JP2018195955A - Wireless communication device and distortion compensation method - Google Patents

Abstract

To improve calibration accuracy among a plurality of antenna elements.SOLUTION: A wireless communication device comprises: a plurality of branches that comprise an antenna element and a power amplifier respectively; a first supplying unit that supplies a first distortion compensation coefficient compensating distortion which occurs when a transmission signal passes through the power amplifier provided in each branch; a second supplying unit that supplies a second distortion compensation coefficient compensating the distortion which occurs when a known signal being used in calibration among the plurality of branches passes through the power amplifier provided in each branch; a distortion compensation unit that performs distortion compensation of the transmission signal using the first distortion compensation coefficient supplied from the first supplying unit and that performs distortion compensation of the known signal using the second distortion compensation coefficient supplied from the second supplying unit; and a calibration unit that performs calibration among the plurality of branches using the known signal in which the distortion is compensated by the distortion compensation unit.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a wireless communication apparatus and a distortion compensation method.

  In recent years, in wireless communication, high-efficiency transmission by digitization is often adopted. For example, when a multi-level phase modulation method is applied for high-efficiency transmission, the transmission side linearizes the amplification characteristics of the power amplifier to suppress non-linear distortion that occurs in the transmission signal and reduce adjacent channel leakage power. There is.

  Further, when power efficiency is improved by using a power amplifier that is inferior in linearity, a technique that compensates for non-linear distortion generated in the power amplifier is used. An example of a technique for compensating for nonlinear distortion is a predistortion (PD) system. The PD method is a technique for reducing nonlinear distortion of a transmission signal amplified by the power amplifier by preliminarily giving a transmission signal an inverse characteristic of nonlinear distortion generated in the power amplifier. In particular, a PD method in which a reverse characteristic of nonlinear distortion is imparted to a transmission signal by digital signal processing is sometimes called DPD (Digital PreDistortion).

  The DPD refers to, for example, a look-up table (LUT) that stores a distortion compensation coefficient in association with the power of the transmission signal, and an LUT that multiplies the transmission signal by a distortion compensation coefficient corresponding to the power of the transmission signal. There is a method. Each distortion compensation coefficient stored in the LUT corresponds to the inverse characteristic of nonlinear distortion in the power amplifier.

  In general, the amplification characteristics of a power amplifier change depending on, for example, temperature and aging. In order to be able to follow this change, in the LUT type DPD, the distortion compensation coefficient stored in the LUT is adaptively controlled so that the error between the reference signal before the power amplifier input and the signal amplified by the power amplifier becomes small. The

  On the other hand, as another technique for high-efficiency transmission, for example, there is beam forming using an adaptive array antenna (AAA). Beam forming is a technique for controlling directivity by providing an array antenna in which a plurality of antenna elements are arranged and controlling the phase of a signal input to each antenna element. When beam forming is performed, calibration may be performed in order to align initial phases at the input ends of a plurality of antenna elements, or to correct a frequency deviation due to an analog circuit for each antenna element.

  In calibration, for example, a known signal is input to a branch for each antenna element, and the amplitude and phase of the known signal fed back from each branch are compared, thereby obtaining a correction coefficient for correcting the analog circuit characteristics for each branch. It is done. That is, the correction coefficient for aligning the characteristics of, for example, a power amplifier and a band pass filter (BPF) provided corresponding to each antenna element is calculated.

JP 2006-94043 A International Publication No. 2003/103163

  By the way, in the calibration described above, the known signal is amplified by the power amplifier and fed back after passing through the BPF. That is, since the known signal for calibration is also amplified by the power amplifier, it is affected by the nonlinearity of the power amplifier and the memory effect. For this reason, it is possible to improve the accuracy of calibration by compensating for distortion of a known signal for calibration using, for example, DPD.

  However, since the known signal for calibration has characteristics different from that of a normal transmission signal, there is a problem that even if the distortion compensation similar to that of the transmission signal is performed, the optimum distortion compensation for the known signal is not always performed. is there. Specifically, the known signal for calibration is characterized in that the frequency characteristics of the amplitude and phase in the frequency band of the transmission signal, for example, are constant so that phase fluctuation and amplitude fluctuation in an analog circuit can be easily obtained. Have Further, since the known signal becomes unnecessary radiation when radiated from the antenna element, the signal level is suppressed to be small.

  Because of these characteristics, if distortion compensation of a known signal for calibration is performed using a distortion compensation coefficient that is adaptively controlled using a transmission signal, the known signal may not be appropriately compensated for distortion. As a result, the accuracy of calibration is lowered, and the accuracy of beam forming may be lowered.

  The disclosed technology has been made in view of the above point, and an object thereof is to provide a wireless communication apparatus and a distortion compensation method that can improve the accuracy of calibration between a plurality of antenna elements.

  In one aspect, a wireless communication apparatus disclosed in the present application compensates for distortion generated when a plurality of branches each including an antenna element and a power amplifier and a transmission signal passes through a power amplifier provided in each branch. A first supply for supplying a first distortion compensation coefficient; and a second for compensating for distortion generated when a known signal used for calibration between the plurality of branches passes through a power amplifier provided in each branch. A second supply unit that supplies the first distortion compensation coefficient and a first distortion compensation coefficient that is supplied from the first supply unit, and performs a distortion compensation of the transmission signal using the first supply unit supplied from the first supply unit. A distortion compensation unit that performs distortion compensation of a known signal using a distortion compensation coefficient of 2, and calibration between the plurality of branches using a known signal that has been distortion compensated by the distortion compensation unit And a calibration unit that executes ® down.

  According to one aspect of the wireless communication apparatus and the distortion compensation method disclosed in the present application, there is an effect that the accuracy of calibration between a plurality of antenna elements can be improved.

1 is a block diagram showing a configuration of a base station apparatus according to Embodiment 1. FIG. FIG. 2 is a block diagram illustrating a configuration of the PD processing unit according to the first embodiment. FIG. 3 is a flowchart showing the calibration process according to the first embodiment. FIG. 4 is a block diagram illustrating a configuration of the PD processing unit according to the second embodiment. FIG. 5 is a flowchart showing the calibration process according to the second embodiment. FIG. 6 is a block diagram illustrating a configuration of the PD processing unit according to the third embodiment. FIG. 7 is a flowchart showing the calibration process according to the third embodiment.

  Hereinafter, embodiments of a wireless communication apparatus and a distortion compensation method disclosed in the present application will be described in detail with reference to the drawings. In addition, this invention is not limited by this embodiment.

(Embodiment 1)
FIG. 1 is a block diagram showing a configuration of base station apparatus 100 according to Embodiment 1. In FIG. 1 includes a processor 110, a memory 120, DA (Digital Analog) converters 130 and 131, power amplifiers 140 and 141, bandpass filters (BPF) 150 and 151, an AD (Analog Digital) converter 160, 161, a switch 170, and an AD converter 180. The DA converters 130 and 131, the power amplifiers 140 and 141, the BPFs 150 and 151, and the AD converters 160 and 161 are provided for each branch corresponding to the antenna element. In FIG. 1, two branches corresponding to two antenna elements are illustrated, but the base station apparatus 100 may include three or more branches corresponding to three or more antenna elements.

  The processor 110 includes, for example, a central processing unit (CPU), a field programmable gate array (FPGA), a digital signal processor (DSP), and the like, and performs overall control of the entire base station apparatus 100. Specifically, the processor 110 includes a transmission signal generation unit 210, a calibration signal (hereinafter abbreviated as “CAL signal”) generation unit 220, multipliers 230 and 231 and a predistortion processing unit (hereinafter “PD processing unit”). 240 and 241 and a calibration unit (hereinafter abbreviated as “CAL unit”) 250.

  The transmission signal generation unit 210 generates a transmission signal transmitted from the base station apparatus 100. That is, the transmission signal generation unit 210 encodes and modulates transmission data to generate a transmission signal. The transmission signal is transmitted from a plurality of antenna elements so that a beam having directivity in a specific direction is formed. For this reason, the transmission signal generation unit 210 generates a transmission signal for each branch corresponding to a plurality of antenna elements.

  The CAL signal generation unit 220 generates a CAL signal that is a known signal used for calibration. Specifically, the CAL signal generation unit 220 generates a CAL signal that has constant amplitude and phase frequency characteristics in the frequency band of the transmission signal and a smaller signal level than the transmission signal. Since the calibration is executed during a period in which the base station apparatus 100 does not transmit or receive a signal, the CAL signal generation unit 220 is a wireless communication system that employs, for example, a TDD (Time Division Duplex) method. Generates a CAL signal during a period of switching from an uplink subframe to a downlink subframe.

  Multipliers 230 and 231 are provided for each branch corresponding to a plurality of antenna elements, respectively, and multiply the transmission signal for each branch by a calibration coefficient. That is, the multipliers 230 and 231 are calibration coefficients for correcting the difference between the amplitude fluctuation and the phase fluctuation between the branches, and multiply the transmission signal for each branch by the calibration coefficient calculated by the CAL unit 250.

  The PD processing units 240 and 241 are provided for each branch, and execute distortion compensation for the transmission signal and the CAL signal for each branch. At this time, the PD processing units 240 and 241 read distortion compensation coefficients from different look-up tables (LUTs) for the transmission signal and the CAL signal, and multiply the transmission signal and the CAL signal by the distortion compensation coefficient. The specific configuration of the PD processing units 240 and 241 will be described in detail later.

  The CAL unit 250 uses the CAL signal fed back from each branch to calculate a calibration coefficient for correcting the difference between the amplitude fluctuation and the phase fluctuation between the branches. That is, the CAL unit 250 acquires a CAL signal fed back through the analog circuit of each branch, and corrects the difference in the characteristics of the analog circuit for each branch based on the amplitude and phase of the CAL signal. Is calculated.

  The memory 120 includes, for example, a RAM (Random Access Memory) or a ROM (Read Only Memory), and stores various types of information when processing is executed by the processor 110.

  The DA converters 130 and 131 are provided for each branch, and DA convert the distortion-compensated transmission signal and CAL signal.

  The power amplifiers 140 and 141 amplify the DA-converted transmission signal and the CAL signal. Although non-linear distortion occurs during amplification by the power amplifiers 140 and 141, the transmission signal and the CAL signal are multiplied by the distortion compensation coefficient in the PD processing units 240 and 241, and thus the non-linear distortion is canceled by the distortion compensation coefficient. .

  The BPFs 150 and 151 filter the transmission signal and the CAL signal after power amplification, and output the transmission signal and the CAL signal in a predetermined frequency band. The transmission signal and the CAL signal are radiated from the respective antenna elements and fed back to the switch 170.

  The AD converters 160 and 161 perform AD conversion on the transmission signal and the CAL signal amplified by the power amplifiers 140 and 141 and feed back to the PD processing units 240 and 241 of the processor 110. That is, the AD converters 160 and 161 feed back the output signals of the power amplifiers 140 and 141 in order to update the distortion compensation coefficients by the PD processing units 240 and 241.

  The switch 170 outputs to the AD converter 180 while switching the CAL signal output from the BPFs 150 and 151 of each branch during the calibration period in which calibration is executed. In other words, the switch 170 sequentially outputs the CAL signal that has passed through the analog circuit of each branch during the calibration period to the AD converter 180. Further, the switch 170 is turned off when it is not the calibration period, and the transmission signal output from the BPFs 150 and 151 is not output to the AD converter 180.

  The AD converter 180 AD converts the CAL signal output from the switch 170 and outputs the CAL signal to the CAL unit 250 of the processor 110. That is, the AD converter 180 outputs the CAL signal from each branch used for calibration to the CAL unit 250.

  FIG. 2 is a block diagram illustrating a configuration of the PD processing unit 240 according to the first embodiment. Since the PD processing unit 241 has the same configuration as the PD processing unit 240, only the PD processing unit 240 will be described here. 2 includes a multiplier 301, a power address generation unit 302, a delay unit 304, a subtractor 305, a switch 306, a transmission lookup table (hereinafter abbreviated as “transmission LUT”) 307, a delay. 308, a coefficient updating unit 309, a calibration lookup table (hereinafter abbreviated as “CAL LUT”) 310, a delay unit 311, a coefficient updating unit 312, and a switch 313.

  Multiplier 301 multiplies the transmission signal or CAL signal input to PD processing section 240 by the distortion compensation coefficient output from switch 313, and gives the inverse characteristic of the nonlinear distortion in power amplifier 140.

  The power address generation unit 302 calculates the power of the transmission signal or CAL signal input to the PD processing unit 240 and generates an address corresponding to the calculated power. That is, the power address generation unit 302 generates the address of the transmission LUT 307 based on the power of the transmission signal, and generates the address of the CAL LUT 310 based on the power of the CAL signal. Then, the power address generation unit 302 outputs an address based on the power of the transmission signal to the transmission LUT 307 and outputs an address Ap based on the power of the CAL signal to the CAL LUT 310.

  The delay unit 304 delays the transmission signal or CAL signal input to the PD processing unit 240 and outputs the delayed signal to the subtracter 305. Specifically, the delay unit 304 transmits the transmission signal and the CAL signal whose distortion has been compensated for only the time until the signal passes through the DA converter 130, the power amplifier 140, and the AD converter 160 and is fed back to the PD processing unit 240. Alternatively, the CAL signal is delayed. That is, the delay unit 304 delays the transmission signal and the CAL signal by a processing delay until the transmission signal and the CAL signal are fed back.

  The subtractor 305 calculates an error between the transmission signal or CAL signal delayed by the delay unit 304 and a feedback signal fed back by the AD converter 160 (hereinafter abbreviated as “FB signal”). That is, the subtractor 305 calculates an error between the transmission signal or CAL signal before being amplified by the power amplifier 140 and the transmission signal or CAL signal after being amplified by the power amplifier 140.

  The switch 306 outputs the error calculated by the subtracter 305 to the coefficient update unit 309 or the coefficient update unit 312. Specifically, the switch 306 outputs an error to the coefficient updating unit 309 during a period in which a signal is transmitted from the base station apparatus 100. On the other hand, the switch 306 outputs an error to the coefficient updating unit 312 during a calibration period in which the base station apparatus 100 does not transmit or receive a signal. That is, the switch 306 switches the output destination of the error depending on whether or not it is the calibration period, and outputs the error of the transmission signal to the coefficient updating unit 309, while outputting the error of the CAL signal to the coefficient updating unit 312. .

  The transmission LUT 307 stores a distortion compensation coefficient for compensating for distortion of the transmission signal in association with an address based on the power of the transmission signal. When the address is input from the power address generation unit 302, the transmission LUT 307 outputs a distortion compensation coefficient corresponding to the address.

  The delay unit 308 delays the distortion compensation coefficient output from the transmission LUT 307 and outputs it to the coefficient update unit 309. Specifically, the delay unit 308 is the time until the transmission signal compensated by the distortion compensation coefficient output from the transmission LUT 307 is fed back to the PD processing unit 240 and the error is calculated by the subtractor 305. Delay distortion compensation factor. That is, the delay unit 308 delays the distortion compensation coefficient by the processing delay until the transmission signal that has been distortion-compensated by the distortion compensation coefficient output from the transmission LUT 307 is fed back.

  The coefficient updating unit 309 updates the distortion compensation coefficient delayed by the delay unit 308 using the error of the transmission signal output from the switch 306, and writes the updated distortion compensation coefficient in the transmission LUT 307.

  The CAL LUT 310 stores a distortion compensation coefficient for compensating for distortion of the CAL signal in association with an address based on the power of the CAL signal. When the address Ap is input from the power address generator 302, the CAL LUT 310 outputs a distortion compensation coefficient corresponding to the address Ap.

  The delay device 311 delays the distortion compensation coefficient output from the CAL LUT 310 and outputs it to the coefficient update unit 312. Specifically, the delay unit 311 is the time until the error is calculated by the subtractor 305 after the CAL signal, which has been distortion-compensated by the distortion compensation coefficient output from the CAL LUT 310, is fed back to the PD processing unit 240. Delay distortion compensation factor. That is, the delay unit 311 delays the distortion compensation coefficient by a processing delay until the CAL signal whose distortion is compensated by the distortion compensation coefficient output from the CAL LUT 310 is fed back.

  The coefficient updating unit 312 updates the distortion compensation coefficient delayed by the delay device 311 using the error of the CAL signal output from the switch 306 and writes the updated distortion compensation coefficient in the CAL LUT 310.

  The switch 313 outputs the distortion compensation coefficient output from the transmission LUT 307 or the CAL LUT 310 to the multiplier 301. Specifically, the switch 313 outputs the distortion compensation coefficient output from the transmission LUT 307 to the multiplier 301 during a period in which a signal is transmitted from the base station apparatus 100. On the other hand, the switch 313 outputs the distortion compensation coefficient output from the CAL LUT 310 to the multiplier 301 during a calibration period when the base station apparatus 100 does not transmit or receive a signal. That is, the switch 313 switches the LUT depending on whether or not it is the calibration period, and outputs the distortion compensation coefficient output from any one of the LUTs to the multiplier 301.

  Next, calibration processing in the base station apparatus 100 configured as described above will be described with reference to the flowchart shown in FIG.

  The PD processing units 240 and 241 of the processor 110 monitor whether or not it is a calibration timing at which calibration is executed (step S101), and when it is not the calibration timing (No in step S101), the calibration is performed. Wait for the timing. Specifically, for example, when the TDD scheme is adopted in a wireless communication system, switching is performed from an uplink subframe in which the base station apparatus 100 receives a signal to a downlink subframe in which the base station apparatus 100 transmits a signal. The timing becomes the calibration timing.

  During a period that is not the calibration timing, the switch 306 switches to output the error output from the subtractor 305 to the coefficient update unit 309, and the switch 313 applies the distortion compensation coefficient output from the transmission LUT 307 to the multiplier 301. Switch to output to. Therefore, for example, in the downlink subframe, the transmission signals input to the PD processing units 240 and 241 are compensated for distortion by the distortion compensation coefficient output from the transmission LUT 307 and before and after amplification by the power amplifiers 140 and 141. The distortion compensation coefficient stored in the transmission LUT 307 is updated so that the transmission signal error is reduced.

  When the calibration timing arrives (step S101 Yes), the switch 306 switches to output the error output from the subtractor 305 to the coefficient updating unit 312 and the switch 313 performs distortion compensation output from the CAL LUT 310. It switches so that a coefficient may be output to the multiplier 301 (step S102). Further, the CAL signal generation unit 220 generates a CAL signal (step S103), and the CAL signal is input to the PD processing units 240 and 241 respectively. The CAL signal input to the PD processing units 240 and 241 may be the same signal.

  When the CAL signal is input to the PD processing units 240 and 241, the power address generation unit 302 calculates the power of the CAL signal and generates an address Ap corresponding to the power (step S104). The generated address Ap is input to the CAL LUT 310, and the distortion compensation coefficient stored in the address Ap is output from the CAL LUT 310. The distortion compensation coefficient is held and delayed by the delay unit 311, while being input to the multiplier 301 via the switch 313. Then, the multiplier 301 multiplies the CAL signal by a distortion compensation coefficient, thereby executing distortion compensation for the CAL signal (step S105).

  As described above, in the distortion compensation of the CAL signal, the distortion compensation coefficient output from the CAL LUT 310 is used. Therefore, a distortion compensation coefficient different from the distortion compensation of the transmission signal is used, and appropriate according to the characteristics of the CAL signal. Distortion compensation can be performed.

  Then, the distortion-compensated CAL signal passes through the DA converters 130 and 131, the power amplifiers 140 and 141, and the BPFs 150 and 151 and is input to the switch 170. The switch 170 sequentially switches the connection state between each branch and the AD converter 180 and sequentially inputs the CAL signal from each branch to the CAL unit 250 via the AD converter 180. Then, the CAL unit 250 calculates the characteristics of the analog circuit for each branch using the CAL signal for each branch, and executes calibration to align the characteristics of the analog circuit between the branches (step S106). That is, for example, the calibration coefficient indicating the characteristics of the analog circuit of another branch with respect to the analog circuit of one branch is calculated. The calibration coefficient calculated by the CAL unit 250 is output to the multipliers 230 and 231 corresponding to the respective branches, and the transmission signal is multiplied by the calibration coefficient in the transmission period after the calibration timing is completed. .

  In the present embodiment, since the calibration coefficient is calculated from the CAL signal that has been distortion compensated using the distortion compensation coefficient output from the CAL LUT 310, the calibration coefficient based on the CAL signal that has been appropriately distortion compensated is calculated. Calculated. As a result, the accuracy of the calibration coefficient calculated from the CAL signal is improved, and the accuracy of calibration between a plurality of antenna elements can be improved.

  In addition, the distortion-compensated CAL signal is fed back by the AD converters 160 and 161 from the output portions of the power amplifiers 140 and 141. The FB signal is input to the PD processing units 240 and 241 respectively, and the subtractor 305 calculates an error from the CAL signal before amplification. The calculated error is input to the coefficient updating unit 312 via the switch 306. On the other hand, the distortion compensation coefficient delayed by the delay unit 311 is also input to the coefficient updating unit 312, and the coefficient updating unit 312 updates the distortion compensation coefficient so as to reduce the error of the CAL signal before and after amplification. Then, the updated distortion compensation coefficient is written into the CAL LUT 310, whereby the CAL LUT 310 is updated (step S107). As a result, the CAL LUT 310 is updated independently of the transmission LUT 307, and the distortion compensation coefficient used for CAL signal distortion compensation can be appropriately updated.

  As described above, according to the present embodiment, the LUT is switched according to whether or not it is the calibration period, and the transmission signal or the CAL is used using the distortion compensation coefficient output from the corresponding LUT at the time of transmission and calibration. Compensate for distortion in the signal. Also, the LUT distortion compensation coefficients corresponding to the transmission and calibration are updated. For this reason, the CAL signal used for calibration can be compensated for distortion independently of the transmission signal, and the accuracy of distortion compensation of the CAL signal can be improved. As a result, the accuracy of calibration using the CAL signal can be improved.

(Embodiment 2)
The feature of the second embodiment is that the distortion compensation coefficient is read out not by the power of the CAL signal but by an address corresponding to the elapsed time from the head.

  Since the configuration of the base station apparatus according to Embodiment 2 is the same as that of base station apparatus 100 (FIG. 1) according to Embodiment 1, description thereof is omitted. In the second embodiment, the configuration of the PD processing units 240 and 241 is different from that of the first embodiment.

  FIG. 4 is a block diagram illustrating a configuration of the PD processing unit 240 according to the second embodiment. Since the PD processing unit 241 has the same configuration as the PD processing unit 240, only the PD processing unit 240 will be described here. In FIG. 4, the same parts as those in FIG. The PD processing unit 240 illustrated in FIG. 4 includes an address generation unit 401 instead of the power address generation unit 302 of FIG.

  The address generation unit 401 calculates the power of the transmission signal input to the PD processing unit 240 during the transmission period, and generates an address corresponding to the calculated power. In addition, the address generation unit 401 generates an address corresponding to the elapsed time from the beginning of the CAL signal input to the PD processing unit 240 during the calibration period. That is, the address generation unit 401 generates the address of the transmission LUT 307 based on the power of the transmission signal during the transmission period, and generates the address of the CAL LUT 310 based on the elapsed time of the CAL signal during the calibration period. Then, the address generation unit 401 outputs an address based on the power of the transmission signal to the transmission LUT 307 and outputs an address At based on the elapsed time of the CAL signal to the CAL LUT 310.

  Since the CAL signal is a known signal unlike the transmission signal, the signal sample corresponding to the elapsed time from the head is known. In addition, for example, due to the memory effect in the power amplifier 140, the nonlinear distortion generated in the power amplifier 140 may be different even for signal samples having the same signal power. For this reason, the address generation unit 401 generates an address based on the elapsed time from the head of the CAL signal, thereby causing the CAL LUT 310 to output a distortion compensation coefficient that can compensate for non-linear distortion fluctuations due to the memory effect. it can.

  Next, calibration processing in the base station apparatus 100 configured as described above will be described with reference to the flowchart shown in FIG. In FIG. 5, the same parts as those in FIG. 3 are denoted by the same reference numerals, and detailed description thereof is omitted.

  As in the first embodiment, until the calibration timing arrives, the address generation unit 401 generates an address based on the power of the transmission signal, and distortion correction of the transmission signal is performed by the distortion compensation coefficient output from the transmission LUT 307. Executed. Further, the distortion compensation coefficient stored in the transmission LUT 307 is updated so that the error of the transmission signal before and after amplification by the power amplifiers 140 and 141 is reduced.

  When the calibration timing arrives (step S101 Yes), the switch 306 switches to output the error output from the subtractor 305 to the coefficient updating unit 312 and the switch 313 performs distortion compensation output from the CAL LUT 310. It switches so that a coefficient may be output to the multiplier 301 (step S102). Further, the CAL signal generation unit 220 generates a CAL signal (step S103), and the CAL signal is input to the PD processing units 240 and 241 respectively. The CAL signal input to the PD processing units 240 and 241 may be the same signal.

  When the CAL signal is input to the PD processing units 240 and 241, the address generation unit 401 generates an address At corresponding to the elapsed time from the head of the CAL signal (step S201). That is, an address At corresponding to the order from the beginning of each signal sample that is known of the CAL signal is generated. The generated address At is input to the CAL LUT 310, and the distortion compensation coefficient stored in the address At is output from the CAL LUT 310. The distortion compensation coefficient is held and delayed by the delay unit 311, while being input to the multiplier 301 via the switch 313. Then, the multiplier 301 multiplies the CAL signal by a distortion compensation coefficient, thereby executing distortion compensation for the CAL signal (step S105).

  As described above, in the distortion compensation of the CAL signal, an address corresponding to the elapsed time from the head of the CAL signal is generated, and the distortion compensation coefficient output from the CAL LUT 310 is used. For this reason, a distortion compensation coefficient different from the distortion compensation of the transmission signal, which can compensate for the effect of the memory effect in the power amplifiers 140 and 141, is used, and appropriate distortion compensation according to the characteristics of the CAL signal is used. Can be executed.

  Then, the distortion-compensated CAL signal is input from the switch 170 to the CAL unit 250 via the AD converter 180, and the CAL unit 250 performs calibration for aligning the analog circuit characteristics between the branches (step S106). ). The calibration coefficient calculated by the CAL unit 250 is output to the multipliers 230 and 231 corresponding to the respective branches, and the transmission signal is multiplied by the calibration coefficient in the transmission period after the calibration timing is completed. .

  In the present embodiment, since the calibration coefficient is calculated by the CAL signal subjected to distortion compensation using the calibration distortion compensation coefficient that can also compensate for the influence of the memory effect, the CAL signal appropriately compensated for distortion is obtained. Based on the calibration coefficient is calculated. As a result, the accuracy of the calibration coefficient calculated from the CAL signal is improved, and the accuracy of calibration between a plurality of antenna elements can be improved.

  In addition, the distortion-compensated CAL signal is fed back by the AD converters 160 and 161 from the output portions of the power amplifiers 140 and 141. By using the FB signal, the coefficient updating unit 312 updates the distortion compensation coefficient. Then, the updated distortion compensation coefficient is written into the CAL LUT 310, whereby the CAL LUT 310 is updated (step S107). As a result, the CAL LUT 310 is updated independently of the transmission LUT 307, and the distortion compensation coefficient used for CAL signal distortion compensation can be appropriately updated.

  As described above, according to the present embodiment, the address based on the power of the transmission signal is generated in the transmission period, while the address based on the elapsed time from the head of the CAL signal is generated in the calibration period. The transmission signal or the CAL signal is subjected to distortion compensation using the distortion compensation coefficient output from the LUT. Therefore, a known CAL signal can be subjected to distortion compensation using a distortion compensation coefficient that takes into account the effect of the memory effect in the power amplifier, and the accuracy of distortion compensation of the CAL signal can be improved. As a result, the accuracy of calibration using the CAL signal can be improved.

(Embodiment 3)
A feature of the third embodiment is that an offset indicating a difference between distortion compensation coefficients of a transmission signal and a CAL signal is calculated, and the offset is added to the distortion compensation coefficient for the transmission signal at the time of distortion compensation of the CAL signal. The distortion compensation coefficient obtained is used.

  Since the CAL signal is a signal having constant frequency characteristics of amplitude and phase within the frequency band of the transmission signal, a transmission signal having non-constant frequency characteristics of amplitude and phase is a signal that passes through the power amplifiers 140 and 141. AM / PM characteristics indicating the relationship between the input power and the output phase are different. That is, there is a difference between the output phase when the transmission signal is input to the power amplifiers 140 and 141 and the output phase when the CAL signal is input to the power amplifiers 140 and 141. It is the difference of the nonlinear distortion with respect to the signal. Therefore, by calculating this difference as an offset, if the offset is added to the distortion compensation coefficient of the transmission signal, the distortion compensation coefficient of the CAL signal can be obtained.

  Since the configuration of the base station apparatus according to Embodiment 3 is the same as that of base station apparatus 100 (FIG. 1) according to Embodiment 1, description thereof is omitted. In the third embodiment, the configuration of the PD processing units 240 and 241 is different from that of the first embodiment.

  FIG. 6 is a block diagram illustrating a configuration of the PD processing unit 240 according to the third embodiment. Since the PD processing unit 241 has the same configuration as the PD processing unit 240, only the PD processing unit 240 will be described here. In FIG. 6, the same parts as those in FIG. The PD processing unit 240 illustrated in FIG. 6 includes a switch 501, an offset calculation unit 502, a switch 503, and an adder 504 instead of the switch 306, the CAL LUT 310, the delay unit 311, the coefficient update unit 312, and the switch 313 illustrated in FIG. Have.

  The switch 501 outputs the FB signal amplified and fed back by the power amplifier 140 to the subtractor 305 or the offset calculation unit 502. Specifically, the switch 501 outputs the FB signal to the subtractor 305 during a period in which the signal is transmitted from the base station apparatus 100. On the other hand, the switch 501 outputs an FB signal to the offset calculation unit 502 during a calibration period when the base station apparatus 100 does not transmit or receive a signal. That is, the switch 501 switches the output destination of the FB signal according to whether or not it is the calibration period, and outputs the FB signal that is feedback of the transmission signal to the subtractor 305, while the FB signal that is feedback of the CAL signal. Is output to the offset calculation unit 502.

  The offset calculation unit 502 calculates an offset by calculating an average value of the FB signal output from the switch 501 during a period in which the switch 503 is off. In other words, the offset calculation unit 502 calculates the average value of the phase of the CAL signal whose distortion is compensated by the distortion compensation coefficient for the transmission signal, and thereby outputs the output phase of the transmission signal from the power amplifier 140 and the output phase of the CAL signal. The offset of is calculated.

  Here, if the transmission signal is distortion-compensated by the distortion compensation coefficient for the transmission signal, the nonlinear distortion in the power amplifier 140 is compensated, and the average value of the output phase when this transmission signal is output from the power amplifier 140 is 0. On the other hand, if the CAL signal is distortion-compensated by the distortion compensation coefficient for the transmission signal, the nonlinear distortion in the power amplifier 140 is not completely compensated, and this CAL signal is output from the power amplifier 140. In this case, the average value of the output phase corresponds to the difference between the distortion compensation coefficients of the transmission signal and the CAL signal. For this reason, the offset calculation unit 502 calculates an offset by calculating an average value of the FB signal.

  The switch 503 is turned off in a predetermined period after the calibration period starts, and turned on after a predetermined period from the start of the calibration period. That is, the switch 503 prevents the offset calculated by the offset calculation unit 502 from being output to the adder 504 immediately after the calibration period starts, and outputs the offset to the adder 504 after a predetermined time has elapsed. To do. Therefore, during a period in which the switch 503 is off, the offset calculation unit 502 calculates an offset from the CAL signal that has been distortion compensated using the distortion compensation coefficient for the transmission signal. When the switch 503 is turned on, the offset calculation unit 502 stops the offset calculation process and supplies the calculated offset to the adder 504.

  When the switch 503 is turned on and an offset is supplied, the adder 504 adds the offset to the distortion compensation coefficient output from the transmission LUT 307. That is, the adder 504 generates a distortion compensation coefficient for the CAL signal by adding an offset to the distortion compensation coefficient for the transmission signal, and outputs the distortion compensation coefficient to the multiplier 301.

  Next, calibration processing in the base station apparatus 100 configured as described above will be described with reference to the flowchart shown in FIG. In FIG. 7, the same parts as those in FIG. 3 are denoted by the same reference numerals, and detailed description thereof is omitted.

  As in the first embodiment, until the calibration timing arrives, an address based on the power of the transmission signal is generated by the power address generation unit 302, and the distortion compensation of the transmission signal is performed by the distortion compensation coefficient output from the transmission LUT 307. Is executed. Further, the distortion compensation coefficient stored in the transmission LUT 307 is updated so that the error of the transmission signal before and after amplification by the power amplifiers 140 and 141 is reduced.

  When the calibration timing comes (step S101 Yes), the switch 501 switches to output the FB signal to the offset calculation unit 502 (step S301), and the switch 503 remains off. Further, the CAL signal generation unit 220 generates a CAL signal (step S103), and the CAL signal is input to the PD processing units 240 and 241 respectively. The CAL signal input to the PD processing units 240 and 241 may be the same signal.

  When the CAL signal is input to the PD processing units 240 and 241, the power address generation unit 302 generates an address corresponding to the power of the CAL signal (step S104), and the generated address is input to the transmission LUT 307. The distortion compensation coefficient for the transmission signal is output from the transmission LUT 307. The distortion compensation coefficient is input to the multiplier 301 via the adder 504. Here, since the switch 503 is off, the adder 504 does not add an offset to the distortion compensation coefficient. The multiplier 301 multiplies the CAL signal by the distortion compensation coefficient for the transmission signal, thereby executing distortion compensation for the CAL signal (step S302).

  Since the distortion compensation here is for compensating the distortion of the CAL signal using the distortion compensation coefficient for the transmission signal, the nonlinear distortion generated in the CAL signal in the power amplifiers 140 and 141 is not sufficiently compensated. Therefore, a CAL signal in which nonlinear distortion remains is fed back to the PD processing units 240 and 241 as an FB signal and input to the offset calculating unit 502 via the switch 501. Then, the offset calculation unit 502 calculates an average value of the FB signals in which the nonlinear distortion remains, thereby calculating an offset (step S303). As described above, the offset between the distortion compensation coefficient for the transmission signal and the distortion compensation coefficient for the CAL signal is calculated from the FB signal fed back while the switch 503 is on.

  When a predetermined time elapses after the calibration timing starts, the switch 503 is turned on, and the offset calculation unit 502 and the adder 504 are connected (step S304). Thereafter, the offset calculation process by the offset calculation unit 502 is stopped, and the already calculated offset is supplied to the adder 504. The adder 504 adds an offset to the distortion compensation coefficient output from the transmission LUT 307 to generate a distortion compensation coefficient for the CAL signal. The distortion compensation coefficient of the CAL signal is executed by multiplying the distortion compensation coefficient by the multiplier 301 to the CAL signal (step S305).

  As described above, after the switch 503 is turned on, the offset is added to the distortion compensation coefficient for the transmission signal to generate the distortion compensation coefficient for the CAL signal. Therefore, sufficient distortion compensation is performed on the CAL signal. Can be applied. Further, if there is one transmission LUT 307, distortion compensation for each of the transmission signal and the CAL signal can be performed independently, so that the circuit scale can be reduced.

  Then, the distortion-compensated CAL signal is input from the switch 170 to the CAL unit 250 via the AD converter 180, and the CAL unit 250 performs calibration for aligning the analog circuit characteristics between the branches (step S106). ). The calibration coefficient calculated by the CAL unit 250 is output to the multipliers 230 and 231 corresponding to the respective branches, and the transmission signal is multiplied by the calibration coefficient in the transmission period after the calibration timing is completed. .

  As described above, according to the present embodiment, the offset between the distortion compensation coefficient for the transmission signal and the distortion compensation coefficient for the CAL signal is calculated and obtained by adding the offset to the distortion compensation coefficient for the transmission signal. CAL signal distortion compensation is performed using the distortion compensation coefficient. For this reason, the distortion compensation of the transmission signal and the CAL signal can be performed using the distortion compensation coefficient stored in one LUT, and the circuit scale can be reduced.

110 processor 120 memory 130, 131 DA converter 140, 141 power amplifier 150, 151 BPF
160, 161, 180 AD converter 170 Switch 210 Transmission signal generation unit 220 CAL signal generation unit 230, 231, 301 Multiplier 240, 241 PD processing unit 250 CAL unit 302 Power address generation unit 304, 308, 311 Delay unit 305 Subtractor 306, 313, 501, 503 Switch 307 Transmission LUT
309, 312 Coefficient update unit 310 CAL LUT
401 Address generation unit 502 Offset calculation unit 504 Adder

Claims (8)

A plurality of branches each comprising an antenna element and a power amplifier;
A first supply unit that supplies a first distortion compensation coefficient that compensates for distortion generated when the transmission signal passes through a power amplifier provided in each branch;
A second supply unit that supplies a second distortion compensation coefficient that compensates for distortion generated when a known signal used for calibration between the plurality of branches passes through a power amplifier provided in each branch;
Distortion compensation of the transmission signal is performed using the first distortion compensation coefficient supplied from the first supply unit, and distortion compensation of the known signal is performed using the second distortion compensation coefficient supplied from the second supply unit. A distortion compensator for performing
A radio communication apparatus comprising: a calibration unit that performs calibration between the plurality of branches using a known signal that has been subjected to distortion compensation by the distortion compensation unit. The distortion compensator is
A multiplier that multiplies the input signal by a distortion compensation coefficient;
During a period when the input signal to the multiplier is a transmission signal, the first distortion compensation coefficient supplied from the first supply unit is input to the multiplier, and the input signal to the multiplier is a known signal. The wireless communication apparatus according to claim 1, wherein the period includes a switch that inputs the second distortion compensation coefficient supplied from the second supply unit to the multiplier. The first supply unit includes:
A first table storing a first distortion compensation coefficient in each of a plurality of addresses generated from the transmission signal;
The second supply unit includes:
The wireless communication apparatus according to claim 1, further comprising: a second table that stores a second distortion compensation coefficient at each of a plurality of addresses generated from the known signal. The first table is:
Storing a first distortion compensation coefficient at each of a plurality of addresses based on the power of the transmission signal;
The second table is:
The wireless communication apparatus according to claim 3, wherein the second distortion compensation coefficient is stored in each of a plurality of addresses based on the power of the known signal. The first table is:
Storing a first distortion compensation coefficient at each of a plurality of addresses based on the power of the transmission signal;
The second table is:
The wireless communication apparatus according to claim 3, wherein a second distortion compensation coefficient is stored in each of a plurality of addresses based on an elapsed time from the head of the known signal. The second supply unit includes:
An adder that adds an offset to the first distortion compensation coefficient supplied by the first supply unit and calculates a second distortion compensation coefficient;
The distortion compensator is
The period during which the transmission signal is input is multiplied by the first distortion compensation coefficient supplied from the first supply unit to the transmission signal, and the period during which the known signal is input is calculated by the adder for the known signal. The wireless communication apparatus according to claim 1, further comprising a multiplier that multiplies the second distortion compensation coefficient. The multiplier is
Part of the period in which the known signal is input is multiplied by the first distortion compensation coefficient supplied from the first supply unit to the known signal, and the remaining period in the period in which the known signal is input. , Multiplying the known signal by the second distortion compensation coefficient calculated by the adder,
The second supply unit includes:
The wireless communication apparatus according to claim 6, further comprising a calculation unit that calculates an offset based on a known signal multiplied by a first distortion compensation coefficient during the partial period. A distortion compensation method executed by a wireless communication apparatus including a plurality of branches each including an antenna element and a power amplifier,
A period during which a transmission signal is transmitted from the plurality of branches is supplied with a first distortion compensation coefficient that compensates for distortion that occurs when the transmission signal passes through a power amplifier provided in each branch.
Performing distortion compensation of the transmission signal using the supplied first distortion compensation coefficient;
During a period in which calibration between the plurality of branches is performed, a second distortion compensation coefficient that compensates for distortion generated when a known signal used for calibration passes through a power amplifier provided in each branch is supplied. And
Performing distortion compensation of the known signal using the supplied second distortion compensation coefficient;
A distortion compensation method, comprising: a process of executing calibration between the plurality of branches using a distortion-compensated known signal.

Images