JP6649305B2 - Wireless communication device and wireless communication method - Google Patents

Description

  The present invention relates to a wireless communication device and a wireless communication method.

[Background surrounding 5th generation mobile communication]
At present, high-performance mobile communication terminals such as smartphones are exploding. With regard to mobile phones, the third generation mobile communication has shifted to the fourth generation mobile communication, and research and development on the further fifth generation mobile communication (commonly known as "5G") are currently underway. One of the studies performed on 5G is the use of macro cells and small cells.

  In conventional mobile phones, one service area is set to a radius of several kilometers, and one macro station area is covered by one base station device. However, there is a very large number of users in such a macrocell. Since the entire system capacity is shared by each user, when accommodating an enormous number of users, the throughput for each user decreases.

  In order to avoid such a decrease in throughput, a technology for setting a small cell, which is a very small service area with a radius of about several tens of meters, in a densely populated area where traffic is concentrated has been developed. In this technology, small traffic is used to offload spot-like traffic to a network without going through a macro cell. Here, it is assumed that the terminal device can simultaneously and concurrently use the communication capability in the small cell and the communication capability in the macro cell. By using such a terminal device, user data is accommodated on the small cell side while exchanging control information using a macro cell. This makes it possible to maximize the advantages of the macro cell and the small cell.

  In 5G described above, the target value of the transmission rate is set to 10 Gbit / s (gigabit per second) or more, and even in this small cell, the same large-capacity communication is performed to realize efficient offload of traffic. There is a need to. In a macro cell, it is assumed that a microwave band having a low frequency is used to allow long-distance propagation. However, considering the current situation in the microwave band where frequency resources are already depleted, small cells that assume relatively short distance communication are expected to use quasi-millimeter wave bands or millimeter wave bands with relatively high frequencies. Have been. The feature of this high frequency band is that the propagation attenuation increases in inverse proportion to the square of the frequency. Therefore, it is preferable that the small cell base station is ideally installed at a location close to the user terminal. For example, in a place where installation is easy, such as on the roof of a building, the distance between the user terminal and the base station is too large, which is not preferable in line design.

  On the other hand, since the small cell is set in a place where traffic is concentrated, even in a place where it is difficult to lay an optical fiber, there is a case where the installation of the base station device is strongly desired. For example, assuming that small cell base station devices are installed in places where there are very many people, such as in front of stations such as Shinjuku and Shibuya, propagation attenuation increases on the rooftop of a building adjacent to such places. . Therefore, installation on a place lower than the roof of the building, for example, on the wall surface of the building, may be required. However, it may be difficult to lay optical fiber on the wall of an existing building, and in such cases, it may be necessary to provide a backhaul line to the base station device using a wireless line. is there.

  When such a backhaul line is provided, it is necessary to perform a large-capacity transmission of 10 Gbit / s or more similarly utilizing a millimeter wave band in order to cope with a large-capacity transmission of 10 Gbit / s or more required in a small cell. There is. In such an environment, since both opposing radio station devices are fixedly installed in a stable place, it is natural that the line of sight is generally secured naturally and the directional antennas are directed toward each other. is there. In this case, although reflected waves between buildings and the like exist to some extent, most of the received signals are line-of-sight components and are expected to be in a state that is difficult to say in a multipath environment. This situation is the same for the access system if the small cell base station device is installed in a high place such as a building wall surface and is generally used in a line-of-sight environment in a manner of looking down on the user from above.

  Next, for large-capacity transmission of 10 Gbit / s or more, which is the transmission rate required for 5G, frequency resources with a very wide bandwidth can be used by utilizing the millimeter wave band. Is growing. For example, if a backhaul line using a millimeter wave band is assumed, the E-band (71 to 76 GHz and 81 to 86 GHz) is used as an example, and if a 1 GHz bandwidth is used, the frequency utilization efficiency is 10 bits. / S / Hz. However, existing radio equipment for achieving a frequency utilization efficiency of 10 bit / s / Hz generally employs spatial multiplexing transmission using a multiple-input multiple-output (MIMO) channel. Spatial multiplexing transmission generally uses a multipath environment, and when a singular value decomposition of a channel matrix H expressing a transfer function of a MIMO channel in a matrix format is performed, a distribution of an absolute value of a singular value obtained as a result is obtained. Represents the characteristics of the spatial multiplexing transmission. Specifically, the square of the absolute value of the singular value is a value proportional to the signal-to-noise ratio (SNR). Communication is not established unless it has a sufficiently large value even after the second singular value. The same applies to large-capacity transmission in small cells that are access systems, but realizing spatial multiplexing transmission in an environment where such line-of-sight waves are dominant is indispensable for realizing a target wireless system. .

  As described above, in 5G, use of the millimeter wave band is expected for both the access system and the backhaul line. As described above, the characteristic of the high frequency band is that the propagation attenuation increases in inverse proportion to the square of the frequency. For example, comparing the 2 GHz band (existing access system) with the 80 GHz band (future system using the millimeter wave band), since the frequency is 40 times, the propagation attenuation is 1600 times and the line gain of 32 dB is insufficient. Will be. Of course, it is not necessary to cover all of 32 dB because it is not necessary to cover a wider area than a macro cell. However, in a high frequency band, a high power amplifier in a transmission stage does not have a high output device. In this case, it is considered that additional line gain must be secured at a level of several tens of dB. Furthermore, since spatial multiplexing transmission is also expected in such an environment, a radio system equipped with a much larger number of antenna elements than in the prior art has been studied in both the base station and the terminal station. . Such a technique is called Massive MIMO. In the following, conventional techniques related to Massive MIMO are introduced.

  In Non-Patent Document 1, the base station side has 256 elements and the terminal station side has 16 elements, and aims at spatial multiplexing transmission of 16 streams using a channel matrix of 256 × 16 size. In this Non-Patent Document 1, in order to transmit a plurality of signal sequences (streams), the directivity is formed by analog beamforming using rotation of a complex phase amount in a wireless analog circuit and digital baseband circuit. Performed in combination with digital beamforming in the digital domain. This avoids the frequent use of analog / digital (A / D) converters and digital / analog (D / A) converters, and reduces power consumption and takes measures against lack of line gain when feeding back channel information. However, for the analog beam formed here, for example, a plurality of fixed patterns are set in the horizontal direction and the vertical direction at a predetermined interval (for example, at intervals of 5 degrees). Select and use as many as you want. Naturally, since the analog beam cannot be accurately directed in the direction of arrival of the line-of-sight wave, selecting a redundant number of analog beams sufficiently compresses the required number of A / D converters and D / A converters. Can not.

[Expansion of Massive MIMO technology when line of sight is dominant]
In the above description of the Massive MIMO technology, since it is mainly assumed to be used in an access system, it is generally assumed that the environment is a multipath environment. However, even in the access system, if the small cell base station is installed at the top, and it is possible to look down and look roughly, it may be necessary to operate in a non-multipath environment. This is particularly remarkable in the case of a backhaul line, and is assumed to be used in an environment in which a so-called Rice coefficient K is 10 dB or more and a multipath component is only about 1/10 or less of a line-of-sight component. In this case, the difference between the line gain corresponding to the first singular value and the line gain corresponding to the second singular value or more is expected to be 20 dB or more, and spatial multiplex transmission of two or more streams is inefficient. It is expected that

  In such an environment, as shown in Non-Patent Document 2, all antenna elements are divided into a plurality of sets by utilizing the high line gain efficiency corresponding to the first singular value, and a sub-array is set for each set. The configuration is effective. By arranging the sub-arrays spatially separated, it is effective to reduce the correlation between the sub-arrays and to transmit the transmission corresponding to the first singular value in parallel with low correlation. Similarly, in Non-Patent Document 3, when the antenna aperture length of the subarray here is narrow, the line of sight wave is dominant, and the correlation between the antenna elements in the subarray is sufficiently strong, the transmission / reception weight of each antenna element is the frequency. A "time-axis beamforming technique" has been proposed, which can be treated as a constant having no dependence, and in this case, weight multiplication can be performed in units of sampling data on the time axis.

This is a relative value of channel information for each antenna element (a relative value of channel information with respect to channel information of an antenna element serving as a reference, when the element spacing is narrow and the correlation of antenna elements is strong, If the complex phase of the channel information of the k-th frequency component of the element is ψ _ref ^(k) , the frequency dependence of the complex phase of information obtained by multiplying each antenna element by Exp {−j _ref ^(k) } This is a method based on the fact that it is almost constant. For example, at the time of reception, the complex phase of the reception weight for synthesizing the received signals of these antenna elements such that the complex phases are the same is substantially constant in the entire frequency band, and Thus, it is possible to use reception weights having the same constant. Generally, a Fourier transform of a function that becomes a constant on the frequency axis becomes a δ function. Therefore, the weight obtained by transforming the reception weight on the frequency axis on the time axis by IFFT only needs to consider the component at t = 0. Become. In other words, since it is not necessary to perform signal processing in consideration of the delay wave component, the time base, which is a predetermined coefficient for each antenna element, is directly added to the sampling data obtained by sampling the analog / baseband received signal by the A / D converter. If the reception weight is multiplied, the directivity can be completely formed only by signal processing on the time axis without converting the reception signal into a signal on the frequency axis by FFT processing or the like.

  A coefficient for rotating the complex phase to be multiplied as a time axis weight is obtained by the following equations (1) to (3).

In the above equation, S _i (n) represents the sampling data of the n-th sample of the i-th antenna in the received training signal, and S _i (n) ^* represents the complex conjugate of S _i (n). Represent. N _FFT assumes a predetermined periodicity, and indicates a periodicity value that is significant in correlation detection, such as the number of FFT points in OFDM. ψ _j is the amount of rotation of the complex phase (on the receiving side) performed by time axis beamforming. The function angle (x) is a function representing the complex phase of the complex number x, and is a value determined by the ratio of the real part of x to the imaginary part of x and the sign of the real part and the imaginary part. Although the in the correlation calculation in the formula (1) performs a correlation operation over N _FFT samples is the number of FFT points in the case of the OFDM signal as a predetermined periodicity, be an integer multiple of example N _FFT Correlation calculations over other sample numbers may be performed so that periodicity is maintained.

Here, as is clear from the equation (2), the complex phase of the complex coefficient c _j given by the above equation (1) and the complex phase of the time axis weight w _j given by the above equation (2) have inverted signs. Has become something. In this sense, in the background art and embodiments of the present invention to be described later, obtaining the complex phase of the complex coefficient c _j given by Expression (1) corresponding to the relative channel information, and calculating the complex phase of the time axis weight w _j Finding the phase is equivalent.

  FIG. 5 is a functional block diagram illustrating a configuration example (sub-array separation type) of a wireless station device using time-axis beamforming in the related art described in Non-Patent Document 3. The wireless station device shown in FIG. 1 includes a baseband signal processing circuit 140 and transmission / reception signal processing circuits 929-1 to 929 -N. The baseband signal processing circuit 140 includes modulators 120-1 to 120-N (MOD # 1 to MOD # N), a signal separation circuit 141, and demodulators 130-1 to 130-N (DEM # 1 to DEM #). N) and a control circuit 460. The transmission / reception signal processing circuit 929-n (n = 1,..., N) includes a time axis transmission weight multiplication circuit 921-n, D / A converters 922-n-1 to 922-n-M, and an up converter 923. -N-1 to 923-n-M, down converter (DC) 924-n-1 to 924-n-M, A / D converters 925-n-1 to 925-n-M, and time axis A reception weight multiplication circuit 926-n and a TDD switch (TDD-SW) 927-n are provided. TDD switch 927-n is connected to antenna elements 928-n-1 to 928-n-M. Here, N corresponds to the multiplexing number (the number of streams) when performing spatial multiplexing, and the transmission / reception signal processing circuits 929-1 to 929 -N are mounted in N systems in total. M represents the number of antenna elements of the sub-array mounted on each of the transmission / reception signal processing circuits 929-1 to 929-N.

  Here, the transmission / reception signal processing circuits 929-1 to 929-N (the transmission / reception signal processing circuits 929-n are accompanied by sub-array antenna elements 928-n-1 to 928-n-M) are described in Non-Patent Document 2. It is assumed that they are installed spatially apart from each other. Further, the baseband signal processing circuit 140 is connected to each of the transmission / reception signal processing circuits 929-1 to 929 -N by wire, and the digital baseband signal is transferred over the wire. The control circuit 460 manages the frame period and the transmission / reception timing, and here also manages the switching of the TDD switches 927-1 to 927-N.

  Furthermore, the time-axis beamforming technology basically presupposes signal processing on the time axis. However, even when a signal on the frequency axis is formed as in the OFDM modulation method, the signal on the frequency axis is subjected to FFT processing and IFFT processing. Can be converted into a signal on the time axis, and by performing signal processing on the time axis signal, the time axis beamforming technique can be similarly applied to the OFDM modulation method together with the single carrier transmission. However, since signal processing on the frequency axis is not assumed here, the modulators 120-1 to 120-N and the demodulators 130-1 to 130-N in FIG. It is considered that the functions of the FFT processing and the IFFT processing are included in N (or the signal separation circuit 141), and these notations are omitted. Therefore, irrespective of the OFDM modulation method or single carrier transmission, the input / output signals from the modulators 120-1 to 120-N and the demodulators 130-1 to 130-N are signals on the time axis. . Further, the signal separation circuit 141 performs signal separation between each signal sequence. Here, it is also possible to perform signal separation on the time axis. Frequency-dependent signal separation may be performed. In this sense, the input to the demodulators 130-1 to 130-N is assumed to be a signal on the time axis or a signal on the frequency axis. The description will be made assuming that the user inputs "."

  The specific signal flow is as follows. First, signal transmission will be described. The modulators 120-1 to 120-N generate transmission signals of the time base digital baseband of the respective streams to be spatially multiplexed, and input these to the time base transmission weight multiplying circuits 921-1 to 921-N. I do. The time-axis transmission weight multiplying circuits 921-n (n = 1,..., N) are each a sub-array for forming the directivity of the digital signal input from the modulator 120-n by the transmission / reception signal processing circuit 929-n. The signal is converted into a digital signal obtained by multiplying the transmission weights corresponding to the antenna elements 928-n-1 to 928-n-M. The D / A converters 922-n-1 to 922-n-M convert the digital signal multiplied by the transmission weight into an analog baseband signal, and the up-converters 923-n-1 to 923-n-M , Convert the analog baseband signal into a signal in a radio frequency band. At the time of transmission, the TDD switch 927-n connects the upconverter 923-nm (m is an integer of 1 or more and M or less) to the antenna element 928-nm. The signals input from the up-converters 923-n-1 to 923-n-M are transmitted from the antenna elements 928-n-1 to 928-n-M, and the transmission / reception signal processing circuits 929-1 to 929 are transmitted. A directional beam is formed every −N.

  Next, reception of a signal will be described. The signals received by the antenna elements 928-n-1 to 928-n-M (n = 1,..., N) are input to the down converters 924-n-1 to 924-n-M via the TDD switch 927-n. Is done. The down converters 924-n-1 to 924-n-M convert radio frequency signals into analog baseband signals. The A / D converters 925-n-1 to 925-n-M convert an analog baseband signal into a digital baseband signal. This digital baseband signal is input to the time axis reception weight multiplying circuit 926-n. The time axis reception weight multiplying circuit 926-n multiplies each of the input signals by the reception weight corresponding to each of the antenna elements 928-n-1 to 928-n-M, and adds and synthesizes the signals after the reception weight multiplication. And converts them into a single signal sequence. That is, the signals are converted into a total of N signal sequences (streams) by the time axis reception weight multiplying circuits 926-1 to 926 -N, and these signals are input to the signal separating circuit 141. The signal separation circuit 141 performs signal separation by suppressing a crosstalk component between the streams, and inputs the separated signals to the corresponding demodulators 130-1 to 130-N. The demodulators 130-1 to 130-N reproduce and output data by a predetermined signal detection process.

  The suppression of the crosstalk component performed by the signal separation circuit 141 can be performed on the time axis, or can be temporarily converted to a frequency axis signal by FFT processing and then performed on the frequency axis. Alternatively, only the signal processing performed by the transmission / reception signal processing circuits 929-1 to 929 -N may be performed, and the signal separation circuit 141 may not perform any processing. However, in any case, the details of the signal separation method here are omitted because they are not directly related to the present application.

  The time axis transmission weights used in the time axis transmission weight multiplying circuits 921-1 to 921-N and the time axis reception weights used in the time axis reception weight multiplication circuits 926-1 to 926-N are shown here. The time axis transmission / reception weight acquisition means obtains it. Similarly, a control circuit (not shown) manages the value of the time axis transmission / reception weight used there. For example, based on the sampling data obtained by the A / D converters 925-n-1 to 925-n-M for a training signal transmitted by a wireless station as a communication partner, the reference antenna element is transmitted over a predetermined number of samples. A correlation value between the antenna elements (for example, 928-n-1) may be determined, and the correlation value may be determined based on the complex phase. The complex phase values of the time axis reception weight and the time axis transmission weight generally do not match because the amount of rotation of the complex phase of a power amplifier and a low noise amplifier (not shown) is different between individual amplifiers. By using the implicit feedback calibration method, it is possible to convert the time axis reception weight to the time axis transmission weight. The transmission / reception weights thus obtained are stored in the memory for each corresponding wireless station device. Then, at the time of transmission and reception, the time axis transmission weight multiplication circuits 921-1 to 921-N and the time axis reception weight multiplication circuits 926-1 to 926-N perform weight multiplication based on the values of these transmission and reception weights. Will be.

  FIG. 6 is a functional block diagram showing a configuration example (shared sub-array type) of a wireless station device using time-axis beam forming in the conventional technology described in Non-Patent Document 3. In the figure, the same functions as those in FIG. 5 are denoted by the same reference numerals. In the figure, a radio station apparatus 942 includes a baseband signal processing circuit 140, transmission / reception signal processing circuits 929-1 to 929-N, distribution couplers (HYB) 941-1 to 941-M, and an antenna element 928- 1 to 928-M. In FIG. 5, the transmission / reception signal processing circuits 929-1 to 929 -N are housed in different housings assuming that the transmission / reception signal processing circuits 929-1 to 929 -N are installed in places spatially separated for each sub-array, 140 shows a configuration in which a wired connection is made between the wireless LAN and the wireless LAN. On the other hand, in FIG. 6, all the transmission / reception signal processing circuits 929-1 to 929 -N and the baseband signal processing circuit 140 are configured as a wireless station device 942 in the same housing, and the antenna elements 928-1 to 928 -M are entirely included. Is shared by For this reason, for example, at the time of transmission, signals from the TDD switches 927-1 to 927-N of the respective transmission / reception signal processing circuits 929-1 to 929-N are combined by the distribution couplers 941-1 to 941-M and combined. The transmitted signal is transmitted from the antenna elements 928-1 to 928-M. Similarly, at the time of reception, the signals received by the antenna elements 928-1 to 928-M are distributed by the distribution couplers 941-1 to 941-M. That is, the distribution coupler 941-m (m = 1,..., M) distributes the signal received by the antenna element 928-m to the TDD switches 927-1 to 927-N and inputs the signal. All other signal processing is common to FIG. 5 and FIG.

  Here, in actual operation, the wireless station device shown in FIG. 5 and the wireless station device shown in FIG. 6 communicate with each other. For example, the base station apparatus has a degree of freedom in installation as on the roof of a building, and a sub-array can be installed at a plurality of locations. On the other hand, when there is a large restriction on the installation of a building wall or the like on the terminal station device side, the terminal station device shares the sub-array with one array antenna by using the configuration of FIG. 5 as the base station device and FIG. 6 as the terminal station device. It is possible to increase the degree of freedom of installation in the form. Alternatively, for example, if the transmission capacity per terminal station apparatus does not require spatial multiplexing, FIG. 6 shows the base station apparatus, and FIG. 5 shows only one of the transmission / reception signal processing circuits 929-1 to 929-N (for example, 5, only the transmission / reception signal processing circuit 929-1) is implemented, and a configuration in which multi-user MIMO transmission is performed by a plurality of terminal stations and one base station is also possible. It is possible.

[Calibration technology in channel information feedback]
In general, when directivity is formed using a plurality of antenna elements on the transmission side, it is necessary to feed back MIMO channel information including the techniques described in Non-Patent Documents 1 to 3. At this time, if the number of antenna elements becomes enormous, the amount of channel information to be fed back becomes enormous, so various measures are required. In the Massive MIMO system as described above, in order to obtain channel information of a forward link in a transmission direction, channel information of a reverse link in a reception direction is used, and a complex phase of a reception signal generated by a circuit such as a low noise amplifier used at the time of reception is used. The relationship between the amount of rotation and the amount of rotation of the complex phase of the transmission signal generated by a circuit such as a high power amplifier used at the time of transmission is converted, and the channel information of the reverse link is multiplied by a predetermined calibration coefficient to convert the It is possible to obtain channel information. Generally, these techniques are known as implicit feedback techniques (for example, see Non-Patent Document 4). Although the details of the calibration process are omitted here, the rotation amount of the complex phase of the received signal generated by a circuit such as a low noise amplifier used at the time of reception using any conventional technique and the circuit generated by a circuit such as a high power amplifier used at the time of transmission are used. The description is made on the assumption that the individual difference and uncertainty of the rotation amount of the complex phase of the transmission signal for each antenna system can be eliminated.

  In the wireless transmission system using a high frequency band such as the millimeter wave band described above, there are problems with the A / D converter and the D / A converter, and the up-converter and the down-converter as described below. I do.

  In the time-axis beamforming technology described in the above “Extension of Massive MIMO technology when line-of-sight waves are dominant”, the number of A / D converters, D / A converters, upconverters, and downconverters is equal to the number of antenna elements. Will be needed. Further, in the configuration example shown in FIG. 6, these are required by the number of antenna elements × the number of streams. In general, the power consumption of A / D converters and D / A converters used in an environment with a very high clock speed due to the wide band tends to increase, and as a result, the heat generation also becomes a non-negligible level. In the case of a device that generates a large amount of heat, a heat radiating plate or the like for radiating the heat is required, and these devices physically need to have a size proportional to the amount of the generated heat. Also, it is difficult to reduce the size of A / D converters and D / A converters that constantly consume power due to the need for heat radiation for expected heat generation during operation in addition to the design requirements of the circuit itself. And become larger.

  Originally, in a system that assumes use of the millimeter wave band, the wavelength of the antenna element itself becomes shorter as the frequency increases, and the size of the antenna element itself can be reduced in a form proportional to this wavelength. For example, if the size of the antenna element can be designed to be sufficiently small with respect to the wavelength, and the antenna elements are arranged at about 1/2 wavelength, the size of the entire array antenna should be designed to be small even if the number of antenna elements is large. Becomes possible. For example, if the center frequency is 80 GHz assuming the E band, one wavelength is 3.75 mm. If an array antenna of 16 × 16 = 256 elements is formed with a wavelength array and a square array structure, the size of the entire array antenna will fall within a size of about 6 cm × 6 cm. However, in order to realize mounting in this size, the A / D converter and the D / A converter themselves are also configured to fit within 3.75 mm square, which is one wavelength, and the amount of heat generation expected during operation is reduced. A heat sink for heat dissipation must also be accommodated in this. If this is not possible, the actual size of the array antenna will depend on the size of the A / D converter and the D / A converter, and the size of the heat sink for heat dissipation. For example, if a wideband signal having a bandwidth of 1 GHz is handled, it is necessary to perform the sampling process performed in the baseband at a super speed of at least 1 GHz. The higher the speed, the higher the power consumption and the larger the amount of heat generated. As described above, the size cannot be reduced due to the large power consumption, despite the use of the Massive MIMO technology which can be reduced in size due to the high frequency band. For these reasons, reducing the power consumed by the A / D converter and the D / A converter is a major issue.

  Furthermore, in addition to the power consumption, an ultra-high-speed A / D converter and a D / A converter are also expensive, so that if the number of antenna elements is in units of 100, the cost is further reduced. There is a need to reduce the total number of A / D converters and D / A converters. The same applies to the up converter and the down converter, and mounting the up converter and the down converter for the number of antenna elements leads to an increase in cost. In addition, local oscillators input to the up-converter and the down-converter need to eliminate the uncertainty of the complex phase in order to form directivity. Therefore, it is required to share local signals. The loss due to distributing an enormous number of these high-frequency signals for each antenna element and the influence of mutual interference between the systems (leakage of signals on the board wiring) cannot be ignored. In particular, to avoid mutual interference between the systems, it is ideal to separate the signals of the respective systems from each other at a physical distance. It leads to conversion.

  Therefore, in order to make the array antenna of each wireless station device smaller, even when utilizing time-axis beamforming, it is consumed by the A / D converter and the D / A converter mounted as much as possible. Power needs to be reduced.

  Further, as described above, in the Massive MIMO technology using a high frequency band, the antenna size can be physically reduced due to the short wavelength, and the overall size of the device can be reduced. In this case, for example, when switching between a transmission system and a reception system using a TDD switch or the like while sharing a transmission / reception antenna, the TDD switch prevents a transmission signal from leaking into the reception system and ensures sufficient isolation. There is a need. However, when the size of the device is reduced, the TDD switch also needs to have a simple configuration. In order to ensure isolation between the transmitting and receiving systems, there is a method of increasing the isolation by using multiple switches. . In order to avoid such a problem, it is assumed that the transmitting antenna and the receiving antenna are physically separated. However, if the transmitting antenna and the receiving antenna are different, the channel of the implicit feedback is used. Since the symmetry is broken, there is a problem that the implicit feedback cannot be used.

  In Non-Patent Document 5 and Non-Patent Document 6, as a technique for solving such a problem, based on the result of channel estimation using some antenna elements, the complex phase of the other entire antenna elements is determined. A method of acquiring rotation amount information or information on transmission / reception weight has been proposed (see Non-Patent Documents 5 and 6).

  In the time axis beamforming technique, for example, at the time of reception, a main path of an incoming wave between a transmitting and receiving station is extracted, and signal processing for directing the directional gain of the antenna element group in the direction is performed. The transmission and reception weights used at this time are constants having no frequency dependence, and as a result, signal processing is reduced at various points. However, since digital signal processing is basically used, transmission / reception weights are multiplied individually for each antenna element in digital signal processing, and an A / D (analog / digital) conversion is performed separately for each antenna system. And a D / A (digital / analog) converter.

  However, multiplication processing of coefficients having no frequency dependence does not necessarily require digital signal processing if it is limited to, for example, only rotation processing of a complex phase without a change in amplitude. Specifically, by using a phase shifter that is an analog circuit and passing the analog signal through this phase shifter set to a predetermined complex phase rotation, signal processing substantially equivalent to weight multiplication processing is performed. Can be realized. In Non-Patent Document 1, directivity control is realized using a phase shifter. This is performed, for example, in a predetermined direction set in horizontal / vertical directions at intervals of 5 degrees. A combination set is determined in advance, and the phase rotation amount of each phase shifter is set according to the direction in which the beam obtained by some control procedure should be directed. However, this combination set of phase rotation amounts is selected from a preset menu, and it is necessary to search for the direction in which the reception level is the highest while individually transmitting a training signal for each direction. . However, in time-axis beamforming, the amount of phase rotation of each antenna element is optimized based on the training signal transmitted from the terminal station apparatus side, so that the calculation of the amount of phase rotation used for forming directivity is simply fed back. In addition to this, the combination set of the amount of rotation of the complex phase can be set with a remarkably high degree of freedom without requiring a menu or the like in advance.

  Therefore, a wireless station device (wireless communication device) in the related art employs digital assist type analog beamforming. That is, the wireless station device performs the calculation of the amount of rotation of the complex phase performed by each phase shifter by digital signal processing, and controls the phase shifter using the complex phase value obtained by the digital signal processing. Thus, a desired complex phase is rotated to form directivity on an analog signal.

  First, time-axis beamforming corresponds to control for directing a directivity to a path where the intensity of an incoming wave is maximum. Therefore, in the plane wave approximation of the arriving wave from the direction of the path having the maximum intensity, the length of the perpendicular drawn from the antenna element arranged in a plane to the wavefront of the plane wave (the plane where the complex phase of the signal is the same) is The path length difference has a geometric periodicity, and is given as a function of coordinates. The transmission / reception weight is a coefficient corresponding to the rotation of the complex phase for canceling the path length difference.

  Of course, the rotation amount of the complex phase for each antenna element slightly deviates from the beautiful periodicity because it actually involves a multipath component and also includes a measurement error. However, approximately, an x-axis and a y-axis are set on a plane on which the antenna is arranged, and three-dimensional notation is performed with respect to a coordinate point of the antenna element as a rotation amount of a complex phase on the z-axis. If the amount of rotation of the complex phase is plotted against the coordinates, all plot points will be on one plane in the plane wave approximation. In addition, since the complex number Exp (−j ｛ψ ± 2π × integer ψ) is completely equivalent even if ± 2π × integer is added to the rotation amount 複 素 of the complex phase, each antenna is large when the whole antenna aperture is large. Depending on the coordinates of the elements, there are cases where it is necessary to consider that they exist on a plane as a value obtained by adding ± 2π × integer. However, considering such a complex phase offset, the rotation amount of the complex phase can be obtained with a small number of antenna elements. The remaining antenna elements can approximately acquire the rotation amount of the complex phase by linear interpolation based on the path length difference.

  Further, even in the case where the uplink / downlink channel asymmetry that separates the transmission antenna and the reception antenna is involved in the design of the apparatus, the rotation amount of the complex phase is approximately obtained by considering the path length difference. It is possible to do. In a system using a plurality of antenna elements, generally, the antenna elements are generally set apart by a half wavelength or more in order to avoid coupling between the antenna elements. Conversely, it is possible to operate at the shortest element spacing of about half a wavelength, but in a high frequency band system with a short wavelength, the physical distance between antenna elements is extremely narrow It is expected to be. For example, assuming an 80 GHz band, the wavelength is about 3.75 mm, and antenna elements may be arranged at intervals of several millimeters. Here, when switching between transmission and reception is performed in a time-division manner and a transmission / reception antenna is shared, a high-power amplifier for transmission and a low-noise amplifier for reception are mounted in this interval, and these amplifiers are mounted. If the signals are close to each other, a signal from the high power amplifier may leak to the low noise amplifier.

  Also, a switch for time-division switching is small enough to fit in this size, but it is necessary to ensure sufficient isolation between the transmission side and the reception side, which is in a severe condition. Furthermore, when introducing a time division switch, the signal amplified by the high power amplifier is input to the antenna element via the time division switch, so the transmission output of the high power amplifier passes through the time division switch Loss due to the passing loss. Similarly, since the signal received by the antenna passes through the time division switch before being input to the low noise amplifier, the signal is also lost by the passage loss caused by passing through the time division switch. SNR characteristics deteriorate. In particular, when a time-division switch is composed of multi-stage switches in order to secure transmission / reception isolation, the passage loss increases, and this effect cannot be ignored. From such a background, there are cases where the transmission antenna and the reception antenna are separately designed in the device, and a measure corresponding to the asymmetry of the uplink / downlink channel is required.

  However, even in such a case, if, for example, the path length difference is ΔL and the wavelength is λ, the amount of rotation of the complex phase for canceling the path length difference is common (2πΔL / λ) in both transmission and reception. Given by At this time, if the amount of rotation of the complex phase of at least three antenna elements is obtained, the amount of rotation of the complex phase of the remaining antenna elements is obtained from the amount of rotation of the complex phase of each antenna element coordinate on the plane including the three points. Can be calculated. If the amount of rotation of the complex phase can be calculated with four or more antenna elements, the z-axis coordinate (the amount of rotation of the complex phase) of the plane in the three-dimensional space given by the coordinates of each antenna element and the amount of rotation of the complex phase is calculated. By using the least squares approach to minimize the error with the amount of rotation of the complex phase, the plane with the smallest sum of the square errors is calculated, and the amount of rotation of the complex phase with respect to each antenna coordinate on the plane is calculated. , May be set in a phase shifter.

  Even if the uplink and downlink (that is, transmission / reception) antenna elements have different configurations as described above, the uplink antenna element and the downlink antenna element are mixed and the overall antenna configuration is reduced. When taking the amount of rotation of the complex phase of the antenna element in the uplink receiving antenna, the amount of rotation of the complex phase of the uplink assuming that the other antenna elements for downlink are also used in the uplink Can be calculated, and a calibration process can be performed based on the phase rotation amount of the antenna element in the uplink, and the rotation amount of the complex phase of the antenna element for the downlink can be calculated.

  Alternatively, even when the uplink antenna element group and the downlink antenna element group are physically separated, the uplink antenna element group and the downlink antenna element group are simply arranged in parallel. In the case of a moving relationship (that is, when the relative positional relationship between the antenna elements does not change between the transmitting and receiving antennas on the same plane), the amount of rotation of the downlink complex phase obtained by the receiving antenna is transmitted as it is. Even if it is used as the complex phase of the corresponding antenna element on the side, if the distance between the transmitting and receiving stations is sufficiently large and it is expected that a common plane wave approximation is possible, a sufficiently approximate effective directivity It is expected that formation is possible.

  As shown in FIGS. 5 and 6, a “sub-array separation type” configuration in which a directional beam is divided into a plurality of sub-arrays and a “sub-array shared” that realizes a plurality of directional beams in one array are provided. Although there is a variation in the configuration according to the “type” (strictly speaking, it is not separated into sub-arrays, it may be understood as an “integrated array”), the “sub-array separated type” will be described here.

  FIG. 7 shows an outline of a prediction of a rotation amount of a complex phase in a conventional linear array in which antenna elements are linearly arranged. FIG. 3 shows a state in which the antenna elements 401-1 to 401-5 are linearly arranged. First, consider a plane wave arriving from the angle θ direction with respect to the front direction of the linear array. The distance between the elements of each antenna is d. Here, the received signal at the antenna element 401-1 is Φ1 (t), the received signal at the antenna element 401-2 is Φ2 (t),..., And the received signal at the antenna element 401-5 is Φ5 (t). Let ΔLs be the path length difference of the s-th antenna element (s is an integer of 2 or more) with respect to antenna element 401-1. ΔL1 is set to 0 for convenience.

  In general, when the wavelength is λ and passes through the path length ΔL, the complex phase rotates by 2πΔL / λ. Assuming plane wave approximation, the path length difference between antenna element 401-1 and antenna element 401-2 is ΔL2 = d · Sin θ. Similarly, the path length difference between antenna element 401-1 and antenna element 401-3 is ΔL3 = 2 × d · Sin θ, and the path length difference between antenna element 401-1 and antenna element 401-4 is ΔL4 = 3 × d · Sin θ, and the path length difference between the antenna element 401-1 and the antenna element 401-5 is ΔL5 = 4 × d · Sin θ.

  Therefore, if attention is paid to the frequency component of a certain wavelength λ, a wave for which the plane wave approximation is established is Φ2 (t) ≒ Exp ｛-2πjΔL2 / λ｝ Φ1 (t), Φ3 (t) ≒ Exp ｛-2πjπL3 / λ｝ The relationship of Φ1 (t), Φ4 (t) ≒ Exp ｛−2πjΔL4 / λ｝ Φ1 (t), Φ5 (t) ≒ Exp ｛−2πjｊL5 / λ｝ Φ1 (t) holds. Using the above relationship of ΔLs = (s−1) d · Sinθ, Φs (t) ≒ Exp ｛−2πj (s−1) d · Sinθ / λ｝ Φ1 (t), and each antenna element has This means that the complex phase is rotated by Exp {-2πjd · Sinθ / λ}. Therefore, for example, channel estimation is performed by the antenna elements 401-1 and 401-5, and based on the rotation amount of the complex phase between them, １ ／ of the difference of the complex phase rotates for each antenna element. It becomes possible to predict.

  Similar prediction is possible not only when the antenna elements are arranged one-dimensionally but also when the antenna elements are arranged two-dimensionally. Next, with reference to FIG. 8, a description will be given of a specific example of a conventional technique for predicting a rotation amount of a complex phase of an array antenna configured in a plane. In FIG. 8, a to z and A to K in the alphabets shown by “O” represent antenna elements. Here, an example of a shape in which equilateral triangles are spread in a close-packed manner is shown, but the shape may be any other configuration.

  For example, it is assumed that the wireless station device performs channel estimation at three points of antenna elements i, m, and q indicated by double circles, and obtains a complex phase of channel information at the three points. Here, for example, the antenna element i is used as a reference antenna, the rotation amount of the complex phase between the antenna element i and the antenna element m is ζ, and the rotation amount of the complex phase between the antenna element i and the antenna element q is η. . In this case, focusing on the antenna element on the straight line connecting the antenna element i and the antenna element m, ζ × １ ／ for the antenna element c, ζ × ２ for the antenna element d, and ζ × 4 / for the antenna element B. 3, the antenna element u can approximate -ζ × １ ／. Similarly, focusing on the antenna element on the straight line connecting the antenna element i and the antenna element q, η × １ ／ for the antenna element b, η × ２ for the antenna element g, and η × 4 / for the antenna element G 3. With the antenna element v, it can be approximated to -η × １ ／. If this is extended, two-dimensional prediction is also possible. For example, for the antenna element a, ζ × １ ／ + η × １ ／, for the antenna element H, −ζ × １ ／ + η × 4/3, For the antenna element E, it can be predicted as ζ × 2/3 + η × 3/3.

  As a condition for enabling the above prediction, the rotation amount of the complex phase needs to be π or less between two antenna elements among the antenna elements for acquiring the rotation amount information of the complex phase. For example, taking the case of FIG. 7 as an example, even if the complex phase difference between the antenna element 401-1 and the antenna element 401-5 is π / 2 or -3π / 2, Exp ｛ j · π / 2} = Exp {−j · 3π / 2} and cannot be distinguished. If the complex phase difference is π / 2, the complex phase difference between the antenna element 401-1 and the antenna element 401-2 should be π / 2 × (1/4). Is −3π / 2, the complex phase difference between antenna element 401-1 and antenna element 401-2 should be −3π / 2 × (１ ／).

  In the situation where there is uncertainty as described above, it is not possible to correctly predict the rotation amount information of the complex phase, so the complex phase difference between the antenna elements that obtains the rotation amount information of the complex phase must be π or less. There is. In an actual measurement, a complex phase may fluctuate due to a measurement error due to noise or an influence of multipath. Therefore, actually, the complex phase difference needs to be a small value with a margin larger than π. The standard value cannot be unconditionally determined because the influence of the reflected wave differs depending on the magnitude of the reflected wave. However, when the antenna element distance d is reduced or the arrival angle θ is sufficiently small, the path length difference ΔL decreases. Therefore, the amount of change in the complex phase due to the path length difference can be made sufficiently small. In the example of FIG. 8, the three points of the antenna elements i, m, and q have a relatively distant positional relationship. However, in order to reduce the complex phase difference between the antenna elements to π or less, the mutual element spacing is small. It is also possible to use adjacent antenna elements (for example, antenna elements a, e, f, etc.). FIG. 8 shows an example in which the channel is estimated by three antenna elements and the complex phases of the remaining antenna elements are approximated in a two-dimensional plane. Is obtained, it is possible to predict the rotation amount information of the complex phase with a little more fineness. Note that, of course, there is no essential restriction on the arrangement of the antenna elements. Therefore, there is no necessity for a densely packed structure as shown in FIG. 8, and for example, a square lattice array can be used.

  Next, a specific example of predicting the rotation amount of the complex phase of a square array antenna formed in a planar shape in the related art will be described with reference to FIGS. Although the details will be described later, the description will be made here assuming a configuration in which the transmitting antenna and the receiving antenna are separated. In FIG. 9, a to r represented by “「 ”represent reception antenna elements, and s to z and A to J represented by“ ● ”represent transmission antenna elements. Here, an example is shown in which the antenna elements are arranged in a square lattice, instead of a shape in which equilateral triangles are spread in a close-packed manner. For example, it is assumed that the wireless station device performs channel estimation at three points of antenna elements a, l, and p indicated by double circles, and determines the complex phase of the channel information at the three points.

  For example, let antenna element a be a reference antenna, let the rotation amount of the complex phase between antenna element a and antenna element 1 be ζ, and let the rotation amount of the complex phase between antenna element a and antenna element p be η. In this case, the linear prediction is slightly more complicated than in FIG. 8, but the basic concept is the same. For example, the square lattice is considered as a lattice point on the xy plane, and the antenna element a is regarded as the origin. At this time, for example, if the coordinates are defined as (1, 0) for the antenna s and (0, 1) for the antenna v (here, for convenience, the y-axis has a downward positive direction). In this case, the antenna 1 corresponds to (5, 3) and the antenna p corresponds to (1, 5).

When the amount of phase rotation is represented by the z-axis and expressed in three dimensions, (0,0,0) for the antenna a, (5,3, は) for the antenna 1, and (1,5, η) for the antenna p. Become. Equation expressing a two-dimensional plane, a, b, the use of coefficients c (this factor not related to the identifier of the antenna _{elements), a (x-x 0} ) + b (y-y 0) + c (z- z ₀ ) = 0. For example, since (0, 0, 0) is a point on the plane, if (x ₀ , y ₀ , z ₀ ) = ( ₀ , ₀ , ₀ ), the relational expression of ax + by + cz = 0 is obtained. Here, (a, b, c) is a normal vector of this plane, and since the absolute value of the vector has no meaning, if a ′ = a / c and b ′ = b / c, the unconstant is In other words, a relational expression of a'x + b'y + z = 0 is obtained. On the other hand, since the coordinates (5, 3, ζ) and (1, 5, η) are on a plane, a two-dimensional linear simultaneous equation for a ′ and b ′ can be established and solved. Then, the values of a 'and b' can be easily obtained.

  When a 'and b' obtained as described above are used, z =-(a'x + b'y). Therefore, by substituting the coordinates of each antenna element into x and y, the complex phase of each antenna element is obtained. Will be found. For example, z = − (3a ′ + b ′) is a desired value for the antenna e because the coordinates are (3, 1), and z = − (for the antenna r because the coordinates are (5, 5). 5a '+ 5b') is a desired value. By doing so, it is possible to estimate the rotation amount information of the complex phase of the remaining antenna elements from the rotation amount information of the complex phase for a small number of antenna elements by the same plane wave approximation without depending on the configuration of the antenna arrangement. It is possible.

It should be noted that the method of utilizing such an equation indicating the relationship between the coordinates of each antenna element and the complex phase of each antenna element is supplemented a little. In the above description regarding FIG. 8, the case where the complex phases of the three antenna elements are obtained and the complex phases of the other antenna elements are obtained by linear approximation therefrom has been described. It is also possible to approximate all the antenna elements a to z and A to K with one plane wave using the multiplication method. For example, in a two-dimensional plane where the antenna elements a to z and A to K exist, arbitrary orthogonal x-axis and y-axis are determined, and the coordinates of the k-th antenna element are (x _k , y _k ). The complex phase φ _k of the k antenna elements is given by the following equation (4) using the coefficients of α, β, and γ.

On the other hand, the complex phase of the k-th antenna element actually estimated using the first antenna element (for example, antenna element a in the figure) as a reference antenna is represented by ~ φ _k (tilde φ indicates “~” above φ). In the following, a combination of (α, β, γ) that minimizes the following evaluation function W (α, β, γ) can be obtained by the least squares method. (Equation (5)).

Based on the combination of (α, β, γ) obtained by the above least squares method, a required complex phase is calculated from the coordinates (x _k , y _k ) of the k-th antenna element using Expression (4). It becomes possible. When the least squares method is used, an arbitrary number of antenna elements for obtaining a complex phase can be selected. Originally, the present invention approximates an incoming wave with a plane wave, but actually includes components other than the line-of-sight wave (plane wave) under the influence of the reflected wave, so that a clean relationship such as equation (4) is obtained. Not be. The error from this plane affects the estimation accuracy of the complex phase.By increasing the number of antenna elements used for the least squares method, it is possible to average the effect of this reflected wave, The accuracy can be improved. On the other hand, when the number of antenna elements used for the least squares method is increased, the circuit scale is increased. Therefore, the number of antenna elements is set in each trade-off.

  In the above description, the procedure for estimating the complex phase of the channel information has been described. Since the amount of rotation corresponds to the amount of rotation, the amount of rotation of the complex phase may be set as a value on the z-axis, and the calculation processing for directly calculating the amount of rotation of the complex phase may be performed.

  In FIG. 9, the receiving antenna elements indicated by “で” corresponding to a to r and the transmitting antenna elements indicated by “●” corresponding to s to z and A to J are nested. For example, if the transmitting antenna and the receiving antenna are separated and each is arranged relatively close, the rotation amount information of the complex phase is acquired by some of the receiving antennas by the above-described method, and based on the information, If information on other antenna elements is predicted, the predicted antenna element may be a transmitting antenna or a receiving antenna. The rotation amount information can be predicted. Note that the transmitting antenna and the receiving antenna do not need to be arranged differently at the re-adjacent lattice points, and may be arranged in another general manner.

  FIG. 10 shows another example of the arrangement of transmitting and receiving antennas of a square array antenna formed in a planar shape in the related art. The difference from FIG. 9 is that, in FIG. 9, the transmitting antennas and the receiving antennas are alternately arranged like Othello cells, whereas in FIG. 10, the transmitting antennas or the receiving antennas are arranged in a vertical column. It is a point that is aligned. In this case, for example, the antenna elements forming a pair of one transmission signal processing circuit and one reception signal processing circuit may be arranged as adjacent antennas such as the antenna elements a and s and the antenna elements d and v in FIG. Will be.

  FIG. 11 shows another example of the arrangement of transmitting and receiving antennas of a square array antenna formed in a planar shape in the related art. The difference from FIG. 10 is that, in FIG. 10, the transmitting antennas and the receiving antennas are arranged in a vertical column, but the transmitting antennas and the receiving antennas are alternately arranged in adjacent columns. In FIG. 11, all the receiving antennas a to r are combined in one place, and similarly, all the transmitting antennas s to z and A to J are also combined in one place, and each is arranged in a different area. ing.

  The advantage of separating the transmission system and the reception system is that, for example, when operating without turning off the power of the high power amplifier of the transmission system even at the time of signal reception, it is possible to prevent the noise of the transmission system from leaking to the reception system. In addition, since the entire transmission system and the entire reception system are separated from each other, mutual isolation can be easily ensured. On the other hand, as shown in FIGS. 9 and 10, in order to estimate the complex phase rotation amount information of an antenna element that has not actually acquired the complex phase rotation amount information by signal reception, using the above-described method. It is not preferable in terms of structure. However, if the antenna arrangements of the transmitting and receiving antennas of the opposing radio station apparatuses # 1 and # 2 have some kind of symmetry, the complex phase rotation amount information (channel On the premise of performing appropriate calibration processing on the information (i.e., the information), it is possible to treat the information as it is as the rotation amount information of the complex phase on the transmission side. As an example, in FIG. 11, the receiving antenna element a and the transmitting antenna element s, the receiving antenna element b and the transmitting antenna element t, the receiving antenna element c and the transmitting antenna element u, the receiving antenna element d and the transmitting antenna element v. Are geometrically translated, and even if the rotation amounts of the complex phases of the reception antennas a to r are directly applied to the transmission antennas s to J in consideration of such symmetry, the plane wave approximation It can be assumed that there is no significant difference where possible.

  It should be noted that the method of selecting the antenna element used for the prediction of the amount of rotation of the complex phase is not limited to using the antenna element at a relatively distant position as shown in FIGS. It is also possible to use. FIG. 12 shows an example of an antenna pattern used for predicting the rotation amount of the complex phase in the background art of the present invention. FIG. 12 shows examples of four patterns (a) to (d) in a 5 × 5 square array. In the drawing, ● and ◎ indicated by symbols a to y represent antenna elements, and based on the rotation amount of the complex phase obtained using the antenna element indicated by ◎, the complex phases of other antenna elements indicated by ● Predict the amount of rotation. Here, a case where five antenna elements are used to determine the amount of rotation of the complex phase is shown as an example, but it goes without saying that the present invention can be implemented with another number of elements equal to or more than three.

  For example, taking FIG. 12A as an example, the rotation amount of the complex phase is obtained using the antenna elements h, l, m, n, and r, and the antenna element m near the center of gravity and the other antenna elements h, l , N, r are calculated, and the complex phase of the correlation value is determined for the antenna elements h, l, n, r. The complex phase obtained in this manner is set on the z-axis, and the least square method similar to that described in Equations (4) and (5) is used. The rotation amount of the complex phase may be estimated. FIG. 12B is the same.

  In addition, in order to expand the spatial spread of antenna elements used for prediction of the amount of rotation of a complex phase, a diagram in which elements other than the first proximity and the second proximity of the reference antenna are used for prediction of rotation amount information of a complex phase. It is also possible to use the patterns 12 (c) and (d). This case is similar to the case of FIGS. 12 (a) and 12 (b), but in the case of (c), for example, the antenna element r is set as the reference antenna, and the antenna element r, the antenna element 1 and the antenna element n After calculating the complex phase of the correlation value, the antenna element f and the antenna element j are directly correlated with the reference antenna element r, but the correlation between the antenna element l and the antenna element f is calculated. The calculation and the correlation calculation between antenna element n and antenna element j may be performed. The sum of the relative value of the complex phase of the antenna element l with respect to the antenna element r and the relative value of the complex phase of the antenna element f with respect to the antenna element l is approximately equal to the relative value of the complex phase of the antenna element f with respect to the antenna element r. It can be considered. Similarly, the sum of the relative value of the complex phase of the antenna element n with respect to the antenna element r and the relative value of the complex phase of the antenna element j with respect to the antenna element n is approximately the complex value of the antenna element j with respect to the antenna element r. It can be considered as a relative value of the phase.

  More specifically, the correlation value between the antenna element r and the antenna elements f and j is directly calculated, while the relative value of the complex phase of the antenna element l with respect to the antenna element r is removed in order to remove the uncertainty of the complex phase of the 2π period. The sum of the relative value of the complex phase of antenna element f with respect to antenna element l and the sum of the relative value of the complex phase of antenna element n with antenna element r and the relative value of the complex phase of antenna element j with antenna element n You may use it in another form. In this case, a value obtained by adding an integral multiple of 2π to a value obtained by directly calculating the correlation value between the antenna element r and the antenna elements f and j, the relative value of the complex phase of the antenna element l with respect to the antenna element r, and the antenna element l , And the sum of the relative value of the complex phase of the antenna element n with respect to the antenna element r and the relative value of the complex phase of the antenna element j with respect to the antenna element n is the closest. Alternatively, the correlation value between the antenna element r and the antenna elements f and j may be corrected.

  The reason for including the process of calculating and adding the complex phase difference in a plurality of stages as described above is that when the complex phase difference between the desired antenna elements is ± π or more, the phase of the complex phase is reduced due to the periodicity of the complex phase. Since the uncertainty of the 2π period cannot be ignored, by adding and using the complex phases of the correlation values between adjacent antenna elements, the complex phase difference between the antenna elements used for estimating the rotation amount of the complex phase becomes π. It is possible to avoid the above, and as a result, it is possible to avoid the uncertainty of the complex phase of the 2π period.

  FIG. 13 is a diagram illustrating a configuration example of a communication system according to the related art. In FIG. 13, reference numerals 450 and 452 denote wireless station devices. Here, for example, a configuration is shown in which a wireless station device 450 corresponding to a base station device communicates with a wireless station device 452 corresponding to a terminal station. It is also possible to perform spatial multiplex transmission on the same frequency.

Here, the wireless station device 450 shown in FIG. 7 includes a baseband signal processing circuit 140 and transmission / reception signal processing circuits 651-1 to 651-N. The baseband signal processing circuit 140 includes modulators 120-1 to 120-N (MOD # 1 to MOD # N), a signal separation circuit 141, and demodulators 130-1 to 130-N (DEM # 1 to DEM #). N) and a control circuit 460. In the wireless station device 450, each of the plurality of transmission / reception signal processing circuits 651-1 to 651-N forms one beam as a sub-array. On the other hand, the wireless station device 452 has a common (sub) array, and one array antenna forms a plurality of beams. In an actual operation, as shown in FIG. 13, the wireless station apparatus 450 faces the wireless station apparatus 452, so that it is possible to spatially multiplex N-system signals. The transmission / reception signal processing circuits 651-1 to 651-N each have a common configuration, receive a signal from the modulator 120-n, and output an output to the signal separation circuit 141.
Hereinafter, the configuration of the transmission / reception signal processing circuits 651-1 to 651-N of the related art will be described with reference to FIGS.

FIG. 14 is a functional block diagram showing a configuration example of a transmission / reception signal processing circuit 651-n (n = 1,..., N) in the related art shown in Non-Patent Document 5.
The transmission / reception signal processing circuit 651-n shown in the figure includes a D / A converter 122-n, an up converter (UC) 123-n, a down converter (DC) 124-n, and an A / D converter 125-n. n, TDD switch (TDD-SW) 127-n, phase shifters 402-n-1 to 402-n-M, switches 403-n-1 to 403-n-M, and distribution coupler 404- n, down converters (DC) 424-n-1 to 424-n-2, A / D converters 425-n-1 to 425-n-2, a correlation calculation circuit 405-n, and phase shift control The circuit includes a circuit 406-n and a complex phase rotation amount prediction circuit 410-n.

  The transmission / reception signal processing circuit 651-n is connected to the antenna elements 401-1 to 401-M and the baseband signal processing circuit 140. Although FIG. 2 shows two configurations of the down converter 424-n and the A / D converter 425-n, the down converter 424-n and the A / D converter 425-n have values smaller than M. Any value may be used. In the following description, there are two down converters 424-n and two A / D converters 425-n (down converters 424-n-1 to 424-n-2 and A / D converters 425-n-1 to 425-n). The case of n-2) will be described as an example.

  A down converter 424-n, an A / D converter 425-n, and the like are arranged in three systems (generally, three systems or more and M-1 systems or less) among the antenna elements 401-1 to 401-M. If the procedure of acquiring the rotation amount information of the complex phase described with reference to FIGS. is there.

  The specific signal flow is shown below. The signal processing is roughly divided into three types: signal processing, signal reception processing, and signal transmission processing when calculating the amount of rotation of the complex phase performed by the phase shifters 402-n-1 to 402-n-M. Will be explained. Further, the control circuit 460 shown in FIG. 13 is a circuit for managing the entire communication. Although not described in detail, the control circuit 460 manages overall timing management and various processes, for example, switching of a TDD switch 127-n and the like, and phase shift. It manages a series of management such as management of setting timing of the devices 402-n-1 to 402-n-M. Therefore, strictly speaking, signals may be input / output to / from each functional block, but are omitted here for simplicity.

  First, the signal processing performed by the phase shifters 402-n-1 to 402-n-M when calculating the amount of rotation of the complex phase will be described. In the signal processing for calculating the amount of rotation of the complex phase, the switch 403-n-1 connects the down converter 424-n-1 to the phase shifter 402-n-1, and the switch 403-n-4 switches down. A converter 424-n-2 and a phase shifter 402-n-4 are connected (actually, a down converter and an antenna system switch provided with an A / D converter not shown here are also similar), The remaining switches are left unconnected. In this state, first, a training signal for channel estimation is transmitted from the wireless station device of the communication partner from which the rotation amount of the complex phase is to be obtained, and the wireless station device receives this signal.

  The signals received by the antenna elements 401-n-1 to M are input to the phase shifters 402-n-1 to 402-n-M, and are analogized by the phase shifters 402-n-1 to 402-n-M. A predetermined complex phase rotation (for example, 0 degree may be applied) is applied on the signal, and each is input to the switches 403-n-1 to 403-n-M. Switches 403-n-1 and 403-n-4 (actually, the same applies to an antenna system switch including a down converter and an A / D converter not shown here, Therefore, the signals input to the two antenna systems will be input to the down converters 424-n-1 to 424-n-2. The down converters 424-n-1 to 424-n-2 convert the input radio frequency signals into analog baseband signals. The A / D converters 425-n-1 to 425-n-2 convert analog signals into digital baseband signals.

  The information converted into digital baseband signals by the A / D converters 425-n-1 to 425-n-2 is input to the correlation calculation circuit 405-n. The correlation calculation circuit 405-n calculates the rotation amount of the complex phase based on the input information by using Expressions (1) to (3). When the calibration process is required as needed, the rotation amount of the complex phase on the transmission side is determined as a value in consideration of the calibration coefficient in Expressions (1) to (3). The rotation amount of the complex phase obtained by the correlation calculation circuit 405-n is input to the complex phase rotation amount prediction circuit 410-n together with the identification number of the wireless station device to communicate with.

  The complex phase rotation amount prediction circuit 410-n predicts complex phase rotation amount information of the remaining antenna elements based on complex phase rotation amount information (or channel information) of a limited antenna system. In the case of FIG. 14, the complex phase rotation amount prediction circuit 410-n includes the antenna elements 401-1 and 401-4 (actually, a down converter and an A / D converter not shown are provided. Based on the complex phase rotation amount information obtained in the same manner, the complex phase rotation amount information of the remaining antenna elements is predicted. The complex phase rotation amount prediction circuit 410-n calculates the amount of rotation of the complex phase to be performed by the phase shifters 402-n-1 to 402-n-M, and together with the identification number of the wireless station device to communicate with, Input to the phase shift control circuit 406-n. In the phase shift control circuit 406-n, the value input here is managed by being stored in a memory or the like. Note that the control circuit 460 performs overall adjustment, timing control, and the like in a series of these processes.

  Next, signal processing when performing actual data communication, that is, when performing transmission or reception processing will be described. First, when performing data communication, all the switches 403-n-1 to 403-n-M are connected to the distribution coupler 404-n. In addition, a control circuit 460 (not shown) grasps a wireless station device to be a communication partner, and instructs a phase shift control circuit 406-n of a rotation amount of a complex phase corresponding to the wireless station device performing communication. Instruct the phase shifters 402-n-1 to 402-nM.

  First, signal processing when transmitting a signal will be described. At the time of transmission, a signal is transmitted while the TDD switch 127-n connects the up-converter 123-n and the distribution coupler 404-n (n = 1,..., N). The modulator 120-n (not shown) generates a time-base digital baseband transmission signal of each stream to be subjected to spatial multiplexing, and inputs the transmission signal to the transmission / reception signal processing circuit 651-n. The transmission / reception signal processing circuit 651-n (n = 1,..., N) receives the transmission signal of the time base digital baseband generated by the modulator 120-n (not shown).

  The D / A converter 122-n of the transmission / reception signal processing circuit 651-n (n = 1,..., N) converts the transmission signal input from the modulator 120-n into an analog baseband signal, and Input to converter 123-n. The up-converter 123-n converts the signal input from the D / A converter 122-n from a baseband signal to a signal in a radio frequency band, and inputs the signal to the TDD switch 127-n. The TDD switch 127-n inputs the signal input from the up converter 123-n to the distribution coupler 404-n.

  The distribution coupler 404-n branches the analog signal input from the TDD switch 127-n into M-system analog signals. The branched analog signals are input to phase shifters 402-n-1 to 402-n-M via switches 403-n-1 to 403-n-M. The phase shifters 402-n-1 to 402-n-M apply a predetermined complex phase rotation on the analog signal. For example, the phase shifters 402-n-1 to 402-n-M apply a predetermined amount of complex phase rotation to a signal in a radio frequency band in an analog signal format. The analog signals subjected to the complex phase rotation by the phase shifters 402-n-1 to 402-n-M are transmitted via the antenna elements 401-n-1 to 401-n-M, respectively. The transmission signal has a predetermined directivity formed by complex phase rotation in the phase shifters 402-n-1 to 402-n-M. Here, one transmission / reception signal processing circuit 651-n has been described, but for example, the antenna elements 401-1-1 to 401-1-M and the antenna elements 401-N-1 to 401-N-M In this case, individual directivity is formed, and communication is performed with a wireless station device in the directivity direction.

  Next, reception of a signal will be described. At the time of reception, the TDD switch 127-n receives a signal with the distribution coupler 404-n connected to the down converter 124-n (n = 1,..., N). The signals received by the antenna elements 401-n-1 to 401-nM (n = 1,..., N) are input to the phase shifters 402-n-1 to 402-nM, respectively. Each of the phase shifters 402-n-1 to 402-n-M applies a predetermined complex phase rotation on an analog signal to the input signal, and outputs the result through switches 403-n-1 to 403-n-M. And input to the distribution coupler 404-n.

  The distribution coupler 404-n combines signals of the respective antenna systems input via the switches 403-n-1 to 403-n-M on an analog signal, and combines the combined signals via the TDD switch 127-n. Input to the down-converter 124-n. The down-converter 124-n converts a radio frequency signal input via the TDD switch 127-n into an analog baseband signal, and inputs the analog baseband signal to the A / D converter 125-n. The A / D converter 125-n converts the signal input from the down converter 124-n from an analog baseband signal to a digital baseband signal, and outputs the signal to a signal separation circuit (not shown). .

  The signal separation circuit 141 in FIG. 13 performs signal separation by suppressing crosstalk components between streams, and inputs the separated signals to the corresponding demodulators 130-1 to 130-N. The suppression of the crosstalk component performed by the signal separation circuit 141 can be performed on the time axis, or can be temporarily converted to a frequency axis signal by FFT processing and then performed on the frequency axis. When suppressing the crosstalk component on the time axis, first, the correlation between the signal sequences input to the signal separation circuit 141 is calculated based on the digital baseband signal corresponding to the received training signal. . Then, based on the MIMO channel matrix given by the calculated correlation, a matrix similar to a general MIMO signal separation process such as ZF (Zero Forcing) type or MMSE (Maximum Mean Square Error) type signal separation is calculated. Then, this matrix may be multiplied by a sampling data unit with respect to the signal vector input to the signal separation circuit 141.

  This is because, in general MIMO signal separation processing such as ZF type or MMSE type signal separation on the frequency axis, different matrices are used for each frequency component, but almost the same matrix on the frequency axis. In the case where is used, it corresponds to the fact that signal separation processing can be performed on the time axis in units of sampling data. Alternatively, only the signal processing performed by the transmission / reception signal processing circuits 651-1 to 651-N may be sufficient, and the signal separation circuit 141 may not perform any processing. However, in any case, the details of the signal separation method here are omitted because they are not directly related to the features of the present embodiment. The demodulators 130-1 to 130-N demodulate the signal in which the crosstalk component has been suppressed in the signal separation circuit 141.

Although not explicitly shown in FIG. 14, if a transmitting-side high-power amplifier (HPA) is disposed, for example, the transmitting-side high-power amplifier (HPA) is disposed at a point indicated by “A” in FIG. if placing LNA) or the like, will be placed at a point with a description of "B" and "C _1" - "C _2" in FIG. Regarding “A” and “B”, since the transmission / reception signal processing circuit 651-n does not assume that they cooperate with each other, a calibration process for removing the uncertainty of the complex phases of the individual high-power amplifiers and low-noise amplifiers Is unnecessary, but regarding the low-noise amplifier at the point where “C ₁ ” to “C ₂ ” are described, the antennas 401-1, 401-4, etc. in the same transmission / reception signal processing circuit 651-n are not connected. It is necessary to remove the uncertainty of the complex phase of the low-noise amplifier of each system by using a prior art implicit feedback calibration method or the like, because the uncertainty of the complex phase between them may be a cause of the uncertainty. .

The present invention can be applied to any method, and the specific method of the calibration process is not limited. Considering this calibration result, for example, assuming that the rotation amounts of the complex phases at “C ₁ ” and “C ₂ ” are +10 degrees and +20 degrees, the complex phases obtained by the equations (1) to (3) are obtained. Is corrected by -10 degrees and -20 degrees to adjust the phase rotation amount. The information on the calibration result is collected by a calibration circuit (not shown), and is collected by the phase shift control circuit 406-n, the complex phase rotation amount prediction circuit 410-n, or the correlation calculation circuit 405-n. The correction is performed using the information.

In the figure, the switches 403-n-2, 403-n-3, 403-n-5 to 403-n-M of the system without the down converter and the A / D converter have the phase shifter 402. −n−1 to 402−n−M, the connection state is set in the signal processing when calculating the amount of rotation of the complex phase, or these switches 403-n−2, 403−n−3, and 403−n− Omitting 5 to 403-nM does not cause any problem in practice.
It should be noted that the phase rotation by the phase shifter normally gives phase rotation by selectively passing through a delay line corresponding to the amount of phase rotation on the device. Therefore, when a phase rotation of x is given as an absolute value, a negative phase rotation (delay) of the complex phase rotation amount is performed with a delay corresponding to the phase x as a signal, and code consistency cannot be obtained. . However, in the following description for convenience, will be referred to as "the rotation amount of complex phase carried out in the phase shifter [psi _alpha" when giving a phase rotation equivalent to multiplication of the coefficient Exp {jψ _α} as a signal with the phase shifter I do.

  FIG. 15 is a functional block diagram showing another configuration example of the transmission / reception signal processing circuit 652-n according to the related art shown in Non-Patent Document 5. The same numbers are assigned to the same functional blocks as those in the above-described drawings.

  The transmission / reception signal processing circuit 652-n shown in the figure includes a D / A converter 122-n, an up converter (UC) 123-n, a down converter (DC) 124-n, and an A / D converter 125-n. n, distribution couplers 414-n, distribution couplers 415-n, phase shifters 402-n-1 to 402-n-M, switches 403-n-1 to 403-n-M, Phasers 409-n-1 to 409-n-M, down converters (DC) 424-n-1 to 424-n-2, A / D converters 425-n-1 to 425-n-2, , A correlation calculation circuit 405-n, a phase shift control circuit 406-n, and a complex phase rotation amount prediction circuit 410-n. Further, antenna elements 401-n-1 to 401-n-M and antenna elements 441-n-1 to 441-n-M are connected to the transmission / reception signal processing circuit 652-n.

  The difference from FIG. 14 is that the antenna elements for transmission 401-n-1 to 401-n-M and the antenna elements for reception 441-n-1 to 441-n-M are separated, and as a result, TDD While the switch 127-n becomes unnecessary, the distribution coupler 404-n shared for transmission and reception is separated into a distribution coupler 414-n for transmission and a distribution coupler 415-n for reception. The shared phase shifters 402-n-1 to 402-n-M are phase shifters 409-n-1 to 409-n-M for transmission and phase shifters 402-n-1 to 402-402 for reception. -N-M. Other configurations are the same.

  As in FIG. 14, the amount of complex processing rotation performed by the phase shifters 402-n-1 to 402-n-M and the phase shifters 409-n-1 to 409-n-M can be roughly divided for signal processing. The calculation will be described separately for three types of signal processing, signal reception processing, and signal transmission processing. Similarly, a control circuit 460 shown in FIG. 13 is a circuit for managing the entire communication, and although not described in detail, manages the entire timing management and various processes. For example, the phase shifters 402-n-1 to 402-n -M and a series of management such as management of the setting timing of the phase shifters 409-n-1 to 409-n-M. Therefore, strictly speaking, signals may be input / output to / from each functional block, but are omitted here for simplicity.

  First, the signal processing performed by the phase shifters 402-n-1 to 402-n-M and the phase shifters 409-n-1 to 409-n-M when calculating the amount of rotation of the complex phase will be described. . In the signal processing for calculating the amount of rotation of the complex phase performed by the phase shifters 402-n-1 to 402-n-M and the phase shifters 409-n-1 to 409-n-M, the switches 403-n- 1 connects the down converter 424-n-1 and the phase shifter 402-n-1 and the switch 403-n-4 connects the down converter 424-n-2 and the phase shifter 402-n-4. (Actually, the same applies to a switch of an antenna system including a down converter and an A / D converter not shown here), and the other switches are not connected.

  These switches are managed by the correlation calculation circuit 405-n under the instruction of the control circuit 460 (not shown), and the phase shifter 402-n- is used except when calculating the rotation amount of the complex phase. 1-402-nM are connected to the distribution coupler 415-n. When performing this process, the phase rotation amounts of the phase shifters 402-n-1 to 402-n-M are set to predetermined values. The amount of rotation of the complex phase obtained in the subsequent processing is set as a difference from an initial predetermined value, as in the above description. Also, other transmission / reception signal processing circuits 652-n (not shown) perform the same processing all at once under the control of the control circuit 460 (not shown).

  In this state, first, a training signal for channel estimation is transmitted from the wireless station device of the communication partner from which the rotation amount of the complex phase is to be obtained, and the wireless station device receives this signal. The signals received by the antenna elements 441-n-1 to 441-n-M are input to the phase shifters 402-n-1 to 402-n-M, and the phase shifters 402-n-1 to 402-n-M are input. At M, a predetermined complex phase rotation is applied on the analog signal, and each is input to the switches 403-n-1 to 403-n-M. Switches 403-n-1 and 403-n-4 (actually, the same applies to an antenna system switch including a down converter and an A / D converter not shown here, Therefore, signals input to these two antenna systems will be input to down converters 424-n-1 to 424-n-2. The down-converters 424-n-1 to 424-n-2 convert the input radio frequency signals into analog baseband signals, and the A / D converters 425-n-1 to 425-n-2 perform analog conversion. Converts a signal to a digital baseband signal.

  The information converted into digital baseband signals by the A / D converters 425-n-1 to 425-n-2 is input to the correlation calculation circuit 405-n. The correlation calculation circuit 405-n calculates the rotation amount of the complex phase by using Expressions (1) to (3). When the calibration process is required as needed, the rotation amount of the complex phase on the transmission side is determined as a value in consideration of the calibration coefficient in Expressions (1) to (3). The rotation amount of the complex phase obtained by the correlation calculation circuit 405-n is input to the complex phase rotation amount prediction circuit 410-n together with the identification number of the wireless station device to communicate with.

  The complex phase rotation amount prediction circuit 410-n predicts complex phase rotation amount information of the remaining antenna elements based on complex phase rotation amount information of a limited antenna system. In the case of FIG. 15, the complex phase rotation amount prediction circuit 410-n calculates the amount of rotation of the complex phase to be performed by the phase shifters 402-n-1 to 402-n-M, and the wireless communication partner The information is input to the phase shift control circuit 406-n together with the identification number of the station device. In the phase shift control circuit 406-n, the value input here is managed by being stored in a memory or the like.

  Further, the rotation amount of the complex phase described above relates to the phase rotation amount of the phase shifters 402-n-1 to 402-n-M in the receiving system, but the rotation amount of the complex phase rotation amount in a low noise amplifier, a high power amplifier, or the like. In order to cancel the individual difference, a calibration process in the implicit feedback of the related art is performed, a rotation amount of the complex phase in the transmission system is converted, and a value set in the phase shifters 409-n-1 to 409-n-M And is similarly managed by being stored in a memory.

  Note that when performing actual data communication, that is, when performing transmission or reception processing, a control circuit 460 (not shown in FIG. 15) grasps a wireless station device to be a communication partner and performs phase shift control circuit 406- For n, the amount of rotation of the complex phase corresponding to the wireless station device performing communication is transmitted to the phase shifters 402-n-1 to 402-n-M and the phase shifters 409-n-1 to 409-n-M. The rotation amount of the complex phase is indicated, and the phase shifters 402-n-1 to 402-n-M and the phase shifters 409-n-1 to 409-n-M set the rotation amount of the complex phase to analog. The above beamforming is realized.

Although not stated in this view, for example, if placing a sender of the high-power amplifier (HPA) or the like, a description of "A" and or "D _1" - "D _M" in FIG. place the point, if placing such reception side low-noise amplifier (LNA), it will be located in a point on the description of "B" and or "E _1" - "E _M" in FIG. As for “A” and “B”, the transmission / reception signal processing circuit 652-n does not assume that they cooperate with each other. Are unnecessary, but the same transmission / reception signal processing circuit is used for the high power amplifier with the description of “D ₁ ” to “D _M ” and the low noise amplifier with the description of “E ₁ ” to “E _M ” 652-n may cause uncertainty in the complex phase between each of the antenna elements 401-n-1 to M and the antenna elements 441-n-1 to 441-n-M. It is necessary to remove the uncertainty of the complex phase of the low-noise amplifier of each system by using the calibration method of the implicit feedback.

The present invention can be applied to any method, and the specific method of the calibration process is not limited. Considering this calibration result, for example, assuming that the amount of rotation of the complex phase at “E ₁ ”, “E ₄ ”, etc. is +10 degrees, +20 degrees, etc., the equations (1) to (3) are obtained. The phase rotation amount is adjusted by correcting the rotation amount of the complex phase by, for example, -10 degrees or -20 degrees. The information on the calibration result is collected by a calibration circuit (not shown), and is collected by the phase shift control circuit 406-n, the complex phase rotation amount prediction circuit 410-n, or the correlation calculation circuit 405-n. The correction is performed using the information.

  Similarly to FIG. 14, switches 403-n-2, 403-n-3, 403-n-5 to 403-n-M of a system without a down converter and an A / D converter are shifted. In the signal processing for calculating the amount of rotation of the complex phase performed by the phasers 402-n-1 to 402-n-M, a connection state is established, or these switches 403-n-2, 403-n-3, and 403 are used. Even if -n-5 to 403-n-M is omitted, there is no practical problem.

  Next, signal processing when performing actual data communication, that is, when performing transmission or reception processing will be described. First, when performing data communication, all the switches 403-n-1 to 403-n-M are connected to the distribution coupler 415-n. Further, a control circuit 460 (not shown) grasps a wireless station device to be a communication partner, and sends a phase shift control circuit 406-n a rotation amount of a complex phase corresponding to the wireless station device performing communication. It instructs the phase shifters 402-n-1 to 402-n-M and / or the phase shifters 409-n-1 to 409-n-M.

  First, signal processing when transmitting a signal will be described. At the time of transmission, as in the case of FIG. 14, the transmission / reception signal processing circuit 652-n (n = 1,..., N) includes the time base digital baseband generated by the modulator 120-n (not shown). The D / A converter 122-n converts the transmission signal input from the modulator 120-n into an analog baseband signal, and inputs the analog / baseband signal to the up-converter 123-n. The up-converter 123-n converts the signal input from the D / A converter 122-n from a baseband signal to a signal in a radio frequency band, and inputs the signal to the distribution coupler 414-n.

  The distribution coupler 414-n branches the analog signal input from the up-converter 123-n into M-system analog signals. The branched analog signals are input to phase shifters 409-n-1 to 409-n-M. The phase shifters 409-n-1 to 409-n-M apply a predetermined complex phase rotation on the analog signal. The analog signals subjected to the complex phase rotation by the phase shifters 409-n-1 to 409-n-M are transmitted via the antenna elements 401-n-1 to 401-n-M, respectively. The transmission signal has a predetermined directivity formed by complex phase rotation in the phase shifters 409-n-1 to 409-n-M. For example, the antenna elements 401-1-1 to 401-1-M and the antenna elements 401-N-1 to 401-NM have individual directivities formed, and a radio station device in the directivity direction is provided. Perform communication.

  Next, signal reception will be described. At the time of reception, the switches 403-n-1 to 403-n-M receive signals while the distribution coupler 415-n and the phase shifters 402-n-1 to 402-n-M are connected (n = 1, ..., N). The signals received by the antenna elements 441-n-1 to 441-n-M (n = 1,..., N) are input to the phase shifters 402-n-1 to 402-n-M, respectively. Each of the phase shifters 402-n-1 to 402-n-M applies a predetermined complex phase rotation on an analog signal to the input signal, and outputs the result through switches 403-n-1 to 403-n-M. And input to the distribution coupler 415-n.

  The distribution coupler 415-n combines signals of the respective antenna systems input via the switches 403-n-1 to 403-n-M on an analog signal, and inputs the combined signal to the down-converter 124-n. I do. The down-converter 124-n converts the input radio frequency signal into an analog baseband signal, and inputs the signal to the A / D converter 125-n. The A / D converter 125-n converts the signal input from the down converter 124-n from an analog baseband signal to a digital baseband signal, and outputs the signal to a signal separation circuit (not shown). . Other processes are the same as those in FIG.

  Note that, in the wireless station device 450 illustrated in FIG. 13, the case where the antennas included in the transmission / reception signal processing circuits 651-1 to 651-N are independent is shown. Of the antenna elements 401-1 to 401-M can be shared. For example, when the transmission / reception signal processing circuits 651-1 to 651-N shown in FIG. 13 take the configuration shown in FIG. 14 (that is, when the transmission antenna and the reception antenna are shared), the transmission / reception signal processing circuit 651-n A signal line from (1 ≦ n ≦ N) to the antenna element 401-nk (1 ≦ k ≦ M) is combined by the distribution coupler 407-k, and the antenna element 401-k is connected before the signal line. A configuration may be used. This configuration is the configuration shown in FIG.

  Further, a configuration in which the transmission / reception signal processing circuits 651-1 to 651-N illustrated in FIG. 13 are replaced with the transmission / reception signal processing circuits 652-1 to 652-N illustrated in FIG. 14 (that is, the transmission antenna and the reception antenna If separated, the signal line from the transmission / reception signal processing circuit 652-n (1 ≦ n ≦ N) to the antenna element 401-nk (1 ≦ k ≦ M) is combined by the distribution coupler 407-k. Then, the antenna element 401-k is connected first, and a signal line from the transmission / reception signal processing circuit 652-n (1 ≦ n ≦ N) to the antenna element 441-nk (1 ≦ k ≦ M) is connected. The configuration may be such that the signals are combined by the distribution coupler 447-k, and the antenna element 441-k is connected to the end. This configuration is the configuration shown in FIG.

  In the description of FIG. 13, the description of the wireless station device 452 is omitted, but the configuration of the wireless station device 452 shown in FIG. It is also possible to configure. In this way, it is possible to form a plurality of directional beams with one (sub) array antenna and perform spatial multiplex transmission on the same frequency at the same time.

  In the above background art of the present invention, the process of estimating the amount of complex phase rotation performed by the variable phase shifter basically corresponding to the transmission / reception weight is performed by digital signal processing, and the actual complex phase rotation processing is performed by analog signal processing. To achieve. For this reason, the modulators 120-1 to 120-N and the demodulators 130-1 to 130-N are premised on signal processing on the frequency axis like the OFDM (Orthogonal Frequency Division Multiplexing) modulation method. In both cases, signal processing on the time axis such as single carrier transmission and SC-FDE (Single-Carrier Frequency Domain Equalization) is applicable to both systems.

  However, the signal separation circuit 141 performs signal separation between the respective signal sequences. Here, it is also possible to perform signal separation on the time axis. Frequency-dependent signal separation may be performed. In this sense, the input to the demodulators 130-1 to 130-N is assumed to be a signal on the time axis or a signal on the frequency axis. The description will be made assuming that the user inputs "." As described above, the concept regarding the variation of the communication scheme such as the OFDM modulation scheme and the SC-FDE is the same in the following description.

In addition, the transmission / reception signal processing circuits 651-1 to 651-N having a sub-array configuration can be installed physically apart from each other, so that the correlation between directional beams formed on analog can be reduced. Are set apart by a predetermined distance or more.
Further, the same applies to all the following descriptions (including the embodiments of the present invention). However, the baseband signal processing circuit 140 and the transmission / reception signal processing circuits 651-1 to 651-N are connected by wire, and Although digital baseband signals are transmitted and received on the line, the D / A converters 122-1 to 122-N and the A / D converters 125-1 to 125-N are mounted on the baseband signal processing circuit 140. In this case, a signal flowing on a wired connection between the baseband signal processing circuit 140 and the transmission / reception signal processing circuits 651-1 to 651-N may be an analog baseband signal.

Satoshi Suyama, Tatsunori Ohara, Kiyun Shin, Yukihiko Okumura, "Effect of Analog Beamforming Configuration in Massive MIMO Using High Frequency Band Hybrid Beamforming", IEICE Technical Report, IEICE, March 2015 , RCS2014-337, p.213-218 Atsushi Ota, Takuto Arai, Hiroshi Shirato, Kazuki Maruta, Tatsuhiko Iwakuni, Masataka Iizuka, "Spatial multiplex transmission technology for millimeter-wave band in the forward-looking environment using parallel transmission of the first eigenmode-scheme proposal and basic performance evaluation results," IEICE Technical Report, The Institute of Electronics, Information and Communication Engineers, August 2015, RCS2015-144, p.73-78 Atsushi Ota, Hiroshi Shirato, Kazuki Maruta, Takuto Arai, Tatsuhiko Iwakuni, Masataka Iizuka, "Efficient Use of First Eigenmode Transmission in Massive MIMO with Line-of-Sight Environment", IEICE Technical Report, IEICE, 2015 November, RCS2015-239, p.293-298 Hayato Fukuzono, Tomonori Murakami, Riichi Kudo, Yasushi Takatori, Masato Mizoguchi, "Experimental evaluation of implicit feedback in downlink multi-user MIMO-OFDM system", IEICE Technical Report, The Institute of Electronics, Information and Communication Engineers, November 2013 , RCS2013-187, p.79-84 Hiroshi Shirato, Kazuki Maruta, Ken Tanaka, Takuto Arai, Tatsuhiko Iwakuni, Satoshi Kurosaki, Atsushi Ota, "Proposal of Digital Assisted Analog Beamforming (DAABF) Technology-Extension Technology to Wireless Entrance with Channel Time Variation-", 2016 IEICE Communication Society Conference, B-5-87, The Institute of Electronics, Information and Communication Engineers, September 2016 Ken Tanaka, Kazuki Maruta, Hiroshi Shirato, Takuto Arai, Tatsuhiko Iwakuni, Satoshi Kurosaki, Atsushi Ota, "Digital Assisted Analog Beamforming (DAABF) Technology Proposal-Extension to FDD System-", IEICE 2016 Communication Society Conference, B-5-89, The Institute of Electronics, Information and Communication Engineers, September 2016

  In FIG. 7, when the difference in the amount of rotation of the complex phase due to the difference in the path length between adjacent antenna elements is π or more, it is not possible to accurately estimate the amount of rotation of the complex phase due to the uncertainty of ± 2nπ of the complex phase. There are cases. For example, consider a case where the antenna element spacing in FIG. 7 is one wavelength and the angle of arrival θ is 30 degrees. In this case, the path length difference between the closest antenna elements is λSin30 = λ / 2, and a rotation of π occurs in terms of a complex phase. The communication environment assumed here is an environment where the line of sight wave is dominant, but is slightly affected by the reflected wave, so that the measured values are slightly different values.

For example, it is assumed that the rotation amount of the complex phase of the line-of-sight wave changes in the antenna elements 401-1, 401-2, and 401-3 as 0 → π → 2π →. At this time, for the small positive errors δ ₁ and δ ₂ , the amount of rotation of the complex phase of π−δ ₁ is measured by the reflected wave at the antenna element 401-2 and 2π + δ ₂ is measured by the reflected wave at the antenna element 401-3. Suppose it was done. At this time, the amount of rotation of the complex phase between 401-2 and the antenna element 401-3 is (2π + δ ₂ ) − (π−δ ₁ ) = π + (δ ₁ + δ ₂ ), and the absolute value exceeds π. On the other hand, if the complex phase rotation amount of the antenna element 401-3 is corrected by 2π + δ ₂ −2π and 2π, the complex phase rotation amount in that case is (2π + δ ₂ −2π) − (π−δ ₁ ) = −π + (δ ₁ + δ ₂ ), and the absolute value of the rotation amount of the complex phase becomes smaller than π. In this case, since it is difficult to judge which of these is correct, assuming that the rotation amount of the complex phase is within ± π, the complex phase rotation amount of the latter antenna element 401-3 is erroneously calculated as (2π + δ ₂ − 2π). As described above, if the least squares estimation process is performed using the wrong amount of rotation of the complex phase due to the uncertainty of the complex phase of the 2π period, the obtained amount of rotation of the complex phase becomes an inappropriate value. Therefore, appropriate directivity control cannot be performed.

  On the other hand, when the size of the entire antenna increases, the beam width of the directional beam decreases as the antenna aperture increases. When forming a narrow directional beam with a large-scale antenna, use more antenna elements, or use distant antenna elements as shown in FIGS. 8 and 9 instead of adjacent or adjacent antenna elements. Is preferred. However, increasing the number of antenna elements used for estimating the direction of arrival and estimating the amount of rotation of the complex phase leads to using a relatively large number of A / D converters having relatively high costs, and from the viewpoint of cost reduction. Not preferred. Also, if a separate antenna element is used, the amount of rotation of the complex phase between the antenna elements has an uncertainty of 2π period as described above, so that the correct amount of rotation of the complex phase cannot be calculated. Therefore, while adopting a configuration in which the A / D converter is mounted on a smaller number of antenna elements, it is possible to avoid the uncertainty of the 2π period of the rotation amount of the complex phase and estimate the rotation amount of the complex phase with high accuracy. A technique for estimating the amount of complex phase rotation is required.

  In view of the above circumstances, an object of the present invention is to provide a technique capable of reducing the cost and estimating the rotation amount of the complex phase with high accuracy.

  One embodiment of the present invention is a wireless communication device that forms directivity using one or a plurality of array antennas including a plurality of antenna elements and wirelessly communicates with another wireless communication device. A signal receiving unit that receives a signal transmitted by the wireless communication device through each of the plurality of antenna elements, and a plurality of systems or some of the plurality of antenna elements different from the plurality of antenna elements. A signal conversion unit that is provided in the antenna element and converts a radio frequency analog signal received by the antenna element into a baseband digital signal, using a training signal transmitted by the other wireless communication device, A correlation calculator for calculating a correlation of a combination of two antenna elements among a plurality of antenna elements provided with a signal converter; and a correlation calculator for directly calculating the correlation. A plurality of antenna elements provided with the signal conversion unit, based on the correlation value or a correlation value obtained by combining the correlation between the antenna elements provided with the signal conversion unit obtained by the correlation calculation unit. An antenna element serving as a reference selected from among the above, a basic correlation information obtaining unit for obtaining correlation information for each combination of the other antenna elements provided with the signal conversion unit, and a basic correlation information obtaining unit for obtaining A first rotation amount prediction unit for predicting a rotation amount of a complex phase to be given to reception signals of the plurality of antenna elements based on the obtained information, and a plurality of antenna elements provided with the signal conversion unit Are grouped into a first partial antenna element and a second partial antenna element, and reception of the second partial antenna element is performed using only a correlation value related to the first partial antenna element. The first rotation amount prediction unit predicts the rotation amount of the complex phase to be given to the signal, and combines the prediction value and the information acquired by the basic correlation information acquisition unit to combine the prediction value with the information of the second partial antenna element. A basic correlation information correction unit that corrects a rotation amount of a complex phase to be given to a received signal, and based on information from the basic correlation information acquisition unit and the basic correlation information correction unit, the received signals of the plurality of antenna elements A second rotation amount prediction unit that predicts a rotation amount of a complex phase to be given to the received signal, and a rotation of the complex phase to be given to the received signal based on the complex phase predicted by the second rotation amount prediction unit. A first phase rotation amount management unit for managing the amount, and a rotation amount of the complex phase managed by the first phase rotation amount management unit, on a received signal for each antenna element used for reception, on an analog signal or Digital A first phase rotator that performs phase rotation on a signal, and an output signal from the first phase rotator, on an analog signal or a digital signal, all or a part thereof used for reception of each array antenna. A wireless communication device including: a signal combining unit that combines a signal over an antenna element; and a signal reproducing unit that reproduces a signal transmitted by the other wireless communication device based on the received signal combined by the signal combining unit. It is.

  One embodiment of the present invention is the above wireless communication device, wherein the first partial antenna element has a difference in path length when a direct wave from a wireless communication device to be a communication partner is approximated by a plane wave. The wavelength is equal to or less than 波長 wavelength with the closest element in the above.

  One embodiment of the present invention is the above wireless communication device, wherein the second partial antenna element has a path length difference when a direct wave from a wireless communication device that is a communication partner is approximated by a plane wave, and The wavelength is equal to or longer than 波長 wavelength with the closest element among the partial antenna elements.

  One embodiment of the present invention is the above wireless communication device, wherein the first phase rotation unit rotates a complex phase by a predetermined value on an analog signal with respect to a reception signal of each of the antenna elements, and The combining unit combines the received signal of the analog signal over all the antenna elements used for reception on the analog signal.

  One embodiment of the present invention is the above wireless communication device, wherein a complex phase of a signal transmitted by the other wireless communication device is changed by a predetermined value on a digital signal or an analog signal for each antenna element. A second phase rotation unit common to or independent of the first phase rotation unit to be rotated, and information acquired by the basic correlation information acquisition unit and / or the basic correlation information correction unit or the first rotation The information predicted by the amount prediction unit and / or the second rotation amount prediction unit, or based on any of information on a new complex phase rotation amount obtained by performing a predetermined calibration process on the information, A third rotation amount prediction unit that calculates a rotation amount of a complex phase to be given by the second phase rotation unit, and the second phase rotation based on the complex phase calculated by the third rotation amount prediction unit. Complex phase given by the part A second phase rotation amount management unit for managing the rotation amount, a transmission signal generation unit for generating a digital signal addressed to the other wireless communication device, and a transmission signal generated by the transmission signal generation unit on a digital signal or an analog signal. A signal distributing unit that branches each antenna element on a signal, and a rotation amount of a complex phase managed by the second phase rotation amount management unit is set in the second phase rotation unit; A signal transmitting unit that transmits, via the antenna element, a signal obtained by performing a predetermined complex phase rotation on the signal branched by the signal distribution unit in the rotating unit as a radio frequency analog signal.

  One embodiment of the present invention is the above wireless communication apparatus, wherein the signal synthesizing unit and the signal distributing unit are individually independent, and the first phase rotating unit and the second phase rotating unit Are independent.

  One embodiment of the present invention is the above wireless communication apparatus, wherein, when a frequency of a received signal is different from a frequency of a transmitted signal, the third rotation amount prediction unit includes a complex phase to be given by a second phase rotation unit. The rotation amount is corrected by multiplying the ratio of the frequency of the reception signal to the transmission signal, and the rotation amount of the corrected complex phase is calculated.

  One aspect of the present invention is a wireless communication method executed by a wireless communication device that forms directivity using one or more array antennas including a plurality of antenna elements and wirelessly communicates with another wireless communication device In, the signal transmitted by the other wireless communication device, a signal receiving step of receiving via each of the plurality of antenna elements, using a training signal transmitted by the other wireless communication device, the plurality of A part of the antenna elements or a plurality of antenna elements different from the plurality of antenna elements is provided in a plurality of systems, and converts a radio frequency analog signal received by the antenna element to a baseband digital signal. A correlation calculating step of calculating a correlation of a combination of two antenna elements among a plurality of antenna elements provided with a signal conversion unit; The signal conversion unit is provided based on the correlation value obtained directly in the step or the correlation value obtained by combining the correlation between the antenna elements provided with the signal conversion unit obtained in the correlation calculation step. A basic correlation information obtaining step of obtaining correlation information for each combination of an antenna element serving as a reference selected from a plurality of antenna elements and another antenna element provided with the signal conversion unit; and A first rotation amount prediction step of predicting a rotation amount of a complex phase to be given to the reception signals of the plurality of antenna elements based on the information acquired in the information acquisition step; and the signal conversion unit. The antenna elements of a plurality of systems are grouped into a first partial antenna element and a second partial antenna element, and a correlation related to the first partial antenna element is obtained. The first rotation amount prediction step predicts the rotation amount of the complex phase to be given to the reception signal of the second partial antenna element using only the second partial antenna element, and obtains the prediction value and the basic correlation information acquisition step. The basic correlation information correction step of correcting the amount of rotation of the complex phase to be given to the reception signal of the second partial antenna element by combining the obtained information and the basic correlation information obtaining step and the basic correlation information correction step. A second rotation amount prediction step of predicting a rotation amount of a complex phase to be given to the reception signals of the plurality of antenna elements based on the obtained information, and a complex phase predicted by the second rotation amount prediction step. A first phase rotation amount management step of managing a rotation amount of a complex phase to be given to the received signal based on the first and second phase rotation amount management steps. A first phase rotation step of performing a phase rotation on an analog signal or a digital signal with respect to a reception signal of each antenna element used for reception, and the first phase rotation step. A signal synthesis step of synthesizing a signal over all antenna elements used for reception of each array antenna on an analog signal or a digital signal, based on the reception signal synthesized by the signal synthesis step A signal reproducing step of reproducing a signal transmitted by the other wireless communication device.

  According to the present invention, it is possible to reduce the cost and to estimate the rotation amount of the complex phase with high accuracy.

It is a figure showing an example of an antenna element in a 1st embodiment. FIG. 3 is a functional block diagram illustrating a configuration example of a transmission / reception signal processing circuit in the same embodiment. FIG. 3 is a diagram illustrating a configuration example of a complex phase rotation amount prediction circuit in the same embodiment. FIG. 9 is a functional block diagram illustrating a configuration of a transmission / reception signal processing circuit according to a second embodiment. FIG. 11 is a functional block diagram illustrating a configuration example of a wireless station device according to the related art. FIG. 11 is a functional block diagram illustrating a configuration example of a wireless station device according to the related art. FIG. 9 is a diagram showing an outline of a complex phase rotation amount prediction in a linear array in which antenna elements are linearly arranged in the related art. FIG. 11 is a diagram illustrating an outline of a prediction of a rotation amount of a complex phase of an array antenna configured in a planar shape in the related art. FIG. 11 is a diagram illustrating an outline of a prediction of a rotation amount of a complex phase of a square array antenna configured in a planar shape in the related art. FIG. 11 is a diagram illustrating an outline of a prediction of a rotation amount of a complex phase of a square array antenna configured in a planar shape in the related art. FIG. 11 is a diagram illustrating an outline of a prediction of a rotation amount of a complex phase of a square array antenna configured in a planar shape in the related art. FIG. 11 is a diagram illustrating an example of an antenna pattern used for predicting a rotation amount of a complex phase according to the related art. FIG. 11 is a diagram illustrating a configuration example of a communication system in a conventional technology. FIG. 9 is a functional block diagram illustrating a configuration example of a transmission / reception signal processing circuit according to a conventional technique. FIG. 9 is a functional block diagram illustrating a configuration example of a transmission / reception signal processing circuit according to a conventional technique. FIG. 11 is a functional block diagram illustrating a configuration example of a wireless station device according to the related art. FIG. 11 is a functional block diagram illustrating a configuration example of a wireless station device according to the related art.

  An embodiment of the present invention will be described in detail with reference to the drawings. Note that the terms “time axis” and “frequency axis” used in this specification may be expressed as “time domain” and “frequency domain”. Will be described.

[First Embodiment]
Hereinafter, the basic principle of the present invention will be described with reference to the drawings. FIG. 1 shows an example of an antenna element according to the first embodiment of the present invention. FIG. 1 shows an arrangement similar to that of FIG. 9, but the largest difference is that antenna elements K to M for measuring the amount of rotation of the complex phase with antenna element o are provided near antenna element o serving as a reference. (K, L, M) is arranged, and antenna elements for measuring the amount of rotation of other complex phases are antenna elements c and m. As in FIG. 9, a to r represented by “O” in the figure represent reception antenna elements used for data reception, and s to z and A to J represented by “●” represent transmission antenna elements used for data transmission. Represent. The antenna elements K to M are used for measuring the amount of rotation of the complex phase, but are not used for transmitting and receiving data.

  As an example, it is assumed that one side of each of the square lattices constituted by a to r represented by “○”, s to z represented by “●”, and A to J is one wavelength interval. Therefore, the squares of the antenna element o and the antenna elements K to M are squares each having a half wavelength. The path length difference as shown in FIG. 7 between the four adjacent antenna elements is, for example, (λ / 2) Sin60 = (√3) / 4 × λ, which is suppressed to a value that is much smaller than a half wavelength. Therefore, among these four elements, the rotation amount of the complex phase surely falls within ± π. Therefore, at these four points, the direction of arrival is roughly estimated by the least squares method or another similar method, and the amount of rotation of the complex phase in the antenna elements c and m is estimated from the estimation result. Based on this, the uncertainty of the 2π period of the complex phase in the antenna element c and the antenna element m is eliminated.

  For example, the antenna element o is set to the origin (0, 0), the antenna element c is set to (0, 4), and the antenna element m is set to (4, 0). At this time, it is assumed that α in the expression corresponding to Expression (4) obtained with the antenna elements c and m is 0.3π, β is 0.7π, and γ is 0. In this case, φ = 0.3πx + 0.7πy, the antenna element c is estimated to be 2.8π, and the antenna element m is estimated to be 1.2π. It is assumed that the rotation amount of the complex phase of the antenna element o and the antenna element c separately obtained is 0.2π, and the rotation amount of the complex phase of the antenna element m is −0.4π. In this case, the amount of rotation of the complex phase of the antenna element c can be, for example,..., -1.8π, 0.2π, 2.2π, 4.2π,. 2.2π is closest to the first predicted complex phase rotation amount of 2.8π, and the result can be regarded as the complex phase rotation amount of the antenna element c with respect to the antenna element o. it can.

  Similarly, the amount of rotation of the complex phase of the antenna element m is, for example,..., -2.4π, -0.4π, 1.6π, 3.6π,. There is a possibility that 1.6π is closest to the first predicted amount of rotation of the complex phase of 1.2π, and this result is regarded as the amount of rotation of the complex phase of the antenna element m with respect to the antenna element o. Can be. In this manner, while eliminating the uncertainty of the 2π period of the rotation amount of the complex phase of the antenna elements c and m, the spatial extent of the antenna element used for estimating the rotation amount of the complex phase is expanded, and High-precision estimation is possible with an antenna element having a wide aperture.

  FIG. 2 is a functional block diagram illustrating a configuration example of a transmission / reception signal processing circuit for realizing the first embodiment. FIG. 2 corresponds to the transmission / reception signal processing circuit 651-n shown in FIG. 13 and mainly describes differences from the transmission / reception signal processing circuit 651-n shown in FIG.

  The transmission / reception signal processing circuit 653-n shown in the figure includes a D / A converter 122-n, an up converter (UC) 123-n, a down converter (DC) 124-n, and an A / D converter 125-n. n, TDD switch (TDD-SW) 127-n, phase shifters 402-n-2 to 402-n-M, switches 403-n-2 to 403-n-M, and distribution coupler 404- n, down converters (DC) 424-n-1 to 424-n-2, A / D converters 425-n-1 to 425-n-2, a correlation calculation circuit 405-n, and phase shift control The circuit includes a circuit 406-n and a complex phase rotation amount prediction circuit 417-n. The transmission / reception signal processing circuit 653-n is connected to the antenna elements 401-1 to 401-M and the baseband signal processing circuit 140. Although FIG. 2 shows two configurations of the down converter 424-n and the A / D converter 425-n, respectively, the down converter 424-n-1 and the A / D converter 425-n-1 There is no restriction on the number, while the down converter 424-n-2 and the A / D converter 425-n-2 may have any value as long as the value is smaller than M. In the following description, there are two down converters 424-n and two A / D converters 425-n (down converters 424-n-1 to 424-n-2 and A / D converters 425-n-1 to 425-n). The case of n-2) will be described as an example. In the figure, the same functional blocks as those in the above-mentioned figures are assigned the same numbers, and the detailed description thereof will be omitted as necessary.

  The difference from FIG. 14 is that a part of the entire antenna element (corresponding to the antenna element 401-1 in FIG. 2 of the present embodiment) is a phase shifter 402-n-1 and a switch 403-n- 1, the signal line is connected to the down converter 124-n-1 and the signal line from the antenna element 401-1 to the distribution coupler 404-n is omitted. Further, the complex phase rotation amount prediction circuit 410-n is replaced with a complex phase rotation amount prediction circuit 417-n.

  In correspondence with FIG. 1 (strictly, FIG. 1 is a diagram assuming that the transmitting and receiving antennas are different, but the following description will be made assuming that the transmitting and receiving antennas are shared). The antenna element 401-1 without the phase shifter 402-n and the switch 403-n corresponds to, for example, the antenna elements K, L, and M in FIG. 1, and the antenna with the phase shifter 402-n and the switch 403-n. The element 401-4 corresponds to, for example, the antenna elements o, c, and m in FIG. Except that the antenna elements K, L, and M are not connected to the distribution coupler 404-n, the signal processing related to transmission and reception of data via the antenna elements 401-2 to 402-M corresponds to the signal processing of the present invention in FIG. This is equivalent to the signal processing of the background art. The difference of the signal processing exists only in the signal processing when calculating the amount of rotation of the complex phase performed by the phase shifters 402-n-2 to 402-n-M. Hereinafter, signal processing performed by the phase shifters 402-n-2 to 402-n-M when calculating the amount of rotation of the complex phase will be described.

  In the signal processing for calculating the rotation amount of the complex phase, the switch 403-n-4 connects the down converter 424-n-2 and the phase shifter 402-n-4 (actually, FIG. The same applies to an antenna system switch including a down converter and an A / D converter not shown), and the remaining switches are not connected. In this state, first, a training signal for channel estimation is transmitted from the wireless station device of the communication partner from which the rotation amount of the complex phase is to be obtained, and the wireless station device receives this signal.

  The signal received by the antenna element 401-n-1 is input to the down converter 424-n-1. The signal received by the antenna element 401-n-4 is input to the phase shifter 402-n-4, and the phase shifter 402-n-4 rotates a predetermined complex phase on the analog signal (for example, 0 degree). May be added) and input to the switch 403-n-4. Switch 403-n-4 (actually, the same applies to an antenna system switch including a down converter and an A / D converter not shown here, but for simplification of description, this one antenna The signal input to only the system is input to the downconverter 424-n-4.

  The down converters 424-n-1 to 424-n-2 convert the input radio frequency signals into analog baseband signals. The A / D converters 425-n-1 to 425-n-2 convert analog signals into digital baseband signals. The information converted into digital baseband signals by the A / D converters 425-n-1 to 425-n-2 is input to the correlation calculation circuit 405-n. The correlation calculation circuit 405-n calculates the rotation amount of the complex phase based on the input information by using Expressions (1) to (3). When the calibration process is required as needed, the rotation amount of the complex phase on the transmission side is determined as a value in consideration of the calibration coefficient in Expressions (1) to (3). The rotation amount of the complex phase obtained by the correlation calculation circuit 405-n is input to the complex phase rotation amount prediction circuit 417-n together with the identification number of the wireless station device to communicate with. At this time, it is assumed that the information on the antenna element set so that the rotation amount of the complex phase for each antenna element is ± π or less and the information on the other antenna elements are inputted in a distinguished manner. In the example of FIG. 1, for example, the antenna elements set such that the rotation amount of the complex phase for each antenna element is ± π or less are the antenna elements o, K, L, and M, and the other antenna elements are the antenna elements. It corresponds to c and m.

  First, the complex phase rotation amount prediction circuit 417-n performs temporary estimation of the rotation amount of the complex phase of all the antenna elements using the antenna elements o, K, L, and M. The coordinates of the antenna elements c and m are set in the provisional estimation result, and the primary prediction of the rotation amount of the complex phase of the antenna elements c and m is performed. Here, for example, as described above, when the antenna element o is the origin (0, 0), the antenna element c is (0, 4), and the antenna element m is (4, 0), the antenna elements c and m If α is 0.3π, β is 0.7π, and γ is 0 in the equation corresponding to equation (4) obtained in the above, φ = 0.3πx + 0.7πy, and the antenna element c is 2.8π. , Antenna element m is provisionally estimated to be 1.2π. The above is the provisional estimation based on the primary prediction.

  Next, the procedure of the secondary estimation will be described. It is assumed that the rotation amount of the complex phase of the antenna element o and the antenna element c separately obtained is 0.2π, and the rotation amount of the complex phase of the antenna element m is −0.4π. In this case, the amount of rotation of the complex phase of the antenna element c can be, for example,..., -1.8π, 0.2π, 2.2π, 4.2π,. 2.2π is closest to the first predicted complex phase rotation amount of 2.8π, and the result can be regarded as the complex phase rotation amount of the antenna element c with respect to the antenna element o. it can. Similarly, the amount of rotation of the complex phase of the antenna element m is, for example,..., -2.4π, -0.4π, 1.6π, 3.6π,. There is a possibility that 1.6π is closest to the first predicted amount of rotation of the complex phase of 1.2π, and this result is regarded as the amount of rotation of the complex phase of the antenna element m with respect to the antenna element o. Can be.

  Based on the corrected complex phase rotation amounts of the antenna elements o, m and the uncertainty of the 2π period that have been determined in this way, and the complex phase rotation amounts of the antenna elements o, K, L, and M, By applying the least-squares method, a final estimation formula corresponding to the formula (4) is calculated. The result is applied to the coordinates of each antenna element, and the amount of rotation of the complex phase of each antenna element is estimated. In this manner, while eliminating the uncertainty of the 2π period of the rotation amount of the complex phase of the antenna elements c and m, the spatial extent of the antenna element used for estimating the rotation amount of the complex phase is expanded, and High-precision estimation is possible with an antenna element having a wide aperture. In this way, the complex phase rotation amount prediction circuit 417-n calculates the rotation amount of the complex phase to be performed by the phase shifters 402-n-2 to 402-n-M, and calculates the rotation amount of the wireless station device to be communicated with. It is input to the phase shift control circuit 406-n together with the identification number. In the phase shift control circuit 406-n, the value input here is managed by being stored in a memory or the like. Note that the control circuit 460 performs overall adjustment, timing control, and the like in a series of these processes.

  FIG. 3 is a diagram illustrating a configuration example of the complex phase rotation amount prediction circuit 417-n according to the first embodiment. In FIG. 3, the complex phase rotation amount prediction circuit 417-n includes a first-order estimation circuit 418-n and a second-order estimation circuit 419-n. Taking the example of FIG. 2 as an example, information on the amount of rotation of the complex phases of the antenna elements o, K, L, and M is input to the first-order estimation circuit 418-n from the correlation calculation circuit. The first-order estimating circuit 418-n first tentatively estimates the amounts of rotation of the complex phases of all the antenna elements using information on the antenna elements o, K, L, and M, and calculates the antenna in accordance with the equation corresponding to equation (4). The coordinates of the elements c and m are set, and the primary prediction of the rotation amount of the complex phase of the antenna elements c and m is performed. The result is input to the second estimation circuit 419-n. The second-order estimation circuit 419-n includes, in addition to the first-order prediction results of the rotation amounts of the complex phases of the antenna elements c and m from the first-order estimation circuit 418-n, the antenna elements o, K, L, and M Is input, and the second-order estimation circuit 419-n uses the input value to determine the uncertainty of the 2π period of the complex phase rotation of the antenna elements c and m as described above. Equivalent to the equation (4) using the least squares method based on the corrected complex phase rotation amounts of the antenna elements c and m and the complex phase rotation amounts of the antenna elements o, K, L and M. Information about the rotation amount of the complex phase of all the antenna elements to be calculated.

  The coordinates of each antenna element are substituted into the equation corresponding to equation (4) obtained in this way, the amount of rotation of the complex phase to be set in the phase shifter of each antenna element is calculated, and this is used for phase shift control. Input to the circuit. As described above, except for the temporary estimation processing performed by the primary estimation circuit 418-n, processing equivalent to the complex phase rotation amount prediction circuit 410-n in FIG. 14 is basically performed.

  As described above, according to the present embodiment, the wireless station device is provided with the antenna element used for the complex phase rotation amount measurement, and performs the complex operation based on the information obtained from the antenna element used for the complex phase rotation amount measurement. Estimate the amount of phase rotation and eliminate the uncertainty of the 2π period regarding the amount of rotation of the complex phase. As a result, the rotation amount of the complex phase can be estimated with high accuracy. Therefore, it is possible to set the directivity with high accuracy while suppressing the number of A / D converters and D / A converters required to acquire the channel information or the complex phase rotation amount information. In addition, the wireless station device can reduce power consumption and the cost of configuring the device.

  In FIG. 2, the transmission / reception signal processing circuit 653-n corresponding to the transmission / reception signal processing circuit 651-n in FIG. 14 in the background art of the present invention has been described, but the transmission / reception signal processing circuit 653-n may be realized by the configuration corresponding to FIG. It is possible. That is, a configuration in which the transmitting antenna and the receiving antenna are separated can be similarly used.

  In FIG. 2, the signal received by the antenna element 401-1 (in FIG. 1, the antenna elements K, L, and M) has been described as not being used for received signal processing in a normal service. May be subjected to the received signal processing. In this case, the circuit configuration connected to the antenna element 401-1 is the same as that in FIG. 14, and is connected to the distribution coupler 404-n via the phase shifter 402-n-1 and the switch 403-n-1. You should do it. The difference from FIG. 4 in this case is that there are antenna elements o, K, L, and M in which the amount of rotation of the complex phase is within ± π as shown in FIG. The point is that the uncertainty of the 2π period of the complex phase rotation amount of the antenna elements c and m can be eliminated by the primary estimation circuit.

  Further, in FIG. 2, it has been described that the signals received by the antenna elements 401-4 (in FIG. 1, the antenna elements o, c, and m) are used for reception signal processing in a normal service. May not be used for the received signal processing. In this case, the phase shifter 402-n-1 and the switch 403-n-1 are omitted, so that the circuit configuration connected to the antenna element 401-4 is the same as the circuit configuration connected to the antenna element 401-1. What is necessary is just to make it the structure which is not connected to 404-n. In this case, the antenna elements used for estimating the amount of rotation of the complex phase are different from the antenna elements used for received signal processing in other normal services, and therefore, as shown in FIG. The antenna may not be embedded in the antenna element, but may be arranged outside the antenna elements a, u, H, r, and the like. In this way, by combining parts having different configurations from the other antenna elements on the outside, it may be preferable from the viewpoint of mounting the device.

[Second embodiment]
FIG. 4 is a functional block diagram showing a configuration of a transmission / reception signal processing circuit for realizing the second embodiment. The difference from FIG. 15 is that a part of the antenna elements (corresponding to the antenna element 441-1 in FIG. 2 of the present embodiment) is a phase shifter as in the case of FIG. It is connected to the down converter 124-n-1 without passing through the switch 442-n-1 and the switch 443-n-1, and a signal line from the antenna element 441-1 to the distribution coupler 444-n is omitted. Further, the complex phase rotation amount prediction circuit 410-n is replaced with a complex phase rotation amount prediction circuit 417-n.

  1, the antenna element 441-1 without the phase shifter 442-n and the switch 443-n corresponds to, for example, the antenna elements K, L, and M in FIG. The antenna element 441-4 including the switch 443-n and the switch 443-n correspond to, for example, the antenna elements o, c, and m in FIG. Except that the antenna elements K, L, and M are not connected to the distribution coupler 404-n, the signal processing related to the reception of data via the antenna elements 441-2 to 442-M is the same as that of FIG. This is equivalent to the signal processing of the background art. Similarly, signal processing related to data transmission via the antenna elements 401-1 to 401-M is equivalent to the signal processing of the background art of the present invention in FIG. The difference in the signal processing is only the signal processing when calculating the amount of rotation of the complex phase performed by the phase shifters 402-n-1 to 402-n-M and the phase shifters 442-n-2 to 442-n-M. Exists. The signal processing for calculating the amount of rotation of the complex phase performed by the phase shifters 402-n-1 to 402-n-M and the phase shifters 442-n-2 to 442-n-M is the first signal processing. This is equivalent to the signal processing of FIG. 3 in the embodiment. Therefore, the complex phase rotation amount prediction circuit 417-n is also realized by the configuration of FIG.

[Third Embodiment]
As described in Non-Patent Document 6, the background art of the present invention describes a method of acquiring a rotation amount of a complex phase set in a phase shifter in the case of FDD in which frequencies used between transmission and reception are different. Here, for example, the values of α, β, and γ relating to the amount of rotation of the complex phase given by Expression (4), and the frequency f _{BW in the} reception direction (backward direction) and the frequency f _{FW in the} transmission direction (forward direction) The conversion formula of the rotation amount of the complex phase in the transmission direction is given by the following Expression 6 with respect to Expression (4) obtained in the reception direction by the following expression.

  Therefore, if a function for performing this conversion is implemented in the complex phase rotation amount prediction circuit 417-n (for example, the secondary estimation circuit 419-n), the present invention can be used in FDD.

  As described above, according to the present embodiment, the wireless station device acquires the complex phase rotation information of some of the antenna elements without acquiring the complex phase rotation information of all the antenna elements. The wireless station device calculates weight information from the acquired complex phase rotation amount information, and calculates weight information of other antenna elements by approximation based on the calculated weight information. At this time, some of the antenna elements are set so as to eliminate the uncertainty of the 2π period of the complex phase, and the complex phase of the entire antenna is temporarily determined using this antenna element. Is temporarily estimated, and the uncertainty of the 2π-period complex phase is eliminated from the rotation amount of the complex phase of the remaining part of the antenna elements using the provisionally estimated value. By removing the uncertainty of the complex phase, it is possible to use a relatively distant antenna element that cannot be set in the background art of the present invention for estimating the rotation amount of the complex phase. Accordingly, the antenna element used for estimating the amount of rotation of the complex phase while suppressing the number of A / D converters and D / A converters required to acquire the channel information or the rotation amount information of the complex phase Can be widened spatially (antenna aperture), and the resulting directivity can be set with high accuracy. Also, as described above, by calculating the weight information of other antenna elements by approximation, even if the transmitting antenna and the receiving antenna are different, implicit feedback can be used. Therefore, it is possible to reduce the size and cost and to use the implicit feedback even when the transmitting antenna and the receiving antenna are different. Further, even when the frequency differs between transmission and reception, directivity control can be performed with high accuracy by using extended implicit feedback.

  In each of the embodiments described above, the complex phase rotation is performed on the radio frequency analog signal. However, the position of the up converter is changed, and the complex phase rotation is performed on the baseband or intermediate frequency analog signal. It is also possible to adopt a configuration in which a frequency conversion with a radio frequency is performed in a subsequent stage or a preceding stage.

[Supplementary information on the embodiment]
The supplementary items relating to the embodiment of the present invention described above are shown below.
The wireless station device in the above-described embodiment may be realized by a computer. In that case, a program for realizing this function may be recorded on a computer-readable recording medium, and the program recorded on this recording medium may be read and executed by a computer system. Here, the “computer system” includes an OS and hardware such as peripheral devices. The “computer-readable recording medium” refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM, and a CD-ROM, and a storage device such as a hard disk built in a computer system. Further, a “computer-readable recording medium” refers to a communication line for transmitting a program via a network such as the Internet or a communication line such as a telephone line. Such a program may include a program that holds a program for a certain period of time, such as a volatile memory in a computer system serving as a server or a client in that case. The program may be for realizing a part of the functions described above, or may be a program that can realize the functions described above in combination with a program already recorded in a computer system, It may be realized using hardware such as a PLD (Programmable Logic Device) or an FPGA (Field Programmable Gate Array).

  As described above, the embodiments of the present invention have been described with reference to the drawings. However, it is apparent that the above embodiments are merely examples of the present invention, and the present invention is not limited to the above embodiments. is there. Therefore, additions, omissions, replacements, and other changes may be made to the components without departing from the technical spirit and scope of the present invention.

  The present invention is applicable to a wireless communication device that transmits and receives a wireless signal using a plurality of antennas.

120-1 to 120-N ... Modulators 122-1 to 122-N, 122-n ... D / A converters 123-1 to 123-N, 123-n ... Up-converters 124-1 to 124-N, 124 -N: Down converters 125-1 to 125-N, 125-n: A / D converters 127-1 to 127-N: TDD switches 130-1 to 130-N: Demodulator 140: Baseband signal processing circuit 141 ... signal separation circuits 401-1 to 401-M ... antenna elements 401-1-1 to 401-NM, 401-n-1 to 401-nM ... antenna elements 402-1-1 to 402-N- M, 402-n-1 to 402-nM ... phase shifters 403-1-1 to 403-NM, 403-n-1 to 403-nM ... switches 404-1 to 404-N ... Distribution couplers 405-1 to 405-N, 40 -N correlation calculation circuits 406-1 to 406-N, 406-n phase shift control circuits 407-1 to 407-M distribution couplers 408-n-1 to 408-n-M TDD switches 409-n -1 to 409-nM phase shifter 410-n complex phase rotation amount prediction circuit 411-n transmission weight calculation circuit 414-n distribution coupler 415-n distribution coupler 417-n complex phase Rotation amount prediction circuit 418-n first-order estimation circuit 419-n second-order estimation circuits 424-1-1 to 424-NM, 424-n-1 to 424-n-M down converter 425- 1-1 to 425-NM, 425-n-1 to 425-nM ... A / D converters 441-1 to 441-M ... antenna elements 441-n-1 to 441-n-M ... antennas Element 450 ... Wireless station devices 451-1 to 451-N ... Reception signal processing circuit 452 wireless station apparatus 453-n transmission / reception signal processing circuit 454-n transmission / reception signal processing circuit 455-n transmission / reception signal processing circuit 460 control circuit 653-n transmission / reception signal processing circuit 654-n transmission / reception The signal processing circuit 901-1~901-N ... modulator 902 ... precoder _903-1~903-N 0 ... IFFT & GI imparting circuit _904-1~904-N 0 ... D / A converter _{905-1~905-N 0} ... upconverter 906-1~906-N 0 _... downconverter _907-1~907-N 0 ... A / D converter _908-1~908-N 0 ... GI removing & FFT circuit 909 ... postcoder 910-1～910 -N ... demodulator 911 ... TDD switch 912-1~912-N 0 _... distributor coupler _{913-1-1~913-N} 0 -M 0 _... phase shifter 9 15-1~915-M 0 _... distributor coupler 916-1~916-M 0 _... antenna elements 921-1~921-N ... time base transmission weight multiplying circuit 922-1~922-M ... D / A converter 922-1-1 to 922-NM ... D / A converters 923-1 to 923-M ... up-converters 923-1-1 to 923-NM ... up-converters 924-1 to 924-M ... down Converters 924-1-1 to 924-NM Downconverters 925-1 to 925-M A / D converters 925-1-1 to 925-NM ... A / D converters 926-1 to 926 -N: time axis reception weight multiplying circuits 927-1 to 927-N: TDD switches 929-1 to 928-M: antenna elements 928-1-1 to 928-NM: antenna elements 929-1 to 929-N ... Transceive signal processing Circuits 941-1 to 941-M Distribution coupler 942 Radio station devices 943-1 to 943-M Adder combiners 944-1 to 944-M Duplicator 945 Radio station devices 951-1 to 951-3 ... High power amplifiers 952-1 to 952-3 ... Low noise amplifiers 953-1 to 953-3 ... TDD switches 954-1 to 954-3 ... Antenna elements 955-1 to 955-3 ... Wireless modules

Claims (8)

A wireless communication device that forms directivity using one or a plurality of array antennas including a plurality of antenna elements and wirelessly communicates with another wireless communication device,
A signal receiving unit that receives the signal transmitted by the other wireless communication device via each of the plurality of antenna elements,
A part of the plurality of antenna elements or a plurality of antenna elements provided separately from the plurality of antenna elements, and a radio frequency analog signal received by the antenna element is converted to a baseband digital signal. A signal conversion unit for converting the
Using a training signal transmitted by the other wireless communication device, a correlation calculating unit that calculates a correlation of a combination of two antenna elements among a plurality of antenna elements provided with the signal conversion unit,
Based on the correlation value obtained directly by the correlation calculation unit or the correlation value obtained by combining the correlation between the antenna elements provided with the signal conversion unit obtained by the correlation calculation unit, the signal conversion unit An antenna element serving as a reference selected from a plurality of antenna elements provided, and a basic correlation information acquisition unit for acquiring correlation information for each combination of the other antenna elements provided with the signal conversion unit,
A first rotation amount prediction unit that predicts a rotation amount of a complex phase to be given to reception signals of the plurality of antenna elements based on information acquired by the basic correlation information acquisition unit;
The plurality of antenna elements provided with the signal conversion unit are grouped into a first partial antenna element and a second partial antenna element, and the second partial antenna element is grouped using only a correlation value related to the first partial antenna element. The first rotation amount prediction unit predicts the rotation amount of the complex phase to be given to the reception signal of the partial antenna element, and combines the predicted value and the information acquired by the basic correlation information acquisition unit to generate the second rotation amount. A basic correlation information correction unit that corrects the amount of rotation of the complex phase to be given to the received signal of the partial antenna element,
A second rotation amount prediction unit that predicts a rotation amount of a complex phase to be given to reception signals of the plurality of antenna elements based on information from the basic correlation information acquisition unit and the basic correlation information correction unit;
A first phase rotation amount management unit that manages a rotation amount of a complex phase to be given to a received signal based on the complex phase predicted by the second rotation amount prediction unit;
A first phase rotation unit that rotates the amount of rotation of the complex phase managed by the first phase rotation amount management unit on an analog signal or a digital signal with respect to a reception signal for each antenna element used for reception; ,
An output signal from the first phase rotation unit, a signal synthesis unit that synthesizes a signal over all or some of the antenna elements used for reception of each array antenna on an analog signal or a digital signal,
A signal reproducing unit that reproduces a signal transmitted by the another wireless communication device based on the received signal synthesized by the signal synthesizing unit,
A wireless communication device comprising:   The first partial antenna element has a path length difference when a direct wave from a wireless communication device serving as a communication partner is approximated to a plane wave by a half wavelength or less between the antenna element and the nearest element among the antenna elements. The wireless communication device according to claim 1.   The second partial antenna element has a difference in path length when a direct wave from a wireless communication device that is a communication partner is approximated to a plane wave by a factor of 1/100 between a nearest one of the first partial antenna elements. The wireless communication device according to claim 1, wherein the wireless communication device has two or more wavelengths. The first phase rotation unit rotates a complex phase by a predetermined value on an analog signal with respect to a reception signal of each of the antenna elements,
The wireless communication device according to any one of claims 1 to 3, wherein the signal combining unit combines the received signal of the analog signal over all or some of the antenna elements used for reception on the analog signal. . For the signal transmitted by the other wireless communication device, the antenna is shared with the first phase rotation unit that rotates a complex phase by a predetermined value on a digital signal or an analog signal for each antenna element, or An independent second phase rotator,
The information acquired by the basic correlation information acquisition unit and / or the basic correlation information correction unit or the information predicted by the first rotation amount prediction unit and / or the second rotation amount prediction unit, or a predetermined calibration is performed on the information. A third rotation amount prediction unit that calculates a rotation amount of a complex phase to be given by the second phase rotation unit based on any of the information on the new complex phase rotation amount obtained by performing the processing;
A second phase rotation amount management unit that manages a rotation amount of a complex phase provided by the second phase rotation unit based on the complex phase calculated by the third rotation amount prediction unit;
A transmission signal generation unit that generates a digital signal addressed to the other wireless communication device,
A signal distribution unit that branches the transmission signal generated by the transmission signal generation unit for each antenna element on a digital signal or an analog signal,
The rotation amount of the complex phase managed by the second phase rotation amount management unit is set in the second phase rotation unit, and the second phase rotation unit applies a predetermined amount to the signal branched by the signal distribution unit. A signal transmitting unit that transmits the signal after performing the complex phase rotation through the antenna element as a radio frequency analog signal,
The wireless communication device according to claim 1, further comprising:   The wireless communication device according to claim 5, wherein the signal synthesizing unit and the signal distribution unit are individually independent, and the first phase rotation unit and the second phase rotation unit are individually independent.   When the frequencies of the reception signal and the transmission signal are different, the third rotation amount prediction unit corrects the multiplication of the rotation amount of the complex phase to be given by the second phase rotation unit by the ratio of the frequency of the reception signal to the transmission signal. The wireless communication apparatus according to claim 5, wherein the correction is performed to calculate a rotation amount of the corrected complex phase. A wireless communication method executed by a wireless communication device that wirelessly communicates with another wireless communication device by forming directivity using one or more array antennas including a plurality of antenna elements,
A signal receiving step of receiving the signal transmitted by the other wireless communication device via each of the plurality of antenna elements,
Using a training signal transmitted by the other wireless communication device, a part of the plurality of antenna elements or provided in a plurality of systems of a plurality of antenna elements different from the plurality of antenna elements, the antenna element A correlation calculating step of calculating a correlation of a combination of two antenna elements among a plurality of antenna elements provided with a signal conversion unit that converts a received radio frequency analog signal into a baseband digital signal,
Based on the correlation value obtained directly in the correlation calculation step or the correlation value obtained by combining the correlation between the antenna elements provided with the signal conversion unit obtained in the correlation calculation step, the signal conversion unit An antenna element serving as a reference selected from a plurality of antenna elements provided, and a basic correlation information obtaining step of obtaining correlation information for each combination of the other antenna elements provided with the signal conversion unit,
A first rotation amount prediction step of predicting a rotation amount of a complex phase to be given to the reception signals of the plurality of antenna elements based on the information acquired in the basic correlation information acquisition step;
The plurality of antenna elements provided with the signal conversion unit are grouped into a first partial antenna element and a second partial antenna element, and the second partial antenna element is grouped using only a correlation value related to the first partial antenna element. A rotation amount of a complex phase to be given to a received signal of the partial antenna element is predicted in the first rotation amount prediction step, and a prediction value and information obtained in the basic correlation information obtaining step are combined to obtain the second rotation amount. A basic correlation information correction step of correcting the amount of rotation of the complex phase to be given to the received signal of the partial antenna element of
A second rotation amount prediction step of predicting a rotation amount of a complex phase to be given to the reception signals of the plurality of antenna elements based on the information obtained from the basic correlation information acquisition step and the basic correlation information correction step; ,
A first phase rotation amount management step of managing a rotation amount of a complex phase to be given to a received signal based on the complex phase predicted in the second rotation amount prediction step;
A first phase rotation step of performing a phase rotation of a complex phase rotation amount managed in the first phase rotation amount management step on an analog signal or a digital signal with respect to a reception signal for each antenna element used for reception; ,
A signal synthesizing step of synthesizing the output signal from the first phase rotation step, on an analog signal or a digital signal, over all antenna elements used for reception of each array antenna;
A signal reproducing step of reproducing a signal transmitted by the other wireless communication device based on the received signal synthesized by the signal synthesizing step,
A wireless communication method comprising:

Images