MIMO synthesis unit that multiplies the reception signal of each antenna by a reception weight matrix and separates it into each stream signal;
The frequency error correction unit estimates the frequency error by performing phase comparison within the first preamble of the received packet, and estimates the frequency error by performing phase comparison between the first preamble and the second preamble. To
A wireless communication apparatus. The transmitter transmits a preamble for frequency error estimation to a packet to be transmitted and a preamble for estimating a channel matrix on the receiver side in a time division manner from each antenna or antenna combination, and from any one antenna The transmitted channel matrix estimation preamble is a second preamble whose phase can be compared with the frequency error estimation preamble,
The frequency error correction unit is
A first frequency error correction unit for estimating and correcting a frequency error of a received signal based on phase comparison in a preamble for frequency error estimation;
In parallel with the channel matrix estimation unit estimating the channel matrix using each preamble for channel matrix estimation, the phase comparison within at least one preamble for channel matrix estimation or between the first and second preambles A second frequency error correction unit that obtains a residual frequency error by the first frequency error correction unit and corrects the residual frequency error by phase comparison of
The wireless communication apparatus according to claim 5, further comprising: The second frequency error correction unit corrects a residual frequency error by rotating a phase of an antenna reception weight matrix obtained by the channel matrix estimation unit;
The wireless communication apparatus according to claim 6. The second frequency error correction unit corrects a residual frequency error by rotating a phase of a channel estimation value of the communication channel synthesized by the MIMO synthesis unit;
The wireless communication apparatus according to claim 6. The second frequency error correction unit corrects the residual frequency error by rotating the phase of the reception data of the communication channel synthesized by the MIMO synthesis unit;
The wireless communication apparatus according to claim 6. The second frequency error correction unit adjusts a correction phase amount for each stream according to the interval between each preamble for channel matrix estimation and data transmitted in a time division manner to correct a residual frequency error.
The wireless communication apparatus according to claim 6. A wireless communication method that includes a plurality of antennas and performs MIMO communication using a plurality of logical channels generated in pairs with a transmitter having a plurality of antennas, wherein the transmitter transmits a first packet to a packet to be transmitted. And a second preamble whose phase can be compared with the first preamble,
A frequency error correction step of estimating and correcting a frequency error using the preamble;
Estimating a channel matrix of a spatially multiplexed channel using the preamble, and obtaining an antenna reception weight matrix from the estimated channel matrix; and
A MIMO combining step of multiplying the reception signal of each antenna by a reception weight matrix and separating it into each stream signal,
In the frequency error correction step, the phase error is estimated within the first preamble of the received packet to estimate the frequency error, and the phase error is estimated between the first preamble and the second preamble to estimate the frequency error. To
A wireless communication method. The transmitter transmits a preamble for frequency error estimation to a packet to be transmitted and a preamble for estimating a channel matrix on the receiver side in a time division manner from each antenna or antenna combination, and from any one antenna The transmitted channel matrix estimation preamble is a second preamble whose phase can be compared with the frequency error estimation preamble,
In the frequency error correction step,
A first frequency error correction for estimating and correcting a frequency error of a received signal based on a phase comparison in a preamble for frequency error estimation;
In parallel with estimating the channel matrix using each preamble for channel matrix estimation in the channel matrix estimation step, phase comparison within at least one preamble for channel matrix estimation or between the first and second preambles A second frequency error correction for obtaining a residual frequency error by the first frequency error correction and correcting the residual frequency error by phase comparison of
The wireless communication method according to claim 11, wherein:

Images