JP2019033387A - Base station, calibration method and calibration apparatus - Google Patents

Abstract

The accuracy of calibration between antennas is improved. A base station includes a plurality of antennas, a plurality of analog circuits provided for each antenna, a signal generation unit, a frequency shift unit, and a coefficient calculation unit. The signal generator 222 generates a first signal including a plurality of subcarriers having a frequency interval wider than that of the plurality of subcarriers included in the data signal. The frequency shift unit 221 generates a second signal obtained by shifting the frequency of the first signal by a predetermined frequency. The coefficient calculation unit 223 is input to the analog circuit 30 corresponding to the antenna 41 for each antenna 41, and is output from each antenna 41 using each of the first signal and the second signal that have passed through the analog circuit 30. Adjust the frequency response of the generated signal. [Selection] Figure 1

Description

  The present invention relates to a base station, a calibration method, and a calibration apparatus.

  Beam forming is a technique for providing directivity in a specific direction by controlling the amplitude and phase of radio waves radiated from a plurality of antenna elements, that is, radiating radio waves concentratedly in a specific direction. For example, if the amplitude and phase of the radio wave radiated from each antenna element to the space are aligned, a beam is formed in the straight direction with respect to the antenna surface, and if the amplitude and phase of each antenna element are shifted, this amplitude and phase shift A beam is formed in a direction according to the above. In beam forming, a beam can be formed with high accuracy in a desired direction by aligning amplitude errors and phase errors between a plurality of antennas. Therefore, antenna calibration for correcting amplitude errors and phase errors between a plurality of antennas is performed.

  Many methods have been studied for antenna calibration since ancient times. For example, in a TDD (Time Division Duplex) system, it is considered to perform calibration of each antenna element using a signal transmitted in a guard time provided for switching between uplink and downlink. Since the guard time is a period in which neither the base station nor the terminal transmits / receives a signal, it is possible to reduce the influence exerted on other devices by executing calibration. Further, if the guard time is too long, the use efficiency of the system resource is lowered, so the guard time is set shorter than the period of the communication frame used for transmission / reception of the data signal.

JP 2010-41269 A

  By the way, in a system that transmits and receives an OFDM (Orthogonal Frequency Division Multiplexing) signal, when transmitting data of the same amount of data, the subcarrier interval becomes wider as the transmission period becomes shorter. Therefore, when the signal used for calibration is an OFDM signal, the subcarrier interval used for a signal transmitted in a short period such as a guard time is wider than the subcarrier interval used for a normal data signal.

  However, beam forming performed using each antenna element after calibration is performed on the data signal. The subcarrier interval used for the data signal is narrower than the subcarrier interval used for the calibration signal. Therefore, the amplitude error and phase error in each subcarrier used in the data signal are estimated by interpolating the amplitude error and phase error in each subcarrier used in the calibration signal. However, since the interpolated amplitude error and phase error are not necessarily close to the amplitude error and phase error in each subcarrier used for the data signal, the calibration accuracy is lowered. There is.

  An object of the technology disclosed in the present application is to provide a base station, a calibration method, and a calibration device that can improve the accuracy of calibration between antennas.

  In one aspect, the base station includes a plurality of antennas, a plurality of analog circuits provided for each antenna, a first generation unit, a second generation unit, and an adjustment unit. The first generation unit generates a first signal including a plurality of subcarriers having a frequency interval wider than that of the plurality of subcarriers included in the data signal. The second generation unit generates a second signal obtained by shifting the frequency of the first signal by a predetermined frequency. For each antenna, the adjustment unit is input to the analog circuit corresponding to the antenna, and the frequency of the signal output from each antenna using each of the first signal and the second signal that passes through the analog circuit. Adjust the response.

  According to one embodiment, the accuracy of calibration between antennas can be improved.

FIG. 1 is a block diagram illustrating an example of a base station according to the first embodiment. FIG. 2 is a diagram for explaining an example of frequency intervals of subcarriers. FIG. 3 is a block diagram illustrating an example of the coefficient calculation unit. FIG. 4 is a diagram illustrating an example of the frequency response of the analog circuit. FIG. 5 is a diagram illustrating an example of the frequency response of the analog circuit. FIG. 6 is a diagram for explaining an example of a synthesis process by the synthesis unit. FIG. 7 is a flowchart illustrating an example of the calibration process. FIG. 8 is a diagram for explaining an example of the frequency response of a FIR (Finite Impulse Response) filter. FIG. 9 is a diagram illustrating an example of the smoothing process. FIG. 10 is a diagram illustrating an example of the smoothing process in the comparative example. FIG. 11 is a diagram illustrating an example of the smoothing process according to the second embodiment. FIG. 12 is a block diagram illustrating an example of a base station according to the third embodiment. FIG. 13 is a diagram illustrating an example of hardware of a base station.

  Hereinafter, embodiments of a base station, a calibration method, and a calibration apparatus disclosed in the present application will be described in detail with reference to the drawings. The following examples do not limit the disclosed technology.

[Configuration of base station 10]
FIG. 1 is a block diagram illustrating an example of the base station 10 according to the first embodiment. The base station 10 includes a digital processing unit 20, a plurality of analog circuits 30-1 to 30-n, a plurality of antennas 41-1 to 41-n, and a plurality of couplers 42-1 to 42-n. The base station 10 includes a selector 43, an amplifier 44, a down converter 45, and an ADC (Analog to Digital Converter) 46. The plurality of antennas 41-1 to 41-n constitute an array antenna. The plurality of antennas 41-1 to 41-n are adjusted by adjusting the amplitude and phase of the transmission signal output to each of the plurality of antennas 41-1 to 41-n by the FIR filter in the digital processing unit 20. A beam having directivity in a predetermined direction is formed.

  Hereinafter, when a plurality of analog circuits 30-1 to 30-n are collectively referred to without being distinguished, they are simply referred to as an analog circuit 30, and each of the plurality of antennas 41-1 to 41-n is distinguished. The generic term “antenna 41” will be used. Hereinafter, the couplers 42-1 to 42-n are simply referred to as couplers 42 when collectively referred to without being distinguished from each other.

  In the base station 10, one analog circuit 30 and one coupler 42 are provided in association with one antenna 41. Each analog circuit 30 performs processing such as up-conversion and amplification on the signal output from the digital processing unit 20 and outputs the processed signal to the corresponding antenna 41. Each antenna 41 radiates the signal output from the corresponding analog circuit 30 to space as a radio wave. Each analog circuit 30 performs processing such as down-conversion and amplification on the received signal received via the corresponding antenna, and outputs the processed received signal to the digital processing unit 20.

  Each coupler 42 feeds back a part of the signal output from the corresponding analog circuit 30 to the antenna 41 to the selector 43. The selector 43 selects one of the signals fed back from the respective couplers 42 and outputs the selected signal to the amplifier 44. Hereinafter, a signal fed back from each coupler 42 is referred to as a feedback signal. The amplifier 44 amplifies the feedback signal output from the selector 43. The down converter 45 performs processing such as quadrature detection and down conversion on the feedback signal amplified by the amplifier 44. The ADC 46 converts the feedback signal subjected to quadrature detection or the like by the down converter 45 from an analog signal to a digital signal and outputs the digital signal to the digital processing unit 20.

  Each analog circuit 30 includes a DAC (Digital to Analog Converter) 31, an up-converter 32, a PA (Power Amplifier) 33, a circulator 34, and a BPF (Band Pass Filter) 35. In addition, each analog circuit 30 includes a reception block that processes a signal received via the antenna 41. Hereinafter, the DAC 31, the up converter 32, the PA 33, the circulator 34, and the BPF 35 are referred to as a transmission block.

  The DAC 31 converts the signal output from the digital processing unit 20 from a digital signal to an analog signal. A signal output from the digital processing unit 20 is a transmission signal or a calibration signal described later. The up-converter 32 performs processing such as orthogonal modulation and up-conversion on the signal converted into the analog signal by the DAC 31. The PA 33 amplifies the signal that has been subjected to quadrature modulation or the like by the up converter 32. The circulator 34 passes the signal amplified by the PA 33 to the BPF 35. In addition, the circulator 34 passes the reception signal received by the antenna 41 and passed through the BPF 35 to the reception block. The BPF 35 limits the frequency band of the signal amplified by the PA 33 and passed through the circulator 34 to a predetermined frequency band. The BPF 35 limits the frequency band of the reception signal received by the antenna 41 to a predetermined frequency band.

  The digital processing unit 20 includes a BB (Base Band) processing unit 21, a calibration unit 22, and a plurality of FIR filters 23-1 to 23-n. The calibration unit 22 includes a plurality of selectors 220-1 to 220-n, a frequency shift unit 221, a signal generation unit 222, and a coefficient calculation unit 223. The coefficient calculation unit 223 is an example of an adjustment unit. In the following, when each of the plurality of FIR filters 23-1 to 23-n is collectively referred to without distinction, it is simply referred to as the FIR filter 23, and each of the plurality of selectors 220-1 to 220-n is distinguished. When referring generically without doing, it is simply referred to as selector 220. Each FIR filter 23 and each selector 220 is provided one by one in association with each antenna 41.

The BB processing unit 21 generates a baseband transmission signal used for data transmission, and outputs the generated transmission signal to each selector 220. FIG. 2 is a diagram for explaining an example of frequency intervals of subcarriers. BB processing unit 21, for example, as shown in FIG. 2 (a), and generates a transmission signal frequency interval is modulated by OFDM using a plurality of subcarriers is Delta] f _sd.

For example, as shown in FIG. 2B, the signal generation unit 222 in the calibration unit 22 generates a calibration signal modulated by OFDM using a plurality of subcarriers having a frequency interval of Δf _sc . Then, the signal generation unit 222 outputs the generated calibration signal to the frequency shift unit 221 and the coefficient calculation unit 223. The frequency interval Δf _sc between the plurality of subcarriers used for the calibration signal is wider than the frequency interval Δf _sd between the plurality of subcarriers used for the transmission signal. In the present embodiment, the frequency interval Δf _sc is, for example, twice the frequency interval Δf _sd . The signal generation unit 222 is an example of a first generation unit.

  The frequency shift unit 221 shifts the frequency of each subcarrier included in the calibration signal output from the signal generation unit 222 by the shift amount specified by the coefficient calculation unit 223. Then, the frequency shift unit 221 outputs a calibration signal in which the frequency of each subcarrier is shifted to each selector 220. When the shift amount instructed from the coefficient calculation unit 223 is 0, the frequency shift unit 221 does not shift the frequency of each subcarrier included in the calibration signal output from the signal generation unit 222. Output to selector 220. Hereinafter, a calibration signal in which the frequency of each subcarrier is not shifted is referred to as a first signal, and a calibration signal in which the frequency of each subcarrier is shifted is referred to as a second signal. The frequency shift unit 221 is an example of a second generation unit.

  Each selector 220 outputs the transmission signal output from the BB processing unit 21 or the calibration signal output from the frequency shift unit 221 to the FIR filter 23. Each FIR filter 23 filters the signal (transmission signal or calibration signal) output from the corresponding selector 220 based on the coefficient set by the coefficient calculation unit 223. Each FIR filter 23 outputs the filtered signal to the corresponding analog circuit 30.

  For each antenna 41, the coefficient calculation unit 223 is input to the transmission block of the analog circuit 30 corresponding to the antenna 41, and uses each of the first signal and the second signal that have passed through the analog circuit 30, respectively. The frequency response of the signal output from 41 is adjusted. Specifically, the coefficient calculation unit 223 selects the antennas 41 one by one, inputs the first signal to the analog circuit 30 corresponding to the selected antenna 41, and outputs the feedback signal via the analog circuit 30. Based on this, the frequency response is calculated. The coefficient calculation unit 223 inputs the second signal to the analog circuit 30 corresponding to the selected antenna 41 and calculates a frequency response based on the feedback signal that has passed through the analog circuit 30. Then, the coefficient calculation unit 223 uses the frequency response calculated from the feedback signal of the first signal and the frequency response calculated from the feedback signal of the second signal, and the FIR corresponding to the selected antenna 41. A coefficient set in the filter 23 is calculated. Then, the coefficient calculation unit 223 sets the calculated coefficient in the FIR filter 23.

  FIG. 3 is a block diagram illustrating an example of the coefficient calculation unit 223. For example, as illustrated in FIG. 3, the coefficient calculation unit 223 includes an adjustment processing unit 2230, a synthesis unit 2231, a calculation unit 2232, and a frequency shift unit 2233.

  The calculation unit 2232 selects one antenna 41 and controls the selector 43 so that the feedback signal output from the coupler 42 corresponding to the selected antenna 41 is selected. Further, the calculation unit 2232 controls the selector 220 corresponding to the selected antenna 41 so that the calibration signal from the frequency shift unit 221 is output to the FIR filter 23. In addition, the calculation unit 2232 sets a setting value (for example, a setting value in which the amplitude of the basic component is 1 and the amplitudes of the delay components are all 0) in which the input signal is output as it is, in each FIR filter 23.

  Then, the calculation unit 2232 instructs the frequency shift unit 221 and the frequency shift unit 2233 to set 0 as the shift amount. The frequency shift unit 221 outputs the first signal to each selector 220 without shifting the frequency of each subcarrier included in the calibration signal output from the signal generation unit 222. Then, the first signal is output from the selector 220 corresponding to the selected antenna 41 to the analog circuit 30 via the FIR filter 23.

  The first signal is output to the coupler 42 as a feedback signal via the transmission block of the analog circuit 30 corresponding to the selected antenna 41. The feedback signal is fed back to the frequency shift unit 2233 in the coefficient calculation unit 223 via the selector 43, the amplifier 44, the down converter 45, and the ADC 46. Since 0 is instructed as a shift amount from the calculation unit 2232, the frequency shift unit 2233 outputs a feedback signal to the calculation unit 2232 without shifting the frequency of each subcarrier.

  Then, the calculation unit 2232 calculates a frequency response in the transmission block of the analog circuit 30 corresponding to the selected antenna 41 based on the feedback signal corresponding to the first signal. FIG. 4 is a diagram for explaining an example of the frequency response of the analog circuit 30. For example, as illustrated in FIG. 4A, the calculation unit 2232 inputs a first signal having flat amplitude characteristics and phase characteristics to the analog circuit 30. The feedback signal corresponding to the first signal that has passed through the analog circuit 30 exhibits an amplitude characteristic and a phase characteristic corresponding to the frequency characteristic 50 of the analog circuit 30, for example, as shown in FIG. In FIG. 4, the amplitude characteristic is shown as an example of the frequency characteristic of the analog circuit 30, but the phase characteristic can be observed in the same manner.

  The calculation unit 2232 demodulates the feedback signal. Then, for each subcarrier, the calculation unit 2232 compares the amplitude and phase of the demodulated feedback signal with the amplitude and phase of the calibration signal generated by the signal generation unit 222, so that the frequency response for each subcarrier is obtained. Is calculated. Then, the calculation unit 2232 outputs the calculated frequency response to the synthesis unit 2231.

Here, the frequency interval Δf _sc of the subcarriers (frequency f _{1 to} f _m in FIG. 4) included in the first signal is wider than the frequency interval Δf _{sd of the} subcarriers included in the data signal. Therefore, the frequency response calculated from the feedback signal corresponding to the first signal includes the frequency response at the frequency of some of the subcarriers included in the data signal (for example, the frequency indicated by the dotted line in FIG. 4B). Is not included.

  Next, the calculation unit 2232 instructs + Δf as the shift amount to the frequency shift unit 221 and instructs -Δf as the shift amount to the frequency shift unit 2233. The frequency shift unit 221 generates a second signal by shifting the frequency of each subcarrier of the calibration signal generated by the signal generation unit 222 by + Δf, and sends the generated second signal to each selector 220. Output. The second signal is output from the selector 220 corresponding to the selected antenna 41 to the analog circuit 30 via the FIR filter 23.

  Then, the second signal is output to the coupler 42 as a feedback signal via the transmission block of the analog circuit 30 corresponding to the selected antenna 41. The feedback signal is fed back to the frequency shift unit 2233 via the selector 43, the amplifier 44, the down converter 45, and the ADC 46.

  The frequency shift unit 2233 performs a frequency shift opposite to the frequency shift performed by the frequency shift unit 221 on the feedback signal corresponding to the second signal. Specifically, the frequency shift unit 2233 shifts the frequency of each subcarrier of the feedback signal corresponding to the second signal by −Δf according to the shift amount of −Δf instructed from the calculation unit 2232. Then, the frequency shift unit 2233 outputs the feedback signal after the frequency shift to the calculation unit 2232. By shifting the frequency of each subcarrier by −Δf, the frequency of each subcarrier of the feedback signal after the frequency shift becomes the same as the frequency of each subcarrier of the first signal.

Then, the calculating unit 2232 calculates the frequency response of the analog circuit 30 corresponding to the selected antenna 41 based on the feedback signal corresponding to the second signal. FIG. 5 is a diagram for explaining an example of the frequency response of the analog circuit 30. For example, as shown in FIG. 5A, the subcarrier frequency (frequency f ₁ ′ to f _m ′ in FIG. 5) included in the second signal is the frequency of the subcarrier included in the first signal (FIG. 5 is shifted by + Δf from the frequency f _{1 to} f _m ). In this embodiment, the shift amount Delta] f of the frequency of each subcarrier, for example, as shown in FIG. 5 (a), smaller than the frequency interval Delta] f _sc of the plurality of sub-carriers used in the calibration signal. For example, as illustrated in FIG. 5A, the calculation unit 2232 inputs a second signal having flat amplitude characteristics and phase characteristics to the analog circuit 30. The feedback signal corresponding to the second signal that has passed through the analog circuit 30 exhibits an amplitude characteristic and a phase characteristic corresponding to the frequency characteristic 50 of the analog circuit 30, as shown in FIG. 5B, for example.

  Specifically, the calculation unit 2232 demodulates a feedback signal corresponding to the second signal. Note that the feedback signal shifted by + Δf by the frequency shift unit 221 is shifted by −Δf by the frequency shift unit 2233, so that the calculation unit 2232 can demodulate the feedback signal without causing interference between subcarriers. .

  The calculation unit 2232 calculates the frequency response for each subcarrier by comparing the amplitude and phase of the demodulated feedback signal with the amplitude and phase of the calibration signal generated by the signal generation unit 222 for each subcarrier. To do. Then, the calculation unit 2232 outputs the calculated frequency response to the synthesis unit 2231.

Here, the frequency interval Δf _sc of the subcarriers included in the second signal (frequency f ₁ ′ to f _m ′ in FIG. 5) is wider than the frequency interval Δf _{sd of the} subcarriers included in the data signal. Therefore, the frequency response calculated from the feedback signal corresponding to the second signal includes the frequency response at the frequency of some of the subcarriers included in the data signal (for example, the frequency indicated by the dotted line in FIG. 5B). Is not included.

  The combining unit 2231 calculates the frequency response for each subcarrier calculated from the feedback signal corresponding to the first signal and the frequency response for each subcarrier calculated from the feedback signal corresponding to the second signal. Get from. Then, the combining unit 2231 combines the acquired frequency responses for each subcarrier.

  FIG. 6 is a diagram for explaining an example of a synthesis process by the synthesis unit 2231. In FIG. 6, the frequency response 51 a indicates the frequency response of each subcarrier calculated from the feedback signal corresponding to the first signal, and the frequency response 51 b is calculated from the feedback signal corresponding to the second signal. The frequency response of each subcarrier is also shown. The frequency shift unit 2233 shifts the frequency of each subcarrier of the feedback signal corresponding to the second signal by −Δf, so that the frequency of each subcarrier after the frequency shift becomes the frequency of each subcarrier of the first signal. It is the same as the frequency.

For example, as illustrated in FIG. 6, the combining unit 2231 shifts the frequency response 51 b of each subcarrier of the feedback signal corresponding to the second signal by + Δf to obtain the frequency response 51 c. Then, the combining unit 2231 includes a frequency response 51a of each subcarrier identified from the feedback signal corresponding to the first signal, and a frequency response 51c of each subcarrier identified from the feedback signal corresponding to the second signal. Is synthesized. Thereby, the frequency response 51d is generated. For example, as shown in FIG. 6, the frequency response 51 d includes a frequency response for each subcarrier having a frequency interval of Δf (Δf _{sd in} this embodiment). The combining unit 2231 outputs the combined frequency response 51d to the adjustment processing unit 2230.

  The adjustment processing unit 2230 calculates the coefficient set in the FIR filter 23 corresponding to the selected antenna 41 using the frequency response 51d synthesized by the synthesis unit 2231. The adjustment processing unit 2230 calculates each coefficient set in the FIR filter 23 so that the frequency response 51d is flat in the frequency band of the data signal, for example. Then, the adjustment processing unit 2230 sets the calculated coefficient in the FIR filter 23.

Here, the frequency interval Δf _sc of the plurality of subcarriers used for the first signal and the second signal is the frequency interval of the plurality of subcarriers used for the data signal as shown in FIGS. 4 and 5, for example. It is wider than Δf _sd . Therefore, if the frequency response of the analog circuit 30 is calculated using only the first signal, some of the frequency components used for the data signal are obtained by interpolation. Therefore, a frequency response different from the actual frequency response of the analog circuit 30 may be calculated at the subcarrier frequency used for the data signal. Therefore, it is difficult to calculate the frequency response of the analog circuit 30 with high accuracy.

On the other hand, in this embodiment, the frequency response of each subcarrier of the first signal and the frequency response of each subcarrier of the second signal shifted from the first signal by a predetermined frequency Δf are combined. . Thereby, it is possible to calculate the frequency response of each subcarrier in a frequency interval narrower than the frequency interval Δf _sc of the plurality of subcarriers used for the first signal. Thereby, compared with the case where the frequency response of the analog circuit 30 is calculated using only the first signal, the frequency range in which the frequency response is interpolated can be narrowed. Thereby, the frequency response of the analog circuit 30 can be calculated more accurately.

In particular, in this embodiment, the frequency interval Δf _sc of the plurality of subcarriers used for the first signal and the second signal is set to a plurality of subcarriers used for the data signal as shown in FIG. 2B, for example. it is twice the frequency interval Δf _sd career. For example, as shown in FIG. 5A, the second signal is shifted from the first signal by ½ of the frequency interval Δf _sc , that is, by the frequency interval Δf _sd . Therefore, the frequency response obtained by combining the frequency response of each subcarrier of the first signal and the frequency response of each subcarrier of the second signal becomes a value close to the frequency response of each subcarrier of the data signal.

  Further, the frequency and the shift amount of each subcarrier of the first signal are selected so that the frequency of each subcarrier after synthesis matches the frequency of each subcarrier of the data signal. Thereby, the frequency of each subcarrier after synthesis matches the frequency of each subcarrier of the data signal. Thereby, the frequency response of each subcarrier used for a data signal can be estimated more accurately by the frequency response of each subcarrier after combining. Also, the frequency response corresponding to the first signal and the frequency response corresponding to the second signal can be calculated separately at different times. Therefore, the frequency response of each subcarrier used for the data signal can be accurately estimated even in a guard time shorter than the communication frame used for data transmission / reception. Therefore, the accuracy of calibration between the antennas 41 can be improved.

[Calibration process]
FIG. 7 is a flowchart illustrating an example of the calibration process. The base station 10 executes the process shown in this flowchart in a guard time provided for switching between uplink and downlink, for example.

  First, the calculation unit 2232 of the coefficient calculation unit 223 selects one antenna 41 (S100). Then, the calculation unit 2232 controls the selector 43 so that the feedback signal output from the coupler 42 corresponding to the selected antenna 41 is selected. Further, the calculation unit 2232 controls the selector 220 corresponding to the selected antenna 41 so that the calibration signal from the frequency shift unit 221 is output to the FIR filter 23. In addition, the calculation unit 2232 sets a setting value for outputting the input signal as it is to each FIR filter 23. The calculation unit 2232 instructs the frequency shift unit 221 and the frequency shift unit 2233 to set the shift amount to 0.

  Next, the signal generator 222 generates a calibration signal (S101). The frequency shift unit 221 outputs the first signal to each selector 220 without shifting the frequency of each subcarrier included in the calibration signal generated by the signal generation unit 222. The first signal is transmitted from the selector 220 corresponding to the antenna 41 selected in step S100 to the analog circuit 30 via the FIR filter 23 (S102).

  The first signal is output to the coupler 42 as a feedback signal via the transmission block of the analog circuit 30 corresponding to the antenna 41 selected in step S100. The feedback signal is fed back to the frequency shift unit 2233 of the coefficient calculation unit 223 via the selector 43, the amplifier 44, the down converter 45, and the ADC 46. The frequency shift unit 2233 receives the feedback signal corresponding to the first signal (S103). Then, frequency shift section 2233 outputs a feedback signal corresponding to the first signal to 2232 without shifting the frequency of each subcarrier.

  Next, the calculation unit 2232 demodulates the feedback signal corresponding to the first signal. Then, the calculation unit 2232 calculates the frequency response of the analog circuit 30 corresponding to the antenna 41 selected in step S100 based on the demodulated feedback signal (S104). Then, the calculation unit 2232 outputs the frequency response calculated for each subcarrier to the synthesis unit 2231.

  Next, the calculation unit 2232 instructs + Δf as the shift amount to the frequency shift unit 221 and instructs -Δf as the shift amount to the frequency shift unit 2233. The frequency shift unit 221 generates a second signal by shifting the frequency of each subcarrier of the calibration signal generated by the signal generation unit 222 by + Δf, and sends the generated second signal to each selector 220. Output. The second signal is transmitted from the selector 220 corresponding to the selected antenna 41 to the analog circuit 30 via the FIR filter 23 (S105).

  The second signal passes through the transmission block of the analog circuit 30 corresponding to the selected antenna 41 and is output to the coupler 42 as a feedback signal. The feedback signal is fed back to the frequency shift unit 2233 via the selector 43, the amplifier 44, the down converter 45, and the ADC 46. The frequency shift unit 2233 receives the feedback signal corresponding to the second signal (S106). Then, the frequency shift unit 2233 shifts the frequency of each subcarrier of the feedback signal corresponding to the second signal by −Δf (S107). Then, the frequency shift unit 2233 outputs the feedback signal after the frequency shift to the calculation unit 2232.

  Next, the calculation unit 2232 demodulates the feedback signal corresponding to the second signal. Then, the calculation unit 2232 calculates the frequency response of the analog circuit 30 corresponding to the antenna 41 selected in step S100 based on the demodulated feedback signal (S108). Then, the calculation unit 2232 outputs the frequency response calculated for each subcarrier to the synthesis unit 2231.

  Next, combining section 2231 shifts the frequency response of each subcarrier of the feedback signal corresponding to the second signal by + Δf. Then, combining section 2231 combines the frequency response of each subcarrier specified from the feedback signal corresponding to the first signal and the frequency response of each subcarrier specified from the feedback signal corresponding to the second signal. (S109). Then, combining section 2231 outputs the frequency response after combining to adjustment processing section 2230.

  Based on the frequency response synthesized by the synthesis unit 2231, the adjustment processing unit 2230 calculates a coefficient set in the FIR filter 23 corresponding to the antenna 41 selected in step S100 (S110).

  Next, the calculation unit 2232 determines whether all the antennas 41 have been selected (S111). When there is an unselected antenna 41 (S111: No), the calculation unit 2232 executes the process shown in step S100 again. On the other hand, when all the antennas 41 have been selected (S111: Yes), the adjustment processing unit 2230 sets the coefficient calculated for each FIR filter 23 in step S110 to the corresponding FIR filter 23 (S112). Then, the base station 10 ends the calibration process shown in this flowchart.

  In addition, although the base station 10 performs the process shown in the above-described flowchart in the guard time, for example, the disclosed technique is not limited to this. For example, the base station 10 performs transmission of the first signal and transmission of the second signal at guard timings at different timings, and performs processing such as frequency response calculation, frequency response synthesis, and coefficient calculation. It may be executed at a time other than the guard time. Thereby, the processing load of the base station 10 can be reduced.

[Effect of Example 1]
As is clear from the above description, the base station 10 of this embodiment includes a plurality of antennas 41, a plurality of analog circuits 30 provided for each antenna 41, a signal generation unit 222, a frequency shift unit 221, and a coefficient. And a calculation unit 223. The signal generator 222 generates a first signal including a plurality of subcarriers having a frequency interval wider than that of the plurality of subcarriers included in the data signal. The frequency shift unit 221 generates a second signal obtained by shifting the frequency of the first signal by a predetermined frequency. For each antenna 41, the coefficient calculation unit 223 is input to the analog circuit 30 corresponding to the antenna 41, and uses each of the first signal and the second signal that have passed through the analog circuit 30 to use the antenna 41. The frequency response of the signal output from is adjusted. Thereby, the base station 10 of a present Example can improve the precision of the calibration between the antennas 41. FIG.

  In the above-described embodiment, the coefficient calculation unit 223 includes an adjustment processing unit 2230, a synthesis unit 2231, a calculation unit 2232, and a frequency shift unit 2233. The frequency shift unit 2233 performs a frequency shift opposite to the frequency shift performed by the frequency shift unit 221 on the second signal that has passed through the analog circuit 30. For each antenna 41, the calculation unit 2232 calculates the frequency response of each subcarrier used for the second signal on which the frequency shift unit 2233 has performed the reverse frequency shift. For each antenna 41, the combining unit 2231 applies the same frequency shift as the frequency shift unit 221 to the subcarriers corresponding to the respective frequency responses calculated by the calculation unit 2232. The combining unit 2231 then combines the frequency response of each subcarrier after the frequency shift and the frequency response of each subcarrier used for the first signal that has passed through the analog circuit 30. The adjustment processing unit 2230 adjusts the frequency response of the signal output from each antenna 41 based on the frequency response for each subcarrier combined by the combining unit 2231 for each antenna 41. Thereby, the base station 10 of a present Example can improve the precision of the calibration between the antennas 41. FIG.

  In the embodiment described above, the predetermined frequency for shifting the frequency of the first signal is smaller than the frequency interval of the plurality of subcarriers used for the first signal. Thus, in the frequency response obtained by combining the frequency response specified using the first signal and the frequency response specified using the second signal, the frequency response specified using the first signal is used. The frequency response of the analog circuit 30 can be specified with higher accuracy. Thereby, the base station 10 of a present Example can improve the precision of the calibration between the antennas 41. FIG.

  In the above-described embodiment, the frequency interval between the plurality of subcarriers used for the first signal is twice the frequency interval between the plurality of subcarriers used for the data signal. The predetermined frequency for shifting the frequency of the first signal is ½ of the frequency interval of the plurality of subcarriers used for the first signal. Thus, the frequency response obtained by combining the frequency response specified using the first signal and the frequency response specified using the second signal is a frequency response in a plurality of subcarriers used for the data signal. A value close to. Thereby, the base station 10 of a present Example can improve the precision of the calibration between the antennas 41. FIG.

The FIR filter 23 has a desired pass characteristic within the band of the data signal, and has a blocking region so that unnecessary signals are not generated outside the band. Therefore, the frequency response C (f) of each FIR filter 23 has a desired pass characteristic within the band Δf _{D of the} data signal, for example, as shown in FIG. 8A, and abruptly outside the band Δf _D. The frequency response is a rectangular shape that attenuates to a point. FIG. 8 is a diagram for explaining an example of the frequency response C (f) of the FIR filter 23.

  When the frequency response C (f) of the FIR filter 23 shown in FIG. 8A is replaced with the time response c (t) of the FIR filter 23 in the time domain, for example, the distribution shown in FIG. 8B is obtained. If the frequency response C (f) of the FIR filter 23 is rectangular, many high-frequency components are included, so the time response c (t) of the FIR filter 23 varies like a sinc function, and the tap of the FIR filter 23 The number increases. In this case, for example, with the number of taps (for example, 32) representing the time response c (t) within the range surrounded by the broken line in FIG. 8B, the rectangular frequency response C (f) shown in FIG. ) Is difficult to reproduce.

On the other hand, for example, as shown in FIG. 9A, when the smoothing process in which the frequency response C (f) attenuates gently is performed outside the band Δf _D of the data signal, the frequency of the FIR filter 23 High frequency components included in the response C (f) are reduced. FIG. 9 is a diagram illustrating an example of the smoothing process. In the example of FIG. 9A, the frequency response C (f) gradually attenuates to 0 in the frequency range of Δf ₁ outside the band Δf _D of the data signal. As a result, the time response c (t) of the FIR filter 23 is concentrated at a time within a predetermined range as shown in FIG. 9B, for example. Therefore, for example, even when the number of taps (for example, 32) expressing the time response c (t) within the range surrounded by the broken line in FIG. 9B is used, the smoothing process after the smoothing process shown in FIG. It becomes possible to reproduce the frequency response C (f). Thereby, the number of taps of the FIR filter 23 can be reduced, and the circuit scale of the FIR filter 23 can be reduced.

In the second embodiment, the adjustment processing unit 2230 in the coefficient calculation unit 223 calculates the coefficient set in the FIR filter 23 based on the frequency response of each subcarrier combined by the combining unit 2231. At that time, for example, as shown in FIG. 9A, the adjustment processing unit 2230 performs the smoothing process so that the frequency response gradually attenuates to 0 outside the band Δf _D of the data signal. Based on the frequency response, the coefficient of the FIR filter 23 is calculated.

Here, for example, when the frequency response of the analog circuit 30 is calculated using only the first signal having subcarriers of the frequencies f _{1 to} f _m as shown in FIG. 10A, for example, in FIG. As shown, the frequency response of the analog circuit 30 is calculated for each subcarrier. FIG. 10 is a diagram illustrating an example of the smoothing process in the comparative example. The frequency response between adjacent subcarriers can be approximated to some extent by interpolating the frequency response of adjacent subcarriers. However, for example, as shown in FIG. 10, the frequency response at frequencies higher than the lower frequency and the frequency f _m than the frequency f _1, it is difficult to interpolate. Furthermore, if the frequency response at frequencies higher than the lower frequency and the frequency f _m than the frequency f ₁ is smoothed, for example, as shown by the dotted line in the region surrounded by the frequency response of FIG. 10 (b), the data signal band Delta] f _D The accuracy of the frequency response deteriorates near the boundary.

On the other hand, in the base station 10 of the present embodiment, for example, as shown in FIG. 11A, the first signal and the second signal shifted by a predetermined frequency Δf (for example, Δf _sd ) are used. The frequency response of the analog circuit 30 is calculated. FIG. 11 is a diagram illustrating an example of the smoothing process according to the second embodiment. In FIG. 11, frequencies f _{1 to} f _m indicate the frequency of each subcarrier in the first signal, and frequencies f ₁ ′ to f _m ′ indicate the frequency of each subcarrier in the second signal. Thus, for example, as shown in FIG. 11B, even when the frequency response at a frequency lower than the frequency f ₁ and a frequency higher than the frequency f _m ′ is smoothed, the frequency near the boundary of the band Δf _D of the data signal The accuracy of response can be kept high.

[Effect of Example 2]
As is clear from the above description, the base station 10 of this embodiment calculates the frequency response of the analog circuit 30 using the first signal, and calculates the frequency response of the analog circuit 30 using the second signal. . Then, the base station 10 shifts the frequency response calculated using the second signal by + Δf and combines it with the frequency response calculated using the first signal. Then, the base station 10 calculates a coefficient set in the FIR filter 23 based on the synthesized frequency response. At that time, the base station 10 calculates the coefficient of the FIR filter 23 so that the component outside the band of the data signal has a frequency response that gently attenuates. Thereby, the base station 10 of a present Example can reduce the number of taps of the FIR filter 23, keeping the calibration precision between the antennas 41 high.

  In the first embodiment, transmission block calibration has been described. In the third embodiment, calibration of a reception block will be described. FIG. 12 is a block diagram illustrating an example of the base station 10 according to the third embodiment. Except for the points described below, in FIG. 12, blocks denoted by the same reference numerals as those in FIG. 1 have the same or similar functions as the blocks shown in FIG.

  The base station 10 includes a digital processing unit 20, a plurality of analog circuits 30-1 to 30-n, a plurality of antennas 41-1 to 41-n, a selector 47, and a plurality of selectors 48-1 to 48-n. Hereinafter, the selectors 48-1 to 48-n are simply referred to as selectors 48 when collectively referred to without being distinguished from each other.

  In the base station 10 of the present embodiment, one analog circuit 30 and one selector 48 are provided in association with one antenna 41. Each selector 48 switches the connection between the corresponding antenna 41 and the corresponding analog circuit 30 and the connection between the selector 47 and the corresponding analog circuit 30.

  Each analog circuit 30 includes a DAC 31, an up converter 32, a PA 33, a circulator 34, a BPF 35, an LNA (Low Noise Amplifier) 36, a down converter 37, and an ADC 38. Hereinafter, the LNA 36, the down converter 37, and the ADC 38 are referred to as reception blocks. Note that a selector 39 is provided in one of the plurality of analog circuits 30 (analog circuit 30-1 in the example of FIG. 12).

  The LNA 36 amplifies the signal output from the circulator 34. The down converter 37 performs processing such as quadrature detection and down conversion on the signal amplified by the LNA 36. The ADC 38 converts the signal subjected to quadrature detection or the like by the down converter 37 from an analog signal to a digital signal and outputs the digital signal to the digital processing unit 20.

  The digital processing unit 20 includes a BB processing unit 21, a calibration unit 22, an FIR filter 23, and a plurality of FIR filters 24-1 to 24-n. The calibration unit 22 includes a selector 220, a frequency shift unit 221, a signal generation unit 222, a coefficient calculation unit 223, a plurality of selectors 225-1 to 225 -n, and a selector 226. In the following, when each of the plurality of FIR filters 24-1 to 24-n is collectively referred to without distinction, it is simply referred to as the FIR filter 24, and each of the plurality of selectors 225-1 to 225-n is distinguished. When referring generically without doing, it is simply referred to as selector 225. Each FIR filter 24 and each selector 225 is provided in association with each antenna 41.

  Each FIR filter 24 filters the signal (reception signal or calibration signal) output from the corresponding analog circuit 30 based on the coefficient set by the coefficient calculation unit 223. Each FIR filter 24 outputs the filtered signal to the corresponding selector 225. Each selector 225 outputs the signal output from the corresponding FIR filter 24 to the BB processing unit 21 or the selector 226. The selector 226 outputs a signal output from any of the plurality of selectors 225 to the coefficient calculation unit 223.

  For each antenna 41, the coefficient calculation unit 223 is input to the reception block of the analog circuit 30 corresponding to the antenna 41, and uses each of the first signal and the second signal that have passed through the analog circuit 30, respectively. The frequency response of the signal received via 41 is adjusted. Specifically, the coefficient calculation unit 223 selects the antennas 41 one by one, inputs the first signal to the reception block of the analog circuit 30 corresponding to the selected antenna 41, and passes through the analog circuit 30. A frequency response is calculated based on the feedback signal. Further, the coefficient calculation unit 223 inputs the second signal to the reception block of the analog circuit 30 corresponding to the selected antenna 41, and measures the frequency response based on the feedback signal that has passed through the analog circuit 30. Then, the coefficient calculation unit 223 uses the frequency response calculated from the feedback signal of the first signal and the frequency response calculated from the feedback signal of the second signal, and the FIR corresponding to the selected antenna 41. A coefficient set in the filter 24 is calculated. Then, the coefficient calculation unit 223 sets the calculated coefficient in the FIR filter 24.

  For example, as illustrated in FIG. 3, the coefficient calculation unit 223 includes an adjustment processing unit 2230, a synthesis unit 2231, a calculation unit 2232, and a frequency shift unit 2233. The calculation unit 2232 controls the selector 39 so that the signal output from the PA 33 in the analog circuit 30-1 is output to the selector 47. In addition, the calculation unit 2232 selects one antenna 41, and the selector 47 and the correspondence so that the signal output from the PA 33 in the analog circuit 30-1 is input to the selector 48 corresponding to the selected antenna 41. The selector 48 to be controlled is controlled. In addition, the calculation unit 2232 controls the selector 220 so that the calibration signal from the frequency shift unit 221 is output to the FIR filter 23. In addition, the calculation unit 2232 sets a set value at which the input signal is output as it is in the FIR filter 23 and each FIR filter 24. Further, the calculation unit 2232 controls the selector 225 corresponding to the selected antenna 41 so that the signal output from the FIR filter 24 is output to the selector 226. In addition, the calculation unit 2232 controls the selector 226 so that the output from the selector 225 corresponding to the selected antenna 41 is output to the coefficient calculation unit 223.

  Then, the calculation unit 2232 instructs the frequency shift unit 221 and the frequency shift unit 2233 to set 0 as the shift amount. As a result, the calibration signal generated by the signal generation unit 222 is output to the selector 220 as the first signal, and is output to the analog circuit 30-1 through the FIR filter 23.

  Then, the first signal is input to the selector 48 corresponding to the selected antenna 41 via the selector 39 and the selector 47 in the analog circuit 30-1. The first signal input to the selector 48 corresponding to the selected antenna 41 is output to the digital processing unit 20 as a feedback signal via the reception block of the corresponding analog circuit 30. The feedback signal is output to the selector 226 via the FIR filter 24 and the selector 225 corresponding to the selected antenna 41, and is output to the coefficient calculation unit 223 via the selector 226. Since 0 is instructed as a shift amount from the calculation unit 2232, the frequency shift unit 2233 outputs a feedback signal to the calculation unit 2232 without shifting the frequency of each subcarrier.

  The calculation unit 2232 demodulates the feedback signal corresponding to the first signal. Then, the calculation unit 2232 compares the amplitude and phase of each demodulated feedback signal in each subcarrier with the amplitude and phase in each subcarrier of the calibration signal generated by the signal generation unit 222, thereby obtaining a frequency response. calculate. Then, the calculation unit 2232 outputs the calculated frequency response to the synthesis unit 2231.

  Next, the calculation unit 2232 instructs + Δf as the shift amount to the frequency shift unit 221 and instructs -Δf as the shift amount to the frequency shift unit 2233. As a result, the calibration signal generated by the signal generator 222 is output to the selector 220 as a second signal in which the frequency of each subcarrier is shifted by + Δf, and is output to the analog circuit 30 via the FIR filter 23. The

  Then, the second signal is input to the selector 48 corresponding to the selected antenna 41 via the selector 39 and the selector 47 in the analog circuit 30-1. The second signal input to the selector 48 corresponding to the selected antenna 41 is output to the digital processing unit 20 as a feedback signal via the reception block of the corresponding analog circuit 30. The feedback signal is output to the selector 226 via the FIR filter 24 and the selector 225 corresponding to the selected antenna 41, and is output to the coefficient calculation unit 223 via the selector 226. Since −Δf is instructed as the shift amount from the calculation unit 2232, the frequency shift unit 2233 shifts the frequency of each subcarrier of the feedback signal corresponding to the second signal by −Δf, and the feedback signal after the frequency shift Is output to the calculation unit 2232.

  Then, the calculation unit 2232 demodulates the feedback signal corresponding to the second signal. Then, the calculation unit 2232 compares the amplitude and phase of each demodulated feedback signal in each subcarrier with the amplitude and phase in each subcarrier of the calibration signal generated by the signal generation unit 222, thereby obtaining a frequency response. calculate. Then, the calculation unit 2232 outputs the calculated frequency response to the synthesis unit 2231.

  The combining unit 2231 calculates the frequency response for each subcarrier calculated from the feedback signal corresponding to the first signal and the frequency response for each subcarrier calculated from the feedback signal corresponding to the second signal. Get from. Then, as described with reference to FIG. 6, for example, the combining unit 2231 combines the acquired frequency responses for each subcarrier. The combining unit 2231 outputs the combined frequency response to the adjustment processing unit 2230.

  The adjustment processing unit 2230 calculates a coefficient set in the FIR filter 24 corresponding to the selected antenna 41 using the frequency response generated by the combining unit 2231. The adjustment processing unit 2230 calculates each coefficient set in the FIR filter 24 so that the synthesized frequency response becomes, for example, flat in the frequency band of the data signal. Then, the adjustment processing unit 2230 sets the calculated coefficient in the FIR filter 24.

[Effect of Example 3]
As is apparent from the above description, the base station 10 of this embodiment calculates the frequency response of the reception block of the analog circuit 30 using the first signal, and uses the second signal to receive the reception block of the analog circuit 30. The frequency response of is calculated. Then, the base station 10 shifts the frequency response calculated using the second signal by + Δf and combines it with the frequency response calculated using the first signal. Then, the base station 10 calculates a coefficient set in the FIR filter 24 based on the synthesized frequency response. Thereby, the base station 10 of a present Example can improve the precision of the calibration between the antennas 41 in a receiving block.

  Also in the present embodiment, as in the second embodiment, when calculating the coefficient set in the FIR filter 24, the FIR filter has a frequency response in which the component outside the band of the data signal gently attenuates. 24 coefficients may be calculated.

[hardware]
The base station 10 in the first to third embodiments described above is realized by hardware as shown in FIG. 13, for example. FIG. 13 is a diagram illustrating an example of hardware of the base station 10. The base station 10 includes an interface circuit 11, a memory 12, a processor 13, a feedback circuit 14, a bypass circuit 15, a plurality of radio circuits 16-1 to 16-n, and a plurality of antennas 41-1 to 41-n. Hereinafter, the plurality of wireless circuits 16-1 to 16-n are simply referred to as the wireless circuit 16 when collectively referred to without being distinguished from each other.

  The interface circuit 11 is an interface for performing wired communication with the core network. Each radio circuit 16 includes an analog circuit 30. One radio circuit 16 is provided corresponding to one antenna 41. Each wireless circuit 16 performs processing such as up-conversion on the signal output from the processor 13 and transmits the processed signal via the corresponding antenna 41. Each radio circuit 16 performs processing such as signal down-conversion received via the corresponding antenna 41 and outputs the processed signal to the processor 13. Each radio circuit 16 includes, for example, an analog circuit 30, a coupler 42, and a selector 48. The radio circuit 16-1 includes a selector 39, for example.

  The feedback circuit 14 feeds back the signal output from each radio circuit 16 to the processor 13. The feedback circuit 14 includes, for example, a selector 43, an amplifier 44, a down converter 45, and an ADC 46. The bypass circuit 15 inputs the signal output from the wireless circuit 16-1 to the selector 48 in any one of the wireless circuits 16. The bypass circuit 15 is a selector 47, for example.

  The memory 12 stores, for example, various programs and data for realizing the functions of the digital processing unit 20. The processor 13 implements each function of the digital processing unit 20, for example, by executing a program read from the memory 12.

  Note that not all programs, data, and the like in the memory 12 need be stored in the memory 12 from the beginning. For example, a program, data, or the like is stored in a portable recording medium such as a memory card inserted into the base station 10, and the base station 10 appropriately acquires and executes the program, data, etc. from such a portable recording medium. It may be. Further, the base station 10 appropriately acquires and executes the program from another computer or server device storing the program, data, etc. via a wireless communication line, public line, Internet, LAN, WAN, or the like. May be.

<Others>
The disclosed technology is not limited to the above-described embodiments, and various modifications are possible within the scope of the gist.

  For example, in each of the embodiments described above, the calibration unit 22 is provided in the digital processing unit 20, but the disclosed technique is not limited to this, and the calibration unit 22 is a calibration provided outside the digital processing unit 20. It may be realized as an application device. In this case, the signal output from the BB processing unit 21 is output to each FIR filter 23 via a calibration device provided outside the digital processing unit 20. Further, the signals output from the FIR filters 24 are output to the BB processing unit 21 via a calibration device provided outside the digital processing unit 20.

  Further, in each of the above-described embodiments, the frequency shift unit 221 generates the second signal by shifting the frequency of each subcarrier of the calibration signal generated by the signal generation unit 222 by + Δf. Technology is not limited to this. The frequency shift unit 221 may generate the second signal by shifting the frequency of each subcarrier of the calibration signal generated by the signal generation unit 222 by −Δf. However, in this case, the frequency shift unit 2233 shifts the frequency of each subcarrier of the feedback signal corresponding to the second signal by + Δf, and outputs the feedback signal after the frequency shift to the calculation unit 2232.

  In each of the above-described embodiments, each processing block of the base station 10 is classified according to the function according to the main processing contents in order to facilitate understanding of the base station 10 in the embodiment. For this reason, the disclosed technique is not limited by the processing block classification method and its name. In addition, each processing block of the base station 10 in each embodiment described above can be subdivided into a larger number of processing blocks according to the processing contents, or a plurality of processing blocks can be integrated into one processing block. You can also. In each embodiment described above, the processing executed by each processing block is realized by software, but at least a part of the processing may be realized by dedicated hardware such as an ASIC (Application Specific Integrated Circuit). Good.

10 base station 11 interface circuit 12 memory 13 processor 14 feedback circuit 15 bypass circuit 16 wireless circuit 20 digital processing unit 21 BB processing unit 22 calibration unit 220 selector 221 frequency shift unit 222 signal generation unit 223 coefficient calculation unit 2230 adjustment processing unit 2231 Synthesis unit 2232 Calculation unit 2233 Frequency shift unit 225 Selector 226 Selector 23 FIR filter 24 FIR filter 30 Analog circuit 31 DAC
32 Upconverter 33 PA
34 Circulator 35 BPF
36 LNA
37 Downconverter 38 ADC
39 Selector 41 Antenna 42 Coupler 43 Selector 44 Amplifier 45 Downconverter 46 ADC
47 Selector 48 Selector 50 Frequency characteristics 51 Frequency response

Claims (6)

In a base station having a plurality of antennas and a plurality of analog circuits provided corresponding to the respective antennas,
A first generator that generates a first signal using a plurality of subcarriers having a frequency interval wider than that of the plurality of subcarriers used for the data signal;
A second generator for generating a second signal obtained by shifting the first signal by a predetermined frequency;
For each of the antennas, a signal that is input to the analog circuit corresponding to the antenna and that is output from the antenna using the first signal and the second signal that pass through the analog circuit. A base station that adjusts a frequency response. The adjustment unit is
A frequency shift unit that performs a frequency shift opposite to the frequency shift performed by the second generation unit on the second signal that has passed through the analog circuit;
For each antenna, a calculation unit that calculates a frequency response of each subcarrier used for the second signal that has been subjected to the reverse frequency shift by the frequency shift unit;
For each antenna, subcarriers corresponding to the respective frequency responses calculated by the calculation unit are subjected to the same frequency shift as the second generation unit, and the frequency responses of the respective subcarriers after the frequency shift A combining unit that combines the frequency response of each subcarrier used for the first signal via the analog circuit;
2. An adjustment processing unit that adjusts a frequency response of a signal output from each antenna based on a frequency response for each subcarrier combined by the combining unit for each antenna. Base station described in.   The base station according to claim 1 or 2, wherein the predetermined frequency is smaller than a frequency interval of a plurality of subcarriers used for the first signal. The frequency interval of the plurality of subcarriers used for the first signal is twice the frequency interval of the plurality of subcarriers used for the data signal,
The base station according to claim 3, wherein the predetermined frequency is ½ of a frequency interval of a plurality of subcarriers used for the first signal. A base station having a plurality of antennas and a plurality of analog circuits provided corresponding to the respective antennas,
Generating a first signal using a plurality of subcarriers having a frequency interval wider than the plurality of subcarriers used for the data signal;
Generating a second signal obtained by frequency-shifting the first signal by a predetermined frequency;
For each of the antennas, a signal that is input to the analog circuit corresponding to the antenna and that is output from the antenna using the first signal and the second signal that pass through the analog circuit. A calibration method characterized by executing a process for adjusting a frequency response. In a calibration apparatus provided in a base station having a plurality of antennas and a plurality of analog circuits provided corresponding to the respective antennas,
A first generator that generates a first signal using a plurality of subcarriers having a frequency interval wider than that of the plurality of subcarriers used for the data signal;
A second generator for generating a second signal obtained by shifting the first signal by a predetermined frequency;
For each of the antennas, a signal that is input to the analog circuit corresponding to the antenna and that is output from the antenna using the first signal and the second signal that pass through the analog circuit. A calibration device comprising: an adjustment unit that adjusts a frequency response.

Images