JP2001036353A - Distortion compensating method and radio communications equipment - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable high-accuracy compensation with the table memory of small capacity and an arithmetic logic in small scale, in the case of compensating the nonlinear distortion of a power amplifier through a predistortion method. SOLUTION: In polynomial arithmetic parts 21 and 22, amplitudes Vi and Vq of baseband signals I and Q from roll-off filters 11 and 12 are respectively substituted to a polynomial g (v) for amplitude distortion compensation, and the arithmetic results g (Vi) and g(Vq) are calculated. A square arithmetic part 23 calculates an arithmetic output Miq2 expressed as Miq2=(g(Vi))2+(g(Vq)2. Corresponding to this arithmetic output Miq2, a phase value θ in one- to-one correspondence with the arithmetic output Miq2 is read out of a table 24 for phase distortion compensation. Synthetic arithmetic parts 25 and 26 calculate baseband signals I' and Q', which are expressed as I'=g(Vi)×cosθ-g(Vq)×sinθ and Q'=g(Vq)×cosθ+g(Vi)×sinθ.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for compensating for non-linear distortion of a power amplifier of a transmission section of a radio communication apparatus, and a radio communication apparatus for compensating for non-linear distortion of a power amplifier of a transmission section.

[0002]

2. Description of the Related Art In a digital radio communication device, the linearity required for a power amplifier for transmission is becoming severer as the speed and capacity of communication are increased. However, when the linearity of the power amplifier is strict, the power efficiency of the power amplifier is reduced, and, for example, in a terminal phone of a digital mobile phone system, it is difficult to prolong the continuous talk time.
Therefore, it has been considered to improve the power efficiency of the power amplifier by compensating for the nonlinear distortion of the power amplifier.

As a method of compensating for nonlinear distortion of a power amplifier, several methods such as a negative feedback method and a feedforward method have been proposed. Recently, however, a method has been proposed in accordance with a nonlinear amplitude characteristic and a phase characteristic of a power amplifier. Therefore, a predistortion method in which a baseband signal is distorted in advance so as to cancel the amplitude distortion and the phase distortion has attracted attention.

[0004] FIG. 11 is a document "1999 IEICE General Conference, paper BS-36" Experimental verification of nonlinear distortion compensation of transmission system using predistortion method "".
Shows a conventional pre-distortion method shown in FIG. 1 and uses a table as means for creating compensation data.

That is, in the method shown in FIG. 11, the power calculator 61 of the distortion compensator 60 calculates Vi ² + V from the amplitudes Vi and Vq of the digital orthogonal baseband signals I and Q before compensation.
An operation output represented by q ² is calculated, and the operation output (V
i ² + Vq ² ), the amplitude distortion compensating data and the phase distortion compensating data corresponding to the operation output (Vi ² + Vq ² ) written in advance in the table 62 are read out from the table 62, and the complex product operation unit 63 Then, the complex product of the baseband signals I and Q and the compensation data read from the table 62 is calculated, and the distortion-compensated digital orthogonal baseband signals I ′ and Q ′ are obtained from the distortion compensator 60. .

The distortion-compensated digital quadrature baseband signals I 'and Q' are supplied to D / A converters 31 and 32, where they are converted into analog signals, respectively.
The high-frequency signal is supplied to the quadrature modulation unit 40 through 3 and 34 and is orthogonally modulated from the quadrature modulation unit 40. The high-frequency signal is amplified by the power amplifier 50.

FIG. 12 shows a document "Spectral Co."
natural by Predistortion
no of OQPSK Signals, Microwa
vejournal, October, 1998, p.
p. 22-37 ", a function approximating a nonlinear amplitude characteristic and a phase characteristic of a power amplifier is defined so as to be fitted to an actual characteristic. An inverse function that cancels the characteristic is obtained, and compensation data is obtained from the inverse function.

That is, in the method shown in FIG. 12, the amplitude inverse function calculator 65 and the phase inverse function calculator 6 of the distortion compensator 60 are used.
In step 6, the digital quadrature baseband signals I and Q before compensation are respectively substituted into the inverse functions of the amplitude and the phase obtained as described above, and the respective inverse functions are calculated. A complex product of the calculation result of the amplitude inverse function and the calculation result of the phase inverse function is calculated, and the
The digital quadrature baseband signals I ′ and Q ′ that have been subjected to distortion compensation are obtained from Subsequent processing is the same as the method of FIG.

[0009]

However, FIG.
According to the conventional pre-distortion method shown in (1), since a large amount of data for compensating for the amplitude distortion and the phase distortion of the power amplifier 50 is written in the table 62, especially the amplitude distortion for which high-precision compensation is required is compensated. In order to write a large amount of data into the table 62, there is a problem that a large-capacity memory is required as a memory constituting the table 62.

In the conventional pre-distortion method shown in FIG. 12, an inverse function operation unit 65 and a phase inverse function operation unit 66 perform an inverse function operation for each of the amplitude and the phase, and a complex product operation unit 67. Therefore, since the complex product operation is performed on the operation results, the amount of operation in the distortion compensation unit 60 becomes enormous, and there is a problem that the distortion compensation unit 60 requires a large-scale operation logic.

In view of the above, the present invention is such that when a non-linear distortion of a power amplifier is compensated for by a pre-distortion method, a high-precision compensation can be performed by a small-capacity table memory and a small-scale arithmetic logic. is there.

[0012]

A distortion compensation method according to the present invention is a method for compensating for non-linear distortion of a power amplifier of a transmission unit of a radio communication apparatus, wherein amplitudes Vi and Vq of orthogonal baseband signals I and Q are determined. , Respectively, a polynomial g for amplitude distortion compensation
(V) to calculate operation results g (Vi) and g (Vq), and from the operation results g (Vi) and g (Vq), Miq ² = (g (Vi)) ² + (g (Vq)) to calculate the arithmetic output Miq ² represented by ^2, the operation output M
From iq ² , the phase value θ corresponding to the operation output Miq ² and the cosine value cos θ and the sine value sin θ are read from the table for phase distortion compensation, one by one,
The phase value θ, or the cosine value cos θ and the sine value si
From nθ and the calculation results g (Vi) and g (Vq), I ′ = g (Vi) × cosθ−g (Vq) × sinθ Q ′ = g (Vq) × cosθ + g (Vi) × sinθ The orthogonal baseband signals I ′ and Q ′ to be calculated are calculated.

In this case, cos θ = 1 and sin θ =
As θ, the orthogonal baseband signals I ′ and Q ′ represented by I ′ = g (Vi) −g (Vq) × θ Q ′ = g (Vq) + g (Vi) × θ can be calculated.

[0014] ^{Alternatively, cosθ = 1-θ 2/} 2, and as ^{sinθ = θ-θ 3/6} , I '= g (Vi) × (1-θ 2/2) -g (Vq) ×
^{(Θ-θ 3/6)} Q '= g (Vq) × (1-θ 2/2) + g (Vi) ×
Quadrature baseband signals I represented by ^{(θ-θ 3/6)} ', Q' can be calculated.

The square root may be calculated from the operation output Miq ² , and the phase value θ or the cosine value cos θ and the sine value sin θ may be read from the phase distortion compensation table based on the calculation result Miq. Needless to say, the present invention includes such a case.

In the above-described distortion compensation method, a small amount of data for compensating the phase distortion of the power amplifier may be written in the table, so that the memory constituting the table may have a small capacity. In addition, the polynomial operation is performed only on the amplitude, and the synthesis operation also obtains cos θ and sin θ from the phase value θ from the table and multiplies the polynomial operation results g (Vi) and g (Vq), or directly from the table. Since only the read cos θ and sin θ are multiplied by the polynomial operation results g (Vi) and g (Vq), the amount of operation for distortion compensation becomes small, and distortion compensation is performed by a small-scale operation logic. Can be.

Further, when an approximate expression consisting of only one or two terms of the Taylor series is used as cos θ and sin θ, the operation for distortion compensation can be simplified without impairing the effect of distortion compensation. Can be.

[0018]

DESCRIPTION OF THE PREFERRED EMBODIMENTS [Configuration of Radio Communication Device] FIG.
An example of a wireless communication device of the present invention, that is, an example of a wireless communication device in which nonlinear distortion of a power amplifier of a transmission unit is compensated for by the distortion compensation method of the present invention, is a case of a terminal phone of a digital mobile phone system.

In this radio communication apparatus, the digital quadrature baseband signals Io and Qo are supplied to the roll-off filters 11 and
12 to remove the respective high frequency components,
The digital quadrature baseband signals I and Q from the roll-off filters 11 and 12 are supplied to a distortion compensating unit 20,
The distortion-compensated digital orthogonal baseband signals I ′ and Q ′ are obtained from the distortion compensator 20 as described later, and the distortion-compensated digital orthogonal baseband signals I ′ and Q ′ are obtained.
Are supplied to D / A converters 31 and 32 and converted into analog signals, respectively, and the analog quadrature baseband signals are supplied to quadrature modulating section 40 through low-pass filters 33 and 34 and quadrature modulating section 40 A modulated high-frequency signal is obtained, and the high-frequency signal is amplified by the power amplifier 50.

[First Example of Distortion Compensation] FIG. 2 shows a first example of the distortion compensation unit 20.

In this example, the roll-off filters 11, 1
2 are supplied to the polynomial arithmetic units 21 and 22, respectively.
The amplitudes Vi and Vq of the baseband signals I and Q are obtained at 1 and 22,
Are respectively substituted into the polynomial g (v) for amplitude distortion compensation, and the calculation results g (Vi) and g (Vq) are calculated.

The polynomial g (v) is defined as follows: a is the linear voltage gain of the power amplifier after distortion compensation, and v is the input voltage of the power amplifier.
When the output voltage of the power amplifier is represented by a function f (v), the power amplifier corresponding to the input voltage g (v) of the power amplifier that satisfies Expression (1) in FIG. Is a function such that the output voltage f (g (v)) is proportional to the input voltage v of the power amplifier before compensation. Note that an asterisk (*) in each equation in FIGS. 3 and 4 means multiplication.

FIG. 5 shows typical characteristics of the power amplifier.
The solid line 1 is the amplitude characteristic. In this power amplifier, when the input power value is A, the output power value is B. However, if the amplitude characteristic of the power amplifier is linear as shown by the thin line 3, the output power value is B ', and If the power value is converted from A to A ', the output power value becomes B', and the original input power value A is converted to the output power value B '. As a result, the amplitude distortion becomes Will not occur. Thus, the polynomial for converting the input power value from A to A ′ is g (v)
It is.

In the distortion compensating section 20 shown in FIG. 2, the squaring section 23 further calculates the operation results g (V
From i) and g (Vq), a calculation output Miq ² represented by equation (2) in FIG. 3 is calculated.

Further, the distortion compensating section 20 has a table 24
And a phase value θ corresponding to a one-to-one operation output Miq ² for compensating for the phase distortion of the power amplifier.
It is written in advance, and the operation output Miq ² of the square operation unit 23 is used to output the operation output M
The phase value θ corresponding to iq ² is read.

The solid line 2 in FIG. 5 shows a typical phase characteristic of the power amplifier. In this power amplifier, when the input power is the value A ′ converted from the original value A as described above, the phase of the output after passing through the power amplifier is D, but the phase characteristic of the power amplifier is a solid line. If it is linear as shown by the line between 2 and broken line 4, the output phase will be D '', and if the input phase of the power amplifier is converted from D to D 'in advance, the output phase will be D'' And the original input phase D becomes the output phase D ''
, And as a result, no phase distortion occurs.

Therefore, the table 24 indicates the operation output Miq
^{It is} assumed that the data having the phase indicated by the broken line 4 in FIG. 5 is provided over the entire input range of the power amplifier as the phase value θ corresponding to ² : 1.

In the distortion compensating section 20 of FIG. 2, the synthesizing operation sections 25 and 26 further calculate the operation results g (Vi) and g (Vq) of the polynomial operation sections 21 and 22 and the phase value θ from the table 24.
Then, the operations represented by equations (3) and (4) in FIG. 3 are performed to calculate the distortion-compensated digital orthogonal baseband signals I ′ and Q ′.

The fact that distortion compensation is realized in the above-described example will be described below. First, it will be shown that the phase distortion of the power amplifier is compensated. Here, it is assumed that there is no amplitude distortion.

Assuming that the phase difference between the output signals I and Q of the roll-off filters 11 and 12 is Φ, this is a value of about 90 degrees, and from the equation (2) in FIG.
Using (Vi) and g (Vq) expressed by equations (5) and (6) in FIG. 3, the output voltage Vout of the power amplifier having phase distortion is expressed by equation (7) in FIG. You. Equation (7)
Vm and ω are the amplitude and angular frequency of the high-frequency signal output from the power amplifier, and h (v) is the phase shift (phase distortion) of the power amplifier.

As shown by the equations (8) and (9) in FIG.
cos (Φ) and Vm * sin (Φ) are Vini and Vin
When replaced by q, the output voltage Vout of equation (7) is represented by equation (10) in FIG.

Further, the distortion compensator 20 is connected to the power amplifier.
The input of the signals I ′ and Q ′ represented by the equations (3) and (4) from the above means that Vini and Vinq in the equation (10) are represented by the equations (11) and (12) in FIG. Therefore, the output voltage Vout of the equation (10) is equal to the output voltage Vout of the equation (1) in FIG.
3) Represented by (14) and (15). Expressions (13) to (1)
H (Vi, Vq) in 5) is h in Formulas (7) and (10).
This is a phase shift (phase distortion) corresponding to (v).

The phase value θ from the table 24 is
Since this phase shift h (Vi, Vq) is canceled out, that is, since there is a relationship shown in the equation (16) in FIG. 4 between the phase value θ and the phase shift h (Vi, Vq), Equation (1
The output voltage Vout of 5) is represented by the equation (17) in FIG. 4, and the phase shift h (Vi, Vq) is canceled.

As described above, the phase distortion of the power amplifier is compensated. Next, it will be described that the amplitude distortion of the power amplifier is compensated.

Output voltage V of power amplifier having amplitude distortion
out is obtained by using the above function f (v).
This is represented by equation (18) in FIG. Phase distortion h (Vi, V
q) is not considered here because it is compensated as described above.

The input voltages Vi and Vq of this power amplifier are
Since the voltages are converted into the voltages g (Vi) and g (Vq) by the polynomial operation, the output voltage Vam of the power amplifier after the conversion is converted.
Is obtained by converting Vi and Vq in equation (18) to g (Vi) and g (V
q), and is represented by equation (19) in FIG.

The function f (v) in equations (18) and (19) is
cos (ωt) and sin (ωt) are considered separately. This is a function f (v) originally representing amplitude distortion.
Is the amplitude of the component of angular frequency ω (DC or a component close to DC)
And cos (ωt), si
This is because n (ωt) should be considered separately.

Then, the function f (g (V (V
i)) and f (g (Vq)) are respectively expressed by equation (1) in FIG.
Thus, the output voltage Vam of the equation (19)
Is represented by Expression (20) in FIG. 4, and the amplitude distortion is canceled.

As described above, the amplitude distortion of the power amplifier is compensated. That is, according to the example of FIG. 2, the amplitude distortion and the phase distortion of the power amplifier are compensated.

Then, according to the example of FIG.
Since only a small amount of data for compensating for the phase distortion of the power amplifier need be written in the table 24, the memory constituting the table 24 may have a small capacity. Moreover, the polynomial operation unit 21,
The polynomial operation at 22 is performed only on the amplitude, and the synthesis operation at the synthesis operation units 25 and 26 is also performed according to Table 2.
Since only cos θ and sin θ are obtained from the phase value θ from 4 and multiplied by the polynomial operation results g (Vi) and g (Vq), the amount of operation for distortion compensation becomes small, and small-scale operation logic Distortion compensation can be performed.

A power amplifier having a maximum output of 29.3 dBm, a small power gain of 26.1 dB, and a maximum phase shift dH of -5.2 degrees having the amplitude and phase characteristics shown by solid lines 1 and 2 in FIG. , QPSK without distortion compensation (Quadr
FIG. 6 shows a result of calculating an output spectrum of a power amplifier including distortion obtained by inputting a ature PSK (quadrature PSK) modulated signal. In this case, the adjacent channel leakage power ratio is -35 dBc.

On the other hand, FIG. 7 shows the result of calculating the output spectrum of the power amplifier obtained by inputting the QPSK modulated signal subjected to the distortion compensation as in the above example to the same power amplifier. In this case, the adjacent channel leakage power ratio is -85 dBc, which is improved by 50 dB as compared with the case where the distortion compensation is not performed in FIG.

[Second and Third Examples of Distortion Compensation] FIG.
Shows a second example of the distortion compensator 20.

In this example, cos θ = 1 and sin θ
Assuming that θ = θ, it is expressed as I ′ = g (Vi) −g (Vq) × θ (21) Q ′ = g (Vq) + g (Vi) × θ (22) Performs an arithmetic operation. Others are the same as the first example of FIG.

FIG. 9 shows a third example of the distortion compensator 20.

[0046] In this ^{example, cosθ = 1-θ 2/} 2, and as ^{sinθ = θ-θ 3/6} , combining unit 25,2
In 6, I '= g (Vi ) × (1-θ 2/2) -g (Vq) ×
^{(Θ-θ 3/6)} Q '= g (Vq) × (1-θ 2/2) + g (Vi) ×
Performs an operation represented by ^{(θ-θ 3/6)} . Others are the same as the first example of FIG.

According to the example shown in FIG. 8 or FIG. 9, the operation in the combining operation sections 25 and 26 can be further simplified.

FIG. 10 shows a power amplifier having the characteristics shown in FIG. 5 when distortion compensation is not performed, when distortion compensation is performed in the second example in FIG. 8, and in the third example in FIG. FIG. 9 shows the results of calculating the effect of distortion compensation when the maximum phase shift dH shown in FIG. 5 is changed as a variable for each of the cases where compensation is performed.

As shown by the broken line in FIG. 10, when the maximum phase shift dH is -5.2 degrees, the distortion is improved by about 25 dB in the second example of FIG. 8 as compared with the case where no distortion compensation is performed. 9, the distortion is improved by about 60 dB in the third example shown in FIG. 9, and the effect of distortion compensation is impaired even when an approximate expression including only one or two terms of the Taylor series is used as cos θ and sin θ. You can see that it is not.

[Other Embodiments] The example of FIG.
4 is a case where the phase value θ corresponding to the operation output Miq ² is read out, but the operation output Miq ² and the cos θ and sin θ corresponding to the one-to-one correspondence are written in the table 24, and the operation output Miq ² cos
θ and sin θ may be read.

In this case, the data amount of the table 24 is increased as compared with the example of FIG. 2, but the data for compensating for the amplitude distortion and the phase distortion as in the conventional example shown in FIG. Is written in the table 62, the amount of data in the table is small, and the operations in the combining operation units 25 and 26 are easier than in the example of FIG.

In the example of FIG. 8 or FIG.
Directly from the table 24 by the operation output Miq ² , co
s θ = 1 and sin θ = θ or cos θ = 1−θ ²
/ 2 and sinθ = θ-θ ^3/6 may be read.

Further, the present invention is not limited to the power amplifier of the transmitting unit of the terminal telephone of the digital portable telephone system, but also the transmitting power amplifier of the base station and the transmitting unit of the base station or the mobile station of other wireless communication systems. The present invention can be widely applied to the case where the nonlinear distortion of the power amplifier is compensated.

[0054]

As described above, according to the present invention, when the non-linear distortion of the power amplifier is compensated by the pre-distortion method, high-precision compensation is performed by a small-capacity table memory and a small-scale arithmetic logic. Can be.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating an example of a wireless communication device according to the present invention.

FIG. 2 is a diagram illustrating a first example of a distortion compensator in FIG. 1;

FIG. 3 is a diagram showing various expressions used for explaining the example of FIG. 2;

FIG. 4 is a diagram showing various formulas used for explaining the example of FIG. 2;

FIG. 5 is a diagram illustrating a typical amplitude-phase characteristic of a power amplifier.

FIG. 6 is a diagram showing an output spectrum calculation result when distortion compensation is not performed in the power amplifier having the characteristics shown in FIG. 5;

7 is a diagram illustrating a result of calculating an output spectrum when distortion compensation is performed in the example of FIG. 2 with the power amplifier having the characteristics of FIG. 5;

FIG. 8 is a diagram illustrating a second example of the distortion compensator in FIG. 1;

FIG. 9 is a diagram illustrating a third example of the distortion compensation unit in FIG. 1;

FIG. 10 is a diagram illustrating the results of calculating the effect of distortion compensation when distortion compensation is not performed and when distortion compensation is performed in the example of FIG. 8 or 9;

FIG. 11 is a diagram showing a first example of a conventional predistortion method.

FIG. 12 is a diagram showing a second example of the conventional predistortion method.

[Explanation of symbols]

11, 12: a roll-off filter, 20: a distortion compensator,
21, 22,... Polynomial operation unit, 23, square operation unit, 24,
Tables, 25, 26 ... Synthesis operation unit, 31, 32 ... D /
A converter, 33, 34 ... low-pass filter, 40 ...
Quadrature modulator, 50 ... power amplifier

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 5J090 AA04 AA24 AA41 CA21 CA26 FA08 FA19 FA20 GN03 GN04 GN07 HN03 HN04 HN17 KA26 KA34 KA42 SA14 TA01 TA02 TA03 5K046 AA05 EE16 EE32 EE55 EE59 EF46

Claims (7)

[Claims] 1. A method for compensating for non-linear distortion of a power amplifier in a transmission unit of a wireless communication apparatus, comprising: calculating amplitudes Vi and Vq of orthogonal baseband signals I and Q by a polynomial g (v) for amplitude distortion compensation; To calculate the operation results g (Vi) and g (Vq). From the operation results g (Vi) and g (Vq), Miq ² = (g (Vi)) ² + (g (Vq) ² ) The operation output Miq ² represented by ² is calculated, and the operation output Miq ² is used to calculate a phase value θ corresponding to the operation output Miq ² on a one-to-one basis or a cosine value cos θ thereof from a table for phase distortion compensation. And the sine value sin
is read, and its phase value θ or cosine value cos θ and sine value si
From nθ and the calculation results g (Vi) and g (Vq), I ′ = g (Vi) × cosθ−g (Vq) × sinθ Q ′ = g (Vq) × cosθ + g (Vi) × sinθ A distortion compensation method for calculating orthogonal baseband signals I ′ and Q ′ to be performed. 2. The distortion compensation method according to claim 1, wherein cos θ = 1 and sin θ = θ, and I ′ = g (Vi) −g (Vq) × θ Q ′ = g (Vq) + g (Vi) × A distortion compensation method for calculating orthogonal baseband signals I ′ and Q ′ represented by θ. 3. A distortion compensation method according to claim ^{1, cosθ = 1-θ 2} /2, and as ^{sinθ = θ-θ 3/6} , I '= g (Vi) × (1-θ 2/2) −g (Vq) ×
^{(Θ-θ 3/6)} Q '= g (Vq) × (1-θ 2/2) + g (Vi) ×
Quadrature baseband signals I represented by ^{(θ-θ 3/6)} ', Q' distortion compensation method for calculating the. 4. An apparatus for compensating for non-linear distortion of a power amplifier in a transmission unit of a wireless communication apparatus, wherein amplitudes Vi and Vq of orthogonal baseband signals I and Q are respectively represented by a polynomial g (v) for amplitude distortion compensation. And polynomial operation means for calculating operation results g (Vi) and g (Vq), and from the operation results g (Vi) and g (Vq), Miq ² = (g (Vi)) ² + ( g (Vq)) and the square calculating means for calculating an operation output Miq ² represented by ^2, by the operation output Miq ^2, the operation output Miq ² and a pair phase value theta, or a cosine value corresponding to 1 cos θ
And a phase distortion compensation table from which the sine value sinθ is read out, and the phase value θ or the cosine value cosθ and the sine value si
From nθ and the calculation results g (Vi) and g (Vq), I ′ = g (Vi) × cosθ−g (Vq) × sinθ Q ′ = g (Vq) × cosθ + g (Vi) × sinθ And a combining operation means for calculating orthogonal baseband signals I ′ and Q ′ to be obtained. 5. The distortion compensator according to claim 4, wherein cos θ = 1 and sin θ = θ, and I ′ = g (Vi) −g (Vq) × θ Q ′ = g (Vq) + g (Vi) A) a distortion compensator for calculating orthogonal baseband signals I ′ and Q ′ represented by × θ. In the distortion compensation apparatus 6. The method of claim ^{4, cosθ = 1-θ 2} /2, and is a ^{sinθ = θ-θ 3/6} , I '= g (Vi) × (1-θ 2 / 2) -g (Vq) ×
^{(Θ-θ 3/6)} Q '= g (Vq) × (1-θ 2/2) + g (Vi) ×
Quadrature baseband signals I represented by ^{(θ-θ 3/6)} ', Q' distortion compensation apparatus is calculated. 7. A wireless communication device comprising the distortion compensation device according to claim 4 as a distortion compensation unit.