KR20140115800A - Wideband signal transmitter including power amplifier model - Google Patents

Abstract

The present invention introduces a wideband signal transmission apparatus including an equivalent model of a power amplifier for linearizing non-linear characteristics of the power amplifier included in the transmitting device by the predistortion of a baseband. The wideband signal transmission apparatus includes a predistorter, a digital-to-analog converter, a modulator, the power amplifier, a demodulator, an analog-to-digital converter, a virtual power amplifier unit, and an adaptive algorithm migration unit.

Description

[0001] The present invention relates to a wideband signal transmitter including a power amplifier model,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a broadband signal transmitting apparatus for transmitting a wideband signal and, more particularly, to a broadband signal transmitting apparatus including an equivalent model of a power amplifier for linearizing a nonlinear characteristic of a power amplifier included in a transmitting apparatus by baseband predistortion ≪ / RTI >

As the demand for high speed data communication increases, the next generation communication systems are gradually changing to use a wider frequency band with an attempt to increase the data density per frequency. In order to increase the data density per frequency, a method of using a higher modulation order such as 256 QAM (Quadrature Amplitude Modulation) or transmitting several data streams at the same frequency simultaneously using several antennas has been proposed . For the frequency band, 802.11n, which is the existing wireless LAN standard, used the maximum 40MHz band, but the next generation 802.11ac uses the band up to 160MHz.

Methods of increasing data density per frequency commonly require higher quality transmitted signals. There are many reasons why the improvement of the quality of the transmitted signal is not effective at present, but the nonlinearity of the power amplifier is a main factor in transmitting at a high power. A variety of linearization techniques have been studied to solve this problem. In recent years, digital predistortion techniques have been attracting attention and are being applied to commercialized products.

1 shows a configuration of a conventional transmitting apparatus.

1, a transmitter 100 includes a predistorter 110, a digital-to-analog converter 120, a modulator 130, a power amplifier 140, a demodulator 150, an analog-to-digital converter 160, And an adaptive algorithm fulfillment unit 170.

The transmission signal x (n) is transmitted to the power amplifier 140 via the digital-to-analog converter 120 and the modulator 130 and is transmitted to the outside. As described above, the predistorter 110 is used to compensate for the nonlinear characteristics of the power amplifier 140, and its characteristics are determined by the adaptive algorithm implementation unit 170. The adaptive algorithm transition unit 170 receives a signal output from the power amplifier 140 having a nonlinear characteristic through the demodulator 150 and the analog-digital converter 160 and outputs the received feedback signal to the predistorter 110 And compares the transmitted signal y (n) with the transmitted signal y (n) to determine the characteristics of the digital predistorter 110.

The thick lines connected between the respective components mean that the signals transmitted and received are complex signals, and the thin lines indicate real signals, and unless otherwise specified, Lt; / RTI >

The signal x (n) whose band is extended at least three times as much as the signal band of the transmission signal x (n) due to the nonlinearity of the power amplifier 140 is generated so that the band extended signal is aliased the sampling rate of the analog-to-digital converter 160 must be very high in order to pass it to the normal adaptation algorithm transition unit 170 without any aliasing. The high sampling rate of the converter is not only complicated in the structure of the converter, but also consumes a lot of power, so there is a serious problem in using such an analog-to-digital converter.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a broadband signal transmitting apparatus including an equivalent model of a power amplifier for linearizing nonlinear characteristics of a power amplifier included in a transmitting apparatus by predistortion of a baseband.

According to one aspect of the present invention, there is provided a wideband signal transmission apparatus including a predistorter, a digital-to-analog converter, a modulator, a power amplifier, a demodulator, an analog-to-digital converter, a virtual power amplifier unit, . The predistorter predistortes the baseband digital transmission data to generate a predistortion signal. The digital-to-analog converter converts the predistortion signal into an analog signal. The modulator modulates an analog signal output from the digital-to-analog converter. The power amplifier amplifies a signal output from the modulator to generate an RF transmission signal. The demodulator demodulates the RF transmission signal. The analog-to-digital converter converts an analog signal output from the demodulator into a digital signal. The virtual power amplifier unit generates the feedback signal using the predistortion signal and the digital signal. The adaptive algorithm assignment unit generates a first control signal for setting the predistorter using the baseband digital transmission data, the predistortion signal, and the feedback signal.

According to another aspect of the present invention, there is provided a broadband signal transmission apparatus including a predistorter, a digital-to-analog converter, a modulator, a power amplifier, a demodulator, an analog-digital converter, a virtual power amplifier unit, Unit. The predistorter predistortes the baseband digital transmission data to generate a predistortion signal. The digital-to-analog converter converts the predistortion signal into an analog signal. The modulator modulates an analog signal output from the digital-to-analog converter. The power amplifier amplifies a signal output from the modulator to generate an RF transmission signal. The demodulator demodulates the RF transmission signal. The analog-to-digital converter converts an analog signal output from the demodulator into a low-IF digital signal that is a signal of a low intermediate frequency band. The virtual power amplifier unit generates the feedback signal using the predistortion signal and the low-IF digital signal. The adaptive algorithm assignment unit generates a first control signal for setting the predistorter using the baseband digital transmission data, the predistortion signal, and the feedback signal.

The broadband signal transmission apparatus according to the present invention has an advantage in that a digital predistorter applied to improve EVM quality and spectral characteristics of an output signal can be implemented at a low cost. In the conventional method, for a digital predistorter, In addition to the analog circuitry, a very high-speed ADC was required, but the proposed approach would allow a very low or even small amount of data collected at any time when using a direct conversion or low-IF analog feedback circuit There is an advantage that a digital predistorter can be designed. It also has the advantage of significantly lowering the cost and power consumption of ADCs, which require significant cost and power consumption.

1 shows a configuration of a conventional transmitting apparatus.
2 is a block diagram of a broadband signal transmitter including a power amplifier model according to an embodiment of the present invention.
3 is another embodiment of a broadband signal transmitting apparatus including a power amplifier model according to the present invention.
Figure 4 illustrates a power amplifier model using finite polynomials.
5 illustrates a power amplifier model using a finite number of polynomials intentionally generating aliasing.
6 is computer simulation data comparing the performance of the broadband signal transmitting apparatus according to the present invention.

In order to fully understand the present invention and the operational advantages of the present invention and the objects achieved by the practice of the present invention, reference should be made to the accompanying drawings, which are provided for explaining exemplary embodiments of the present invention, and the contents of the accompanying drawings.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. Like reference symbols in the drawings denote like elements.

In order to design a conventional predistorter, it is necessary to sample transmission and reception signals that are not aliased. For a predistorter that takes distortion into the third order, the bandwidth of the transmit and feedback signals can be extended to three times the original signal. For example, when the signal band is 80 MHz band, the bandwidth is extended to 240 MHz by predistortion and distortion of the power amplifier. In order to express it in the digital domain, a minimum 240 MHz analog to digital converter (ADC) A digital to analog converter (DAC) is required. If further consideration is given to the transition of the analog filter, a faster ADC / DAC is needed.

In the case of a DAC, there is no room for decreasing the sampling rate because the predistorted signal needs to be transmitted to the power amplifier. However, in the case of the ADC, it is ensured that information about the distortion occurring in the power amplifier can be sufficiently obtained even at a low sampling rate If possible, a predistorter can be designed with a low sampling rate.

The present invention proposes a method of estimating a power amplifier model using a feedback signal collected at a low sampling rate and then using a virtual power amplifier output signal generated by using the estimated power amplifier model, And a distortion algorithm design algorithm. Since the conventional predistorter technique can be applied directly after the model of the power amplifier is obtained, the power amplifier model estimation will be described in detail here.

First, an example of the proposed broadband signal transmission apparatus will be described, and the theoretical description of the broadband signal transmission apparatus will be described later. The broadband signal transmission apparatus proposed by the present invention can be largely divided into a direct conversion system and a low-IF system, which will be described in order.

2 is a block diagram of a broadband signal transmitter including a power amplifier model according to an embodiment of the present invention.

Referring to FIG. 2, the direct conversion type wideband signal transmission apparatus 200 includes a predistorter 210, a DAC 220, a modulator 230, a power amplifier 240, a demodulator 250, an ADC 260 An adaptive algorithm implementation unit 270, and a virtual power amplifier unit 280. [

The predistorter 210 pre-distorts the baseband digital transmission data x (n) to generate the predistortion signal y (n). The DAC 220 converts the predistortion signal y (n) into an analog signal. The modulator 230 modulates an analog signal output from the DAC 220. The power amplifier 240 amplifies the signal output from the modulator 230 to generate an RF transmission signal.

)를 생성한다. 적응알고리즘 이행 유닛(270)은 기저대역 디지털 송신 데이터(x(n)), 전치왜곡신호(y(n)) 및 피드백신호(

)를 이용하여 전치왜곡기(210)를 설정하는 제1제어신호(CON1)를 생성한다. The demodulator 250 demodulates an RF transmission signal output from the power amplifier 240. The ADC 260 converts the analog signal output from the demodulator 250 into a digital signal z (kL). The virtual power amplifier unit 280 uses the predistortion signal y (n) and the digital signal z (kL)

). The adaptive algorithm implementation unit 270 receives the baseband digital transmission data x (n), the predistortion signal y (n)

) To generate the first control signal CON1 for setting the predistorter 210. [

The virtual power amplifier unit 280 includes a power amplifier model 281, a down sampler 282, and a model estimator 283.

)를 생성한다. 내부 파라미터에 대해서는 후술한다. 다운샘플러(282)는 피드백신호(

)를 다운 샘플링(down sampling)한 신호(

)를 생성한다. 모델추정기(283)는 전치왜곡신호(y(n)), 다운샘플러의 출력신호(

) 및 디지털 신호(z(kL)) 중 적어도 2개의 신호를 이용하여 제2제어신호(CON2)를 생성한다. The power amplifier model 281 determines an internal parameter in response to the second control signal CON2 and outputs the feedback signal y (n) using the predistortion signal y (n)

). The internal parameters will be described later. The down sampler 282 receives the feedback signal

) Down-sampled

). The model estimator 283 receives the predistortion signal y (n), the down sampler output signal

) And a digital signal z (kL), to generate a second control signal CON2.

The signals transmitted and received between the modulator 230 and the power amplifier 240 and between the power amplifier 240 and the demodulator 250 are real signals and the signals transmitted and received between the remaining components are complex signals.

As described above, the sampling rate of the ADC 260 is lower than the sampling rate of the DAC 220.

) 및 디지털 신호(z(kL))에 최소자승기법을 적용하여 제2제어신호(CON2)를 생성한다. The power amplifier model 281 is set based on a plurality of polynomials, and the second control signal CON2 determines the values of a plurality of coefficients included in the polynomial. Details will be described later. The model estimator 283 generates a second control signal CON2 by applying a method of least squares to the predistortion signal y (n) and the digital signal z (kL) The signal y (n), the output signal of the down sampler 282

) And a digital signal z (kL) to generate a second control signal CON2.

Here, the demodulator 250 and the ADC 260 are not used after the power amplifier model 281 constituting the virtual power amplifier unit 280 is set. By doing so, power consumption can be minimized, and the output of the power amplifier model 281 is used instead of using the output signal of the actual power amplifier 240 for the subsequent predistorter setting.

FIG. 2 shows a direct conversion scheme implemented. In the following description, the Low-IF scheme will be described.

3 is another embodiment of a broadband signal transmitting apparatus including a power amplifier model according to the present invention.

3, a low-IF broadband signal transmission apparatus 300 includes a predistorter 310, a DAC 320, a modulator 330, a power amplifier 340, a demodulator 350, an ADC 360, an adaptive algorithm transition unit 370, and a virtual power amplifier unit 380.

The predistorter 310 predistortes the baseband digital transmission data x (n) to produce a predistortion signal y (n). The DAC 320 converts the predistortion signal y (n) into an analog signal. The modulator 330 modulates the analog signal output from the DAC 320. The power amplifier 340 amplifies a signal output from the modulator 330 to generate an RF transmission signal.

)를 생성한다. 적응알고리즘 이행 유닛(370)은 기저대역 디지털 송신 데이터(x(n)), 전치왜곡신호(y(n)) 및 피드백신호(

)를 이용하여 전치왜곡기(310)를 설정하는 제1제어신호(CON1)를 생성한다. The demodulator 350 demodulates the RF transmission signal. The ADC 360 converts the analog signal output from the demodulator 350 into a low-IF digital signal z (kL), which is a signal of a low intermediate frequency band. The virtual power amplifier unit 380 uses the predistortion signal y (n) and the low-IF digital signal z (kL)

). The adaptive algorithm implementation unit 370 receives the baseband digital transmission data x (n), the predistortion signal y (n)

) To generate the first control signal CON1 for setting the predistorter 310. [

The virtual power amplifier unit 380 includes a power amplifier model 381, a down sampler 382, a model estimator 383 and a multiplier 384.

)를 생성한다. 곱셈기(384)는 피드백신호(

)에 중간주파수 신호(f_IF(n))를 곱하여 중간주파수 피드백신호(

)를 생성한다. 다운샘플러(382)는 중간주파수 피드백신호(

)를 다운 샘플링 한 신호(

)를 생성한다. 모델추정기(383)는 전치왜곡신호(y(n)), 다운샘플러의 출력신호(

) 및 low-IF 디지털 신호(z(kL)) 중 적어도 2개의 신호를 이용하여 제2제어신호(CON2)를 생성한다. The power amplifier model 381 determines an internal parameter in response to the second control signal CON2 and outputs the feedback signal y (n) using the predistortion signal y

). The multiplier 384 multiplies the feedback signal < RTI ID = 0.0 >

) By the intermediate frequency signal f _IF (n) to obtain an intermediate frequency feedback signal (

). The down sampler 382 receives the intermediate frequency feedback signal < RTI ID = 0.0 >

) Down-sampled signal (

). The model estimator 383 receives the predistortion signal y (n), the output signal of the down sampler

) And a low-IF digital signal z (kL), to generate a second control signal CON2.

Here, it is assumed that between the modulator 330 and the power amplifier 340, between the power amplifier 340 and the demodulator 350, between the demodulator 350 and the ADC 360, and between the ADC 360 and the virtual power amplifier unit 380 The signals transmitted and received are real signals, and the signals transmitted and received between the remaining components are complex signals.

Here, the power amplifier model 381 is set based on a plurality of polynomials, and the second control signal CON2 determines values of a plurality of coefficients included in the polynomial.

) 및 low-IF 디지털 신호(z(kL))에 최소 자승 기법을 적용하여 제2제어신호(CON2)를 생성한다. The model estimator 383 generates the second control signal by applying a least squares technique to the predistortion signal y (n) and the low-IF digital signal z (kL) ), The output signal of the down sampler 382 (

) And a low-IF digital signal z (kL) to generate a second control signal CON2.

The demodulator 350 and the ADC 360 are not used after the power amplifier model 381 constituting the virtual power amplifier unit 380 is set.

Hereinafter, modeling of the power amplifier will be described.

Figure 4 illustrates a power amplifier model using finite polynomials.

Referring to FIG. 4, when a polynomial-based model such as a memory polynomial is used for modeling a power amplifier, the relationship between the input signal and the output signal of the power amplifier is relatively small G ₀ , g ₁ , g ₂ , g ₃ ) of the parameters (φ ₀ , φ ₁ ) and the coefficients of these parameters (coefficient, g ₀ , g ₁ , g ₂ , g ₃ ).

The parameters shown on the left side of FIG. 4 are samples of first order, third order, one sample delayed first order, and one sample delayed third order from top to bottom. Multiply them by the coefficients g ₀ , g ₁ , g ₂ , and g ₃ , respectively, and then combine them.

That is, the characteristics of the power amplifier are defined by coefficients of finite polynomials generated from the input signal. Therefore, the actual number of feedback samples required to extract the model of the power amplifier is completely modeled by the model order assumed beforehand by the power amplifier, and the set of polynomials of the model generated from the input signal corresponding to the collected feedback signal is the number of parameters The power amplifier can be modeled only if the number of minimum parameters is secured. There is no restriction on the sampling rate of the output signal of the power amplifier even when the ideal case is not present.

For more detail, the power amplifier using a memory polynomial with a third order polynomial and a memory depth of 1 as a model of the power amplifier is expressed by Equation 1 and Equation Can be displayed together.

이다. here,

to be.

In Equation (1), the baseband equivalent input signal of the power amplifier model is y (n) and the baseband equivalent output signal is z (n). An example of a typical power amplifier model estimation using this model is shown in FIG. The input and output equations as shown in Equation (2) can be obtained by generating a distorted polynomial using the transmission signal according to the model order and then collecting the feedback signals.

N is a natural number, and the degree of each polynomial component in the feedback signal is estimated by using an algorithm such as least squares using Equation 2, and the result is output to a power amplifier model . Algorithms for estimating coefficients from Equation (2) vary, and there is no problem in extracting coefficients even if any method is used.

5 illustrates a power amplifier model using a finite number of polynomials intentionally generating aliasing.

The parameters shown on the left side of FIG. 5 are samples of first order, third order, one sample delayed first order and one sample delayed third order from top to bottom. Multiply them by the coefficients g ₀ , g ₁ , g ₂ , and g ₃ , respectively, and then combine them.

Referring to FIG. 5, the feedback signal sampled by low-frequency or random sampling is distorted as shown in the right-hand side of FIG. 5 where the distorted signal is superimposed on the frequency axis. In order to extract the model coefficient of the power amplifier from this signal, the value of each distortion component polynomial generated from the transmission signal is also distorted using the same low-frequency sampling or random sampling, and then these distorted polynomials are compared with the same distorted feedback signal The equation for estimating the power amplifier characteristics can be obtained as in the conventional case. L (L is a natural number) times lower sampling rate can be expressed by Equation (3).

Even when random sampling is used, an input / output relation expression can be obtained as shown in Equation (4). Since the form of equations is the same, coefficients can also be estimated using a normal algorithm.

here

이다.

to be.

After estimating the model of the power amplifier from the feedback signal obtained using the low-speed sampling or random sampling as described above, the output of the virtual power amplifier generated inside the power amplifier model, rather than the feedback signal obtained from the actual power amplifier output By using indirect training technique, predistorter can be designed. Since this power amplifier model output is generated using the transmission samples contained in the power amplifier model, it is possible to obtain the undistorted sample data by the feedback sampling, and thus the conventional predistorter algorithm can be applied.

So far, we have examined the case where the feedback signal is obtained by the direct conversion method, which is the most widely used and currently used transmission / reception method shown in FIG.

The low-IF scheme shown in FIG. 3 is used to reduce the number of ADCs used for digital conversion of the feedback signal. In the conventional case, when an analog signal converted into a low-IF signal is obtained by an ADC and then a predistorter is designed, the signal is converted into a baseband signal through digital signal processing. There is only difference in whether it is digital or not, and it is exactly the same as direct conversion method. However, if you are using low-speed sampling or random sampling earlier, some formulas need to be converted.

When estimating the model using a third-order memory polynomial with a delay of one tap, the low-IF feedback signal sampled at the period of Ts can be expressed as Equation (5).

It can be seen that the low-IF signal sampled from Equation (5) is a linear combination of signals obtained by converting the nonlinear polynomial terms generated from the input signal to low-IF. That is, in the conventional method, the feedback signal is lowered to the baseband and compared with the baseband distortion terms. Conversely, it can be seen that the power amplifier model coefficient can be obtained even using the low-IF modulated baseband distortion terms. Using Equation (5), the relationship between the input signal and the feedback signal is derived as shown in Equation (6).

here,

 to be.

From Equation (6), the real and imaginary values of the power amplifier model coefficients can be obtained and used to generate the output in the virtual power amplifier model shown in FIG.

6 is computer simulation data comparing the performance of the broadband signal transmitting apparatus according to the present invention.

6 shows the spectrum in the frequency domain of the transmission signal, and when the predistorter is not used (red color), the band of the original transmission signal (dotted line) is extended by about five times. Because of this effect, the transmitted signal may be difficult to meet the frequency spectrum mask requirements, and the earned value management (EVM) characteristics of the transmitted signal may also deteriorate. It can be seen that the application of the predistorter is similar to the spectrum of the original transmission signal without distortion. Even when the proposed method is applied to the direct conversion feedback and the low-IF feedback method, It can be seen that the linearization performance is almost the same. In other words, almost the same performance can be obtained even with much relaxed ADC conditions.

The proposed scheme is a form of adding a virtual power amplifier model and an additional analog distortion effect to a conventional predistorter, and it is possible to implement it directly in digital hardware or in software to model this effect. The proposed method is somewhat different depending on the technique applied to the analog circuit of the feedback path. It can be applied to both the direct conversion method and the low-IF method. If direct conversion is used, a decimator or a random sampler that reflects the sampling of the actual ADC is added to the output of the virtual predistorter. In the case of the low-IF scheme, a virtual predistorter output and a decimator Or a portion that converts the virtual predistorter output to low-IF between random samplers is added.

Since the proposed method can use the output signal of the virtual power amplifier as the actual feedback signal after estimating the model of the virtual power amplifier, the predistorter design technique using the conventional high speed sampled feedback signal can be applied arbitrarily have. In the previous description, a simple model of low order is used for conceptual understanding of the proposed method, but the proposed method for both parameter-based models such as random order volterra model, memory polynomial model, and simple polynomial model Can be applied. As the algorithm for estimating the model, it is possible to apply least squares (LS), LMS, RLS and various other derivation algorithms based on the equations described in the description of the present invention.

While the present invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not limited to the disclosed embodiments. It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the scope of the present invention.

110, 210, 310: predistorter 120, 220, 320: DAC
130, 230, 330: Modulator 140, 240, 340: Power amplifier
150, 250, 350: Demodulator 160, 260, 360: ADC
170, 270, 370: Adaptive algorithm fulfillment unit
280, 380: virtual power amplifier unit

Claims (10)

A predistorter for predistorting the baseband digital transmission data to generate a predistortion signal;
A digital-analog converter for converting the predistortion signal into an analog signal;
A modulator for modulating the analog signal output from the digital-to-analog converter;
A power amplifier for amplifying a signal output from the modulator to generate an RF transmission signal;
A demodulator for demodulating the RF transmission signal;
An analog-to-digital converter for converting an analog signal output from the demodulator into a digital signal;
A virtual power amplifier unit for generating a feedback signal using the predistorted signal and the digital signal; And
And an adaptive algorithm transferring unit for generating a first control signal for setting the predistorter using the baseband digital transmission data, the predistortion signal, and the feedback signal.
2. The power amplifier according to claim 1,
A power amplifier model for determining an internal parameter in response to a second control signal and generating the feedback signal using the predistortion signal;
A down-sampler for down-sampling the feedback signal; And
And a model estimator for generating the second control signal using at least two of the predistortion signal, the downsampler output signal, and the digital signal.
3. The method of claim 2,
Wherein the sampling rate of the analog-to-digital converter is lower than the sampling rate of the digital-to-analog converter.
A predistorter for predistorting the baseband digital transmission data to generate a predistortion signal;
A digital-analog converter for converting the predistortion signal into an analog signal;
A modulator for modulating the analog signal output from the digital-to-analog converter;
A power amplifier for amplifying a signal output from the modulator to generate an RF transmission signal;
A demodulator for demodulating the RF transmission signal;
An analog-to-digital converter for converting an analog signal output from the demodulator into a low-IF digital signal that is a signal of a low intermediate frequency band;
A virtual power amplifier unit for generating a feedback signal using the predistortion signal and the low-IF digital signal; And
And an adaptive algorithm transferring unit for generating a first control signal for setting the predistorter using the baseband digital transmission data, the predistortion signal, and the feedback signal.
5. The apparatus of claim 4, wherein the virtual power amplifier unit comprises:
A power amplifier model for setting an equivalent model of the power amplifier that generates the feedback signal using the predistortion signal and the second control signal;
A multiplier for multiplying the feedback signal by an intermediate frequency signal to generate an intermediate frequency feedback signal;
A down-sampler for down-sampling the intermediate frequency feedback signal; And
And a model estimator for generating the second control signal using at least two signals of the predistortion signal, the downsampler output signal, and the low-IF digital signal.
The method according to claim 3 or 5,
Wherein the power amplifier model is set based on a plurality of polynomials and the second control signal determines a value of a plurality of coefficients included in the polynomial.
6. The apparatus of claim 3 or 5,
Generating a second control signal by applying a least squares technique to the predistortion signal and the digital signal or applying a least squares technique to the predistortion signal and the low IF digital signal to generate the second control signal Wherein the wideband signal transmission apparatus comprises:
6. The apparatus of claim 3 or 5,
And generating a second control signal by applying a least squares technique to the predistortion signal, the downsampler output signal, and the digital signal, or applying the least squares technique to the predistortion signal, the downsampler output signal, Wherein the second control signal is generated by applying a least squares technique.
6. The apparatus of claim 3 or 5, wherein the demodulator and the analog-
And the power amplifier model constituting the virtual power amplifier unit is not used after the power amplifier model is set.
10. The apparatus of claim 9,
After the power amplifier model constituting the virtual power amplifier unit is set,
And applying a least squares technique to the output signal of the predistortion signal and the downsampler to generate the second control signal by applying a least squares technique to the predistortion signal and the output signal of the downsampler, Wherein the wideband signal generator generates the wideband signal.

Images