JP6190635B2 - Predistorter and control method of predistorter - Google Patents

Description

  The present invention relates to a predistorter that compensates for distortion components generated in a power amplifier and a predistorter control method.

One method for compensating for nonlinear distortion components (hereinafter also referred to as distortion components) generated in a power amplifier is a method using a predistorter. The predistorter generates a distortion compensation signal according to a nonlinear characteristic model in the power amplifier, and adds the distortion compensation signal to the signal input to the power amplifier. For example, a power series predistorter (hereinafter simply referred to as a predistorter) according to Non-Patent Document 1 and Non-Patent Document 2 uses a nonlinear characteristic model (hereinafter referred to as a power series model) based on a power series, and has a frequency. A distortion compensation signal capable of compensating for the dependent distortion component is generated. According to Non-Patent Document 1 and Non-Patent Document 2, the input signal of the predistorter is x, and the model order of the power series model (however, the model order is the order of the power series model expressed by a polynomial, that is, the polynomial The predistorter output signal y can be expressed as in Equation (1), where 3 is the highest order among the orders of the included terms.

Here, the second term on the right side of Equation (1) is the distortion compensation signal, | x | ² x represents the third-order distortion signal generated by the third-order distortion generator of the predistorter, and a ₁ is the linear gain. A ₃ represents a complex coefficient given to the third-order distortion vector adjuster for adjusting the amplitude and phase of the third-order distortion signal, and h ₃ represents a frequency characteristic compensator that gives a frequency characteristic to the output of the third-order distortion vector adjuster. The symbol || represents an absolute value, and the symbol * (asterisk) represents a convolution operation.

This predistorter generates a distortion compensation signal by digital signal processing. At this time, a ₃ and h ₃ are appropriately adjusted so that distortion components having frequency dependency generated in the power amplifier are reduced.

S. Mizuta, Y. Suzuki, T. Hirota, and Y. Yamao, “Digital predistortion linearizer for compensating frequency-dependent IM distortion,” in Proc. 34th European Microwave Conference, pp. 1053-1056, Oct. 2004. S. Mizuta, Y. Suzuki, S. Narahashi, and Y. Yamao, “A New Adjustment Method for the Frequency-Dependent IMD Compensator of the Digital Predistortion Linearizer,” IEEE Radio and Wireless Symposium 2006, pp. 255-258, Jan . 2006.

  In LTE-Advanced, there is a technology called carrier aggregation that uses a plurality of frequency bands (hereinafter also referred to as component carriers) simultaneously. Carrier aggregation is roughly classified into a case where a plurality of adjacent frequency bands are used and a case where discrete frequency bands are used.

  When using carrier aggregation, depending on the nonlinear characteristics of the power amplifier, the frequency band to be used, and the like, a predistorter adapted to a single frequency band may not sufficiently reduce the distortion component. Hereinafter, the case where two frequency bands are used will be described in detail.

Assuming that a signal to be transmitted using the first frequency band is s ₁ , a signal to be transmitted using the second frequency band is s ₂ , and an input signal of the predistorter is x = s ₁ + s ₂ , The output signal y of the tota can be expressed as in equation (2). Here, the symbol ^* (superscript asterisk) represents a complex conjugate.

In Equation (2), if the third-order distortion signal generated by the third-order distortion generator of the predistorter is d _C (t), d _C (t) can be expressed as Equation (3).

Here, 2 | s ₂ | ² s ₁ is an intermodulation distortion signal of s ₁ and s ₂ . This distortion signal is generated in the same frequency band as | s ₁ | ² s ₁ . 2 | s ₁ | ² s ₂ is also an intermodulation distortion signal of s ₁ and s ₂ . This distortion signal is generated in the same frequency band as | s ₂ | ² s ₂ .

  It is known that when the power amplifier is operated with high efficiency, the nonlinear characteristic of the power amplifier becomes complicated. At this time, if the phase and amplitude differ for each distortion component in multiple distortion components generated in the same frequency band, the predistorter adapted for the single frequency band cannot adjust the phase and amplitude for each distortion signal. Therefore, the distortion component generated in the power amplifier remains.

From the above, when using carrier aggregation, the predistorter adapted to a single frequency band cannot sufficiently reduce the intermodulation distortion components of the signals s ₁ and s ₂ . In order to solve this problem, a predistorter 900 having the following configuration can be considered as a predistorter adapted to a plurality of frequency bands (reference patent document: Japanese Patent Application No. 2011-269882).

FIG. 1 shows the configuration of the predistorter 900 and its peripheral devices. The peripheral devices of this example are two signal generators A and B, an amplifier 950, and a feedback signal generator 960. In this example, the signal generated by the signal generator A is a signal s ₁ transmitted using the first frequency band, and an I-phase signal and a Q-phase signal (hereinafter also referred to as I / Q signal). A digital signal consisting of The signal generated by the signal generator B is a signal s ₂ transmitted using a second frequency band different from the first frequency band, and an I-phase signal and a Q-phase signal (hereinafter referred to as I / Q signal). Digital signal.

  The predistorter 900 receives the signals from the linear transmission path 901A including a delay unit, the linear transmission path 901B including a delay unit, the distortion compensation signal generation path 909 for generating a distortion compensation signal, and the signal from the signal generator A. A distributor 902A for distributing the signal to the linear transmission path 901A and the distortion compensation signal generation path 909, a distributor 902B for distributing the signal from the signal generator B to the linear transmission path 901B and the distortion compensation signal generation path 909, and linear transmission A signal adder 903A that combines the output of the path 901A and the output of the distortion compensation signal generation path 909, a signal adder 903B that combines the output of the linear transfer path 901B and the output of the distortion compensation signal generation path 909, and a signal adder 903A Digital-to-analog converter (DA) that converts the output (digital I / Q signal with distortion compensation component added) into an analog I / Q signal ) 904A, a digital-analog converter (DAC) 904B that converts the output of the signal adder 903B (digital I / Q signal to which a distortion compensation component is added) into an analog I / Q signal, and one of the outputs of the amplifying device 950 An analog-to-digital converter (ADC) 905 that converts the output (analog I / Q signal) of the feedback signal generator 960 that captures the signal as a feedback signal into a digital I / Q signal, and distortion observation that measures a distortion component from the output of the ADC 905 906 and a controller 907 for determining an adjustment amount such as a vector coefficient (amplitude and phase) and a frequency characteristic compensator coefficient (amplitude and phase) used in the distortion compensation signal generation path 909 based on the output of the distortion observer 906. including. Note that the delay devices included in the linear transmission paths 901A and 901B delay the input signal to the linear transmission path by the delay time generated in the distortion compensation signal generation path 909.

  The amplifying device 950 includes a quadrature modulator 951A that quadrature modulates the analog I / Q signal that is the output of the DAC 904A, a quadrature modulator 951B that quadrature modulates the analog I / Q signal that is the output of the DAC 904B, and a quadrature modulator 951A. Upconverter 952A for converting the frequency of the modulation output into the carrier frequency, the upconverter 952B for converting the frequency of the modulation output from the quadrature modulator 951B into a carrier frequency different from the previous carrier frequency, and the output of the upconverter 952A A power combiner 953 that combines the signal and the output signal of up-converter 952B, and a power amplifier 954 that amplifies the output signal of power combiner 953 are included. The power-amplified high frequency signal is supplied from the output terminal 970 to the antenna via a duplexer (not shown), for example.

  The feedback signal generation device 960 includes a directional coupler 961 that extracts a part of the output of the amplification device 950 as a feedback signal, a frequency down converter 962 that converts the frequency of the feedback signal, and an orthogonal signal that orthogonally demodulates the down-converted feedback signal. A demodulator 963 is included.

FIG. 2 shows a functional block diagram of the distortion compensation signal generation path 909.
The distortion compensation signal generation path 909, the third-order distortion signal | s ₁ | adjustment of vector coefficients of ² s ₁ (amplitude and phase) | a ² s ₁ generated third-order distortion generator 9092A1 and third-order distortion signal | s ₁ A distortion signal generator 9092A including a third-order distortion vector adjuster 9092A2, a third-order distortion generator 9092B1 that generates a third-order distortion signal | s ₂ | ² s _2, and a third-order distortion signal | s ₂ | ² s _2. A distortion signal generation unit 9092B including a third-order distortion vector adjuster 9092B2 that adjusts vector coefficients (amplitude and phase), and intermodulation distortion signals 2 | s ₂ | ² s ₁ and 2 | s ₁ | ² s ₂ The generated intermodulation distortion generator 90950, the submodulation distortion vector adjuster 90951 for adjusting the vector coefficients (amplitude and phase) of the intermodulation distortion signal 2 | s ₂ | ² s ₁ , and the intermodulation distortion signal 2 | s ₁ | Secondary third order distortion that adjusts the vector coefficient (amplitude and phase) of ² s ₂ The sub-distortion signal generation unit 9095 including the vector adjuster 90952, the signal distribution unit 9091A that distributes the signal from the distributor 902A to the distortion signal generation unit 9092A and the sub-distortion signal generation unit 9095, and the signal from the distributor 902B. Distortion signal generation unit 9092B and sub-distortion signal generation unit 9095 respectively distribute signal distribution unit 9091B, signal addition unit 9093A that combines the output of distortion signal generation unit 9092A and the output of sub-distortion signal generation unit 9095, distortion signal generation A signal adder 9093B that combines the output of the unit 9092B and the output of the sub-distortion signal generator 9095. The output of the signal adder 9093A is input to the signal adder 903A, and the output of the signal adder 9093B is input to the signal adder 903B. Vector coefficient values adjusted by the third-order distortion vector adjuster 9092A2, the third-order distortion vector adjuster 9092B2, the sub-third-order distortion vector adjuster 90951, and the sub-third-order distortion vector adjuster 90952 are based on instructions from the controller 907. .

  The operations of the strain observer 906 and the controller 907 will be described with reference to FIG. FIG. 3 schematically shows an output spectrum of the amplification device 950. The signal band Ab corresponds to the output signal of the signal generator A, and the signal band Bb corresponds to the output signal of the signal generator B. This example shows a case of carrier aggregation using a discrete frequency band.

  The distortion observer 906 measures the power in each band in the third-order distortion component lower bands Ab1, Bb1, and the third-order distortion component upper bands Ab2, Bb2.

  The controller 907 includes a third-order distortion vector adjuster 9092A2, a third-order distortion vector adjuster 9092B2, a sub-third-order distortion included in the distortion compensation signal generation path 909 so as to reduce the power of the distortion component measured by the distortion observer 906. A vector coefficient adjustment amount to be given to each of the vector adjuster 90951 and the sub-third-order distortion vector adjuster 90952 is determined. At this time, the distortion components generated in the third-order distortion component lower band Ab1 and the third-order distortion component upper band Ab2 are compensated by adjusting the vector coefficients given to the third-order distortion vector adjuster 9092A2 and the sub-third-order distortion vector adjuster 90951. it can. The distortion components generated in the third-order distortion component lower band Bb1 and the third-order distortion component upper band Bb2 can be compensated by adjusting the vector coefficients given to the third-order distortion vector adjuster 9092B2 and the sub-third-order distortion vector adjuster 90952.

When the predistorter based on the reference patent document is used, there is a concern that a certain amount of time is required until the distortion component generated in the power amplifier is compensated. The reason for this will be described below.
In the predistorter based on the reference patent document, in order to compensate for distortion components, for example, a third-order distortion vector adjuster 9092A2, a sub-third-order distortion vector adjuster 90951, a third-order distortion vector adjuster 9092B2, a sub-third-order distortion vector The adjustment amount of each vector coefficient is determined in the order of the adjuster 90952. At this time, assuming that the time required for determining the vector coefficient to be applied to each vector adjuster (hereinafter also referred to as adjustment time) is the same and the adjustment time is t ₁ second, the vector coefficients of all the vector adjusters are determined. It takes 4t ₁ seconds to complete. When the number of component carriers used in carrier aggregation increases, the number of third-order distortion vector adjusters and the number of sub-third-order distortion vector adjusters may increase, so that the adjustment time further increases compared to this example. . That is, the adjustment time increases with the number of vector adjusters. This increase in the adjustment time increases the time until the distortion component is compensated.

  In the above-described example, it is assumed that a signal using two different frequency bands cannot be generated by one signal generator. The above concerns are not expected only in such cases. Even when one signal generator generates a signal that uses a plurality of different frequency bands at the same time, it performs each process of generating and adjusting a distortion signal for each band after demultiplexing the signal, Concerns similar to those in the above example are expected.

  From this point of view, the present invention provides a predistorter capable of compensating a distortion component generated in a power amplifier in as short a time as possible even when using carrier aggregation, and a control method therefor. Objective.

  A predistorter according to the present invention is a predistorter that preliminarily adds a distortion compensation signal that cancels a distortion component generated in a power amplifier to an input signal including a plurality of component carriers. (2) Self-distortion signal obtained using only component carrier signals for each component carrier, and intermodulation distortion obtained using two or more different component carrier signals Distortion signal generating means for generating at least a part of the signal (hereinafter, the self-distortion signal and the intermodulation distortion signal are collectively referred to as a distortion signal), and corresponding to each of the distortion signals. Distortion compensation signal generation that outputs a distortion compensation signal, including a vector adjuster that adjusts the amplitude and phase of the distortion signal A path, (3) a signal adder that combines the output of the linear transmission path and the output of the distortion compensation signal generation path, and (4) a power measuring instrument that measures the power of each distortion signal using the signal of the component carrier. (5) a distortion observer that observes distortion components included in the output of the power amplifier that amplifies the power with respect to the output of the signal adder; and (6) the amount of adjustment of the amplitude and / or phase of each vector adjuster. And a controller to set. The controller selects at least some vector adjusters from among the plurality of vector adjusters using at least the measurement result of the power measuring instrument, and observes distortion of the amount of adjustment of the amplitude and / or phase to be given to the selected vector adjuster. Further, the determined adjustment amount or the adjustment amount based on the adjustment amount is set as the adjustment amount of the amplitude and / or phase of the vector adjuster that has not been selected.

  Alternatively, the predistorter according to the present invention is a predistorter that preliminarily adds a distortion compensation signal that cancels a distortion component generated in a power amplifier to an input signal including a plurality of component carriers. (2) a self-distortion signal obtained by using only the component carrier signal for each component carrier, and a mutual transmission obtained by using two or more different component carrier signals. Distortion signal generation means for generating at least a part of the modulation distortion signal (hereinafter, the self-distortion signal and the intermodulation distortion signal are collectively referred to as a distortion signal), and corresponding to each of the distortion signals. And a vector adjuster for adjusting the amplitude and phase of the corresponding distortion signal, and outputting distortion compensation signals. A signal generation path; (3) a signal adder that combines the output of the linear transfer path and the output of the distortion compensation signal generation path; and (4) an output of a power amplifier that performs power amplification on the output of the signal adder. A distortion observer for observing the included distortion component; and (5) a controller for setting an adjustment amount of the amplitude and / or phase of each vector adjuster. The controller selects a vector adjuster that compensates for a distortion component generated in a predetermined frequency band from among the distortion components included in the output of the power amplifier, and provides an amplitude and / or an amplitude to be supplied to the selected vector adjuster. The amount of phase adjustment is determined based on the observation result of the strain observer, the amplitude and / or phase adjustment amount of the vector adjuster that was not selected is estimated based on the determined amount of adjustment, and the estimated amount of adjustment Is set as the amount of adjustment of the amplitude and / or phase of the vector adjuster not selected.

  According to the present invention, some vector adjusters are selected from among a plurality of vector adjusters, amplitude and phase adjustment amounts to be given to the selected vector adjusters are determined, and further, the determined adjustment amounts are selected. Since it is set as the adjustment amount of the amplitude and phase of the vector adjuster that did not exist, the adjustment time can be shortened compared to the case where the adjustment amount of amplitude and phase of all vector adjusters is determined individually. The distortion component generated in the amplifier can be compensated for in a short time.

The block diagram which shows the structural example of the series type predistorter which should be adapted to a some frequency band. The block diagram which shows the structural example of a distortion compensation signal generation path | route. The schematic diagram of the output spectrum of an amplifier. The block diagram which shows the structural example of the predistorter of 1st Embodiment. The processing flowchart of 1st Embodiment. The processing flowchart of the modification 2 of 1st Embodiment. The block diagram which shows the structural example of the predistorter of 2nd Embodiment. The processing flowchart of 2nd Embodiment. The block diagram which shows the structural example of the predistorter of 3rd Embodiment. The block diagram which shows the structural example of the distortion compensation signal generation path | route contained in the predistorter of 3rd Embodiment. The processing flowchart of 3rd Embodiment. The schematic diagram of the output spectrum of an amplifier. The schematic diagram of the output spectrum of an amplifier.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In each embodiment, constituent elements having the same functions as constituent elements included in the predistorter 900 shown in FIG. 1 are assigned the reference numerals of the constituent elements in the predistorter 900, and redundant description is omitted. Therefore, refer to the description about the predistorter 900 shown in FIG.

<Summary of the invention>
According to the power series model, various distortion signals can be assumed according to the number of component carriers and the model order of the power series model. Such a distortion signal is a self-distortion signal obtained by using only the component carrier signal for each component carrier (s _{k is} the signal of the k-th component carrier, and m is the order not exceeding the model order of the power series model. , | S _k | ^m s _k ) and an intermodulation distortion signal obtained by using signals of two or more component carriers different from each other (an example of a power series model whose model order is 3) , | S ₁ | ² s ₂ and s ₂ ² s ₁ ^* ).

  From the viewpoint of compensating for distortion components generated in power amplifiers operating near saturation power, the predistorter adapted for carrier aggregation preferably generates distortion signals that overlap at least the frequency band of the output signal of the linear transfer path. Let That is, a predistorter adapted for carrier aggregation generates at least all self-distortion signals and intermodulation distortion signals that overlap with the output frequency band of the linear transmission path. The vector adjuster is provided corresponding to each of the plurality of distortion signals generated by the predistorter.

  In the present invention, a part of the plurality of vector adjusters is selected, an amplitude and phase adjustment amount to be given to the selected vector adjuster is determined, and further, from the determined adjustment amount In accordance with a predetermined rule, the adjustment amount of the amplitude and phase of the vector adjuster not selected is set.

  In the embodiment described below, the number of component carriers is two for convenience of description. However, even if the number of component carriers is three or more, it can be easily and obviously expanded with reference to the above summary and the following embodiments. In the embodiment described below, the model order is set to 3 for convenience of description. However, even if the model order is 5 or more, it can be easily and trivially expanded with reference to the above summary and the following embodiment. Of course, even when the number of component carriers is 3 or more and the model order is 5 or more, it can be easily and obviously expanded with reference to the above summary and the following embodiments.

<First Embodiment>
FIG. 4 shows the configuration of the predistorter 100 of the first embodiment and its peripheral devices. The peripheral devices in this example are two signal generators A and B, an amplifier 950, and a feedback signal generator 960. Further, in this example, the signal generated by the signal generator A is a complex baseband signal s ₁ transmitted using the first frequency band, and an I-phase signal and a Q-phase signal (hereinafter referred to as I / Q signal). Digital signal. The signal generated by the signal generator B is a complex baseband signal s ₂ that is transmitted using a second frequency band different from the first frequency band, and includes an I-phase signal and a Q-phase signal (hereinafter, I-phase signal). / Q signal).

  Unlike the predistorter 900 shown in FIG. 1, the predistorter 100 includes a power measuring device 101. Further, instead of the distributor 902A, the predistorter 100 transmits a signal from the signal generator A to the linear transmission path 901A and distortion compensation. It includes a distributor 102A that distributes the signal generation path 909 and the power measuring instrument 101. Instead of the distributor 902B, the signal from the signal generating apparatus B is transferred to the linear transmission path 901B, the distortion compensation signal generating path 909, and the power measuring instrument 101. In addition to the controller 907, based on the output of the distortion observer 906, the vector coefficient (amplitude and phase) and the frequency characteristic compensator coefficient (amplitude) used in the distortion compensation signal generation path 909 are replaced with the controller 907. And a controller 107 for determining an adjustment amount such as phase). The controller 107 includes a vector adjuster control unit 1071, a management unit 1072, and a selection unit 1073.

In the first embodiment, when a distortion compensation signal is generated from a signal input to the predistorter 100, each component constituting the distortion compensation signal (specifically, the third-order distortion signal | s ₁ | ² s ₁ , third-order distortion signal | s ^₂ | ₂ s _2, intermodulation distortion signal ^₂ | s 2 | 2 s ₁ and intermodulation distortion signal ^{_{2 | s 1 | 2 s 2}} ) to calculate the power. Adjust the amplitude and phase of the component with the largest power to reduce the distortion component generated by the power amplifier, and set the amplitude and phase obtained here as the amplitude and phase of each remaining component in the distortion compensation signal To do. As a result, the distortion component generated in the power amplifier can be reduced by adjusting the amplitude and phase of one component without adjusting the amplitude and phase of the other components, so that the adjustment time can be shortened.

  Hereinafter, a control method of the predistorter 100 will be described (see FIG. 5).

(Step S1)
The management unit 1072 includes four vector adjusters (third-order distortion vector adjuster 9092A2, third-order distortion vector adjuster 9092B2, sub-third-order distortion vector adjuster 90951, and sub-third-order distortion vector included in the distortion compensation signal generation path 909. The selection unit 1073 is controlled so as to select a vector adjuster to be controlled among the adjusters 90952). The selection unit 1073 instructs the power measuring device 101 to measure power.

The power measuring device 101 includes a third-order distortion signal (| s ₁ | ² s ₁ ) of the third-order distortion vector adjuster 9092A2, and a third-order distortion signal (| s ₂ | ² s ₂ ) of the third-order distortion vector adjuster 9092B2. The power of the intermodulation distortion signal (2 | s ₂ | ² s ₁ ) of the sub-third-order distortion vector adjuster 90951 and the power of the inter-modulation distortion signal (2 | s ₁ | ² s ₂ ) of the sub-third-order distortion vector adjuster 90952 _They are obtained by calculation using the digital signals s ₁ and s ₂ , respectively. The power measuring device 101 notifies the selection unit 1073 of each obtained power. The power is not measured using components input to each vector adjuster (third-order distortion vector adjuster 9092A2, third-order distortion vector adjuster 9092B2, sub-third-order distortion vector adjuster 90951, and sub-third-order distortion vector adjuster 90952). The reason is that the selection unit 1073 selects the vector adjuster before the component is input to each vector adjuster.

(Step S2)
The selection unit 1073 compares the measurement results of the power measuring device 101, selects a vector adjuster to which the component with the largest power is input, and notifies the management unit 1072 of the selection result.

(Step S3)
The management unit 1072 instructs the vector adjuster control unit 1071 to control the vector adjuster selected by the selection unit 1073. The vector adjuster control unit 1071 is a perturbation method (reference document 1 “T. Nojima and T. Konno,“ Cuber Predistortion Linearizer for Relay Equipment in 800 MHz Band Land Mobile Telephone System, ”IEEE Transactions on vehicular technology, Vol. 34, Issue 4, pp.169-177, 1985.) or a calculation method using quadratic function approximation (Ref. 2, “J. Ohkawara, Y. Suzuki, and S. Narahashi,“ Fast Calculation Scheme for Frequency Characteristic Compensator of Compensate distortion components by controlling the amplitude and phase given to the vector adjuster notified from the management unit 1072 by a known method such as Digital Predistortion Linearizer, "IEEE Vehicular Technology Conference Spring 2009, proceedings, Apr. 2009." The amount of adjustment of the amplitude and phase for obtaining the value is obtained. At this time, it is preferable that the amplitude given to the unselected vector adjuster is zero. This is because by setting it to 0, the influence of the vector adjuster not selected on the selected vector adjuster may be reduced. The vector adjuster control unit 1071 notifies the management unit 1072 of the obtained adjustment amount.

  In the following, a calculation method using the perturbation method and quadratic function approximation will be described by taking as an example the case where the third-order distortion vector adjuster 9092A2 is selected.

(In case of perturbation method)
The vector adjuster control unit 1071 designates one of the third-order distortion component upper band Ab2 and the third-order distortion component lower band Ab1 to the distortion observer 906 and instructs its power measurement, and this measurement Based on the result, the phase and amplitude applied to the third-order distortion vector adjuster 9092A2 are adjusted so as to reduce the power in the band. First, the power PD in the designated band is measured before and after the arbitrarily set phase XP, the phase is changed by a predetermined offset value ΔXP in the direction in which the power PD decreases, and the power PD is changed by the distortion observer 906. taking measurement. By repeating the phase change and the measurement of the power PD, the phase XP _MIN at which the power PD is equal to or less than a predetermined threshold value TH is obtained. When the power PD does not become equal to or less than the predetermined threshold TH due to the phase adjustment, the amplitude is adjusted by the same process. The phase and amplitude are repeatedly adjusted, and the phase XP _MIN and the amplitude XA _{MIN at} which the power PD is less than or equal to a predetermined threshold TH are obtained. The designated bandwidth may be a predetermined bandwidth. In some cases, the processing of the strain observer 906 can be simplified by narrowing the bandwidth.

(Calculation method using quadratic function approximation)
In-band power (PD ₁ , PD ₂ ,..., PD _R ) designated by the phases (XP ₁ , XP ₂ ,..., XP _R ) of different R points (hereinafter R is an integer of 3 or more) , And from the used phase (XP ₁ , XP ₂ ,..., XP _R ) and the measured power (PD ₁ , PD ₂ ,..., PD _R ) The coefficients (a ₂ , a ₁ , a ₀ ) of the quadratic function (PD = a ₂ XP ² + a ₁ XP + a ₀ ) indicating the dependency are obtained. A phase XP _MIN (= −a ₁ / 2a ₂ ) that minimizes the power PD in the coefficients (a ₂ , a ₁ , a ₀ ) is set in the third-order distortion vector adjuster 9092A2. Although a quadratic function has been described as an example here, a coefficient (b ₂ ) of a trigonometric function (PD = b ₂ cos (b ₁ −XP) + b ₀ ) is used as the dependence of the specified in-band power on the phase in the calculation of the phase. , B ₁ , b ₀ ) may be obtained by the method of least squares. In the obtained trigonometric function, the phase XP _MIN (= b ₁ −π) that minimizes the power PD (that is, b ₁ −XP = π) is set in the third-order distortion vector adjuster 9092A2. In the calculation method using quadratic function approximation, when the coefficient a ₂ is 0 or less or the coefficient of the quadratic function cannot be obtained, the measured power (PD ₁ , PD ₂ ,..., PD _R ) The phase with the lowest power may be XP _MIN . With respect to the amplitude, an amplitude XA _MIN that minimizes the power PD is obtained by the same processing, and is set in the third-order distortion vector adjuster 9092A2.
In order to specify a quadratic function using the least square method, R needs to be at least 3. However, since R can be approximated with higher accuracy by increasing R, it is within the range allowed by calculation time and other requirements. And R may be larger than 3. It is simplest to set the same number of R in the setting of the phase value and the amplitude value. However, when it is necessary to improve the approximation accuracy of either the phase value or the amplitude value, R is set in the setting of the phase value and the amplitude value. May be different values.

  In the above two examples, the amplitude value is set after setting the phase value. However, when the sensitivity of the distortion component generated in the power amplifier is higher than the phase value, the amplitude value is set. May be performed first.

(Step S4)
The management unit 1072 instructs the vector adjuster control unit 1071 to set the amplitude and phase of each vector adjuster that is not selected in the process of step S2.
The vector adjuster control unit 1071 sets the amplitude and phase adjustment amounts obtained in step S3 as the amplitude and phase adjustment amounts of each vector adjuster that has not been selected.

  Thus, in the first embodiment, the distortion signal amplitude and phase adjustment amounts are determined for only one selected vector adjuster, and this adjustment amount is used as the amplitude and phase adjustment amounts of the other vector adjusters. To reflect. Thereby, it is not necessary to individually determine the vector coefficients of the four vector adjusters, so that the adjustment time can be shortened.

The reason for selecting the vector adjuster to which the component with the largest power is input will be described below.
In the spectrum shown in FIG. 3, two components (| s ₁ | ² s ₁ and 2 | s ₂ | ² s ₁ ) are synthesized as distortion components generated in the transmission band Ab. Therefore, the distortion signal (| s ₁ | ² s ₁ ) input to the third-order distortion vector adjuster 9092A2 and the distortion signal (2 | s ₂ | ² s ₁ ) input to the sub-third-order distortion vector adjuster 90951 are used. Thus, it is considered that the two distortion components are individually compensated. At this time, when there is a difference in power between the two distortion components because, for example, there is a large power difference between the signal s ₁ and the signal s ₂ , the amplitude given to the vector adjuster to compensate for the distortion component with the smaller power Even if the phase is adjusted, the power of the distortion component hardly changes, and the amplitude and phase for compensating the distortion component may not be obtained accurately. For this reason, even if the obtained amplitude and phase adjustment amounts are set as the amplitude and phase adjustment amounts of the other vector adjusters, the distortion component may not fall below the threshold value TH. In the first embodiment, in order to avoid such inconvenience, a vector adjuster to which a distortion signal with the largest power is input is selected.

  In step S4, an adjustment amount obtained by multiplying the amplitude adjustment amount obtained in step S3 by a predetermined magnification α (0 <α <1) is selected in step S2. It may be set as an amplitude adjustment amount of each vector adjuster that has not been present. A ratio based on the highest power may be obtained using the power obtained in the process of step S1, and a magnification that is not necessarily the same may be set for each vector adjuster according to the power ratio. The reason why such an adjustment method is permitted is that, depending on the characteristics of the power amplifier, the distortion component generated in the power amplifier may be further reduced by weighting.

  A characteristic that the difference between the phase and amplitude adjustment amount given to the reference vector adjuster and the amplitude and phase adjustment amount given to another vector adjuster is a constant value with respect to a certain vector adjuster. In the case of using a power amplifier, the amplitude difference and the phase difference are measured in advance, and the difference is obtained by adjusting these differences with respect to the amplitude and phase adjustment amounts obtained in step S3. The amplitude and phase adjustment amounts to be obtained may be set as the amplitude adjustment amounts of the respective vector adjusters not selected in the process of step S2.

(Modification 1 of the first embodiment)
In the first modification of the first embodiment, the vector coefficient of the vector adjuster is set as follows.
The distortion signal (| s ₁ | ² s ₁ ) input to the third-order distortion vector adjuster 9092A2 has the highest power and the distortion signal and the distortion signal (2 | s ₂ | input to the sub-third-order distortion vector adjuster 90951). ² s ₁ ) is within a predetermined range (however, it is preferably within a sufficiently small range), the third-order distortion vector adjuster 9092A2 and the sub-third-order distortion are processed in step S3. Vector adjuster 90951 may be adjusted together. At this time, the amplitude and phase given to the third-order distortion vector adjuster 9092A2 and the sub-third-order distortion vector adjuster 90951 are set to the same value. This is because, when the power of the distortion component is increased by setting different values, it is difficult to determine which of the two amplitudes (or phases) has caused the increase of the distortion component power. . When the power difference between the distortion signal input to the third-order distortion vector adjuster 9092A2 and the distortion signal input to the sub-third-order distortion vector adjuster 90951 is small, by adjusting the vector coefficients for the two vector adjusters together, There are cases where the adjustment amount of the amplitude and phase for compensating the distortion component with higher accuracy than the case of selecting one vector adjuster may be obtained.

  The power of the distortion signal input to the sub third-order distortion vector adjuster 90951 is the highest, and the power difference between the distortion signal and the distortion signal input to the third-order distortion vector adjuster 9092A2 is within a predetermined range (however, sufficiently The same applies to the case where it is preferably within a small range.

  When the power of the distortion signal input to the third-order distortion vector adjuster 9092B2 is the highest, the power difference between the distortion signal and the distortion signal input to the sub-third-order distortion vector adjuster 90952 is compared and within a predetermined range ( However, it is preferable that the third order distortion vector adjuster 9092B2 and the sub third order distortion vector adjuster 90952 are adjusted together. The power of the distortion signal input to the sub third-order distortion vector adjuster 90952 is the highest, and the power difference between the distortion signal and the distortion signal input to the third-order distortion vector adjuster 9092B2 is within a predetermined range (however, sufficiently The same applies to the case where it is preferably within a small range.

  For the reasons described above, the sum of the distortion signal input to the third-order distortion vector adjuster 9092A2 and the distortion signal input to the sub-third-order distortion vector adjuster 90951, the distortion signal input to the third-order distortion vector adjuster 9092B2, and the sub-3 The power sums of the distortion signals input to the second distortion vector adjuster 90952 may be compared, and the third distortion vector adjuster and the sub third distortion vector adjuster having the higher power sum may be selected. At this time, in the process of step S3, the same amplitude value and phase value are given to the selected third-order distortion vector adjuster and sub-third-order distortion vector adjuster to adjust together.

(Modification 2 of the first embodiment)
In the second modification of the first embodiment, the vector coefficient of the vector adjuster is set as follows.
Depending on the characteristics of the power amplifier, the distortion component power may not be smaller than the threshold value TH unless the amplitude and phase applied to each vector adjuster are set to different values. For this reason, you may control the predistorter 100 in the procedure shown in FIG.

(Step S5)
After each processing from step S1 to step S4 described above is performed, the management unit 1072 notifies the respective powers of the third-order distortion component lower bands Ab1 and Bb1 and the third-order distortion component upper bands Ab2 and Bb2. The vector adjuster control unit 1071 is instructed.
The vector adjuster control unit 1071 instructs the distortion observer 906 to measure the respective powers of the third-order distortion component lower bands Ab1 and Bb1 and the third-order distortion component upper bands Ab2 and Bb2, respectively. The management unit 1072 is notified of the obtained measurement result.

(Step S6)
The management unit 1072 determines whether the notified power of each band is smaller than a threshold value. The management unit 1072 ends the process when the power of all bands is smaller than the threshold value. When there is a band whose power is equal to or greater than the threshold, the management unit 1072 notifies the vector adjuster control unit 1071 of the band whose power is equal to or greater than the threshold. The threshold value may be determined individually for each band. In some cases, the adjustment time can be shortened or the performance required for the power amplifier can be alleviated by individually determining.

(Step S7)
The vector adjuster control unit 1071 determines amplitude and phase adjustment amounts to be given to the vector adjuster corresponding to the band notified from the management unit 1072. Here, the amplitude and phase adjustment amounts are determined one by one for the corresponding vector adjuster. Depending on the characteristics of the power amplifier, the power of the distortion component can be made smaller than the threshold without the amplitude and phase adjustment amount given to each vector adjuster having a large difference with respect to the amplitude value and phase value instructed from the management unit 1072. There is a case. For this reason, the adjustment time may be shortened by using a perturbation method with a small offset value. When adjusting the amplitude applied to the vector adjuster using the perturbation method, the amplitude may be decreased. This may prevent an increase in distortion component power generated in the power amplifier. This is because the power of the component input to the vector adjuster for adjusting the amplitude is smaller than the power of the component input to the vector adjuster selected in step S2.

  When both the distortion component power in the vicinity of the transmission band Ab and the distortion component power in the vicinity of the transmission band Bb are equal to or greater than the threshold, the third-order distortion vector adjuster 9092A2 and the third-order distortion vector adjuster 9092B2 are adjusted together. Also good. At this time, the amplitude and phase applied to each vector adjuster may be different values. This is because the bands where distortion components are generated are different. Therefore, a combination of vector adjusters to be adjusted together may be a third-order distortion vector adjuster 9092A2 and a sub-third-order distortion vector adjuster 90952, or a sub-third-order distortion vector adjuster 90951 and a third-order distortion vector adjuster 9092B2. Also good. It is preferable to measure the characteristics of the power amplifier in advance and select an optimum combination from the result.

  After the adjustment of the selected vector adjuster is completed, the process returns to step S5.

(Modification 3 of the first embodiment)
Depending on the characteristics of the power amplifier, the management unit 1072 may instruct the vector adjuster control unit 1071 to set only the phase of the vector adjuster in the process of step S4. In this case, the management unit 1072 instructs the vector adjuster control unit 1071 to individually adjust the amplitude of the vector adjuster. This setting method is effective when there is no significant difference in the phase of distortion components generated in the power amplifier and there is a difference in amplitude.

  Depending on the characteristics of the power amplifier, the management unit 1072 may instruct the vector adjuster control unit 1071 to set only the amplitude of the vector adjuster in the process of step S4. In this case, the management unit 1072 instructs the vector adjuster control unit 1071 to individually adjust the phase of the vector adjuster. This setting method is effective when there is no significant difference in amplitude of distortion components generated in the power amplifier and there is a difference in phase.

  In the first embodiment, there is one directional coupler, one down converter, one quadrature demodulator, and one ADC. However, if each circuit does not support carrier aggregation, each circuit may be prepared for each component carrier. Good. At this time, as many strain observers as the number of component carriers may be prepared, and the output of each ADC may be switched over time.

  In consideration of the case where only one of the third-order distortion component lower bands Ab1 and Bb1 and the third-order distortion component upper bands Ab2 and Bb2 does not satisfy the threshold value in the process of step S5, the predistorter 100 performs the frequency characteristic compensator. May be provided. The frequency characteristic compensator converts the output of the vector adjuster into the frequency domain, adjusts the amplitude and phase so that the distortion component is reduced for each predetermined bandwidth, and then converts it into the time domain. The output of the frequency characteristic compensator is input to the signal adder. Since the frequency characteristic can be given to the distortion compensation signal by using the frequency characteristic compensator, there is a case where the power of the distortion component can be made smaller than the threshold value. When the predistorter 100 includes a plurality of frequency characteristic compensators, the frequency characteristic compensator is selected in the same manner as described above, and the frequency characteristic in which the amplitude and phase determined by the selected frequency characteristic compensator are not selected. It may be set as the amplitude and phase of the compensator.

  When using a power amplifier that has the characteristics that the phase and amplitude differ for each distortion component, but the difference does not increase among multiple distortion components that occur in the same frequency band, the difference between the phase and amplitude becomes a constant value. When a power amplifier having features is used, if the predistorter described in the present embodiment is used when the characteristics of the power amplifier cannot be grasped in advance, the pre-disclaimer described in the above reference patent document (Japanese Patent Application No. 2011-269882) is used. Distortion components can be reduced in a shorter time than a tota. This is because the predistorter of the above-mentioned reference patent document sequentially controls the vector adjusters one by one and adjusts the amplitude and phase given to each vector adjuster from the initial values. The predistorter selects one vector adjuster, adjusts the amplitude and phase applied to this vector adjuster, and then transfers it to the other vector adjuster based on the amplitude and phase applied to the selected vector adjuster. This is to set the given amplitude and phase.

Second Embodiment
A predistorter 200 according to the second embodiment will be described with reference to FIG. The predistorter 200 of the second embodiment is that the selection unit 1073 of the controller 107 selects not only the power measuring instrument 101 but also a vector adjuster to be adjusted using the measurement result of the distortion observer 906. Different from form. By adding the distortion component power generated in the power amplifier to the selection index, the vector adjuster is selected after measuring the distortion component power. However, the vector can easily compensate for the distortion component compared to the first embodiment. You may be able to select a regulator.

  Hereinafter, a control method of the predistorter 200 will be described with reference to FIG.

(Step S21)
The process of step S21 in the second embodiment is the same as the process of step S1 in the first embodiment.

(Step S21a)
The selection unit 1073 instructs the distortion observer 906 to measure the distortion component power. The distortion observer 906 measures each power of the third-order distortion component lower bands Ab1 and Bb1 and the third-order distortion component upper bands Ab2 and Bb2, and notifies the selection unit 1073 of the measurement results.

(Step S22)
The selection unit 1073 compares the measurement results of each power from the distortion observer 906 and selects the band with the largest power. Next, the measurement result of each power from the power measuring device 101 is compared, and the distortion signal having the higher power among the distortion signal power corresponding to the selected band is selected. The management unit 1072 is notified of the selection result of the band and the selection result of the vector adjuster to which the selected distortion signal is input.

(Step S23)
The management unit 1072 instructs the vector adjuster control unit 1071 to control the vector adjuster selected by the selection unit 1073. The vector adjuster control unit 1071 controls the amplitude and phase given to the vector adjuster notified from the management unit 1072 by a known method such as the above-described perturbation method or calculation method using quadratic function approximation, and compensates for distortion components. The amount of adjustment of the amplitude and phase for obtaining the value is obtained. In this process, the band measured by the strain observer 906 is the band selected by the selection unit 1073 in the process of step S22. As a result, the result of the band having the largest distortion component power can be used, and therefore the amplitude and phase for reducing the distortion component may be easily obtained. The vector adjuster control unit 1071 notifies the management unit 1072 of the obtained adjustment amount.

(Step S24)
The process of step S24 in the second embodiment is the same as the process of step S4 in the first embodiment.

  In the second embodiment, the selection unit 1073 selects the vector adjuster using not only the distortion signal power but also the distortion component power. Thereby, from the viewpoint that the band in which the power of the distortion component is the largest can be designated in the adjustment of the vector adjuster, it may be easy to obtain the adjustment amount of the amplitude and phase of the vector adjuster for compensating the distortion component.

<Third Embodiment>
FIG. 9 shows the configuration of the predistorter 300 of the third embodiment and its peripheral devices. The peripheral devices in this example are two signal generators A and B, an amplifier 980, and a feedback signal generator 960. Further, in this example, the signal generated by the signal generator A is a complex baseband signal s ₁ transmitted using the first frequency band, and an I-phase signal and a Q-phase signal (hereinafter referred to as I / Q signal). Digital signal. The signal generated by the signal generator B is a complex baseband signal s ₂ that is transmitted using a second frequency band different from the first frequency band, and includes an I-phase signal and a Q-phase signal (hereinafter, I-phase signal). / Q signal).

The predistorter 300 includes a signal adder 308 that synthesizes the input signals, a distortion compensation signal generation path 309 for generating a distortion compensation signal, and a signal adder 308 that performs distortion compensation on the signal from the signal generator A. A divider 302A for distributing to the signal generation path 309, a distributor 302B for distributing the signal from the signal generator B to the signal adder 308 and the distortion compensation signal generation path 309, and a delay time generated in the distortion compensation signal generation path 309. The linear transfer path 301 including a delay device that delays the output signal of the signal adder 308, the signal adder 303 that combines the output of the linear transfer path 301 and the output of the distortion compensation signal generation path 309, and the output of the signal adder 303 A digital-to-analog converter (DAC) 304 for converting (digital I / Q signal to which a distortion compensation component has been added) into an analog I / Q signal; An analog-to-digital converter (ADC) 305 that converts the output (analog I / Q signal) of the feedback signal generation device 960 that takes a part of the output of 980 as a feedback signal into a digital I / Q signal, and a distortion component from the output of the ADC 305 And a distortion observer 306 for measuring the distortion, and an adjustment amount such as a vector coefficient (amplitude and phase) and a frequency characteristic compensator coefficient (amplitude and phase) used in the distortion compensation signal generation path 309 based on the output of the distortion observer 306. A controller 307 for determining is included. The controller 307 includes a vector adjuster control unit 3071 and a vector coefficient estimation unit 3075. When the signals s ₁ and s _{2 of} the signal generators A and B are baseband signals centered on a direct current, the distributors 302A and 302B have a frequency between the center frequency of the signal band Ab and the center frequency of the signal band Bb. The frequency of the signal from each signal generator is shifted so as to be a difference. This is because the signals s ₁ and s ₂ are generated at the same center frequency when the frequency is not shifted. The frequency shift can be performed by using, for example, FFT and IFFT. For example, the signals s ₁ and s ₂ are converted into the frequency domain by FFT, and the respective frequencies are set to desired frequencies (for example, −ΔF / 2 for the signal s ₁ and + ΔF / 2 for the signal s ₂ ). This is done by using IFFT after shifting to.

  The amplifying device 980 includes a quadrature modulator 981 that quadrature modulates the analog I / Q signal that is the output of the DAC 304, an up converter 982 that converts the frequency of the modulation output from the quadrature modulator 981 into a carrier frequency, and an up converter 982 A power amplifier 984 that amplifies the output signal is included. The power-amplified high frequency signal is supplied from the output terminal 970 to the antenna via a duplexer (not shown), for example.

  The feedback signal generation device 960 includes a directional coupler 961 that extracts a part of the output of the amplification device 950 as a feedback signal, a frequency down converter 962 that converts the frequency of the feedback signal, and an orthogonal signal that orthogonally demodulates the down-converted feedback signal. A demodulator 963 is included.

FIG. 10 shows a functional block diagram of the distortion compensation signal generation path 309.
The distortion compensation signal generation path 909, the third-order distortion signal | s ₁ | adjustment of vector coefficients of ² s ₁ (amplitude and phase) | a ² s ₁ generated third-order distortion generator 9092A1 and third-order distortion signal | s ₁ A distortion signal generator 9092A including a third-order distortion vector adjuster 9092A2, a third-order distortion generator 9092B1 that generates a third-order distortion signal | s ₂ | ² s _2, and a third-order distortion signal | s ₂ | ² s _2. A distortion signal generation unit 9092B including a third-order distortion vector adjuster 9092B2 that adjusts vector coefficients (amplitude and phase), and intermodulation distortion signals 2 | s ₂ | ² s ₁ and 2 | s ₁ | ² s ₂ An intermodulation distortion generator 30950 that generates s ₂ ² s ₁ ^* and s ₁ ² s ₂ ^* and a secondary third order that adjusts vector coefficients (amplitude and phase) of the intermodulation distortion signal 2 | s ₂ | ² s ₁ Distortion vector adjuster 90951 and intermodulation distortion signal 2 | s ₁ | ² s ₂ vector coefficients (amplitude and level) Sub-third-order distortion vector adjuster 90952 for adjusting the phase), sub-third-order distortion vector adjuster 30953 for adjusting the vector coefficients (amplitude and phase) of the intermodulation distortion signal s ₁ ² s ₂ ^* , and the inter-modulation distortion signal. A sub-distortion signal generation unit 3095 including a sub-third-order distortion vector adjuster 30954 that adjusts vector coefficients (amplitude and phase) of s ₂ ² s ₁ ^* , and a signal from the distributor 302A as a distortion signal generation unit 9092A A signal distribution unit 9091A that distributes to the sub-distortion signal generation unit 3095, a signal distribution unit 9091B that distributes the signal from the distributor 302B to the distortion signal generation unit 9092B and the sub-distortion signal generation unit 3095, respectively, and a distortion signal generation unit 9092A , A sub-distortion signal generation unit 3095, and a distortion signal generation unit 9092A. The output of the signal synthesis unit 3093 becomes the input of the signal adder 303. Third-order distortion vector adjuster 9092A2, third-order distortion vector adjuster 9092B2, sub-third-order distortion vector adjuster 90951, sub-third-order distortion vector adjuster 90952, sub-third-order distortion vector adjuster 30953, and sub-third-order distortion vector adjuster The value of the vector coefficient to be adjusted by each of 30954 is based on an instruction from the controller 307.

The distortion compensation signal generation path 309 also generates distortion signals s ₂ ² s ₁ ^* and s ₁ ² s ₂ ^* for compensating distortion components generated in bands other than the vicinity of the signal bands Ab and Bb, and also uses the distortion signals. This is different from the above-described embodiment in that the amplitude and phase given to the vector adjuster are obtained. Also, the third embodiment differs from the above-described embodiment in that the amplitude and phase given to each vector adjuster are estimated, and depending on the characteristics of the power amplifier, the distortion component is further reduced compared to the above-described embodiment. There are cases where it is possible.

  Hereinafter, a control method of the predistorter 300 will be described with reference to FIG.

(Step S31)
The vector coefficient estimation unit 3075 instructs the vector adjuster control unit 3071 to control the sub third-order distortion vector adjuster 30953. The vector adjuster control unit 3071 determines the amplitude and phase to be given to the sub third-order distortion vector adjuster 30953 notified from the vector coefficient estimation unit 3075 by a known method such as the above-described perturbation method or calculation method using quadratic function approximation. The amount of adjustment of amplitude and phase for controlling and compensating for distortion components is obtained. At this time, the distortion observer 306 measures the power of the intermodulation distortion component lower band Cb shown in FIG. 12, and notifies the vector adjuster control unit 3071 of the measurement result. The vector adjuster control unit 3071 notifies the vector coefficient estimation unit 3075 of the obtained adjustment amount.

(Step S32)
The vector coefficient estimation unit 3075 instructs the vector adjuster control unit 3071 to control the sub third-order distortion vector adjuster 30954. The vector adjuster control unit 3071 determines the amplitude and phase to be given to the sub third-order distortion vector adjuster 30954 notified from the vector coefficient estimation unit 3075 by a known method such as the above-described perturbation method or calculation method using quadratic function approximation. The amount of adjustment of amplitude and phase for controlling and compensating for distortion components is obtained. At this time, the distortion observer 306 measures the power of the intermodulation distortion component upper band Db shown in FIG. 12, and notifies the vector adjuster control unit 3071 of the measurement result. The vector adjuster control unit 3071 notifies the vector coefficient estimation unit 3075 of the obtained adjustment amount.

  Here, the adjustment amount of the vector coefficient is determined in the order of the sub-third-order distortion vector adjuster 30953 and the sub-third-order distortion vector adjuster 30954, but this order may be reversed, or the sub-third-order distortion vector adjuster. 30953 and the sub-third-order distortion vector adjuster 30954 have different distortion component bands to be compensated for and may be processed in parallel. When performing parallel processing, the adjustment time may be shortened depending on the characteristics of the power amplifier.

(Step S33)
The vector coefficient estimator 3075 gives the third-order distortion vector adjusters 9092A2 and 9092B2 and the sub-third-order distortion vector adjusters 90951 and 90952 using the amplitude and phase adjustment amounts obtained in the processes of steps S31 and S32. Estimate the amount of amplitude and phase adjustment. An example of estimating the amplitude is shown below, but the phase estimation is the same. The amplitude adjustment amount for compensating for the distortion component obtained in step S31 is YA ₅ , and the amplitude adjustment amount for compensating for the distortion component obtained in step S32 is YA ₆ .
The center frequency f _d3 of the distortion component compensated by the sub-third-order distortion vector adjuster 30953, the center frequency f _d4 of the distortion component compensated by the sub-third-order distortion vector adjuster 30954, and the third-order distortion vector adjuster 9092A2 and the sub-third order. The center frequency f _{d1 of} the distortion component compensated by the distortion vector adjuster 90951 is set to the center frequency f _{d2 of} the distortion component compensated by the third-order distortion vector adjuster 9092B2 and the sub-third-order distortion vector adjuster 90952.
The amount of amplitude adjustment given to the third-order distortion vector adjuster 9092A2 and the sub-third-order distortion vector adjuster 90951 is expressed as YA ₅ + (f _d1 −f _d3 ) (YA ₆ −YA ₅ ) / (f _d4 −f _d3 ) To do. The amount of adjustment of each amplitude given to the third-order distortion vector adjuster 9092B2 and the sub-third-order distortion vector adjuster 90952 is YA ₅ + (f _d2 −f _d3 ) (YA ₆ −YA ₅ ) / (f _d4 −f _d3 ) To do.

  When the distortion component generated in the power amplifier has frequency characteristics, the distortion coefficient can be compensated with higher accuracy than when the same value is set in each vector adjuster by estimating the vector coefficient by linear interpolation. There are cases where it is possible.

(Step S34)
The vector coefficient estimation unit 3075 instructs the vector adjuster control unit 3071 to set the amplitude and phase adjustment amounts obtained in the process of step S33 in each vector adjuster. The vector adjuster control unit 3071 sets the amplitude and phase adjustment amounts notified from the vector coefficient estimation unit 3075 as the amplitude and phase adjustment amounts of each vector adjuster.

In the above example, the example in which the component carriers are discontinuously arranged is shown, but the operation when the component carriers are continuous as shown in FIG. 13 will be described below.
In the process of step S31, the power in the third-order distortion component lower band Ab1 is measured, and the adjustment amounts of the amplitude and phase of the sub-third-order distortion vector adjuster 30953 are obtained. In the process of step S32, the power in the third-order distortion component upper band Bb2 is measured, and the adjustment amount of the amplitude and phase of the sub-third-order distortion vector adjuster 30954 is obtained. In the process of step S33, the following process is performed. The amplitude obtained by the sub-third-order distortion vector adjuster 30953 is arranged at the center frequency of the band where the distortion component to be compensated by the sub-third-order distortion vector adjuster 30953 is generated, and obtained by the sub-third-order distortion vector adjuster 30954 The third-order distortion vector adjuster 9092A2 and the sub-third-order distortion vector adjustment are arranged at the center frequency of the band where the distortion component generated by the sub-third-order distortion vector adjuster 30954 is generated and linearly interpolated based on these amplitude values. The amount of adjustment of the amplitude at the center frequency of the band where the distortion component to be compensated by the unit 90951 is generated, and the center frequency of the band where the distortion component to be compensated by the third order distortion vector adjuster 9092B2 and the sub third order distortion vector adjuster 90952 is generated. Find the amount of amplitude adjustment. The phase estimation is the same as the amplitude estimation.

  Even in the case where the component carriers are continuously arranged, the method described in other embodiments as well as the present embodiment may be used.

In the above-mentioned embodiment, the case where the signal which uses two different frequency bands simultaneously cannot be generated with one signal generator was assumed. The embodiment of the present invention is not limited to such a case. In the case where one signal generator generates a signal that uses a plurality of different frequency bands at the same time, each of the above embodiments is performed by performing each process of generating and adjusting a distortion signal for each band after demultiplexing the signal. Can be reduced to

Claims (4)

A predistorter that pre-adds a distortion compensation signal that cancels a distortion component generated in a power amplifier to an input signal including a plurality of component carriers,
A linear transmission path for delay transmission of the component carrier signal;
At least a part of the self-distortion signal obtained by using only the signal of the component carrier for each component carrier and the intermodulation distortion signal obtained by using two or more components carrier signals different from each other (hereinafter, A self-distortion signal and at least a part of the intermodulation distortion signal are collectively referred to as a distortion signal), and a distortion signal generation means for generating each of these distortion signals. A vector adjuster for adjusting the amplitude and phase, and a distortion compensation signal generation path for outputting the distortion compensation signal;
A signal adder for combining the output of the linear transfer path and the output of the distortion compensation signal generation path;
A power meter that measures the power of each of the distortion signals using the signal of the component carrier;
A distortion observer for observing a distortion component included in the output of the power amplifier that performs power amplification on the output of the signal adder;
At least some of the plurality of vector adjusters are selected using the measurement result of the power measuring device, and the amount of amplitude and / or phase adjustment given to the selected vector adjuster is distorted. A controller that determines based on the observation result of the observer, and further sets the determined adjustment amount or the adjustment amount based on the adjustment amount as the amplitude and / or phase adjustment amount of the vector adjuster that has not been selected; Including predistorter. The predistorter according to claim 1,
The controller includes a selection unit and a vector adjuster control unit,
The selection unit compares the measurement results of the power measuring device, selects a vector adjuster to which a distortion signal having the largest power is input,
The vector adjuster control unit controls the amplitude and / or phase applied to the vector adjuster selected by the selection unit so as to reduce the distortion component generated by the power amplifier based on the observation result of the distortion observer. A predistorter that determines an adjustment amount, and further sets the determined adjustment amount or an adjustment amount based on the adjustment amount as an amplitude and / or phase adjustment amount of an unselected vector adjuster . The predistorter according to claim 1,
The controller includes a selection unit and a vector adjuster control unit,
The selection unit compares the observation results of the distortion observer, selects a band in which a distortion component with the largest power is generated among distortion components generated in the power amplifier, and further, a measurement result of the power measurement instrument Comparing the power of the distorted signal corresponding to the selected band, and selecting a vector adjuster to which the distorted signal having the largest power is input,
The vector adjuster control unit controls the amplitude and / or phase applied to the vector adjuster selected by the selection unit so as to reduce the distortion component generated by the power amplifier based on the observation result of the distortion observer. A predistorter that determines an adjustment amount, and further sets the determined adjustment amount or an adjustment amount based on the adjustment amount as an amplitude and / or phase adjustment amount of an unselected vector adjuster . A predistorter control method for adding in advance a distortion compensation signal that cancels a distortion component generated in a power amplifier to an input signal including a plurality of component carriers,
The predistorter is
A linear transmission path for delay transmission of the component carrier signal;
At least a part of the self-distortion signal obtained by using only the signal of the component carrier for each component carrier and the intermodulation distortion signal obtained by using two or more components carrier signals different from each other (hereinafter, A self-distortion signal and at least a part of the intermodulation distortion signal are collectively referred to as a distortion signal), and a distortion signal generation means for generating each of these distortion signals. A vector adjuster for adjusting the amplitude and phase, and a distortion compensation signal generation path for outputting the distortion compensation signal;
A signal adder for combining the output of the linear transfer path and the output of the distortion compensation signal generation path;
A power meter that measures the power of each of the distortion signals using the signal of the component carrier;
A distortion observer for observing a distortion component included in the output of the power amplifier that performs power amplification on the output of the signal adder;
A controller for setting the amount of adjustment of the amplitude and phase of each of the vector adjusters,
The above control method is
The power meter measures the power of each distorted signal using the component carrier signal;
The controller selecting a part of the vector adjusters among the plurality of vector adjusters using at least the measurement result of the power meter;
Determining the amount of amplitude and / or phase adjustment that the controller gives to the selected vector adjuster based on the observation result of the distortion observer;
A method for controlling the predistorter, comprising: the controller setting the determined adjustment amount or an adjustment amount based on the adjustment amount as an amplitude and / or phase adjustment amount of the vector adjuster that has not been selected.

Images