KR20100013613A - Digital predistortion apparatus and method for signal in harmonic bands - Google Patents

Abstract

The present invention relates to a digital predistortion apparatus and method for harmonic band signals. The present invention relates to a harmonic signal by calculating a predistortion parameter for each of a magnitude shift and a phase shift between a source signal and a harmonic signal for the source signal. The present invention provides a digital predistortion apparatus and method for harmonic band signals to transmit signals in harmonic (harmonic) bands as well as source frequency bands.

To this end, the present invention, in the digital pre-distortion apparatus, after extracting the magnitude shift and phase shift between the source signal and the harmonic signal for the source signal, the predistortion parameter for each of the magnitude shift and the phase shift is obtained. Digital signal processing means for calculating; And predistortion means for predistorting the harmonic signal using the predistortion parameter calculated by the digital signal processing means.

Description

Digital predistortion device of harmonic band signal and its method {DIGITAL PREDISTORTION APPARATUS AND METHOD FOR SIGNAL IN HARMONIC BANDS}

The present invention relates to a digital predistortion apparatus and method for harmonic band signals, and more particularly, to calculate a predistortion parameter for each magnitude shift and phase shift between a source signal and a harmonic signal for the source signal. The present invention relates to a digital predistortion apparatus and method for harmonic band signals for pre-distorting harmonic signals, so that signals can be transmitted not only at the source frequency band but also at the harmonic (harmonic) band.

In general, the RF (Radio Frequency) power amplifier has a nonlinear characteristic, and the harmonic (harmonic) signal is generated by the nonlinear characteristic. Harmonic means a signal component occurring in a frequency band corresponding to a multiple of the source frequency. Normally, harmonics are generated when passing through a nonlinear system. They are unnecessary components in terms of source frequency and operate as noise components for other signals.

Recently, technologies such as Software Defined Radio (SDR) and Cognitive Radio (CR) have come into the spotlight as next generation wireless communication technologies. SDR is a technology that minimizes the number of RF devices or systems, and transmits the transmission signal in multimode / multiband by changing software in the digital band. CR is a technology that recognizes the frequency environment around the user and improves the frequency usage efficiency by transmitting a signal from the secondary user in a band not used by the primary user.

On the other hand, there are various techniques for linearizing a nonlinear transmission signal in a transmission system. In particular, the Digital PreDistortion (DPD) technique is a technique for linearizing the characteristics of the overall transmission system by implementing the inverse function of the transmitter's nonlinear characteristics in the digital baseband. The digital predistortion technique is in the spotlight compared to other linearization techniques due to economical efficiency and ease of implementation.

1 is a block diagram of an embodiment of a conventional digital predistortion apparatus.

As shown in FIG. 1, a conventional digital predistortion device predistorts a baseband digital input signal by using a precompensation gain calculated through polynomial coefficients modeling the inverse nonlinear distortion characteristics of the power amplifier 104. Pre-distorter 101, digital-to-analog converter 102, and digital-to-analog converter 102 for converting the digital input signal distorted by the pre-distorter 101 into an analog signal of intermediate frequency IF. Frequency upconverter 103 for converting the analog signal converted into a high frequency signal in a corresponding frequency band, a power amplifier 104 for amplifying the high frequency signal converted in the frequency upconverter 103, and the power amplifier ( A filter 105 for removing the out-of-band signal from the high frequency signal amplified by 104, and the frequency of the signal filtered by the filter 105 A frequency down converter 106 for lowering the number, an analog / digital converter 107 for converting the analog signal converted in the frequency down converter 106 into a digital signal, and digital converted in the analog / digital converter 107. A digital signal processor for determining coefficients for calculating a precompensation polynomial using a signal and a baseband digital input signal, and for calculating precompensation gains for all possible magnitudes of the input signal using the determined polynomial coefficients. 108).

Here, the frequency up-converter 103 has a reference clock (Ref. CLK) from a phase locked loop (PLL) and a transmit local oscillation signal generated by an oscillator. : Mixer (LOTX) to the filtered intermediate frequency output to produce a desired frequency (Mixer).

Such a conventional digital predistortion device can only linearize the output components of the source frequency band, and has a problem in that the harmonic signal for the source frequency signal cannot be linearized.

That is, the conventional digital predistorter can only linearize the output components of the source frequency band by extracting the nonlinear characteristics of the signal near the output center frequency which is the same as the center frequency of the input frequency, and calculating the inverse function for this function. However, there is a problem in that the harmonic signal with respect to the source frequency signal cannot be linearized.

In addition, the conventional digital predistorter cannot linearize the harmonic signal with respect to the source frequency signal, and thus transmits the signal through the 1.6 GHz or 2.4 GHz (IMS) band as well as the 800 MHz (3G) source frequency band. In order to be applied to a multiband transmitter, there is a problem in that an additional RF element must be added to the multiband transmitter.

It is an object of the present invention to solve this problem.

Accordingly, the present invention calculates a pre-distortion parameter for each of the magnitude shift and phase shift between the source signal and the harmonic signal for the source signal, and predistorts the harmonic signal, thereby not only the source frequency band but also the harmonic (harmonic) band. It is an object of the present invention to provide a digital pre-distortion apparatus and method of harmonic band signals for transmitting a signal in.

The objects of the present invention are not limited to the above-mentioned objects, and other objects and advantages of the present invention which are not mentioned above can be understood by the following description, and will be more clearly understood by the embodiments of the present invention. Also, it will be readily appreciated that the objects and advantages of the present invention may be realized by the means and combinations thereof indicated in the claims.

The apparatus of the present invention for achieving the above object, in the digital pre-distortion apparatus, after extracting the magnitude shift and phase shift between the source signal and the harmonic signal for the source signal, respectively, the magnitude shift and the phase shift Digital signal processing means for calculating a predistortion parameter for the signal; And predistortion means for predistorting the harmonic signal using the predistortion parameter calculated by the digital signal processing means.

In addition, the method of the present invention for achieving the above object, in the digital pre-distortion method, a variation extraction step of extracting the magnitude shift and phase shift between the source signal and the harmonic signal for the source signal; Calculating a pre-distortion parameter for each of the extracted magnitude shift and phase shift; And a predistortion step of predistorting the harmonic signal using the calculated predistortion parameter.

On the other hand, the present invention extracts the characteristics of the source frequency band and harmonic signals in the field of extracting the nonlinear model of the RF power amplifier, the design of the digital predistorter based on the same, and based on the nonlinear characteristics of the harmonic signal By linearizing, the signal is transmitted not only at the source frequency band but also at the harmonic (harmonic) band.

In addition, the present invention proposes a technique for effectively modeling the characteristics of the source frequency and the harmonic signal and an implementation technique of the digital predistorter using the same, thereby reducing the distortion of the source frequency signal and simultaneously transmitting the signal to the harmonic frequency without distortion. Applicable to mode / multiband transmission systems.

In addition, the present invention can effectively build an SDR or CR system by transmitting a high-quality linearized signal in the harmonic signal band that is an unwanted component of the noise signal.

In addition, the present invention can pre-process the harmonic band signal in which the noise component distorted by the nonlinear component is generated at the baseband using digital technology, thereby constructing a multiband / multimode system without the addition of RF hardware in the radio transmitter. To make sure.

In addition, the present invention, in order to build a system for transmitting a high quality signal in the harmonic band using a general wireless transmission system, after sampling the non-linear signal of the source frequency band and harmonic frequency band and measured through an analog / digital converter, Using the measured signal, nonlinear characteristics of the source frequency and harmonic band of the power amplifier are extracted.

In addition, the present invention is applicable to a multiband / multimode system capable of transmitting signals through the source frequency band and the harmonic band by using nonlinear characteristics of the source frequency and the harmonic band.

In addition, the present invention is applied to a multi-band / multi-mode system, it is helpful to use the frequency utilizing the frequency of emergency or unused. For example, 800 MHz (3G) signals can be sent in harmonic bands such as 1.6 GHz, 2.4 GHz (ISM band) and 3.2 GHz band without changing the transmitter. In addition, the signals related to the 4G system (700MHz band) can be sent to the 3G system (2.1GHz band) currently used.

The present invention as described above, by calculating the pre-distortion parameter for each of the magnitude shift and phase shift between the source signal and the harmonic signal for the source signal to pre-distorted the harmonic signal, harmonics as well as the source frequency band ( Harmonic) has the effect of transmitting a signal in the band.

The above objects, features, and advantages will become more apparent from the detailed description given hereinafter with reference to the accompanying drawings, and accordingly, those skilled in the art to which the present invention pertains may share the technical idea of the present invention. It will be easy to implement. In addition, in describing the present invention, when it is determined that the detailed description of the known technology related to the present invention may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted. Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

2 is a block diagram of a digital predistortion apparatus of a harmonic band signal according to the present invention.

As shown in FIG. 2, the digital predistortion apparatus of the harmonic band signal according to the present invention extracts a magnitude shift and a phase shift between a source signal and a harmonic signal for the source signal, and then the magnitude shift and the phase shift. A digital signal processor (DSP) 210 for calculating a pre-distortion parameter for each variation, and a pre-distorter for pre-distorting the harmonic signal using the pre-distortion parameter calculated by the digital signal processor 210 ( DPD) 220.

Here, the digital signal processor 210 corrects the magnitude and phase of the harmonic signal with respect to the source signal.

In addition, the controller 230 sets the order of the harmonic signals processed by the predistorter 220, the filter 240, and the frequency downconverter 250, thereby digitally pre-translating the signals in the source frequency band. Distortion and digital pre-distortion can be selected to transmit signals in the harmonic band.

That is, the predistorter 220 sets the phase of the source signal and the phase of the harmonic signal to 1 / M, respectively, in which the M value is set by the controller 230. In addition, the filter 240 filters the M-order harmonic signal from the output signal of the power amplifier, where the M value is set by the controller 230. In addition, the frequency down converter 250 lowers the frequency of the M-th harmonic signal filtered by the filter 240 by Mω, where the M value is set by the controller 230.

Hereinafter, the nonlinear model between the source frequency and the harmonics applied to the digital signal processor 210 will be described in detail.

The real Taylor model, which exhibits nonlinear characteristics, is a non-memory form that considers only odd term components and baseband complex envelope signals in the Taylor series to analyze nonlinear effects in the source frequency band.

The real Taylor series has an AM / AM (Amplitude Modulation to AM) characteristic in which the output signal scales according to the magnitude of the input signal, and AM / PM (AM to Phase Modulation) in which the phase of the output signal changes in accordance with the magnitude of the input signal. To characterize together, we can expand to a complex Taylor series with complex coefficients and complex envelopes. At this time, the nonlinearity between the input signal and the harmonic output signal is as shown in FIG.

In the nonlinear model between the source frequency and the harmonics, the input signal x (t) of the pass band and the output signal y (t) of the wide band are expressed by Equation 1 below.

는

를 만족하고,

와

는 x(t)와

의 복소수 포락선 신호를 나타내며, r(t)와

는 입력 포락선 신호의 크기와 위상을 각각 나타내는 실수 신호이다. 또한, y(t)에서 n은 비선형성의 최대 차수를 나타낸다.here,

Is

Satisfy

Wow

Is x (t)

Represents a complex envelope signal of r (t) and

Is a real signal representing the magnitude and phase of the input envelope signal, respectively. In addition, n in y (t) represents the maximum degree of nonlinearity.

Equation 1 is an example of a nonlinear model having no memory. This can be applied to nonlinear models with memory. That is, a model having a non-memory may be represented by a Taylor series as shown in Equation 2 below.

는

를 만족하고, 따라서 y(t)는

를 만족한다.here,

Is

, So y (t) is

Satisfies.

를 짝수항 m과 홀수항 m으로 각각 나타낼 수 있다.At this time, as shown in [Equation 3] and [Equation 4] below using the [Equation 2],

Can be represented by an even term m and an odd term m, respectively.

Harmonic component for even terms: m = 2p (p = 0,1,2,3,…)

Harmonic component for odd term: m = 2p + 1 (p = 0,1,2,3,…)

From Equations 3 and 4, the relationship between the input complex envelope signal and the m-th harmonic band output envelope signal can be summarized as in Equation 5 below.

를 만족한다. 즉, 테일러 시리즈 형식의 작은 메모리 효과를 가진 AM/PM 특성을 나타내기 위해 복소 상관계수

를 도입하였다. 이때, m차 고조파 대역 출력신호의 실제 측정값은 AM/PM 왜곡으로 나타나므로, 일반적으로 복소 상관계수

를 고려할 수 있다.Where n 'represents the quotient of (nm) divided by 2,

Satisfies. That is, complex correlation coefficient to represent AM / PM characteristic with small memory effect of Taylor series format.

Was introduced. In this case, since the actual measured value of the m-th harmonic band output signal is represented by AM / PM distortion, it is generally a complex correlation coefficient.

May be considered.

In fact, when extracting AM / AM _m and AM / PM _m from the measurement signal, AM / AM _m can be measured simply due to the source frequency input band signal and the m-th harmonic band signal. However, the extraction of AM / PM _m is related differently. This is because the phase must be corrected.

Here, in order to extract AM / AM _m and AM / PM _m , the input signal of the source frequency band and the output signal of the m-th harmonic band are used as the complex (inphase) and imaginary (quadature) signals.

와 같이 조절한다.Since the input signal and the output signal are composed of m-th harmonics, the magnitude and phase of the signal are not synchronized. Therefore, in order to match the magnitude and phase of the m-th harmonic signal,

Adjust as shown.

과

의 크기나 복소 신호(I/Q) 사이의 상관도를 사용하여 맞춘다. 여기서,

는 시간 지연을 나타내고,

는 시간 지연의 상관성이 가장 높은 시간 보정 값을 나타낸다. 입력신호와 출력신호의 시간 조절 후 AM/AM_m을 추출한다.And the delay of input signal and output signal

and

Adjust using the magnitude and the correlation between the complex signal (I / Q). here,

Represents a time delay,

Denotes a time correction value having the highest correlation of time delay. After adjusting the time of input signal and output signal, extract AM / AM _m .

의 변경을 통해 I/Q의

와

의 위상을 조절한다. 여기서, 위상 상관성이 가장 높은 값을 위상 보정 값

라 한다. 이때, 복소 신호 실수값(I)에 대해 조절하였다면 복소 신호 허수값(Q)에 대입하여 출력 값을 확인할 수 있다.Phase correction is then adjusted using the complex signal (I / Q). Ie phase

Through the change of I / Q

Wow

Adjust the phase of Here, the phase correction value is the highest phase correlation value.

It is called. In this case, if the complex signal real value I is adjusted, the output value may be checked by substituting the complex signal imaginary value Q.

Here, in order to reduce the phase extraction time, a correlation is calculated for a predetermined number of phase values having the same interval (for example, seven), and a phase value having the highest correlation is selected. The phase correction value is detected by calculating a correlation between the selected phase value and each phase value between the two nearest phase values.

For example, in order to detect the phase correction value while changing by 1 degree for the entire 360 degree, 360 steps are required. Therefore, as an example, 0 degrees, 45 degrees, 90 degrees, 135 degrees, 180 degrees, 225 degrees, and 270 degrees are set as the initial target phases, and the correlations for each are selected to select a phase representing the maximum value.

The phase correction value is detected for each phase within a range of values smaller and larger than the selected phase in the initial target phase. At this time, if the selected phase is 180 degrees, the phase correction value having the highest correlation is detected while changing by 1 degree in the range of 135 degrees and 225 degrees.

As a result, the AM / AM _m function and the AM / PM _m function whose magnitude and phase are corrected are shown in Equation 6 below.

과 출력신호

의 AM/AM 및 AM/PM을 의미한다. 예를 들어, AM/AM₁과 AM/PM₁을 알 경우 AM/AM₃과 AM/PM₃, AM/AM₅와 AM/PM₅를 알 수 있다.Where AM / AM _m And AM / PM _m are input signals

And output signal

Means AM / AM and AM / PM. For example, if AM / AM ₁ and AM / PM ₁ are known, AM / AM ₃ and AM / PM ₃ and AM / AM ₅ and AM / PM ₅ can be known.

Hereinafter, the operation of the predistorter 220 will be described in more detail.

이며, 원천 주파수 대역의 통과 대역 신호는 하기의 [수학식 7]과 같다.The output complex envelope signal of the predistorter 220

The passband signal of the source frequency band is expressed by Equation 7 below.

의 크기와 위상을 각각 나타내는 실제 신호이다.Where b (t) and d (t)

The actual signal representing the magnitude and phase of each.

와 같게 되는 m차 계수 대역의 복소 포락선 신호가 필요하다. G는 선형이득을 나타낸다.At this time,

There is a need for a complex envelope signal of the order-order coefficient band equal to < RTI ID = 0.0 > G stands for linear gain.

가 필요하다.Therefore, satisfying Equation 8 below

Is needed.

The digital predistorter satisfying Equation 8 uses a numerical analysis method and a look up table (LUT) method for linearization. At this time, a digital predistorter for linearizing the m-th harmonic band is defined as DPD _m .

For example, DPD ₂ represents a state of linearizing an output signal of a second harmonic band in a digital predistorter, DPD ₃ represents a state of linearizing an output signal of a third harmonic band in a digital predistorter, and DPD ₁ Denotes a state in which the digital predistorter linearizes the output signal (source signal) of the first harmonic band.

Hereinafter, the numerical method will be described.

In order to obtain an analytical solution, a polynomial for b (t) is defined as shown in Equation 9 below.

Therefore, the following [Equation 10] is satisfied.

Equation 10 is the same as Equation 11 below.

Here, the relationship as shown in [Equation 12] can be obtained by the multiple of the inverse matrix.

At this time, b (t) from [Equation 12] should satisfy the following [Equation 13].

[Equation 13] can be solved using a numerical method such as 'Newton', 'Secant'. Through the root of Equation 13, d (t) is extracted as shown in Equation 14 below.

Finally, DPD _m plays the same role as the following [Equation 15].

Hereinafter, a method based on the LUT will be described with reference to FIG. 4.

Numerical methods are available for low order nonlinear models. However, if the order is high, it takes a long time to find b (t). Therefore, LUT (Loop up table) method is suitable.

를

으로 변환하며, LUT_AM_m과 LUT_PM_m은 AM/AM_m과 AM/PM_m의 역함수(사전 왜곡 파라미터)로 동작한다.That is, using the functions AM / AM _m and AM / PM _m ,

To

LUT_AM _m and LUT_PM _m operate as inverse functions (pre-distortion parameters) of AM / AM _m and AM / PM _m .

5 is an exemplary view showing the performance of the digital predistortion apparatus of the harmonic band signal according to the present invention.

As shown in FIG. 5, (a) represents an input signal of a source frequency band, (b) represents an output signal of a third harmonic band extracted from a signal (a) that has passed through a power amplifier, and (c) Denotes an input signal of a second source frequency band in which the output signal of the third harmonic band of (b) is pre-distorted when the third harmonic is selected by the controller 230, and (d) denotes an input signal passing through the power amplifier. The input signal of the source frequency band of 2, that is, the output signal of the 3rd harmonic band.

Figure 6 is a harmonic signal having a triple in the case As another example also, when not applied the DPD ₃ and applying the DPD _3, the source frequency of performance of the digital pre-distortion system of the harmonic band signal in accordance with the present invention 16-QAM and 64-QAM signals are shown in complex coordinates.

The (a) as shown in Figure 6 represents the DPD ₃ unapplied 16-QAM signal, (b) the DPD ₃ application shows a 16-QAM signal, (c) denotes a DPD ₃ unapplied 64-QAM signal, ( d) shows DPD ₃ applied 64-QAM signal.

Here, (a) and (c) can not recognize the shape of the signal due to the non-linear distortion occurs in the third harmonic signal, but (b) and (d) can determine the shape of the signal by linearizing the third harmonic signal. .

Further, the effective value (rms) of the EVM (Error Vector Magnitude) of the 16-QAM signal is not applied to 93% when the DPD ₃ and is 6.4% when applied to the DPD _3. The effective value of the EVM for the 64-QAM is 88% when unapplied the DPD _3, and is 6.5% when applied to the DPD _3.

In this case, since the EVM has a high noise of the 3rd high frequency output signal and is not perfect due to various memory effects, it is preferable to use the "Wiener-Hammerstein model" and the "Voterra model" in consideration of the memory effect.

7 is a flowchart illustrating a digital predistortion method of a harmonic band signal according to the present invention.

First, a magnitude shift and a phase shift between a source signal and a harmonic signal for the source signal are extracted (701).

Thereafter, a pre-distortion parameter for each of the extracted magnitude shift and phase shift is calculated (702).

Thereafter, the harmonic signal is predistorted using the calculated predistortion parameter (703).

On the other hand, the method of the present invention as described above can be written in a computer program. And the code and code segments constituting the program can be easily inferred by a computer programmer in the art. In addition, the written program is stored in a computer-readable recording medium (information storage medium), and read and executed by a computer to implement the method of the present invention. The recording medium may include any type of computer readable recording medium.

The present invention described above is capable of various substitutions, modifications, and changes without departing from the technical spirit of the present invention for those skilled in the art to which the present invention pertains. It is not limited by the drawings.

The present invention can be used for a multiband / multimode transmission system and the like.

1 is a configuration diagram of an embodiment of a conventional digital predistortion apparatus;

2 is a block diagram of an embodiment of a digital predistortion apparatus of a harmonic band signal according to the present invention;

3 is an example diagram illustrating nonlinearity between an input signal and a harmonic output signal according to the present invention;

4 is a configuration diagram of an embodiment of a predistorter according to the present invention;

5 is an exemplary view illustrating performance of a digital predistortion apparatus of a harmonic band signal according to the present invention.

6 is another exemplary diagram showing the performance of the digital predistortion apparatus of the harmonic band signal according to the present invention;

7 is a flowchart illustrating a digital predistortion method of a harmonic band signal according to the present invention.

* Explanation of symbols for the main parts of the drawings

210: digital signal processor (DSP) 220: predistorter (DPD)

230: controller

Claims (8)

In the digital predistortion apparatus, Digital signal processing means for extracting a magnitude shift and a phase shift between a source signal and a harmonic signal for the source signal and calculating a pre-distortion parameter for each of the magnitude shift and the phase shift; And Predistortion means for predistorting the harmonic signal using the predistortion parameter calculated by the digital signal processing means Digital pre-distortion device of the harmonic band signal comprising a. The method of claim 1, Further comprising control means for setting the phase adjustment value (M) of the pre-distortion means, The pre-distortion means, And a phase of the source signal and a phase of the harmonic signal, respectively, 1 / M. The method according to claim 1 or 2, The digital signal processing means, The digital predistortion device of the harmonic band signal, characterized in that for correcting the magnitude and phase of the harmonic signal with respect to the source signal. The method of claim 3, wherein The digital signal processing means, Correlation is calculated for a predetermined number of phase values having the same interval to select the most correlated phase value, and then each phase value between two phase values located closest to the selected phase value among the predetermined number of phase values. Digital pre-distortion apparatus of harmonic band signal, characterized in that the phase correction value is detected by calculating the correlation with each other. In the digital predistortion method, A shift extraction step of extracting a magnitude shift and a phase shift between a source signal and a harmonic signal for the source signal; Calculating a pre-distortion parameter for each of the extracted magnitude shift and phase shift; And Pre-distortion step of pre-distorting the harmonic signal using the calculated pre-distortion parameter Digital pre-distortion method of the harmonic band signal comprising a. The method of claim 5, wherein The pre-distortion step, And dividing the phase of the source signal and the phase of the harmonic signal by a predetermined phase adjustment value. The method according to claim 5 or 6, The mutation extraction step A correction step of correcting the magnitude and phase of the harmonic signal with respect to the source signal Digital pre-distortion method of the harmonic band signal comprising a. The method of claim 7, wherein The correction step, Calculating correlations with a predetermined number of phase values having the same interval to select the phase values having the highest correlation; Calculating a correlation between the phase values between two phase values closest to the selected phase value among the predetermined number of phase values; And Detecting a phase value having the largest correlation as the phase correction value Digital pre-distortion method of the harmonic band signal comprising a.

Images