JP2014107806A - Amplification device, distortion compensation method and distortion compensation program - Google Patents

Abstract

Provided are an amplification device, a distortion compensation method, and a distortion compensation program capable of improving the amplification efficiency of a wideband signal and appropriately performing distortion compensation with a simple configuration in a configuration employing a DPD and an envelope tracking method. To do.
An amplifier device includes a power amplifier, a distortion compensation unit that performs distortion compensation of the power amplifier, and a power supply voltage VDD (t) or a power supply current IDD (t) supplied to the power amplifier. And an ET power source 111 that changes in accordance with an envelope tracking signal V [n] that is a signal corresponding to the envelope of the signal x [n]. The distortion compensator 101 is a linear combination of a plurality of first control functions having the baseband signal x [n] and the power supply voltage VDD (t), the power supply current IDD (t), or the envelope tracking signal V [n] as variables. Based on the above, distortion compensation is performed.
[Selection] Figure 1

Description

  The present invention relates to an amplification device, a distortion compensation method, and a distortion compensation program, and more particularly, to an amplification device, a distortion compensation method, and a distortion compensation program that perform distortion compensation using a function.

  In a base station in a mobile communication system, for example, a high power amplifier (HPA) for amplifying and outputting an RF (Radio Frequency) transmission signal is provided. This HPA generally has high amplification efficiency but low linearity of input / output characteristics. For this reason, a transmission signal having desired characteristics may not be obtained due to distortion of the output signal of the HPA.

  DPD (Digital Pre-Distortion) has been proposed as one of distortion compensation methods for compensating such amplifier distortion (for example, Lei Ding, “Digital predistortion of Power Amplifiers for Wireless Applications”, Georgia Institute). of Technology, March 2004. (Non-Patent Document 1)). That is, in DPD, a model representing the input / output characteristics of an amplifier is estimated, and a model having characteristics opposite to the model is generated by digital signal processing. The distortion in the amplifier is compensated by adding an inverse model to a target signal to be amplified by the amplifier, specifically, a digital signal before analog conversion.

  In addition, with the increase in frequency and bandwidth of mobile communication systems, there is an increasing need for low power consumption and high efficiency of radio transmission amplifiers regardless of whether the base station or terminal is used. For example, in LTE-Advanced, band expansion up to a maximum of 100 MHz is scheduled.

  Furthermore, for example, an ET (Envelope Tracking) drain modulation method or a collector modulation method has been developed as a technique for improving the efficiency of an amplifier. In the ET system, unnecessary power consumption is suppressed by changing the power supply voltage or power supply current of an amplification element such as an FET (Field Effect Transistor) in an amplifier in accordance with the envelope or envelope of a high-frequency input signal. Improve power efficiency.

  Here, in the ET system, a switching power supply is generally used as an ET power supply that changes a power supply voltage or power supply current supplied to an amplifier in accordance with an envelope signal.

  However, when a wide-band signal as described above is amplified, a switching power supply used as an ET power supply cannot follow a high-frequency envelope, and the efficiency of the amplifier may not be improved.

  In order to solve such a problem, a method of generating a signal that is narrower than the original envelope signal and does not fall below the voltage level of the original envelope signal can be considered.

  For example, US Patent Application Publication No. 2010/0295613 (Patent Document 1) discloses the following method. That is, the RF (Radio Frequency) amplifier input signal envelope signal is low-pass filtered to generate a residual signal indicating the difference between the filtered envelope signal and the original envelope signal, and the residual signal is low-pass filtered. The remaining signal after filtering is added to the envelope signal after filtering. These operations are repeated until the voltage of the signal indicating the addition result becomes equal to or higher than the original envelope signal.

  Further, when the DPD and ET methods as described above are employed and an envelope tracking signal having a narrower band than the original envelope signal is used, the following problems occur.

  That is, since the original envelope signal is the envelope of the target signal, when the power supply voltage or power supply current corresponding to the original envelope signal is supplied to the amplifier, the target signal and the power supply voltage or power supply current have a certain relationship. It will be. In this case, distortion in the amplifier can be handled as being dependent on the instantaneous voltage of the target signal.

  On the other hand, the relationship between the amplitude of the envelope tracking signal and the amplitude of the target signal varies depending on the content of the processing when generating the envelope tracking signal. For this reason, when a power supply voltage or power supply current corresponding to the envelope tracking signal is supplied to the amplifier, the target signal and the power supply voltage or power supply current do not have a certain relationship.

  That is, in the configuration in which the power supply voltage or power supply current corresponding to the envelope tracking signal is supplied to the amplifier, the distortion in the amplifier is not only the instantaneous voltage of the target signal but also the power supply voltage or power supply current supplied from the ET power supply to the amplifier. It depends.

  In order to solve such a problem, it is necessary to perform distortion compensation on the signal input to the amplifier based on the power supply voltage or power supply current supplied from the ET power supply to the amplifier and the target signal. In the technique described in Patent Document 1, distortion compensation is performed using a two-dimensional look-up table (LUT) using the amplitude of the input signal of the RF amplifier and the power supply voltage from the ET power supply as indexes.

US Patent Application Publication No. 2010/0295613

  However, in the technique described in Patent Document 1, when high-precision distortion compensation is performed, the memory amount of the lookup table becomes enormous and the circuit scale increases.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to improve the amplification efficiency of a wideband signal and appropriately perform distortion compensation in a simple configuration in a configuration employing a DPD and an envelope tracking method. An amplification device, a distortion compensation method, and a distortion compensation program that can be performed in the same manner are provided.

  (1) In order to solve the above-described problem, an amplification device according to an aspect of the present invention includes an amplifier, a distortion compensation unit that performs distortion compensation of the amplifier, and a power supply voltage or power supply current supplied to the amplifier as a target signal. A power supply that changes in accordance with an envelope tracking signal that is a signal corresponding to the envelope tracking signal, and the distortion compensator includes a plurality of variables that use the target signal and the power supply voltage, the power supply current, or the envelope tracking signal as variables. The distortion compensation is performed based on the linear combination of the first control functions.

  With such a configuration, it is possible to perform amplifier distortion compensation without using a two-dimensional lookup table that requires a large storage capacity. Therefore, in the configuration employing the envelope tracking method, the amplification efficiency of the wideband signal can be improved and the distortion compensation can be appropriately performed with a simple configuration.

  (2) Preferably, the distortion compensation unit includes the power supply voltage, the power supply current, or the envelope tracking signal at a distortion compensation timing that is a timing for performing the distortion compensation, and the target signal at a timing corresponding to the distortion compensation timing. The distortion compensation is performed on the basis of a linear combination of the plurality of first control functions, where.

  With such a configuration, distortion compensation of the amplifier can be performed based on the target signal at an appropriate timing and the power supply voltage, power supply current, or envelope tracking signal.

  (3) More preferably, the distortion compensator is at least one of the target signal at a timing before the timing corresponding to the distortion compensation timing and the target signal at a timing after the timing corresponding to the distortion compensation timing. The distortion compensation is performed based on a linear combination of the plurality of first control functions, one of which is further a variable.

  With such a configuration, distortion compensation of the amplifier can be appropriately performed even in the case where a memory effect in which the amplification characteristic of the amplifier changes depending on the difference or direction of the change of the target signal with time is generated in the amplifier.

  (4) More preferably, the distortion compensator includes the power supply voltage at a timing before the distortion compensation timing, the power supply current or the envelope tracking signal, and the power supply voltage at a timing after the distortion compensation timing, The distortion compensation is performed on the basis of a linear combination of the plurality of first control functions having at least one of a power supply current and the envelope tracking signal as a variable.

  With such a configuration, even when a memory effect occurs in the amplifier in which the amplification characteristic of the amplifier changes depending on the difference or direction of the temporal change in the power supply voltage, the power supply current, or the envelope tracking signal, the amplifier distortion compensation is appropriately performed. Can be done.

  (5) More preferably, the distortion compensator performs the distortion compensation at the distortion compensation timing to generate a distortion compensation signal from the target signal, and the distortion compensator includes the plurality of first control functions. The distortion compensation is performed based on the linear combination, the plurality of second control functions at the plurality of distortion compensation timings, and the distortion compensation signal at the plurality of distortion compensation timings, and the plurality of second control functions are The output signal output from the amplifier at a timing corresponding to the timing of the target signal, and the power supply voltage, the power supply current, or the envelope tracking signal are used as variables.

  With such a configuration, it is possible to grasp the amplification distortion characteristics of the amplifier based on the distortion compensation signal, the output signal, the power supply voltage, the power supply current, or the envelope tracking signal.

  Thereby, it is possible to generate a distortion compensation signal capable of appropriately compensating for the amplification distortion characteristics of the amplifier.

  (6) More preferably, the distortion compensation unit is configured to linearly combine the plurality of second control functions such that a difference between the distortion compensation signal and the linear combination of the plurality of second control functions satisfies a predetermined condition. The coefficient of the second control function in is determined, and the coefficient is set as the coefficient of the first control function in the linear combination of the plurality of first control functions.

  With such a configuration, the coefficient of the first control function in the linear combination of the first control functions can be specifically and appropriately determined.

  (7) Preferably, the amplifying apparatus further includes a correspondence between the power supply voltage, the power supply current, or the envelope tracking signal and a coefficient of the first control function in a linear combination of the plurality of first control functions. A storage unit that stores the relationship, and the distortion compensation unit obtains, from the storage unit, the coefficient corresponding to the power supply voltage, the power supply current, or the envelope tracking signal at a distortion compensation timing that is a timing for performing the distortion compensation. Then, a linear combination of the plurality of first control functions is generated based on the coefficient, and the distortion compensation is performed based on the linear combination of the plurality of first control functions.

  With such a configuration, the contribution of the power supply voltage, the power supply current, or the envelope tracking signal to distortion compensation can be included in the coefficient, so the load of calculation processing in the distortion compensation unit can be reduced.

  Further, for example, when the amplifying apparatus is started, distortion compensation can be started immediately without determining the coefficient in the amplifying apparatus.

  (8) Preferably, the first control function and the second control function are Gaussian functions.

  As described above, the configuration using the Gaussian function as the kernel function makes it possible to appropriately and quantitatively calculate the degree of similarity between the two vectors having the first control function and the second control function as arguments.

  (9) In order to solve the above problems, a distortion compensation method according to an aspect of the present invention changes a power supply voltage or a power supply current supplied to an amplifier according to an envelope tracking signal that is a signal corresponding to an envelope of a target signal. And a step of performing distortion compensation of the amplifier based on a linear combination of a plurality of first control functions having the target signal and the power supply voltage, the power supply current, or the envelope tracking signal as variables. .

  With such a configuration, it is possible to perform amplifier distortion compensation without using a two-dimensional lookup table that requires a large storage capacity. Therefore, in the configuration employing the envelope tracking method, the amplification efficiency of the wideband signal can be improved and the distortion compensation can be appropriately performed with a simple configuration.

  (10) In order to solve the above-described problem, a distortion compensation program according to an aspect of the present invention is a distortion compensation program used in an amplifying apparatus, and supplies a power supply voltage or a power supply current supplied to an amplifier to a computer as a target signal. A plurality of first control functions having a variable of the target signal and the power supply voltage, the power supply current, or the envelope tracking signal. And a step of performing distortion compensation of the amplifier based on coupling.

  With such a configuration, it is possible to perform amplifier distortion compensation without using a two-dimensional lookup table that requires a large storage capacity. Therefore, in the configuration employing the envelope tracking method, the amplification efficiency of the wideband signal can be improved and the distortion compensation can be appropriately performed with a simple configuration.

  According to the present invention, in the configuration employing the DPD and the envelope tracking method, the amplification efficiency of the wideband signal can be improved and the distortion compensation can be appropriately performed with a simple configuration.

It is a figure which shows the structure of the amplifier which concerns on the 1st Embodiment of this invention. It is a figure for demonstrating the problem of the broadening of an envelope signal. It is a figure for demonstrating the narrowing of an envelope signal. It is a figure which shows the power supply voltage control by ET power supply in the amplifier which concerns on the 1st Embodiment of this invention. It is a figure which shows an example of the distortion characteristic of amplification in the power amplifier which concerns on the 1st Embodiment of this invention. It is a figure which shows the relationship between the level of the distortion compensation signal u [n] and the level of the output signal y [n] in the amplifier which concerns on the 1st Embodiment of this invention. It is a figure which shows the relationship between the level of the baseband signal x [n] in the amplifier which concerns on the 1st Embodiment of this invention, and the level of the distortion compensation signal u [n]. It is a figure which shows the relationship between the level of the baseband signal x [n] in the amplifier which concerns on the 1st Embodiment of this invention, and the level of the output signal y [n]. It is a figure which shows a specific example of a two-dimensional lookup table. It is a figure which shows an example of the timing relationship of each signal in the amplifier which concerns on the 1st Embodiment of this invention. It is a figure which shows another example of the timing relationship of each signal in the amplifier which concerns on the 1st Embodiment of this invention. It is the flowchart which defined the operation | movement procedure at the time of the amplification apparatus which concerns on the 1st Embodiment of this invention determines a coefficient in a learning phase. It is a figure which shows an example of the learning data which the amplifier which concerns on the 1st Embodiment of this invention accumulate | stores in a learning phase. 5 is a flowchart defining an operation procedure when the amplifying apparatus according to the first embodiment of the present invention performs distortion compensation in the DPD calculation phase. 3 is a flowchart defining an operation procedure when the amplifying apparatus according to the first embodiment of the present invention performs distortion compensation. It is a figure which shows the structure of the modification of the amplifier which concerns on the 1st Embodiment of this invention. It is a figure which shows the structure of the modification of the amplifier which concerns on the 1st Embodiment of this invention. It is a figure which shows the structure of the modification of the amplifier which concerns on the 1st Embodiment of this invention. It is a figure which shows the structure of the modification of the amplifier which concerns on the 1st Embodiment of this invention. It is a figure which shows an example of a structure of the amplifier which concerns on the 2nd Embodiment of this invention. It is a figure which shows an example of the table which the memory | storage part in the amplifier which concerns on the 2nd Embodiment of this invention memorize | stores.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals and description thereof will not be repeated.

<First Embodiment>
[Configuration and basic operation]
FIG. 1 is a diagram showing a configuration of an amplifying apparatus according to the first embodiment of the present invention.

  Referring to FIG. 1, an amplifying apparatus 201 includes a distortion compensation unit 101, an ET power supply 111, a power amplifier (amplifier) 121, a narrowband ET signal generation unit 152, a digital analog converter (DAC) 154, and an up-converter. A converter 155, a coupler 156, an oscillator 157, a down converter 158, an analog / digital converter (ADC) 159, and an attenuator 160 are provided.

  When the narrowband ET signal generation unit 152 receives a target signal, for example, a baseband signal x [n] that is a digital signal to be amplified by the power amplifier 121, the narrowband ET signal generation unit 152 generates a digital signal corresponding to the envelope of the baseband signal x [n]. An envelope tracking signal V [n] is generated.

  Here, the signal will be described. In the amplification device 201, a digital signal and an analog signal are processed. When the signal is a digital signal, hereinafter, the signal may be represented as * [n]. * The index n in [n] represents the order of the digital signals generated according to the sampling interval of the digital signals, that is, the clock period T. Specifically, for example, the first generated digital signal is represented by * [1], and the fifth generated digital signal is represented by * [5].

  Further, when the signal is an analog signal, the signal may be represented as * (t) below. * The argument t in (t) represents time.

  Note that “*” in the description of the digital signal and the analog signal means a wild card. Specifically, it is transmitted / received between the amplification device 201 and another device and between each functional block in the amplification device 201. Represents a code for identifying a signal.

  FIG. 2 is a diagram for explaining the problem of widening the envelope signal.

  More specifically, referring to FIG. 2, when the band of the target signal to be amplified by the amplifier is 20 MHz, for example, the envelope indicating the voltage level of this signal has a relatively gentle waveform, and the ET power supply 111 Can follow this envelope.

  Here, as compared with the case where the ET method is not employed, that is, the case where the power supply voltage of the amplifier is fixed, the efficiency corresponding to the shaded portion in the figure can be realized by employing the ET method.

  On the other hand, when the band of the target signal is a wide band, for example, 100 MHz, the envelope indicating the voltage level of this signal has a steep waveform, and it becomes difficult for the ET power source to follow the envelope.

  FIG. 3 is a diagram for explaining the narrowing of the envelope signal.

  Referring to FIG. 3, amplifying apparatus 201 generates a signal corresponding to the original envelope, more specifically, envelope tracking signal V [n] obtained by narrowing the band of the original envelope. Envelope tracking is made easier.

  More specifically, referring to FIG. 1 again, the amplifying apparatus 201 adopts the ET method. The narrowband ET signal generation unit 152 detects Sqrt (XI [n] ^ 2 + XQ [n] ^ 2) that is the amplitude of the target signal, for example, the baseband signal x [n], that is, the voltage value, and the time of the detected voltage value An envelope signal EV [n] indicating an envelope or envelope that is a change is generated. Here, “Sqrt (*)” represents calculating the square root of *. Further, “* ^ 2” represents calculating the square of *.

  The envelope signal EV [n] indicates, for example, the positive envelope in FIG. The envelope signal EV [n] is a digital signal and is equal to the envelope of the baseband signal x [n].

  Then, the narrowband ET signal generation unit 152 generates an envelope tracking signal V [n] that is a signal obtained by narrowing the envelope signal EV [n] and does not fall below the level of the envelope signal EV [n]. Generate.

  The narrowband ET signal generation unit 152 outputs the generated envelope tracking signal V [n] to the ET power supply 111 and the distortion compensation unit 101. Note that the timing at which the narrowband ET signal generation unit 152 outputs the envelope tracking signal V [n] to the ET power supply 111 and the distortion compensation unit 101 will be described later.

  The ET power supply 111 supplies power to the power amplifier 121 based on the envelope tracking signal V [n] received from the narrowband ET signal generation unit 152. More specifically, the ET power supply 111 generates the power supply voltage VDD (t) or the power supply current IDD (t) corresponding to the envelope tracking signal V [n] at the timing when the envelope tracking signal V [n] is received.

  In this case, the level of the power supply voltage VDD (t) or the level of the power supply current IDD (t) is proportional to the level of the envelope tracking signal V [n], for example.

  The ET power supply 111 supplies the generated power supply voltage VDD (t) or power supply current IDD (t) to the power amplifier 121.

  There may be a delay between the time when the ET power supply 111 receives the envelope tracking signal V [n] and the time when the power supply voltage VDD (t) or the power supply current IDD (t) is supplied to the power amplifier 121. The delay in this case is sufficiently shorter than the clock cycle T.

  Hereinafter, to simplify the description, the timing at which the ET power supply 111 receives the envelope tracking signal V [n] and the ET power supply 111 supplies the power supply voltage VDD (t) or the power supply current IDD (t) to the power amplifier 121. The timing is the same timing.

  The distortion compensation unit 101 performs distortion compensation of the power amplifier 121 by performing DPD (Digital Pre-Distortion), for example. More specifically, the distortion compensation unit 101 generates a distortion compensation signal u [n] based on the baseband signal x [n] and the envelope tracking signal V [n] received from the narrowband ET signal generation unit 152. To do.

  That is, the distortion compensation unit 101 generates a distortion compensation signal u [n] by two-dimensional DPD using the baseband signal x [n] and the envelope tracking signal V [n] as variables.

  The distortion compensation unit 101 outputs the generated distortion compensation signal u [n] to the digital / analog converter 154. Details of the distortion compensation signal u [n] will be described later.

  The digital-analog converter 154 converts the distortion compensation signal u [n] received from the distortion compensation unit 101 into an analog signal, and outputs the analog signal to the up-converter 155.

  The up-converter 155 multiplies the analog signal received from the digital-analog converter 154 by the RF-band local signal generated by the oscillator 157 to up-convert the analog signal into an RF-band signal, and the up-converted signal Output to the power amplifier 121.

  FIG. 4 is a diagram showing power supply voltage control by the ET power supply in the amplifying apparatus according to the first embodiment of the present invention.

  Referring to FIG. 4, power amplifier 121 includes, for example, an N-channel MOSFET (Metal Oxide Semiconductor Field Effect Transistor), and ET power supply 111 receives envelope tracking signal V [n] received from narrowband ET signal generation unit 152. Based on the level, the power supply voltage VDD (t) or the power supply current IDD (t) supplied to the drain of the N-channel MOSFET is changed. For example, the ET power supply 111 changes the level of the drain voltage Vd between, for example, Vd0 to Vd5.

  Referring again to FIG. 1, power amplifier 121 amplifies the distortion compensation signal from up-converter 155 converted to an analog signal in the RF band, and signal Y (t), which is the amplified analog signal, to coupler 156. Output.

  Coupler 156 outputs signal Y (t) received from power amplifier 121 to, for example, an antenna (not shown). At this time, the coupler 156 outputs a signal Y (t) having a level corresponding to the amount of coupling to the attenuator 160.

  Attenuator 160 attenuates signal Y (t) received from coupler 156 and outputs the attenuated signal to down converter 158. At this time, the attenuator 160 attenuates the signal Y (t) with an attenuation factor of 1 / k that is the reciprocal of the amplification factor k of the power amplifier 121. That is, attenuator 160 outputs a signal obtained by attenuating signal Y (t) received from coupler 156 to 1 / k to down converter 158.

  The down converter 158 multiplies the signal Y (t) received from the attenuator 160 by the RF band local signal generated by the oscillator 157 to down-convert the signal Y (t) into a baseband analog signal. And output to the analog-digital converter 159.

  The analog-digital converter 159 converts the analog signal received from the down converter 158 into a digital signal, and outputs an output signal y [n] that is the converted digital signal to the distortion compensation unit 101.

  The distortion compensation unit 101 accumulates, for example, an envelope tracking signal V [n], a distortion compensation signal u [n], and an output signal y [n] as learning data. Then, the distortion compensation unit 101 constructs a learning network that calculates an output with respect to a two-dimensional input, that is, a two-dimensional DPD model based on, for example, learning data.

[Amplification characteristics]
FIG. 5 is a diagram illustrating an example of distortion characteristics of amplification in the power amplifier according to the first embodiment of the present invention.

  Referring to FIG. 5, the horizontal axis indicates the level of signal U (t) received by power amplifier 121, and the vertical axis indicates the level of signal Y (t) output from power amplifier 121. When the signal U (t) received by the power amplifier 121 is efficiently amplified by the power amplifier 121, the relationship between the level of U (t) and the level of the signal Y (t) output from the power amplifier 121 is generally nonlinear. It becomes a relationship.

  In addition, this non-linear relationship changes according to the power supply voltage VDD (t) or the power supply current IDD (t) supplied from the ET power supply 111 to the power amplifier 121. For this reason, the level of the signal Y (t) is represented by the function F (U (t), VDD (t)) or the function F (U (t), IDD (t)).

  Hereinafter, a case where the ET power supply 111 supplies the power supply voltage VDD (t) to the power amplifier 121 will be described. The same argument can be applied to the case where the ET power supply 111 supplies the power supply current IDD (t) to the power amplifier 121.

  For example, when the power supply voltage VDD (t) is 20, the relationship between the level of U (t) and the level of Y (t) is expressed by a function F (U (t), 20) shown in FIG. . The relationship between the level of U (t) and the level of Y (t) when the power supply voltage VDD (t) is 30 is expressed by a function F (U (t), 30).

  Thus, the nonlinear relationship between the level of the signal U (t) received by the power amplifier 121 and the level of the signal Y (t) output from the power amplifier 121 is referred to as amplification distortion characteristics in the power amplifier 121.

  The distortion compensation unit 101 compensates for amplification distortion characteristics in the power amplifier 121 by performing, for example, two-dimensional DPD.

  FIG. 6 is a diagram illustrating a relationship between the level of the distortion compensation signal u [n] and the level of the output signal y [n] in the amplifying apparatus according to the first embodiment of the present invention.

  Referring to FIG. 6, output signal y [n] is a digital signal in the baseband of a signal obtained by multiplying signal Y (t) output from power amplifier 121 by 1 / k. The distortion compensation signal u [n] is a digital signal in the baseband that is the basis of the signal U (t) received by the power amplifier 121. The envelope tracking signal V [n] is a digital signal that is the basis of the power supply voltage VDD (t).

  Therefore, the function f (u [n], V [n]) indicating the relationship between the level of the distortion compensation signal u [n] and the level of the output signal y [n] is the function F (U (t (t)) shown in FIG. ), VDD (t)) is equivalent to a function obtained by multiplying 1 / k.

  Specifically, the function f (u [n], 20) and the function f (u [n], 30) are changed into the function F (U (t), 20) and the function F (U (t), 30). Each corresponds.

  The distortion compensation unit 101 generates the distortion compensation signal u [n] based on, for example, the inverse characteristics of the function f (u [n], V [n]).

  FIG. 7 is a diagram showing a relationship between the level of the baseband signal x [n] and the level of the distortion compensation signal u [n] in the amplifying apparatus according to the first embodiment of the present invention.

  Referring to FIG. 7, the function g (x [n], V [n]) indicating the relationship between the level of the baseband signal x [n] and the level of the distortion compensation signal u [n] is shown in FIG. This corresponds to the inverse function of the function f (u [n], V [n]).

  Specifically, the function g (x [n], 20) and the function g (x [n], 30) are the same as the function f (u [n], 20) and the function f (u [n], 30). Each corresponds to an inverse function.

  For example, when receiving the baseband signal x [n] and the envelope tracking signal V [n], the distortion compensation unit 101 generates the distortion compensation signal u [n] based on the function g (x [n], V [n]). Generate.

  In other words, the baseband signal x [n] is converted by the two-dimensional DPD in the distortion compensator 101 into a distortion compensation signal u [n] given an inverse characteristic corresponding to the envelope tracking signal V [n].

  Specifically, when the index n is a, the baseband signal x [a] is compared with the level of the baseband signal x [a] by the two-dimensional DPD in the distortion compensation unit 101 when V [a] is 30. For example, it is converted into a distortion compensation signal u [a] that is larger by Δu shown in FIG.

  Referring to FIG. 5 again, when distortion compensation signal u [a] is converted into signal U (t = b), which is an analog signal in the RF band, at time t = b, power amplifier 121 receives signal U ( b) is amplified.

  The signal U (t = b) is larger than Ubase, which is a level corresponding to the level of the baseband signal x [a], by ΔU, which is a level corresponding to Δu.

  The power amplifier 121 generates the signal Y (b) by amplifying the signal U (b). The signal Y (b) corresponds to a level obtained by multiplying Ubase, which is a level corresponding to the level of the signal of the baseband signal x [a], by k.

  In other words, the power amplifier 121 amplifies a signal U (t) corresponding to the distortion compensation signal u [n], thereby a signal Y (t) obtained by multiplying the signal corresponding to Ubase, that is, the baseband signal x [n] by k. And the generated signal Y (t) is output to the coupler 156.

  FIG. 8 is a diagram illustrating a relationship between the level of the baseband signal x [n] and the level of the output signal y [n] in the amplifying apparatus according to the first embodiment of the present invention.

  Referring to FIG. 8, the level of output signal y [n] is a signal obtained by multiplying Ubase, that is, a signal Y (t) having a level obtained by multiplying a signal corresponding to baseband signal x [n] by 1 / k. It is a digital signal in the baseband. Therefore, the relationship between the level of the baseband signal x [n] and the level of the output signal y [n] is y [n] = x [n], that is, a linear relationship.

  This is because the baseband signal x [n] is converted by the two-dimensional DPD in the distortion compensator 101 into a distortion compensation signal u [n] having an inverse characteristic corresponding to the envelope tracking signal V [n]. This means that the relationship between the level of the baseband signal x [n] and the level of the signal Y (t) can be a linear relationship.

[About 2D lookup table]
FIG. 9 is a diagram showing a specific example of a two-dimensional lookup table.

  Referring to FIG. 9, in the technique described in Patent Document 1, distortion compensation is performed using a two-dimensional lookup table using the input signal amplitude x and the power supply voltage VDD supplied from the ET power supply to the amplifier as indexes. .

  Specifically, the predistortion unit 20 described in Patent Document 1 calculates xp23 from the two-dimensional lookup table when, for example, the amplitude x of the input signal is 3 and the power supply voltage VDD from the ET power supply is 30 volts. read out.

  Then, the predistortion unit 20 distorts the signal received from the baseband processor 21 based on the read xp23. Thereby, the distortion characteristic in the RF power amplifier 10 is compensated.

  However, in such a configuration, when high-precision distortion compensation is performed, there is a problem that the memory amount of the two-dimensional lookup table becomes enormous and the circuit scale increases.

  Therefore, the amplification device 201 according to the embodiment of the present invention enables highly accurate distortion compensation without increasing the circuit scale by the following configuration and operation.

[Generation of distortion compensation signal]
The distortion compensator 101 compensates for distortion of the power amplifier 121 based on a linear combination of a plurality of first control functions with the baseband signal x [n] as the target signal and the envelope tracking signal V [n] as variables. I do.

  The first control function is, for example, a Gaussian function. Note that the first control function may be a non-linear function such as a power series function or a sigmoid function. Further, when evaluating the value of the Gaussian function, the value may be evaluated using a one-dimensional lookup table. Thereby, the calculation load required for the evaluation of the value can be reduced.

For example, a variable vector x [n] composed of a baseband signal x [n] and an envelope tracking signal V [n], which are variables, is defined by the following equation (1).

  In the following, a vector or a matrix is indicated in bold in the expression, and “vector” or “matrix” is clearly indicated in a sentence other than the expression.

Here, the baseband signal x [n] will be described. The baseband signal x [n] is, for example, a complex number. When the baseband signal x [n] is a complex number, the baseband signal x [n] is represented by the following equation (2).

  Here, j is an imaginary unit. Further, xI [n] and xQ [n] are the amplitude of the I component and the Q component in the baseband signal, respectively. Note that the baseband signal x [n] may be a real number.

  The index n in the envelope tracking signal V [n] means that the envelope tracking signal corresponds to the envelope signal EV [n] indicating the envelope of the baseband signal x [n] having the index n.

The linear combination of, for example, KI first control functions having the variable vector x [n] as a variable is expressed by, for example, the following expression (3) for the I component.

  Here, the vector tIk is a predetermined vector and has the same length and the same dimensions as the variable vector x [n]. The component of the vector tIk is, for example, a complex number. The vector tIk represents the position where the k-th first control function in the linear combination is arranged.

  The vector tIk is set to be the same as the variable vector x [pI] having an arbitrary index pI, for example. Further, the vector tIk may be set to an arbitrary vector.

  ΣIk is a predetermined scalar. A value obtained by multiplying σIk by 2Sqrt (2ln2) represents the full width at half maximum of the k-th first control function in the linear combination. σIk may be set to an arbitrary value. Here, “ln2” represents the natural logarithm of 2.

  CIk is a scalar determined by a procedure described later. cIk represents a coefficient of the k-th first control function in the linear combination.

The linear combination of, for example, KQ first control functions having the variable vector x [n] as a variable is expressed by, for example, the following expression (4) for the Q component.

  Here, the vector tQk is a predetermined vector and has the same length and the same dimension as the variable vector x [n]. The component of the vector tQk is, for example, a complex number. The vector tQk represents a position where the kth first control function in the linear combination is arranged.

  The vector tQk is set to be the same as the variable vector x [qI] having an arbitrary index qI, for example. The vector tQk may be set to an arbitrary vector.

  ΣQk is a predetermined scalar. A value obtained by multiplying σQk by 2Sqrt (2ln2) represents the full width at half maximum of the k-th first control function in the linear combination. σQk may be set to an arbitrary value.

  Further, cQk is a scalar determined by a procedure described later. cQk represents a coefficient of the kth first control function in the linear combination.

  Note that the first control function is also called a kernel function. A kernel function quantifies the relationship between two vectors.

  Specifically, for example, in Equation (3), when the square of the magnitude of the difference vector between the variable vector x [n] and the vector tIk is smaller than σIk, the kernel function gives a value close to 1. In other words, the kernel function gives a value close to 1 when the variable vector x [n] and the vector tIk are similar. Here, “similar” means that the directions and sizes of two vectors are not greatly different.

  Further, when the square of the magnitude of the difference vector between the variable vector x [n] and the vector tIk is larger than σIk, the kernel function gives a value close to 0. In other words, the kernel function gives a value close to 0 when the variable vector x [n] and the vector tIk are not similar. Here, “not similar” means that at least one of the directions and sizes of two vectors is greatly different.

  By changing the magnitude of σIk, the change in the value of the kernel function with respect to the degree of similarity between the variable vector x [n] and the vector tIk can be adjusted.

The distortion compensation unit 101 generates a distortion compensation signal u [n] based on a linear combination of a plurality of first control functions having a variable vector x [n] as a variable. The distortion compensation signal u [n] is expressed by the following equation (5), for example.

  Here, j is an imaginary unit. The first and second terms on the right side of Equation (5) are the amplitude of the I component and the Q component of the distortion compensation signal, respectively. Note that the distortion compensation signal u [n] may be a real number.

  The index n in the distortion compensation signal u [n] means that the distortion compensation signal u [n] is generated in the distortion compensation unit 101 based on the variable vector x [n] having the index n.

  Equation (5) represents the inverse characteristic of the characteristic shown in FIG.

[Signal timing]
Here, the timing at which the distortion compensation signal u [n], the baseband signal x [n], the envelope tracking signal V [n], and the output signal y [n] are generated or output will be described.

  FIG. 10 is a diagram illustrating an example of a timing relationship of each signal in the amplifying apparatus according to the first embodiment of the present invention.

  Referring to FIG. 10, the horizontal axis indicates time t. The baseband signal x [n] is data of the clock period T received by the distortion compensator 101. Here, x [n−L2] to x [n + L1] are arranged in time series.

  The distortion compensation signal u [n] is data of the clock period T output from the distortion compensation unit 101 to the power amplifier 121. Here, u [n-2] to u [n + 2] are arranged in time series.

  The envelope tracking signal V [n] is data of the clock period T output from the narrowband ET signal generation unit 152, and here, V [n−M2] to V [n + M1] are arranged in time series.

  The output signal y [n] is data of the clock period T received by the distortion compensation unit 101 via the analog-digital converter 159, and here, y [n-2] to y [n + 2] are arranged in time series.

  The distortion compensation unit 101 receives the generated distortion compensation signal u [n] after a calculation time TCalc necessary for generating the distortion compensation signal u [n] has elapsed after receiving the baseband signal x [n]. Output to the amplifier 121.

  Here, the timing at which the distortion compensation unit 101 outputs the generated distortion compensation signal u [n] to the power amplifier 121 is defined as a distortion compensation timing TComp that is a timing for performing distortion compensation.

  The timing before the distortion compensation timing TComp when the distortion compensation unit 101 receives the baseband signal x [n] is defined as a timing corresponding to the distortion compensation timing, that is, a distortion compensation corresponding timing TCorr.

  The narrowband ET signal generation unit 152 outputs the envelope tracking signal V [n] to the ET power supply 111 at the distortion compensation timing TComp.

  The ET power supply 111 receives the envelope tracking signal V [n] output from the narrowband ET signal generation unit 152 at the distortion compensation timing TComp, and the power supply voltage VDD (t) or the power supply current corresponding to the envelope tracking signal V [n]. IDD (t) is output to the power amplifier 121.

  In addition, the narrowband ET signal generation unit 152 generates an envelope tracking signal V [m] having an index m in a range from (n−M2) to (n + M1), for example, at a timing before the distortion compensation timing TComp. 101. Here, M1 and M2 are integers of zero or more.

  The distortion compensation unit 101 uses the envelope tracking signal V [m] having an index m in the range from (n−M2) to (n + M1) when generating the distortion compensation signal u [n].

  Further, the distortion compensation unit 101 receives the output signal y [n] in the baseband corresponding to the signal Y (t) output from the power amplifier 121 via the analog-digital converter 159 at the distortion compensation timing TComp.

[Index n of variable contained in variable vector x [n]]
The variable vector x [n] may follow the definitions shown in the following formulas (6), (7), and (8) in addition to the definitions shown in formula (1).

In the variable vector x [n], for example, the baseband signal x [m] at the pre-distortion-corresponding timing TPreCorr before the distortion compensation-corresponding timing TCorr, and the base at the post-distortion-corresponding timing TPostCorr after the distortion compensation-corresponding timing Tcorr. At least one of the band signals x [m] is further set as a variable. Specifically, the variable vector x [n] is defined by the following formula (6), for example.

  Here, L1 and L2 are integers greater than or equal to zero and never become zero at the same time.

  A baseband signal x [m] having an index m in the range from (n-1) to (n-L2) corresponds to the baseband signal x [m] at the timing TPreCorr before distortion compensation as shown in FIG. To do.

  Further, the baseband signal x [m] having the index m in the range from (n + 1) to (n + L1) corresponds to the baseband signal x [m] at the timing TPostCor after distortion compensation as shown in FIG.

In the variable vector x [n], for example, an envelope tracking signal V [m] at a pre-distortion timing TPreComp before the distortion compensation timing TComp and an envelope tracking signal at a post-distortion compensation timing TPostComp after the distortion compensation timing TComp. At least one of V [m] is further set as a variable. Specifically, the variable vector x [n] is defined by the following equation (7), for example.

  Here, M1 and M2 are integers greater than or equal to zero and never become zero at the same time.

  The envelope tracking signal V [m] having an index m in the range from (n−1) to (n−M2) corresponds to the envelope tracking signal V [m] at the pre-distortion compensation timing TPreComp as shown in FIG. .

  Further, the envelope tracking signal V [m] having an index m in the range from (n + 1) to (n + M1) corresponds to the envelope tracking signal V [m] at the distortion-compensated timing TPostComp as shown in FIG.

In the variable vector x [n], for example, the variable added in Expression (6) and the variable added in Expression (7) are further variables. Specifically, the variable vector x [n] is defined by the following equation (8), for example.

  Here, L1 and L2 are integers greater than or equal to zero and never become zero at the same time.

  Further, M1 and M2 are integers equal to or greater than zero, and never become zero at the same time.

  In addition, since the calculation load increases as the dimension of the variable vector x [n] increases, dimensional compression may be performed by, for example, principal component analysis.

  Further, the dimension of the variable vector x [n] may be reduced by compressing the component of the variable vector x [n] into a linear prediction coefficient or the like.

[Determination of coefficient of first control function in linear combination]
The distortion compensation unit 101 includes a linear combination of a plurality of first control functions shown in Expression (5), a plurality of second control functions at a plurality of distortion compensation timings, and a distortion compensation signal at a plurality of distortion compensation timings. Based on this, distortion compensation is performed.

  More specifically, the distortion compensation unit 101 performs the following processing based on the envelope tracking signal V [n], the distortion compensation signal u [n], and the output signal y [n] included in the accumulated learning data.

  That is, the distortion compensation unit 101 determines the coefficient cIk and the coefficient cQk in Expression (5) based on a plurality of second control functions at a plurality of distortion compensation timings and distortion compensation signals at a plurality of distortion compensation timings.

  Then, the distortion compensation unit 101 constructs a two-dimensional DPD model by substituting the determined coefficient cIk and coefficient cQk into the equation (5). The distortion compensation unit 101 performs distortion compensation using the constructed two-dimensional DPD model.

  Hereinafter, a procedure for determining the coefficient cIk in the procedure for determining the coefficient cIk and the coefficient cQk will be described. For the coefficient cQk, the coefficient cQk can be determined in the same manner as the coefficient cIk by replacing “I” with “Q” in the procedure.

  For example, u [n], u [n−1],..., U [n−P] are distortion compensation signals at a plurality of distortion compensation timings. Here, P is a non-zero integer.

  When P is positive, for example, the distortion compensation signals at a plurality of distortion compensation timings represent (| P | +1) distortion compensation signals before u [n].

  When P is negative, for example, the distortion compensation signals at a plurality of distortion compensation timings represent (| P | +1) distortion compensation signals after u [n].

When the distortion compensation signal u [n] is, for example, a complex number, U [n] is defined by the following equation (9) that generalizes the equation (5).

  Here, j is an imaginary unit. UI [n] and uQ [n] are the I component of the distortion compensation signal u [n] and the Q component of the distortion compensation signal u [n], respectively.

A vector uI tilde whose component is the I component of the distortion compensation signal u [n] at (| P | +1) distortion compensation timings is defined by the following equation (10), for example.

  Here, T in Equation (10) represents transposition of a vector.

  The plurality of second control functions are the same Gaussian functions as the first control function. The plurality of second control functions output corresponding to the signal Y (t) output from the power amplifier 121 at a timing corresponding to the timing of the baseband signal x [n] included in the variable vector x [n]. The envelope tracking signal V [n] included in the signal y [n] and the variable vector x [n] is a variable.

  FIG. 11 is a diagram illustrating another example of the timing relationship of each signal in the amplifying apparatus according to the first embodiment of the present invention.

  Referring to FIG. 11, the horizontal axis indicates time t. The baseband signal x [n], the envelope tracking signal V [n], and the output signal y [n] are arranged in a time series defined in FIG.

  For example, when the variable vector x [n] conforms to the definition in Expression (8), the timing corresponding to the timing of the baseband signal x [n] included in the variable vector x [n] and the baseband signal x [n]. The correspondence relationship between the output signal y [n] corresponding to the signal Y (t) output from the power amplifier 121 is as follows with respect to timing.

  That is, for example, the baseband signal x [n−L2] at the timing of (L2) clock cycles T before the distortion compensation corresponding timing TCorr is the timing of (L2) clock cycles T before the distortion compensation timing TComp. Corresponds to the output signal y [n−L2].

  Further, for example, the baseband signal x [n] at the distortion compensation corresponding timing TCorr corresponds to the output signal y [n] at the distortion compensation timing TComp.

  Further, for example, the baseband signal x [n + L1] at a timing after (L1) clock cycles T from the distortion compensation timing TCorr is output at a timing after (L1) clock cycles T from the distortion compensation timing TComp. This corresponds to the signal y [n + L1].

  For example, when the variable vector x [n] conforms to the definition in Expression (8), the envelope tracking signal V [n] included in the variable vector x [n] has an index m of (n−M2) to (n + M1). ) (M1 + M2 + 1) envelope tracking signals V [m].

  For example, the timing at which the narrowband ET signal generation unit 152 outputs the envelope tracking signal V [n−M2] to the ET power supply 111 corresponds to the timing of (M2) clock cycles T before the distortion compensation timing TComp.

  For example, the timing at which the narrowband ET signal generation unit 152 outputs the envelope tracking signal V [n] to the ET power supply 111 corresponds to the distortion compensation timing TComp.

  For example, the timing at which the narrowband ET signal generation unit 152 outputs the envelope tracking signal V [n + M1] to the ET power supply 111 corresponds to the timing after (M1) clock cycles T from the distortion compensation timing TComp.

  That is, the plurality of second control functions have the output signal y [n] and the envelope tracking signal V [n] at the timing as variables.

  In other words, the plurality of second control functions are included in the output signal y [n] having the same index n as the baseband signal x [n] included in the variable vector x [n], and the variable vector x [n]. The envelope tracking signal V [n] to be used is a variable.

Specifically, the plurality of second control functions have the output signal y [n] and the envelope tracking signal V [n] as variables when the variable vector x [n] is defined by Expression (1). . A variable vector y [n] constituted by the variables is defined by the following equation (11), for example.

Here, the output signal y [n] will be described. The output signal y [n] is, for example, a complex number. When the output signal y [n] is a complex number, the output signal y [n] is represented by the following equation (12).

  Here, j is an imaginary unit. Further, yI [n] and yQ [n] are the amplitudes of the I component and Q component in the output signal, respectively. Note that the output signal y [n] may be a real number.

When the variable vector x [n] is defined by the equation (6), the variable vector y [n] is defined by the following equation (13), for example.

  Here, L1 and L2 follow the definition in Equation (6).

When the variable vector x [n] is defined by the equation (7), the variable vector y [n] is defined by the following equation (14), for example.

  Here, M1 and M2 follow the definition in Equation (7).

When the variable vector x [n] is defined by the equation (8), the variable vector y [n] is defined by the following equation (15), for example.

  Here, L1, L2, M1, and M2 follow the definition in Expression (8).

The plurality of second control functions at (| P | +1) distortion compensation timings are expressed by, for example, the following expression (16) for the I component.

  Here, the vectors tIk and σIk in the equation (3) are used as the vectors tIk and σIk. In addition, in the Q components in the plurality of second control functions at (| P | +1) distortion compensation timings, the vectors tQk and σQk use the vectors tQk and σQk in Expression (4).

  The distortion compensation unit 101 determines that a difference between a distortion compensation signal at (| P | +1) distortion compensation timings and a linear combination of a plurality of second control functions at (| P | +1) distortion compensation timings is a predetermined condition. The coefficient of the second control function in the linear combination of the plurality of second control functions is determined so as to satisfy the above.

  Then, the distortion compensation unit 101 sets the determined coefficient as the coefficient of the first control function in the linear combination of the plurality of first control functions.

More specifically, the I component of the linear combination of the plurality of second control functions is expressed by the following equation (17), for example.

  Here, KI in Equation (3) is used as KI. i is an integer from zero to P. The coefficient cIk tilde of the second control function in the linear combination is determined by the procedure described later.

  Note that the Q component of the linear combination of the plurality of second control functions is configured by KQ second control functions. As KQ in this case, KQ in Expression (4) is used.

The difference between the I component of the distortion compensation signal at (| P | +1) distortion compensation timing and the I component of the linear combination of the KI second control functions at (| P | +1) distortion compensation timing is predetermined. Satisfying the condition specifically corresponds to satisfying the following expression (18).

  Here, the difference corresponds to the difference between uI [n−i], which is the I component of the distortion compensation signal, and the I component of the linear combination of the plurality of second control functions shown in Expression (17).

  In Equation (18), the square of the absolute value of the difference indicates a square error, and i is (| P | +1) from zero to P. The argmin cIk tilde in equation (18) is the coefficient cIk tilde of the second control function so that the sum of the squares of the absolute values of (| P | +1) differences is minimum, that is, the square error is minimized. This means that a vector cI tilde whose components are the determined coefficient cIk tilde is generated.

  Here, the component of the vector cI tilde is the coefficient cIk of the first control function.

  In other words, the argmin cIk tilde means that the coefficient cIk tilde of the second control function is determined by the least square method, and a vector cI tilde whose components are the determined coefficient cIk tilde is generated.

Equation (18) is calculated by the following equation (19).

Here, the matrix K is a matrix of (| P | +1) rows KI columns. The component of p rows and k columns in the matrix K is expressed by the following equation (20). K + is a pseudo inverse matrix.

  Moreover, the following meaning is included in the calculation from the third line to the fourth line in Expression (19). That is, in the third row, the square of the vector magnitude, which is the difference between the vector uI tilde and the product of the matrix K and the vector cI tilde, is calculated. The square of the magnitude is expressed by a quadratic expression for KI coefficients cIk tilde.

  Therefore, when the quadratic expression has an extreme value, the square of the magnitude is minimized. More specifically, in the KI coefficient cIk tilde in which KI equations obtained by differentiating the quadratic equation with respect to each coefficient cIk tilde are simultaneously zero, the quadratic equation has an extreme value. Specifically, KI coefficients cIk tilde when the quadratic expression has an extreme value is uniquely calculated by the linear least square method by the expression on the fourth line in Expression (19).

  Also, the distortion compensation unit 101 increases the number of the plurality of distortion compensation timings, for example, more than the number of the plurality of first control functions. Specifically, (| P | +1) that is the number of the plurality of distortion compensation timings is set larger than KI that is the number of the plurality of first control functions.

  When (| P | +1) is smaller than KI, (| P | +1), which is the number of known numbers, is smaller than KI, which is the number of unknowns, and therefore KI coefficient cIk tilde can be uniquely determined. Disappear.

  Specifically, since there is no inverse matrix of the product of the transposed matrix of the matrix K in the fourth row and the matrix K in Equation (19), KI coefficient cIk tildes cannot be calculated.

  When (| P | +1) and KI have the same value, the number of known numbers (| P | +1) and the number of unknown numbers KI are the same, so KI coefficient cIk tilde. Can be calculated.

  In this case, the difference between the first term and the second term included in the absolute value in Equation (18) is zero when i takes any value from zero to P.

  For this reason, when the number of known numbers (| P | +1) and the number of unknown numbers KI have the same value, there is no point in performing the least squares method for minimizing the square error.

  Therefore, by making (| P | +1), which is the number of the plurality of distortion compensation timings, larger than KI, which is the number of the plurality of first control functions, the KI coefficient cIk tilde is appropriately set by the least square method. Can be calculated.

  The distortion compensator 101 performs the same calculation for KQ coefficients cQk tilde. Then, distortion compensation section 101 substitutes KI coefficient cIk tilde and KQ coefficient cQk tilde into KI coefficient cIk and KQ coefficient cQk shown in Equation (5), respectively.

  The distortion compensation unit 101 generates a distortion compensation signal u [n] based on Expression (5), and outputs the generated distortion compensation signal u [n] to the power amplifier 121, thereby obtaining a two-dimensional lookup table. Distortion compensation can be performed without using it.

  As a result, a configuration of the amplifying apparatus 201 that does not use a two-dimensional look-up table that requires an enormous memory capacity can be achieved, and thus the circuit scale in the amplifying apparatus 201 can be reduced.

  In addition, when the KI that is the number of the plurality of first control functions is smaller than the number of the plurality of distortion compensation timings (| P | +1), the distortion compensation unit 101 has a small number of first control functions. Since the distortion compensation signal u [n] can be generated based on the above, the calculation load when generating the distortion compensation signal u [n] can be reduced.

  Expression (18) is an expression for determining a coefficient value necessary for representing the characteristic shown in FIG. 6, that is, the distortion characteristic of amplification in the power amplifier 121 by the two-dimensional DPD.

  Expression (5) into which the value of the coefficient determined in this way is substituted can appropriately represent the characteristic shown in FIG. 7, that is, the inverse characteristic of the distortion characteristic of amplification in the power amplifier 121, by two-dimensional DPD.

  That is, the distortion compensation unit 101 can appropriately compensate for the distortion characteristics of amplification in the power amplifier 121 based on Expression (5) in which the coefficient determined by Expression (19) is substituted. ] Can be generated.

[Example of operation of amplification device in learning phase]
The amplification device according to the embodiment of the present invention reads and executes a program including each step of the following sequences from a memory (not shown). This program can be installed externally. The installed program is distributed in a state stored in a recording medium, for example.

  FIG. 12 is a flowchart defining an operation procedure when the amplification device according to the first embodiment of the present invention determines a coefficient in the learning phase.

  FIG. 13 is a diagram illustrating an example of learning data accumulated in the learning phase by the amplifying apparatus according to the first embodiment of the present invention.

  Referring to FIG. 12, first, distortion compensator 101 accumulates learning data necessary for determining coefficient cIk and coefficient cQk of the first control function shown in Expression (5) in the learning phase (step) S102).

  Specifically, the distortion compensation unit 101 accumulates, for example, the distortion compensation signal u [n], the output signal y [n], and the envelope tracking signal V [n] illustrated in FIG.

  Next, the distortion compensation unit 101 generates a variable vector y [n] based on the accumulated output signal y [n] and the envelope tracking signal V [n]. Then, the distortion compensation unit 101 substitutes the accumulated (| P | +1) number of distortion compensation signals u [n] and the variable vector y [n] into the equation (18), thereby calculating the coefficient of the second control function. The cIk tilde and the coefficient cQk tilde are determined (step S104).

  Next, the distortion compensation unit 101 substitutes the coefficient cIk tilde and the coefficient cQk tilde of the second control function into the coefficient cIk and the coefficient cQk of the first control function in the equation (5), so that the first control function The coefficient is updated (step S106).

  With the above operation, the distortion compensation unit 101 determines the coefficient of the first control function in the learning phase.

  When the amplification device 201 determines a coefficient while actually operating, for example, an appropriate distortion compensation signal is generated in the range of the envelope tracking signal V [n] and the distortion compensation signal u [n] that the amplification device 201 does not use much. You may not be able to.

  In contrast, in step S102, the distortion compensation unit 101 sets, for example, the envelope tracking signal V [n] to a predetermined value, and changes the distortion compensation signal u [n] while changing the corresponding output signal y. Receive [n].

  Specifically, when the envelope tracking signal V [n] is set to 20, for example, the distortion compensation unit 101 responds while changing the distortion compensation signal u [n] to 1.1 and 2.3. The output signals y [n] 1 and 2 are received.

  Next, when the envelope tracking signal V [n] is set to 30, for example, the distortion compensation unit 101 changes the distortion compensation signal u [n] to 1.2 and 2.4, and the corresponding output signal. Receive 1 and 2 which are y [n].

  At this time, for example, the distortion compensation unit 101 accumulates learning data in the range of the envelope tracking signal V [n] and the distortion compensation signal u [n] in a range that can be used in the own amplification device 201, for example.

  Accordingly, the coefficient cIk and the coefficient cQk of the first control function that can appropriately perform distortion compensation are determined regardless of the values of the envelope tracking signal V [n] and the distortion compensation signal u [n]. it can.

[Example of operation of amplification device in DPD calculation phase]
FIG. 14 is a flowchart defining an operation procedure when the amplifying apparatus according to the first embodiment of the present invention performs distortion compensation in the DPD calculation phase.

  Referring to FIG. 14, first, narrowband ET signal generation section 152 receives baseband signal x [n], and generates envelope tracking signal V [n] based on baseband signal x [n] ( Step S202).

  Next, the ET power supply 111 supplies the power supply voltage VDD (t) or the power supply current IDD (t) to be supplied to the power amplifier 121 according to the envelope tracking signal V [n] generated by the narrowband ET signal generation unit 152. Change (step S204).

  Next, the distortion compensation unit 101 generates the distortion compensation signal u [n] based on the baseband signal x [n] and the envelope tracking signal V [n] generated by the narrowband ET signal generation unit 152, for example. (Step S206).

  Next, the distortion compensation unit 101 performs distortion compensation by outputting the generated distortion compensation signal u [n] to the power amplifier 121 (step S208).

  With the above operation, the distortion compensation unit 101 performs distortion compensation in the DPD calculation phase.

  In step S206, the distortion compensator 101 configures the variable vector x [n] from, for example, the baseband signal x [n] and the envelope tracking signal V [n], and configures the variable vector x [n], that is, the expression (1) The variable vector x [n] shown in FIG.

  Thereby, the distortion compensation signal u [n] can be generated based on the baseband signal x [n] and the envelope tracking signal V [n] at appropriate timing.

  Further, the distortion compensation unit 101 configures, for example, a variable vector x [n] shown in Expression (6), and substitutes the configured variable vector x [n] into Expression (5).

  Thereby, even when the memory effect that the amplification characteristic in the power amplifier 121 changes due to the difference or direction of the temporal change of the baseband signal x [n] occurs, the power amplifier 121 can be appropriately compensated. A distortion compensation signal u [n] can be generated.

  Further, the distortion compensation unit 101 configures, for example, a variable vector x [n] shown in Expression (7), and substitutes the configured variable vector x [n] into Expression (5).

  Thereby, even when the memory effect that the amplification characteristic in the power amplifier 121 changes due to the difference or direction of the temporal change of the envelope tracking signal V [n] occurs, the power amplifier 121 can be appropriately compensated. A distortion compensation signal u [n] can be generated.

[Example of operation of amplification device]
FIG. 15 is a flowchart defining an operation procedure when the amplifying apparatus according to the first embodiment of the present invention performs distortion compensation.

  Here, it is assumed that the amplification device 201 is newly activated.

  Referring to FIG. 15, first, distortion compensation unit 101 determines a coefficient cIk tilde and a coefficient cQk tilde of the second control function based on Expression (18) in the learning phase. Then, the distortion compensation unit 101 substitutes the coefficient cIk tilde and the coefficient cQk tilde for the coefficient cIk and the coefficient cQk in the equation (5), respectively (step S302).

  Next, the distortion compensation unit 101 generates a distortion compensation signal u [n] based on Expression (5) in the DPD calculation phase. Then, the distortion compensation unit 101 performs distortion compensation by outputting the generated distortion compensation signal u [n] to the power amplifier 121 (step S304).

  Next, for example, when the operation of the amplifying apparatus 201 is continued (YES in step S306) and there is no change in environmental temperature (NO in step S308), the distortion compensation unit 101 continues distortion compensation in the DPD calculation phase. Is performed (step S304).

  On the other hand, for example, when the environmental temperature changes (YES in step S308), the distortion compensator 101 performs the first operation based on Expression (18) in the learning phase in order to cope with the change in the amplification characteristic of the power amplifier 121 due to the temperature change. A coefficient cIk tilde and a coefficient cQk tilde of a control function of 2 are determined. Then, the distortion compensation unit 101 updates the coefficient by substituting the determined coefficient cIk tilde and coefficient cQk tilde into the coefficient cIk and the coefficient cQk in Equation (5) (step S310).

  Further, for example, when the operation of the amplifying apparatus 201 is stopped by the operator (NO in step S306), the amplifying apparatus 201 ends the operation.

  Through the above operation, the amplification device 201 performs the compensation operation for the distortion characteristics of amplification in the power amplifier 121.

  In step S308, when the environmental temperature changes, the distortion compensation unit 101 proceeds to the learning phase and updates the coefficient. However, for example, the distortion compensation unit 101 may enter the learning phase and update the coefficient every predetermined time.

  Further, the distortion compensation unit 101 may update the coefficient while performing distortion compensation in step S310. More specifically, the distortion compensator 101 generates a distortion compensation signal u [n] based on the coefficient before being updated, outputs the generated distortion compensation signal u [n] to the power amplifier 121 and learning data. Accumulate.

  At this time, for example, when the envelope tracking signal V [n] is 20, the distortion compensation unit 101 accumulates learning data corresponding thereto, and when the envelope tracking signal V [n] is 30, learning corresponding thereto. Accumulate data.

  Then, the distortion compensation unit 101 determines the coefficient cIk tilde and the coefficient cQk tilde of the second control function based on the accumulated learning data.

  Then, the distortion compensation unit 101 updates the coefficient by substituting the determined coefficient cIk tilde and coefficient cQk tilde into the coefficient cIk and the coefficient cQk in Equation (5).

  As a result, the distortion compensation unit 101 can perform coefficient update processing while performing distortion compensation even when the amplification characteristic of the power amplifier 121 changes due to a change in environmental temperature, for example.

  As described above, the amplifying apparatus 201 is configured to perform distortion compensation based on the baseband signal x [n] and the envelope tracking signal V [n]. However, the present invention is not limited to this. The amplifying apparatus may be configured to perform distortion compensation as follows.

[When distortion compensation is performed based on the power supply voltage VDD (t)]
FIG. 16 is a diagram illustrating a configuration of a variation of the amplifying apparatus according to the first embodiment of the present invention.

  Referring to FIG. 16, amplifying apparatus 202 is an amplifying apparatus that performs distortion compensation based on power supply voltage VDD (t), as compared with amplifying apparatus 201. The contents other than those described below are the same as those of the amplification device 201.

  As compared with the amplification device 201, the amplification device 202 includes a distortion compensation unit 102 instead of the distortion compensation unit 101, and further includes an analog-digital converter (ADC) 171.

  The analog-digital converter 171 converts the power supply voltage VDD (t) received from the ET power supply 111 into a digital signal, and outputs the power supply voltage signal VDD [n] that is the converted digital signal to the distortion compensation unit 102.

  The distortion compensation unit 102 generates a distortion compensation signal u [n] based on the baseband signal x [n], the power supply voltage signal VDD [n], and the output signal y [n].

  Since the power supply voltage VDD (t) is a signal generated based on the envelope tracking signal V [n] in the ET power supply 111, it is between the envelope tracking signal V [n] and the power supply voltage signal VDD [n]. Have a certain relationship.

  Specifically, the timing at which the envelope tracking signal V [n] is output from the narrowband ET signal generation unit 152 to the ET power supply 111 is the same as the timing at which the distortion compensation unit 102 receives the power supply voltage signal VDD [n]. .

  Further, there is a proportional relationship between the magnitude of the envelope tracking signal V [n] and the magnitude of the power supply voltage signal VDD [n], for example.

  Note that the distortion compensation unit 102 receives the power supply voltage signal VDD [m] generated based on the envelope tracking signal V [m] at the post-distortion compensation timing TPostComp at a timing after the distortion compensation timing TComp shown in FIG. be able to. Here, the index m is, for example, from (n + 1) to (n + M1).

  That is, the power supply voltage signal VDD [m] corresponding to the envelope tracking signal V [m] at the distortion compensated timing TPostComp is not included in the variable vector x [n].

Therefore, the variable vector x [n] used when the distortion compensation unit 102 generates the distortion compensation signal u [n] is generally defined by, for example, the following equation (21).

  Here, L1 and L2 are integers of zero or more. M2 is an integer greater than or equal to zero.

When the variable vector x [n] is defined by the equation (21), the variable vector y [n] is defined by the following equation (22), for example.

  Therefore, the distortion compensator 102 replaces “envelope tracking signal V [n]” with “power supply voltage signal VDD [n]” in the processing performed by the distortion compensator 101, the variable vector x [n], and the variable By using the vector y [n], the distortion compensation signal u [n] can be generated by the same processing as the distortion compensation unit 101.

  The distortion compensation unit 102 performs distortion compensation by outputting the generated distortion compensation signal u [n] to the power amplifier 121.

  As described above, the amplifying apparatus 202 is configured to perform distortion compensation based on the variable vector x [n] shown in Expression (21) and the variable vector y [n] shown in Expression (22). It is not a thing. The amplifying apparatus may be configured to perform the following distortion compensation.

  FIG. 17 is a diagram showing a configuration of a modification of the amplifying device according to the first embodiment of the present invention.

  Referring to FIG. 17, amplifying apparatus 205 includes variable vector x that further includes power supply voltage signal VDD [m] corresponding to envelope tracking signal V [m] at post-distortion-compensated timing TPostComp as a variable, as compared with amplifying apparatus 202. This is an amplifying apparatus that performs distortion compensation based on [n]. The contents other than those described below are the same as those of the amplification device 202.

  As compared with the amplification device 202, the amplification device 205 includes a distortion compensation unit 105 instead of the distortion compensation unit 102, and further includes a memory 172, a digital / analog converter 173, and an output buffer 174.

  The ET power supply 111 outputs a power supply voltage VDD (t) corresponding to the envelope tracking signal V [n] to the analog / digital converter 171.

  The analog-digital converter 171 converts the power supply voltage VDD (t) received from the ET power supply 111 into a digital signal, and outputs the power supply voltage signal VDD [n], which is the converted digital signal, to the distortion compensator 105 and the memory 172.

  When the distortion compensation unit 105 receives the power supply voltage signal VDD [n] from the analog-digital converter 171, the compensation signal u [n] at a timing after, for example, (M1) or more clock cycles T have elapsed from the received timing. Is output to the power amplifier 121.

  More specifically, the distortion compensation unit 105 receives from the analog-digital converter 171 the power supply voltage signal VDD [n] to the power supply voltage signal VDD [n + M1]. Then, distortion compensation section 105 outputs compensation signal u [n] to power amplifier 121 at distortion compensation timing TComp after elapse of, for example, (M1) or more clock cycles T from the received timing.

  The memory 172 receives the power supply voltage signal VDD [n] from the analog / digital converter 171. Then, the memory 172 outputs the power supply voltage signal VDD [n] to the digital / analog converter 173 at the distortion compensation timing TComp after elapse of, for example, (M1) or more clock cycles T.

  Digital-analog converter 173 converts power supply voltage signal VDD [n] received from memory 172 into an analog signal, and outputs the converted analog signal to output buffer 174.

  When receiving the analog signal from the digital-analog converter 173, the output buffer 174 supplies the buffer power supply voltage BSV (t) corresponding to the received analog signal to the power amplifier 121.

  In this case, the buffer power supply voltage BSV (t) has the same level as the power supply voltage VDD (t) on which the power supply voltage signal VDD [n] is based, for example.

  The power amplifier 121 receives the compensation signal u [n] and the buffer power supply voltage BSV (t) at the distortion compensation timing TComp.

The variable vector x [n] used when the distortion compensation unit 105 generates the distortion compensation signal u [n] has a power supply voltage signal VDD [m] corresponding to the envelope tracking signal V [m] at the post-distortion compensation timing TPostComp. Is included. For this reason, the variable vector x [n] is generally defined by the following equation (23), for example.

  Here, L1 and L2 are integers of zero or more. M1 and M2 are integers greater than or equal to zero.

When the variable vector x [n] is defined by the equation (23), the variable vector y [n] is defined by the following equation (24), for example.

  Therefore, the distortion compensation unit 105 can generate the distortion compensation signal u [n] by the same processing as the distortion compensation unit 102 by using the variable vector x [n] and the variable vector y [n]. .

  Then, the distortion compensation unit 105 performs distortion compensation by outputting the generated distortion compensation signal u [n] to the power amplifier 121.

[When distortion compensation is performed based on the power supply current IDD (t)]
FIG. 18 is a diagram showing a configuration of a modification of the amplifying device according to the first embodiment of the present invention.

  Referring to FIG. 18, amplifying apparatus 203 is an amplifying apparatus that performs distortion compensation based on power supply current IDD (t) as compared with amplifying apparatus 201. The contents other than those described below are the same as those of the amplification device 201.

  Compared with the amplification device 201, the amplification device 203 includes a distortion compensation unit 103 instead of the distortion compensation unit 101, and further includes an analog-digital converter (ADC) 181.

  The analog-digital converter 181 converts the power supply current IDD (t) received from the ET power supply 111 into a digital signal, and outputs the power supply current signal IDD [n] that is the converted digital signal to the distortion compensation unit 103.

  The distortion compensation unit 103 generates a distortion compensation signal u [n] based on the baseband signal x [n], the power supply current signal IDD [n], and the output signal y [n].

  Since the power supply current IDD (t) is a signal generated based on the envelope tracking signal V [n] in the ET power supply 111, the power supply current IDD (t) is between the envelope tracking signal V [n] and the power supply current signal IDD [n]. Have a certain relationship.

  Specifically, the timing at which the envelope tracking signal V [n] is output from the narrowband ET signal generation unit 152 to the ET power supply 111 and the timing at which the distortion compensation unit 102 receives the power supply current signal IDD [n] are the same. .

  Further, there is a proportional relationship between the magnitude of the envelope tracking signal V [n] and the magnitude of the power supply current signal IDD [n], for example.

  Note that the distortion compensation unit 103 receives the power supply current signal IDD [m] generated based on the envelope tracking signal V [m] at the post-distortion compensation timing TPostComp at a timing after the distortion compensation timing TComp shown in FIG. be able to. Here, the index m is, for example, from (n + 1) to (n + M1).

  That is, the power supply current signal IDD [m] corresponding to the envelope tracking signal V [m] at the distortion compensated timing TPostComp is not included in the variable vector x [n].

Therefore, the variable vector x [n] used when the distortion compensation unit 103 generates the distortion compensation signal u [n] is generally defined by the following equation (25), for example.

  Here, L1 and L2 are integers of zero or more. M2 is an integer greater than or equal to zero.

When the variable vector x [n] is defined by the equation (25), the variable vector y [n] is defined by the following equation (26), for example.

  Therefore, the distortion compensator 103 replaces “envelope tracking signal V [n]” with “power supply current signal IDD [n]” in the processing performed by the distortion compensator 101, the variable vector x [n], and the variable By using the vector y [n], the distortion compensation signal u [n] can be generated by the same processing as the distortion compensation unit 101.

  The distortion compensation unit 103 performs distortion compensation by outputting the generated distortion compensation signal u [n] to the power amplifier 121.

  As described above, the amplifying apparatus 203 is configured to perform distortion compensation based on the variable vector x [n] shown in Expression (25) and the variable vector y [n] shown in Expression (26). It is not a thing. The amplifying apparatus may be configured to perform distortion compensation as follows.

  FIG. 19 is a diagram showing a configuration of a modification of the amplifying device according to the first embodiment of the present invention.

  Referring to FIG. 19, amplifying apparatus 206 further includes a variable vector x that includes power supply current signal IDD [m] corresponding to envelope tracking signal V [m] at post-distortion compensated timing TPostComp as a variable, as compared with amplifying apparatus 203. This is an amplifying apparatus that performs distortion compensation based on [n]. The contents other than those described below are the same as those of the amplification device 203.

  As compared with the amplification device 203, the amplification device 206 includes a distortion compensation unit 106 instead of the distortion compensation unit 103, and further includes a memory 182, a digital / analog converter 183, and an output buffer 184.

  The ET power supply 111 outputs a power supply current IDD (t) corresponding to the envelope tracking signal V [n] to the analog / digital converter 181.

  The analog-digital converter 181 converts the power supply current IDD (t) received from the ET power supply 111 into a digital signal, and outputs the power supply current signal IDD [n] that is the converted digital signal to the distortion compensation unit 106 and the memory 182.

  When receiving the power supply current signal IDD [n] from the analog-digital converter 181, the distortion compensator 106 receives the compensation signal u [n] at a timing after elapse of, for example, (M1) clock cycles T from the received timing. Is output to the power amplifier 121.

  More specifically, the distortion compensator 106 receives from the analog-digital converter 181 the power supply current signal IDD [n] to the power supply current signal IDD [n + M1]. Then, the distortion compensation unit 106 outputs the compensation signal u [n] to the power amplifier 121 at the distortion compensation timing TComp after elapse of, for example, (M1) or more clock cycles T from the received timing.

  The memory 182 receives the power supply current signal IDD [n] from the analog-digital converter 181. Then, the memory 182 outputs the power supply current signal IDD [n] to the digital / analog converter 183 at the distortion compensation timing TComp after elapse of, for example, (M1) or more clock cycles T.

  Digital-analog converter 183 converts power supply current signal IDD [n] received from memory 182 into an analog signal, and outputs the converted analog signal to output buffer 184.

  When the output buffer 184 receives an analog signal from the digital-analog converter 183, the output buffer 184 supplies the buffer power supply current BSI (t) corresponding to the received analog signal to the power amplifier 121.

  The buffer power supply current BSI (t) in this case has the same level as the power supply current IDD (t) on which the power supply current signal IDD [n] is based, for example.

The variable vector x [n] used when the distortion compensation unit 106 generates the distortion compensation signal u [n] has a power supply current signal IDD [m] corresponding to the envelope tracking signal V [m] at the post-distortion compensation timing TPostComp. Is included. For this reason, the variable vector x [n] is generally defined by the following equation (27), for example.

  Here, L1 and L2 are integers of zero or more. M1 and M2 are integers greater than or equal to zero.

When the variable vector x [n] is defined by the equation (27), the variable vector y [n] is defined by the following equation (28), for example.

  Therefore, the distortion compensation unit 106 can generate the distortion compensation signal u [n] by the same processing as the distortion compensation unit 103 by using the variable vector x [n] and the variable vector y [n]. .

  Then, the distortion compensation unit 106 performs distortion compensation by outputting the generated distortion compensation signal u [n] to the power amplifier 121.

  By the way, in the technique described in Patent Document 1, since distortion compensation is performed using a two-dimensional lookup table, when trying to perform highly accurate distortion compensation, the memory amount of the lookup table becomes enormous, and the circuit scale is large. Will become bigger.

  On the other hand, in the amplifying device according to the first embodiment of the present invention, the power amplifier 121, the distortion compensation unit 101, 102 or 103 that performs distortion compensation of the power amplifier 121, and the power supplied to the power amplifier 121 And an ET power source 111 that changes the voltage VDD (t) or the power source current IDD (t) according to an envelope tracking signal V [n] that is a signal corresponding to the envelope of the baseband signal x [n]. Then, the distortion compensator 101, 102, or 103 includes a plurality of first band signals having the baseband signal x [n], the power supply voltage VDD (t), the power supply current IDD (t), or the envelope tracking signal V [n] as variables. Distortion compensation is performed based on a linear combination of control functions of one.

  With such a configuration, distortion compensation of the power amplifier 121 can be performed without using a two-dimensional lookup table that requires a large storage capacity. Therefore, in the configuration employing the envelope tracking method, the amplification efficiency of the wideband signal can be improved and the distortion compensation can be appropriately performed with a simple configuration.

  Further, in the amplifying apparatus according to the first embodiment of the present invention, the distortion compensator 101, 102, or 103 has the power supply voltage VDD (t) and the power supply current IDD (at the distortion compensation timing TComp that is a timing for performing distortion compensation. t) or an envelope tracking signal V [n] and a linear combination of a plurality of first control functions having a variable corresponding to the timing corresponding to the distortion compensation timing, that is, the baseband signal x [n] at the distortion compensation corresponding timing TCorr. To compensate for distortion.

  With such a configuration, distortion compensation of the power amplifier 121 is performed based on the baseband signal x [n] at an appropriate timing and the power supply voltage VDD (t), the power supply current IDD (t), or the envelope tracking signal V [n]. It can be performed.

  In the amplifying apparatus according to the first embodiment of the present invention, the distortion compensator 101, 102, or 103 includes the baseband signal x [n] at the pre-distortion-corresponding timing TPreCorr before the distortion-compensation-corresponding timing TCorr. Distortion compensation is performed based on a linear combination of a plurality of first control functions having at least one of the baseband signals x [n] at a post-distortion compensation timing TPostCorr after the distortion compensation correspondence timing TCorr as a variable.

  With such a configuration, even when a memory effect occurs in the power amplifier 121 due to the difference or direction of the baseband signal x [n] with respect to time, the distortion of the power amplifier 121 occurs. Compensation can be performed appropriately.

  In the amplifying apparatus according to the first embodiment of the present invention, the distortion compensator 101, 102, or 103 includes the power supply voltage VDD (t) and the power supply current IDD at the pre-distortion compensation timing TPreComp before the distortion compensation timing TComp. (T) or the envelope tracking signal V [n] and / or the power supply voltage VDD (t), the power supply current IDD (t), or the envelope tracking signal V [n] at the post-distortion compensation timing TPostComp after the distortion compensation timing TComp. Distortion compensation is performed based on a linear combination of a plurality of first control functions with one of them as a variable.

  With such a configuration, there is a memory effect that the amplification characteristic in the power amplifier 121 changes depending on the difference or direction of the temporal change of the power supply voltage VDD (t), the power supply current IDD (t), or the envelope tracking signal V [n]. Even when it occurs in the power amplifier 121, distortion compensation of the power amplifier 121 can be performed appropriately.

  Further, in the amplifying apparatus according to the first embodiment of the present invention, the distortion compensator 101, 102, or 103 performs distortion compensation at the distortion compensation timing to convert the distortion compensation signal u [n] from the baseband signal x [n]. ] Is generated. The distortion compensation unit 101, 102, or 103 includes a linear combination of a plurality of first control functions, a plurality of second control functions at a plurality of distortion compensation timings, and a distortion compensation signal u [n] at a plurality of distortion compensation timings. Based on the above, distortion compensation is performed. The plurality of second control functions include the output signal y [n] output from the power amplifier 121 at the timing corresponding to the timing of the baseband signal x [n], the power supply voltage VDD (t), and the power supply current IDD. Let (t) or envelope tracking signal V [n] be a variable.

  With such a configuration, based on the distortion compensation signal u [n], the output signal y [n], the power supply voltage VDD (t), the power supply current IDD (t), or the envelope tracking signal V [n], the power amplifier 121 The distortion characteristics of amplification can be grasped.

  As a result, it is possible to generate a distortion compensation signal u [n] that can appropriately compensate for amplification distortion characteristics of the power amplifier 121.

  In the amplifying apparatus according to the first embodiment of the present invention, the distortion compensator 101, 102, or 103 has a predetermined difference between the distortion compensation signal u [n] and the linear combination of the plurality of second control functions. The coefficient of the second control function in the linear combination of the plurality of second control functions is determined so as to satisfy the condition. Then, the distortion compensation unit 101, 102, or 103 sets the coefficient as the coefficient of the first control function in the linear combination of the plurality of first control functions.

  With such a configuration, the coefficient of the first control function in the linear combination of the first control functions can be specifically and appropriately determined.

  Further, in the amplifying apparatus according to the embodiment of the present invention, the distortion compensator 101, 102, or 103 increases the number of the plurality of distortion compensation timings more than the number of the plurality of first control functions.

  With such a configuration, the coefficient of the first control function, which is an unknown number, can be appropriately determined based on a known number greater than the number of unknown numbers.

  Further, when the number of the plurality of first control functions is smaller than the number of the plurality of distortion compensation timings, the distortion compensation unit 101, 102, or 103 is based on the small number of first control functions. Since [n] can be generated, the calculation load when generating the distortion compensation signal u [n] can be reduced.

  In the amplifying apparatus according to the first embodiment of the present invention, the first control function and the second control function are Gaussian functions.

  As described above, the configuration using the Gaussian function as the kernel function makes it possible to appropriately and quantitatively calculate the degree of similarity between the two vectors having the first control function and the second control function as arguments.

  In the amplifying device according to the first embodiment of the present invention, when a plurality of baseband signals x [n] or envelope tracking signals V [n] are included in the variable vector x [n], the index n is continuous. However, the present invention is not limited to this. For example, the baseband signal x [n] or the envelope tracking signal V [n] having an index n for each predetermined interval may be included in the variable vector x [n]. In this case, the variable vector x [n] includes data sampled at predetermined intervals.

  Next, another embodiment of the present invention will be described with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals and description thereof will not be repeated.

<Second Embodiment>
The present embodiment relates to an amplifying apparatus that performs distortion compensation based on stored coefficients as compared with the amplifying apparatus according to the first embodiment. The contents other than those described below are the same as those of the communication apparatus according to the first embodiment.

  FIG. 20 is a diagram illustrating an example of the configuration of the amplifying apparatus according to the second embodiment of the present invention.

  Referring to FIG. 20, the amplifying apparatus 204 includes a distortion compensating unit 104 instead of the distortion compensating unit 101 as compared with the amplifying apparatus 201 according to the first embodiment, and further includes a coupler 156, a down converter 158, A storage unit 162 is provided instead of the analog-digital converter (ADC) 159 and the attenuator 160.

  FIG. 21 is a diagram illustrating an example of a table stored in the storage unit in the amplifying apparatus according to the second embodiment of the present invention.

  Referring to FIGS. 20 and 21, storage unit 162 is a linear combination of, for example, power supply voltage VDD (t), power supply current IDD (t) or envelope tracking signal V [n] and a plurality of first control functions. The correspondence relationship between the coefficient cIk and the coefficient cQk of the first control function is stored.

  Hereinafter, a case where the storage unit 162 stores the correspondence between the envelope tracking signal V [n] and the coefficient cIk and the coefficient cQk of the first control function in the linear combination of the plurality of first control functions will be described.

  By replacing “envelope tracking signal V [n]” with “power supply voltage VDD (t)” or “power supply current IDD (t)”, the following description will be given by the description of power supply voltage VDD (t) or power supply current IDD (t). A similar argument can be applied to this case.

  More specifically, the storage unit 162 stores the coefficient cIk and the coefficient cQk of the first control function in the linear combination of the plurality of first control functions optimized with respect to the envelope tracking signal V [n], as envelope tracking. Stored for each signal V [n].

  Specifically, the storage unit 162 is optimized for the envelope tracking signal V [n] and the envelope tracking signal V [n] when the envelope tracking signal V [n] is 20, 30, and 40, for example. The table 10 including the correspondence relationship between the coefficient cIk and the coefficient cQk is stored.

  In this case, the numbers of I and Q components of the first control function, that is, KI and KQ are 5 and 4, respectively.

  For example, the narrowband ET signal generation unit 152 outputs the envelope tracking signal V [n] to the distortion compensation unit 104 and the ET power supply 111 at the distortion compensation timing TComp.

  The distortion compensation unit 104 receives the envelope tracking signal V [n], for example, 30 from the narrowband ET signal generation unit 152 at a timing before the distortion compensation timing TComp. Then, the distortion compensation unit 104 acquires the coefficient cIk and the coefficient cQk corresponding to the case where V [n] is 30 from the table 10 in the storage unit 162.

  In this case, the coefficients cIk, that is, cI1, cI2, cI3, cI4, and cI5 are 1.5, 0.8, 1.9, 0.1, and 3.3, respectively. The coefficients cQk, that is, cQ1, cQ2, cQ3, and cQ4 are 3.2, 1.4, 1.9, and 0.8, respectively.

  Since the coefficient is a coefficient optimized with respect to the envelope tracking signal V [n], the envelope tracking signal V [n] as a variable is included in the expression (5) through the coefficient. That is, when generating a linear combination of the first control functions shown in Expression (5) based on the coefficient, the envelope tracking signal V [n] is not included in the variable vector x [n].

  Thus, the envelope tracking signal V [n] is not included in the variable vector x [n], but is included in a linear combination of a plurality of first control functions via the coefficients. Therefore, the arithmetic processing in the present embodiment is based on a linear combination of a plurality of first control functions whose variables are the baseband signal x [n] and the envelope tracking signal V [n], that is, the variable vector x [n]. This corresponds to a calculation process for performing distortion compensation.

  Specifically, the distortion compensation unit 104 generates a variable vector x [n] represented by Expression (1), Expression (6), or Expression (7) that does not include the envelope tracking signal V [n], for example.

  Then, the distortion compensation unit 104 generates a linear combination of the first control functions shown in Expression (5) based on the variable vector x [n] and the coefficient, and based on the generated linear combination of the first control functions. Thus, the distortion compensation signal u [n] is generated. The distortion compensation unit 104 performs distortion compensation by outputting the distortion compensation signal u [n] to the power amplifier 121 at the distortion compensation timing TComp.

  Thereby, the distortion compensator 104 appropriately performs two-dimensional DPD based on the envelope tracking signal V [n] included in the coefficient and the baseband signal x [n] included in the variable vector x [n]. A distortion compensation signal can be generated.

  In addition, since the dimension of the variable vector x [n] can be reduced, that is, the amount of data that must be processed in the distortion compensator 104 can be reduced, the calculation processing load in the distortion compensator 104 can be reduced.

  The table 10 stores a correspondence relationship between the envelope tracking signal V [n] generated at the time of factory shipment according to the characteristics of the power amplifier 121, the coefficient cIk, and the coefficient cQk, for example.

  Further, the table 10 may store, for example, a correspondence relationship between the envelope tracking signal V [n] generated by another device such as a computer, the coefficient cIk, and the coefficient cQk.

  Further, the distortion compensation unit 104 may acquire, for example, the correspondence relationship between the envelope tracking signal V [n], the coefficient cIk, and the coefficient cQk from another device by communication.

  Since other configurations and operations are the same as those of the amplification device according to the first embodiment, detailed description thereof will not be repeated here.

  As described above, in the amplifying apparatus according to the second embodiment of the present invention, the storage unit 162 includes the power supply voltage VDD (t), the power supply current IDD (t), or the envelope tracking signal V [n] and a plurality of The correspondence relationship between the coefficient cIk and the coefficient cQk of the first control function in the linear combination of the first control functions is stored. The distortion compensation unit 104 stores a coefficient cIk and a coefficient cQk corresponding to the power supply voltage VDD (t), the power supply current IDD (t), or the envelope tracking signal V [n] at the distortion compensation timing TComp, which is a timing for performing distortion compensation. 162. Then, the distortion compensation unit 104 generates a linear combination of a plurality of first control functions based on the coefficient, and performs distortion compensation based on the linear combination of the plurality of first control functions.

  With such a configuration, the contribution of the power supply voltage VDD (t), the power supply current IDD (t) or the envelope tracking signal V [n] to the distortion compensation can be included in the coefficient, so that the calculation processing load in the distortion compensation unit 104 is increased. Can be reduced.

  Further, for example, when the amplifying device 204 is activated, distortion compensation can be started immediately without determining the coefficient cIk and the coefficient cQk in the amplifying device 204.

  The above embodiment should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

101, 102, 103, 104 Distortion compensation unit 111 ET power supply 121 Power amplifier (amplifier)
152 Narrowband ET signal generator 159,171,181 Analog to digital converter (ADC)
154,173,183 Digital-to-analog converter (DAC)
155 Upconverter 156 Coupler 157 Oscillator 158 Downconverter 160 Attenuator 162 Storage unit 174, 184 Output buffer 201, 202, 203, 204 Amplifier

Claims (10)

An amplifier;
A distortion compensation unit that performs distortion compensation of the amplifier;
A power supply that changes a power supply voltage or a power supply current supplied to the amplifier according to an envelope tracking signal that is a signal corresponding to an envelope of a target signal;
The amplification device, wherein the distortion compensation unit performs the distortion compensation based on a linear combination of a plurality of first control functions having the target signal and the power supply voltage, the power supply current, or the envelope tracking signal as variables.   The distortion compensator includes the power supply voltage, the power supply current, or the envelope tracking signal at a distortion compensation timing that is a timing for performing the distortion compensation, and the target signal at a timing corresponding to the distortion compensation timing as a variable. The amplification device according to claim 1, wherein the distortion compensation is performed based on a linear combination of a plurality of first control functions.   The distortion compensator further includes at least one of the target signal at a timing before a timing corresponding to the distortion compensation timing and the target signal at a timing after a timing corresponding to the distortion compensation timing as a variable. The amplification device according to claim 2, wherein the distortion compensation is performed based on a linear combination of a plurality of first control functions.   The distortion compensation unit includes the power supply voltage, the power supply current, or the envelope tracking signal at a timing before the distortion compensation timing, and the power supply voltage, the power supply current, or the envelope tracking signal at a timing after the distortion compensation timing. 4. The amplification device according to claim 2, wherein the distortion compensation is performed based on a linear combination of the plurality of first control functions, wherein at least one of the plurality of first control functions is a variable. 5. The distortion compensation unit generates the distortion compensation signal from the target signal by performing the distortion compensation at the distortion compensation timing,
The distortion compensator is based on a linear combination of the plurality of first control functions, a plurality of second control functions at the plurality of distortion compensation timings, and the distortion compensation signals at the plurality of distortion compensation timings. Performing the distortion compensation,
3. The plurality of second control functions have an output signal output from the amplifier at a timing corresponding to a timing of the target signal, and the power supply voltage, the power supply current, or the envelope tracking signal as variables. The amplification device according to any one of claims 1 to 4.   The distortion compensation unit includes the second control function in a linear combination of the plurality of second control functions such that a difference between the distortion compensation signal and the linear combination of the plurality of second control functions satisfies a predetermined condition. The amplifying apparatus according to claim 5, wherein the coefficient is determined, and the coefficient is a coefficient of the first control function in a linear combination of the plurality of first control functions. The amplification device further includes:
A storage unit that stores a correspondence relationship between the power supply voltage, the power supply current, or the envelope tracking signal and a coefficient of the first control function in a linear combination of the plurality of first control functions;
The distortion compensation unit acquires the coefficient corresponding to the power supply voltage, the power supply current, or the envelope tracking signal at a distortion compensation timing that is a timing for performing the distortion compensation from the storage unit, and based on the coefficient, 4. The amplification device according to claim 1, wherein a linear combination of the first control functions is generated, and the distortion compensation is performed based on the linear combination of the plurality of first control functions. 5. .   The amplifying apparatus according to claim 1, wherein the first control function and the second control function are Gaussian functions. Changing the power supply voltage or power supply current supplied to the amplifier in accordance with an envelope tracking signal that is a signal corresponding to the envelope of the target signal;
A distortion compensation method comprising: compensating distortion of the amplifier based on a linear combination of a plurality of first control functions having the target signal and the power supply voltage, the power supply current, or the envelope tracking signal as variables. . A distortion compensation program used in an amplification device,
Changing the power supply voltage or power supply current supplied to the amplifier in accordance with an envelope tracking signal that is a signal corresponding to the envelope of the target signal;
Performing the distortion compensation of the amplifier based on a linear combination of a plurality of first control functions using the target signal and the power supply voltage, the power supply current, or the envelope tracking signal as variables. Distortion compensation program.

Images