JP4576347B2 - Distortion compensated quadrature modulator - Google Patents

Description

  The present invention relates to a wireless transmission device, and more particularly to a distortion-compensating quadrature modulator for outputting a highly accurate quadrature modulated wave with an analog quadrature modulator.

  In a digital modulation system used in a mobile communication system such as a W-CDMA (Wideband-Code Division Multiple Access) system and a PDC in which a use band of a radio frequency is limited, AM, FM, etc. Compared with an analog modulation system, a highly accurate modulator is required. This is not a problem when a quadrature modulator is configured with a digital circuit, but when the quadrature modulator is configured with an analog element, it is important to compensate for the following three linear distortions.

  First, the output signals I (t) and Q (t) of the D / A converter used for the input of the quadrature modulator completely adjust the offset of the DC component generated with respect to the original zero balance modulation signal. Difficult to do. Further, a shift occurs due to a temperature change and a secular change, and a carrier leak component is superimposed on the modulated wave due to the shift of the DC offset.

  Second, the gain ratio between the in-phase component I and the quadrature component Q of the analog modulation signal is shifted from the original 1, and a distortion component is superimposed on the image frequency region.

  Third, it is difficult to accurately manufacture a π / 2 phase shifter necessary for a quadrature modulator, and a distortion component is superimposed on an image frequency region as a deviation of the orthogonality.

  In particular, in the direct conversion method in which a circuit scale is expected to be reduced by directly converting a baseband signal into an RF frequency, it is becoming more and more important to solve problems caused by analog elements.

  Conventionally, as a method for solving these problems, a method using affine transformation is known (see, for example, Non-Patent Document 1). A schematic configuration of the method using the affine transformation is shown in FIG.

  In FIG. 8, 11 is a distortion compensation circuit, 22 is an analog quadrature modulator, 12 is a detector that detects an RF transmission signal, 13 is an LPF that averages the detection output, and 14 is a distortion compensation coefficient and a test pattern. It is a control unit. The affine transformer 111 in the distortion compensation circuit 11 has a configuration as shown in FIG. 2, and DC offsets inherent in the analog quadrature modulator 22 by giving four affine transformation coefficients DCI, DCQ, α, θ, It functions to correct the amplitude ratio (IQ gain ratio) and orthogonality.

  As a specific control method, the switches (SW2) 114 and (SW3) 115 are switched to the control unit 14 side, a test pattern signal is output, and the output of the LPF 13 is observed. Next, switches (SW2) 114 and (SW3) 115 are set to the affine converter 111 side, switches (SW0) 112 and (SW1) 113 are switched to the control unit 14 side, and a test pattern signal is transmitted. However, α and θ are obtained by observing the output level of the LPF 13.

In addition, as a method for calculating the affine transformation coefficient without using the test pattern, a method of separating the first to third linear distortions by calculation and updating the affine transformation count by an update equation and a perturbation method were used. There is a way. However, the method using the perturbation method is used as a means for compensating for distortion caused by a shift in DC offset (see, for example, Patent Document 1).
Hiroshi Suzuki, 1 other, "Affine Transform Linear Distortion Compensation-Application to Linear Signal Transmission including Equalization in Mobile Radio Communication", IEICE Transactions B-II, IEICE, 1992 January, Vol. J75-B-II, No. 1, p. 1-9 JP 2006-50331 A

  However, when distortion compensation is performed by an affine transformer, the method of using an update equation for calculating an affine transformation coefficient is lower in transmission power, or when transmitting 0 data, the distortion component is higher than when transmission power is high. The detection accuracy is lowered, and it becomes impossible to accurately compensate for the linear distortion.

  On the other hand, the method using the perturbation method updates the affine transformation coefficients one by one, so that there is a problem that convergence is slower than the method using the update equation, particularly when the transmission power is high.

  The present invention has been made to solve the above-described problem. By using the characteristics of the update equation and the perturbation method in calculating the affine transformation coefficient, the DC offset and the I of the analog modulation signal can be obtained regardless of the transmission power level. An object of the present invention is to provide a distortion-compensating quadrature modulator that can compensate for distortion caused by a shift in the / Q gain ratio and distortion caused by a shift in orthogonality of the quadrature modulator.

The distortion-compensated quadrature modulator according to the first aspect of the present invention affine-transforms the input complex amplitude signals I (t) and Q (t) based on the affine transformation coefficient, and compensates the signal a (t ) And b (t) to output affine transformation means;

A quadrature modulator that quadrature modulates a local oscillation signal based on the compensated signals a (t) and b (t) and outputs a modulated wave signal that is a real signal, and frequency conversion to the modulated wave signal or the modulated wave signal Alternatively, quadrature detection means for outputting complex feedback signals I ′ (t) and Q ′ (t) from a signal subjected to at least one of amplification, and complex feedback signal I ′ (t) output from the quadrature detection means, A controller that updates an affine transformation coefficient of the affine transformation means based on Q ′ (t),
The control unit extracts a linear distortion remaining in the complex feedback signals I ′ (t) and Q ′ (t) as a distortion coefficient, and calculates an affine transformation coefficient of the affine transformation means according to an update expression including the distortion coefficient. First control means for updating to a new affine transformation coefficient, linear distortion remaining in the complex feedback signals I ′ (t) and Q ′ (t) are extracted as distortion coefficients, and the affine transformation means is obtained using a perturbation method. Second control means for updating the affine transformation coefficients of the complex feedback signals I ′ (t) and Q ′ (t) to a new affine transformation coefficient, and the first control means according to the average power of a predetermined section of the complex feedback signals I ′ (t) and Q ′ (t), And selecting means for selecting one of the second control means.

の式を用いて、線形歪を検出し、

の式を用いてアフィン変換係数を更新し、
前記平均振幅値がある一定の閾値よりも低い場合は、直交変調器の直交度のずれにより発生する線形歪と複素振幅信号Ｉ(ｔ)及びＱ(ｔ)の平均振幅比のずれによって発生する線形歪の補償に関わるアフィン変換係数の更新を停止し、

の式を用いてＤＣオフセット成分を検出し、摂動法を用いてＤＣオフセットパワーが小さくなる方向に補正係数ＤＣＩ、ＤＣＱの値を更新することを特徴する。 According to a second aspect of the present invention, in the distortion-compensated quadrature modulator according to the first aspect of the invention, the control unit includes the complex amplitude signals I (t) and Q (t) or the complex feedback signals I ′ (t) and Q 'has a level detection means for calculating an average amplitude value in N sample intervals of (t),
If the average amplitude value is higher than a certain threshold,

The linear distortion is detected using the equation of

Update the affine transformation coefficient using the formula
When the average amplitude value is lower than a certain threshold value, the average amplitude value is generated due to a linear distortion caused by a deviation in orthogonality of the quadrature modulator and a deviation in average amplitude ratio of the complex amplitude signals I (t) and Q (t). Stop updating affine transformation coefficients related to linear distortion compensation,

The DC offset component is detected using the equation (2), and the values of the correction coefficients DCI and DCQ are updated in a direction in which the DC offset power decreases using the perturbation method.

  According to the present invention, when calculating the affine transformation coefficient, the affine transformation device 21 uses a first CLC (carrier leak canceller) algorithm that uses an update equation according to the level or a second CLC algorithm that uses a perturbation method. By updating the compensation coefficient, the DC offset, the distortion caused by the deviation of the I / Q gain ratio of the analog modulation signal, and the distortion caused by the deviation of the orthogonality of the quadrature modulator are compensated regardless of the transmission power level. be able to.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
(First embodiment)
FIG. 1 is a block diagram showing a configuration example of a distortion compensating quadrature modulator according to the first embodiment of the present invention.

  Complex amplitude signals I (t) and Q (t) output from the preceding digital modulation section (not shown) are input to the affine converter 21. The affine converter 21 has a function of compensating for each linear distortion of a DC offset (carrier leak), an I / Q gain ratio shift, and an orthogonality shift of the analog modulation signal. The affine transformer 21 has the same configuration as that shown in FIG. 2 shown in the prior art, and is provided in an analog unit such as the analog quadrature modulator 22 by giving affine transformation coefficients DCI, DCQ, tanθ, IQGain / (cosθ). Functions to correct DC offset, amplitude ratio (IQ gain ratio), and orthogonality.

That is, the affine transformer 21 performs affine transformation shown in the following equation based on the input complex amplitude signals I (t) and Q (t) based on the affine transformation coefficient, and sends the compensated signal to the analog quadrature modulator 22. Output.

  The analog quadrature modulator 22 performs quadrature modulation on the local oscillation signal based on the compensated signal, up-converts it to an RF (Radio Frequency) frequency band, and outputs a modulated wave signal that is a real signal.

  The analog quadrature modulator 22 includes a local oscillator 31, a phase shifter 32, multipliers 33 and 34, and an adder 35. The local oscillator 31 generates an RF band sine wave as a carrier. The phase shifter 32 shifts the output phase of the local oscillator 31 by π / 2. Multipliers 33 and 34 multiply the I phase and Q phase of the input signal by the output of local oscillator 31 and the output of phase shifter 32, respectively. The adder 35 additively synthesizes the outputs of the multipliers 33 and 34 to obtain the output of the analog quadrature modulator 22. The signal up-converted to the RF band by the analog quadrature modulator 22 is output as an RF transmission signal.

  Further, a part of the signal output from the analog quadrature modulator 22 is input to the frequency converter 23 constituting the feedback circuit and down-converted to the IF band. The signal down-converted to the IF band is input to the digital quadrature detector 24 and subjected to quadrature detection. The digital quadrature detector 24 removes a carrier wave component from the input modulated wave signal or a signal obtained by subjecting the modulated wave signal to at least one of frequency conversion and amplification, and an IF frequency complex feedback signal I ′ (t) and Q ′ (t) is output. The IF feedback complex feedback signals I ′ (t) and Q ′ (t) are stored in the memory 25. The memory 27 stores the input complex amplitude signals I (t) and Q (t).

  The control unit 26 extracts linear distortion remaining in the complex feedback signals I ′ (t) and Q ′ (t) as a distortion coefficient, and depends on the levels of the complex feedback signals I ′ (t) and Q ′ (t). The current affine transformation coefficient is updated to a new affine transformation coefficient by using an update formula including a distortion coefficient or a perturbation method, and is reset in the affine transformer 21.

  That is, the control unit 26 measures the level of the feedback signal from the data stored in the memory 25, and when the signal level is equal to or higher than a certain threshold, the carrier leak detection algorithm (hereinafter referred to as the first algorithm) using the update equation. DC offset, gain ratio shift, and orthogonality shift are detected by a CLC algorithm), the compensation coefficient of the affine transformer 21 is updated, and each linear distortion is compensated. Further, when the level of the feedback signal is less than the threshold value, the control unit 26 detects only the DC offset by a carrier leak detection algorithm using a perturbation method (hereinafter referred to as a second CLC algorithm), and affine transforms. The compensation coefficient of the device 21 is updated.

  As shown in FIGS. 3A and 3B, as a characteristic of linear distortion, the DC offset does not vary greatly depending on the magnitude of transmission power, but IQ average amplitude ratio deviation and orthogonality deviation. If the transmission power decreases, the absolute level of distortion decreases accordingly. Therefore, if the transmission power is sufficiently low, there is no problem even if the detection and update of the IQ average amplitude ratio shift and the orthogonality shift are stopped, and only the DC offset component may be detected to update the compensation coefficient. Therefore, when the transmission power is low, only the DC offset is detected by the carrier leak detection algorithm using the perturbation method, that is, the second CLC algorithm, and the compensation coefficient is updated. When the transmission power is high, the carrier using the update equation is used. The leak detection algorithm, that is, the first CLC algorithm detects DC offset, gain ratio shift, and orthogonality shift, and updates the compensation coefficient.

Next, a detailed control operation of the control unit 26 will be described with reference to a flowchart shown in FIG.
The control unit 26 performs level detection by calculating an average amplitude value in N sample sections of the complex feedback signals I ′ (t) and Q ′ (t) stored in the memory 25 (step A1), and detects the detected signal. It is determined whether or not the level is equal to or greater than a certain threshold (step A2).

  If it is determined in step A2 that the levels of the complex feedback signals I ′ (t) and Q ′ (t) are equal to or greater than a certain threshold value, the DC offset and gain are obtained by the first CLC algorithm using the update equation. The deviation of the ratio and the deviation of the orthogonality are detected and the compensation coefficient is updated (step A3).

の式を用いて線形歪を検出する。 That is,

The linear distortion is detected using the following equation.

の式を用いてアフィン変換器２１のアフィン変換係数を更新する。 Then

The affine transformation coefficient of the affine transformer 21 is updated using the following equation.

の式を用いてＤＣオフセット成分を検出し、摂動法を用いてＤＣオフセットパワー（Offset）が小さくなる方向に補正係数ＤＣＩ、ＤＣＱの値を更新する。 If it is determined in step A2 that the levels of the complex feedback signals I ′ (t) and Q ′ (t) are less than the threshold value, only the DC offset is obtained by the second CLC algorithm using the perturbation method. It detects and updates the compensation coefficient of the affine transformer 21. That is, when the levels of the complex feedback signals I ′ (t) and Q ′ (t) are less than the threshold value, the linear distortion caused by the deviation of the orthogonality of the analog quadrature modulator 22 and I (t) and Q ( the update of the affine transformation coefficient related to the compensation of the linear distortion caused by the deviation of the average amplitude ratio of t) is stopped,

The DC offset component is detected using the equation (2), and the values of the correction coefficients DCI and DCQ are updated in a direction in which the DC offset power (Offset) is reduced using the perturbation method.

  FIG. 5 is an operation explanatory diagram of the second CLC algorithm using the perturbation method. P0 is a point indicating correction coefficients DCI and DCQ currently set in the affine transformer 21. As shown in FIG. 5A, points P1 to P4 in which positive and negative perturbations corresponding to the step size are respectively set to the correction coefficients DCI and DCQ using P0 as a reference value of the affine transformation coefficient, and P1 to P4 are set. DC offset power is calculated respectively, and the point at which the minimum DC offset power is detected (P1 in this example) is set as a new reference value, as shown in FIG. Is moved by the determined step size.

  As shown in the first embodiment, when the affine transformation coefficient is calculated, when the signal level is small, by switching to the second CLC algorithm using the perturbation method that does not require the calculation of the phase difference φ, for example, the power supply Even in the case where a non-transmission (no signal) state continues immediately after the input, at least distortion caused by the DC offset can be compensated. The second CLC algorithm has been described in which the perturbation method is applied to a set of DCI and DCQ. Similarly, IQGain and θ are set to a set, and the sum of | 1-α | and | sinθ ′ | is minimized. Perturbation control may be performed as follows.

(Second Embodiment)
Next, a second embodiment of the present invention will be described.
The distortion-compensating quadrature modulator according to the second embodiment is the same as the distortion-compensating quadrature modulator according to the first embodiment, when the control unit 26 detects a signal level by comparing with a threshold value. The complex amplitude signals I (t) and Q (t) sent from the modulation unit, that is, the transmission signal data are stored in the memory 27 and used. The control unit 26 selects the CLC algorithm by comparing the average amplitude of the complex amplitude signals I (t) and Q (t) stored in the memory 27 with a threshold value. In this case, the threshold value may be set to a transmission signal level at which the detected value of the orthogonality deviation or the amplitude ratio deviation does not converge by gradually lowering the transmission signal level.

  Since other parts have the same configuration as that of the first embodiment, detailed description thereof is omitted.

  As shown in the second embodiment, the levels of the complex amplitude signals I (t) and Q (t) sent from the digital modulation unit are detected, and the first CLC using the update equation according to the levels is detected. Even if the compensation coefficient of the affine converter 21 is updated by the second CLC algorithm using the algorithm or the perturbation method, the DC offset, the analog modulation signal, regardless of the transmission power level, as in the case of the first embodiment. It is possible to compensate for the distortion caused by the deviation of the average amplitude ratio of I (t) and Q (t) and the distortion caused by the deviation of the orthogonality of the orthogonal modulator.

(Third embodiment)
In this embodiment, the distortion-compensating quadrature modulator shown in the first embodiment is applied to a transmission device (wireless base station) that transmits, for example, a 4-carrier W-CDMA (Wide-Code Division Multiple Access) signal. An example is shown.

  FIG. 6 is a block diagram showing a configuration of a transmission apparatus according to the third embodiment of the present invention, and shows only a transmission system after baseband processing.

  In FIG. 6, the digital modulation unit 41 performs band limitation on the four input baseband signals, up-sampling to a desired sampling frequency, digital quadrature modulation to each carrier frequency, and multi-carrier synthesis. The signal digitally modulated by the digital modulation unit 41 is subjected to distortion compensation including affine transformation by a distortion compensation unit 62. The distortion compensation unit 62 is configured by connecting a DPD (Digital PreDistortion) unit 61 and the affine converter 21 in series as will be described in detail later with reference to FIG. The signal compensated for distortion by the distortion compensator 62 is converted to an analog signal by the D / A converter 42 and then sent to the analog quadrature modulator 44 via a low-pass filter (LPF) 43. The analog quadrature modulator 44 is realized by, for example, an MMIC (Microwave Monolithic Integrated Circuit).

  The analog quadrature modulator 44 up-converts the analog signal D / A converted by the D / A converter 42 to a target RF frequency band and outputs an RF band modulated wave signal. This RF band modulated wave signal is amplified by the power amplifier 45 and transmitted to the outside from an antenna (not shown).

  A part of the signal amplified by the power amplifier 45 is extracted by the coupler 51 and input to the mixer 52 which is a frequency converter of the feedback system. The mixer 52 down-converts the signal amplified by the power amplifier 45 to the IF frequency band. The IF signal down-converted by the mixer 52 is band-limited by the band-limiting filter 53, and frequency components other than the target IF frequency are removed. The analog signal band-limited by the band-limiting filter 53 is converted into a digital signal by the A / D converter 54 and sent to the digital quadrature detection unit 55. The digital quadrature detection unit 55 performs digital quadrature detection on the digital signal converted by the A / D converter 54, and complex feedback signals I ′ (t) and Q ′ (t) having IF frequencies substantially equal to the digital modulation unit 41. ) Is output.

  The output signal of the digital quadrature detection unit 55 is sent to the control unit 26 '. As described in the first embodiment, the control unit 26 ′ updates each distortion compensation coefficient used in the affine transformer 21 based on the signal modulated by the digital modulation unit 41 and the signal detected by the digital quadrature detection unit 55. I do.

Next, a specific configuration example of the control unit 26 ′ will be described with reference to FIG.
In this case, the affine converter 21 controlled by the control unit 26 ′ configures a distortion compensation unit 62 by connecting a DPD (Digital PreDistortion) unit 61 in series. The DPD unit 61 calculates the instantaneous power of the input IF signal, reads the predistortion corresponding to the power from the distortion compensation table, and multiplies the input IF signal. The distortion compensation table stores reverse characteristics of nonlinear distortion generated by the power amplifier 45 and the like. The distortion compensation table of the DPD unit 61 and the four affine transformation coefficients DCI, DCQ, tanθ, IQGain / (cosθ) of the affine transformer 21 are updated according to the signal level by the control unit 26 ′ so that the distortion is reduced. . Input / output of the DPD unit 61 and the affine converter 21 is performed in real time, but the distortion compensation table and the coefficient may be updated by batch processing.

  The memories 27 and 25 provided in the control unit 26 ′ are input from the digital modulation unit 41 (IF signals I (t) and Q (t)) and input from the digital quadrature detection unit 55 (complex feedback signal I '(T) and Q' (t) are temporarily stored.

  The threshold determination unit 71 measures the average level (power) of the complex feedback signal read from the memory 25, and compares the determination level with a threshold to determine whether the determination signal is equal to or higher than the threshold. The linear distortion detection unit 75 and the offset power detection unit Output to 76.

  When a determination signal indicating that the average level is equal to or greater than the threshold value is input, the linear distortion detection unit 75 includes the I-phase and Q-phase DC offsets included in the complex feedback signal read from the memory 25, I / Q gain ratio and orthogonality are detected separately and output to the coefficient updating unit 77. The IF signal read from the memory 27 is also input to the linear distortion detector 75 as a reference for compensating for the phase rotation φ of the complex feedback signal.

  The offset power detection unit 76 detects the power of the DC offset (carrier leak) included in the complex feedback signal read from the memory 25 when a determination signal indicating that the average level is not equal to or greater than the threshold is input. It outputs to the coefficient update part 77.

  When the linear distortion detected by the linear distortion detector 75 is input, the coefficient updating unit 77 uses the update formula that is the first CLC algorithm, and the linear distortion detected by the offset power detector 76 is input. In some cases, the affine transformation coefficient is updated using the perturbation method which is the second CLC algorithm.

  The adaptive equalizer 72 adaptively equalizes the complex feedback signal read from the memory 25 using the IF signal read from the memory 27 as a reference signal, and an equalization error (difference between the equalization output and the reference signal). The data is output to the DPD algorithm unit 73. Equalization is performed by linear calculation, and the equalization error represents a component that cannot be equalized by linear calculation, that is, nonlinear distortion.

  The DPD algorithm unit 73 updates each coefficient of the polynomial that reproduces the inverse characteristics of the nonlinear distortion one by one so as to minimize the average power of the equalization error.

  The polynomial calculation unit 74 calculates a polynomial using the coefficient updated by the DPD algorithm unit 73 and interpolates the distortion compensation table of the DPD unit 61. The adaptive equalizer 72, the DPD algorithm unit 73, and the polynomial calculation unit 74 are collectively referred to as a DPD control unit 78. In this example, the complex feedback signal is commonly used for both linear distortion compensation and nonlinear distortion compensation.

  As a result of measuring the level of the transmission signal in which the first CLC algorithm and the second CLC algorithm can be operated in the transmission apparatus, the first CLC algorithm maintains the same accuracy with a transmission power of at least 15.5 [dBm]. It was confirmed that detection was possible while maintaining the same accuracy with the transmission power of up to 46 [dBm]. At this time, the threshold value for switching between the first CLC algorithm and the second CLC algorithm may be set to a level between 46 [dBm] and 15.5 [dBm]. However, it is desirable to set the threshold value to 15.5 [dBm] in the case where a large dynamic range for detecting and updating the amplitude ratio deviation and orthogonality deviation is taken.

  In the third embodiment shown in FIG. 6, as the D / A converter 42 and the A / D converter 54 that greatly affect the accuracy of detection and compensation of distortion components, for example, a 16-bit D / A converter, a 12-bit A / D A converter is used.

  In the analog quadrature modulator that performs distortion compensation using an affine transformer as shown in the above embodiment, when calculating the affine transformation coefficient, an update formula or a perturbation CLC algorithm is selected according to the signal level. Regardless of transmission power level, at any transmission level, DC offset, distortion due to deviation of average amplitude ratio of analog modulation signals I (t) and Q (t), and deviation of orthogonality of quadrature modulator It is possible to accurately compensate for linear distortion such as distortion.

  Note that the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying constituent elements without departing from the scope of the invention in the implementation stage.

It is a block diagram which shows the structure of the distortion compensation quadrature modulator which concerns on 1st Embodiment of this invention. It is a block diagram of this invention and the conventional affine converter. It is a figure which shows the relationship between the level of transmission power, and linear distortion. It is a flowchart which shows operation | movement of the control part in 1st Embodiment of this invention. It is operation | movement explanatory drawing of the 2nd CLC algorithm using the perturbation method in the embodiment. It is a block diagram which shows the structure of the transmitter which concerns on 3rd Embodiment of this invention. It is a specific block diagram of the control part in the embodiment. It is a block diagram of the conventional distortion compensation quadrature modulator.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 21 ... Affine converter, 22 ... Analog quadrature modulator, 23 ... Frequency converter, 24 ... Digital quadrature detector, 25, 27 ... Memory, 26 ... Control part, 31 ... Local oscillator, 32 ... Phase shifter, 33, 34 ... multiplier, 35 ... adder, 41 ... digital modulator, 42 ... D / A converter, 43 ... low pass filter (LPF), 44 ... analog quadrature modulator, 45 ... power amplifier, 51 ... coupler, 52 ... Mixer, 53 ... band limiting filter, 54 ... A / D converter, 55 ... digital quadrature detection unit, 61 ... DPD unit, 62 ... distortion compensation unit, 71 ... threshold decision unit, 72 ... adaptive equalizer, 73 ... DPD algorithm 74, polynomial calculation unit, 75 ... linear distortion detection unit, 76 ... offset power detection unit, 77 ... coefficient update unit, 78 ... DPD control unit

Claims (2)

Affine transformation means for performing affine transformation on the input complex amplitude signals I (t) and Q (t) based on affine transformation coefficients and outputting compensated signals a (t) and b (t) represented by the following equations: ,

A quadrature modulator that quadrature modulates a local oscillation signal based on the compensated signals a (t) and b (t) and outputs a modulated wave signal that is a real signal;
Quadrature detection means for outputting complex feedback signals I ′ (t) and Q ′ (t) from the modulated wave signal or a signal obtained by performing frequency conversion or amplification on the modulated wave signal;
A controller that updates the affine transformation coefficients of the affine transformation means based on the complex feedback signals I ′ (t) and Q ′ (t) output from the quadrature detection means,
The control unit extracts a linear distortion remaining in the complex feedback signals I ′ (t) and Q ′ (t) as a distortion coefficient, and calculates an affine transformation coefficient of the affine transformation means according to an update expression including the distortion coefficient. First control means for updating to new affine transformation coefficients;
Second, linear distortion remaining in the complex feedback signals I ′ (t) and Q ′ (t) is extracted as a distortion coefficient, and the affine transformation coefficient of the affine transformation means is updated to a new affine transformation coefficient using a perturbation method. Control means,
Selecting means for selecting one of the first control means and the second control means in accordance with an average power of a predetermined section of the complex feedback signals I ′ (t) and Q ′ (t); A distortion-compensating quadrature modulator comprising: The control unit includes level detection means for calculating an average amplitude value in N sample intervals of the complex amplitude signals I (t) and Q (t) or the complex feedback signals I ′ (t) and Q ′ (t). Have
If the average amplitude value is higher than a certain threshold,

The linear distortion is detected using the equation of

Update the affine transformation coefficient using the formula
When the average amplitude value is lower than a certain threshold value, the average amplitude value is generated due to the linear distortion caused by the deviation of the orthogonality of the quadrature modulator and the deviation of the average amplitude ratio of the complex amplitude signals I (t) and Q (t). Stop updating affine transformation coefficients related to linear distortion compensation,

2. The distortion-compensated quadrature modulator according to claim 1, wherein the DC offset component is detected using the following equation, and the values of the correction coefficients DCI and DCQ are updated in a direction in which the DC offset power decreases using the perturbation method. .

Images