KR20020070572A - Linear power amplifier having a linearizer using delay line - Google Patents

Abstract

PURPOSE: A linearized power amplifier having a linearization device using a delay line is provided to perform a realtime phase control operation by using a delay line. CONSTITUTION: The first power divider(2) divides a carrier signal received from a carrier input terminal(1) into a main path and an auxiliary path. The first variable attenuator(4) and the first variable phase shifter(5) change the intensity and a phase of the carrier signal of the main path. A power amplifier(6) is connected with each rear end of the first variable attenuator(4) and the first variable phase shifter(5). A subtraction circuit(10) outputs only cross modulation distortion signals. The first directional coupler(7) extracts the carrier signals and the cross modulation distortion signals from the power amplifier(6) and applies a comparative signal to the subtraction circuit(10). The first delay line(3) delays the carrier signal of the auxiliary path. The second variable attenuator(14) and the second variable phase shifter(15) changes the intensity and phases of the cross modulation distortion signals. An auxiliary amplifier(16) is connected with each rear end of the second variable attenuator(14) and the second variable phase shifter(15). The second directional coupler(18) extracts and outputs the carrier signals.

Description

Linear power amplifier with linearizer using delay line {LINEAR POWER AMPLIFIER HAVING A LINEARIZER USING DELAY LINE}

The present invention relates to a linear power amplifier having a linearizer, and more particularly, to a linear power amplifier having a feedforward linearizer using a delay line.

Correction amplifiers, such as high power amplifiers (HPAs) for base stations, are required to satisfy the linearity required by the system without the influence of pressure and external environment with minimal power. It is important to ensure excellent efficiency characteristics and linearity at maximum output conditions and standby conditions.

However, in general, due to the nonlinear characteristics of active devices used in RF and microwave circuits, these amplifiers generally have nonlinear circuit characteristics. In particular, as shown in FIG. 1, a large power amplifier operates a transistor, which is a main active element, in a saturation region having a strong nonlinear characteristic to extract maximum power, and in the case of a large power amplifier operated in a saturation region, gain and phase are different. It is common to be distorted.

In addition, in the case of digital mobile communication and satellite communication using multiple channels, when two or more carriers are input and co-amplified to a large power amplifier, intermodulation signals are generated due to nonlinear characteristics near the saturation region, as shown in FIG. Done.

That is, when two carriers f1 and f2 are amplified through the HPA, various intermodulation signals 3f1-2f2, 2f1-f2, 2f2-f1, and 3f2-2f1 appear in addition to the amplified carrier. . These intermodulation signals act as cross-talk or noise, causing a drop in transmission quality.

In this case, when a linearizer is used together with the power amplifier to compensate for the nonlinear characteristics of the saturation region operation of the large power amplifier, the desired output can be greatly reduced while significantly reducing the intermodulation signals generated during the co-amplification of multiple carriers in the saturated region. Power can be obtained. That is, when using a linear power amplifier (LPA) using a linearizer, as shown in Figure 3, only the desired carrier is amplified.

Linearization is the minimization of the effects of intermodulation distortion on the signals of adjacent channels due to the nonlinear nature of the transistors near the saturation region when a multichannel signal is applied to a large power amplifier. Can greatly affect the call quality.

In the meantime, the proposed linearization methods can be classified into four types, which include 'input power back-off', 'negative feedback', 'predistortion' and 'feedforward'. The feedforward method was applied.

The feedforward method removes signals other than the main signal (IMD) by creating a signal with the same spectral shape as the IMD generated in the large power amplifier and having a 180 ° phase difference and combining the same at the output of the large power amplifier. This method is widely used for satellite earth stations or land mobile communication base stations, and has an improvement effect compared to other methods. By the way, the typical example of a feedforward scheme is to use pilot tones, which must use a plurality of pilot to maintain the broadband characteristics, and the equalizer and the monitoring part also have several blocks that are responsible for a specific frequency. Since the circuit is complicated, there is a problem that the pilot tone generates unwanted signal distortion as it passes through the amplifier.

The present invention is to solve the above problems, an object of the present invention, it is possible to minimize the change of the input signal power and distortion of various signals without using the pilot tone, by using a delay line It is to provide a linear power amplifier having a linearizer capable of phase control in real time.

Further objects and effects of the present invention will become more apparent from the following detailed description of the invention described with reference to the accompanying drawings.

1 is a graph illustrating the nonlinearity of a large power amplifier.

2 shows input / output spectrum characteristics of a large power amplifier.

3 illustrates input and output spectral characteristics of the linear power amplifier.

4 is a conceptual diagram of a feedforward linearization scheme according to the present invention.

5 is a configuration diagram of a feedforward linearization method using a delay line according to an exemplary embodiment of the present invention.

6 shows a signal power calculation method of the linearizer according to the present invention.

7 is a block diagram of a linearization system according to the present invention.

8 is a configuration of a delay line of the linearization system according to the present invention.

9A to 9H are circuit diagrams of the A to H modules of Fig. 7, respectively.

10 is an intensity and phase control automation circuit in the system of the present invention.

11 shows the linearizer power measurement points.

12A and 12B are graphs showing output signals for each characteristic of an amplifier having a linearizer and an amplifier having no linearizer according to the present invention, respectively.

13A and 13B are diagrams and circuitry of the intensity control circuit, respectively.

14A and 14B are diagrams and circuitry of the phase control circuit, respectively.

<Description of the symbols for the main parts of the drawings>

1: carrier input stage 2: first power divider

3: first delay line 4: first variable attenuator

5: first variable phase converter 6: main amplifier

7: First Directional Coupler 8: Third Directional Coupler

9: second power divider 10: subtraction circuit

11: carrier output stage

13: second delay line 14: second variable attenuator

15: second variable phase converter 16: auxiliary amplifier

17: second directional coupler 18: fifth directional coupler

19: fourth power distributor

21: amplifier 22: synchronous checker

24: first gain controller 25: first phase controller

28: fourth directional coupler 29: third power splitter

31: amplifier 32: isolator

33: third delay line 34: second gain controller

35 second phase controller 36 amplifier

37: limiter 38: sixth directional coupler

39: fifth power splitter

100: A block 200: B block

300: C block 400: D block

500: E block 600: F block

700: G block 800: H block

201: RF switch 202: oscillator

203: Divider 204: Detector

205: DC switch 206: Automatic Level Control (ALC)

207: attenuator

701: power divider 702: 90 coupler

703: RF Switch 704: Oscillator

705: Divider 706: Mixer

707: power divider 708: DC switch

709: Divider 710: ALC

711: phase shifter

That is, the linear power amplifier having a linearizer for achieving the above object of the present invention,

A first power divider (2) for distributing the carrier signal input through the carrier input terminal (1) to the main path and the auxiliary path; A first variable attenuator (4) and a first variable phase converter (5) for changing the strength and phase of the signal in the main path; A large power amplifier (6) connected to the rear end of the first variable attenuator (4) and the first variable phase converter (5) and amplifying to a desired output level; A subtraction circuit 10 outputting only intermodulation distortion signals among the carrier wave and intermodulation distortion signals generated in the main path; A first directional coupler (7) which partially extracts carrier and intermodulation distortion signal components from the output of the large power amplifier (6) and applies them as a comparison signal to the subtraction circuit (10); A first delay line (3) for delaying the carrier signal in the auxiliary path and applying it as a reference signal to the subtraction circuit (10); A second variable attenuator 14 and a second variable phase converter 15 for changing the intensity and phase of the intermodulation distortion signal components obtained by the subtraction circuit 10; An auxiliary amplifier 16 connected to the rear end of the second variable attenuator 14 and the second variable phase converter 15 and amplifying to a desired output level; A carrier wave by combining the intermodulation distortion signal components on the auxiliary path amplified to an appropriate size by the auxiliary amplifier 16 and a sum signal of the carrier wave and the intermodulation distortion signals passing through the second delay line 13 on the main path; A second directional coupler 18 for extracting and outputting only the signal; Wherein the carrier signal component extracts the comparison signal information such that the spectrum of the reference signal and the comparison signal coincide with each other so that the first magnitude control circuit (AGC1) 24 and the first phase control circuit (APC1) ( 25), and extracts the reference signal information to be applied to the first magnitude control circuit (AGC1) 24 and the first phase control circuit (APC1) 25, thereby causing the main path and the auxiliary path. The first amplitude control circuit (AGC1) 24 and the first phase control circuit (APC1) 25 are the first variable attenuator 4 and the first variable phase so that the carrier signal components are in the same spectrum with each other. A B block 200 for controlling the converter 5 to appropriately change the magnitude and phase of the comparison signal; And extracting output signal information of the subtraction circuit so that, when the output signal of the subtraction circuit is coupled back to the main path, the output signal information of the subtraction circuit is extracted so as to have the same magnitude as that of the intermodulation distortion component on the main path and have an inverse (180 °) phase. The second magnitude control circuit AGC2 34 to be applied to the magnitude control circuit AGC2 34 and the second phase control circuit APC2 35 and to extract the output signal information of the auxiliary amplifier 16. ) And the second phase control circuit (AGC2) 34 and the second phase control circuit (APC2) 35 are applied to the main path and auxiliary by being applied to the second phase control circuit (APC2) 35. G for controlling the second variable attenuator 14 and the second variable phase converter 15 to appropriately change the magnitude and phase of the output signal of the subtraction circuit so that the intermodulation distortion signal components in the path are in opposite spectra. Block 700; By further including, the final output is characterized in that intermodulation signal components are removed and only pure carriers remain.

Preferably, a third directional coupler 8 is coupled to the input signal of the subtraction circuit 10, and the extracted signal is passed through the second power divider 9 to the first magnitude control circuit AGC1. And a fourth directional coupler 28 coupled to the signal on the auxiliary path of the first power divider 2, which is applied to the first phase control circuit APC1 25 and the signal extracted therefrom. Is applied to the first magnitude control circuit (AGC1) 24 and the first phase control circuit (APC1) 25 via a third power divider (29), thereby providing the first magnitude control circuit (AGC1) 24. And the first phase control circuit (APC1) 25 controls the first variable attenuator 4 and the first variable phase converter 5 to appropriately change the magnitude and phase of the input signal.

Also preferably, the fifth directional coupler 18 is coupled to the amplified signal of the auxiliary amplifier 16, and the extracted signal is passed through the fourth power divider 19 to the second magnitude control circuit ( AGC2) 34 and a second phase control circuit (APC2) 35, which are coupled to the sixth directional coupler 38 with respect to the output signal of the subtraction circuit 10, wherein the extracted signal is The second magnitude control circuit is applied to the second magnitude control circuit (AGC2) 34 and the second phase control circuit (APC2) 35 through the third delay line 33 and the fifth power divider 39. (AGC2) 34 and the second phase control circuit (APC2) 35 control the second variable attenuator 14 and the second variable phase converter 15 to appropriately change the magnitude and phase of the distortion signal. It is characterized by.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

First, the feedforward linearizer according to the present invention will be described with reference to FIG.

In general, when a single frequency input signal is applied to a large power amplifier, the output gain decreases and the phase delay occurs as the size of the input signal increases due to the effects of AM-to-AM and AM-to-PM. In addition, when an input signal having multiple frequencies is input, intermodulated distortion signal components having different nonlinear propagation characteristics are generated in addition to an output signal having the same frequency as the input signal. The signal loop is used to extract only the distortion signal generated from the output of the large power amplifier, and it is called an error loop to adjust the distortion signal and couple it to the output of the large power amplifier.

That is, as shown in FIG. 4, the two carriers applied as inputs through the carrier input stage 1 have the same size and phase divided by the power divider 2 into the main path and the auxiliary path (point A and See spectrum of point A '). In the main path (point A), a large power amplifier 6 is amplified to a desired output level, and intermodulation distortion signals are also generated (see spectrum at point B).

The carrier and intermodulation distortion signal components at the output of the main amplifier 6 are partially extracted by the directional coupler 7 and applied to the subtraction circuit 10, while the pure carrier at the auxiliary path (point A ') The delay line 3 is applied to the subtraction circuit 10 via the variable attenuator 4 and the variable phase converter 5.

The subtraction circuit outputs only the intermodulation distortion signals among the carrier wave and intermodulation distortion signals generated in the main path. The intermodulation distortion signal components (spectrum of point C) obtained in the subtraction circuit are again amplified to an appropriate magnitude in the auxiliary amplifier 16 through the variable attenuator 14 and the variable phase converter 15 (see the spectrum of point D). ), Which is applied to the directional coupler 17 on the main path, wherein the signal amplified by the main amplifier 6 is also delayed for a predetermined time by the delay line 13 and then intermodulated distortion signal from the auxiliary path in the directional coupler. Combined with the components. Therefore, the intermodulation distortion signal components on the main path and the auxiliary path cancel each other out, and only the amplified carrier signal components (spectrum of point E) are output to the output terminal.

Here, the variable attenuator (A) is for adjusting the level of the intermodulation distortion signal appearing in the large power amplifier, the variable phase converter (Φ) is to adjust the phase of the inverse (180 °) when coupled back to the main path For fine tuning. Finally, just before the coupler 17 at the output of the main path is coupled, the magnitude and phase of the intermodulation signal level on the main path are reversed, so that the final output of the linearizer is the same as for point E. The components are removed and only pure carriers remain.

As shown in FIG. 5, the overall configuration of the linearizer according to the present invention includes a variable attenuator for adjusting the magnitude, phase, and time delay of a signal, a variable phase shifter, a subtraction circuit, a delay line, a control unit for adjusting these circuits, and the like. Separate by.

To do this, first, the output characteristics of the large power amplifier should be measured as the initial work in designing the linearizer. The maximum output powers of the main signal and the third-order distortion signal of the MAU 6 of FIG. 4 should be measured and applied to the design so that normal operation can be performed even when the magnitude of the input signal is attenuated. The distortion signal amplifier (CAU) 16 amplifies only the IM signal, so it must maintain linearity to avoid additional distortion in the output signal. In addition, the operating range of the variable input signal power of the magnitude control circuit and the phase control circuit should be measured and considered.

6 is a simplified overall circuit diagram of signal power calculated in consideration of operating characteristics of each module. The intensity of the output power in each device is indicated in dBm.

In addition, in FIG. 6 and below, the same member number is attached | subjected to the element which has the same function as FIG. 5, and detailed description is abbreviate | omitted.

5 is a configuration diagram for each module of the feedforward linearizer entire system of the present invention. First of all, the operating signal is divided into two signals through the power divider (2), and one signal is converted into the large power amplifier (6) through the variable attenuator (4) and the variable phase converter (5). The output signal at this time has a distortion component. This signal is sent as a comparison signal to the subtraction circuit 10 via the directional coupler 7. Meanwhile, another signal distributed by the power divider is applied to the subtraction circuit 10 as a reference signal via the delay line 3. On the other hand, the reference signal is another input signal of the subtraction circuit 10 whose initial input signal compensates for the delay time of the comparison signal via the delay line 3.

In this process, the main signal component is adjusted by the magnitude control circuit AGC1 24 and the phase control circuit APC1 25 so that the spectrum of the reference signal and the comparison signal coincide. For this purpose, the directional coupler 8 is coupled to the comparison signal, and the extracted signal is applied to the magnitude control circuit AGC1 24 and the phase control circuit APC1 25 via the power distributor 9. The directional coupler 28 is coupled to the reference signal, and the extracted signal is applied to the magnitude control circuits AGC1 24 and the phase control circuits APC1 25 through the power divider 29. The magnitude control circuit (AGC1) 24 and the phase control circuit (APC1) 25 allow the variable attenuator 4 and the variable phase converter 5 to appropriately change the magnitude and phase of the comparison signal.

Since the main signal components are in the same spectrum, they are canceled by the subtraction circuit 10 so that only the distortion signal appears as an output. This signal is adjusted by the magnitude and phase control circuits 14 and 15 and the distortion signal amplifier 16 in the signal path to a signal having the same magnitude and 180 degrees out of phase with the distortion signal of the large power amplifier.

Also, adjustment is made by the magnitude control circuit (AGC2) 34 and the phase control circuit (APC2) 35 so that the spectra of the distortion signal of the auxiliary circuit and the distortion signal components of the main circuit are equal in magnitude and 180 ° out of phase. do. To this end, the directional coupler 38 is coupled to the distortion signal, and the extracted signal is passed through the delay line 33 and the power divider 39 to the magnitude control circuit AGC2 34 and the phase control circuit APC2. The directional coupler 18 is applied to the amplified signal, and the extracted signal is passed through the power divider 19 to the magnitude control circuit AGC2 34 and the phase control circuit APC2. By the magnitude control circuit (AGC2) 34 and the phase control circuit (APC2) 35 so that the variable attenuator 14 and the variable phase converter 15 are magnitude and phase of the distortion signal. Make the appropriate changes.

As a result, in the directional coupler 17 before the final output stage 11, the signal passing through the delay line 13 of the main path and the signal whose size and phase of the auxiliary path are appropriately combined are combined so that the distortion signals cancel each other and Since only the signal components remain, it will function as a linearizer.

The system is divided into eight modules in consideration of the efficiency of the circuit, the A-H block (100-800) is shown in FIG. Reference numerals 21, 31 and 36 are amplifiers, 37 are limiters, 22 are synchronizers, and 32 are isolators.

9A to 9H are examples of circuit diagrams of blocks A to H in FIG.

According to the feedforward linearizer design of the present invention, first, phase control is possible in real time by using a delay line. The delay line is intended to prevent the narrowing of the bandwidth resulting from the inconsistency of the time delay by matching the time delay that occurs as the signal passes through the various elements in the linearizer. In general, the phase control method compares the previous signal with the current signal and applies a signal that exhibits excellent characteristics. Therefore, real-time signal processing is not possible, thereby allowing a certain error range. However, although the present invention has a disadvantage in that the volume becomes large, the accurate phase control function is implemented in real time using a method of adjusting the length of the delay line. The delay line used is MICRO-COAX's 141A-TP type, whose internal diameter is 0.92mm and its external diameter is 2.9mm, and the delay time characteristics of about 5ns and the insertion loss characteristics of about 0.73dB were measured per meter.

In the present invention, as shown in FIG. 8, the delay line DL1 is used for phase compensation in the signal loop, the delay line DL2 for phase compensation of the error loop, and the delay line DL3 for phase control. Used. In particular, the function of DL3, a phase control delay line, is a main theory of the present invention. In this regard, the DL2 used between the main signal sections B and D1 performs a linear operation as a passive element, but is a distortion signal amplifier located in the auxiliary signal section. Since the nonlinear output signal characteristic of the input signal is exhibited by the operation in the nonlinear region of the transistor, the feedforward linearization function is required to compensate for this so that the gain of the distortion signal amplifier is maintained at a constant value. For this purpose, a function for automatically controlling the magnitude and phase of the signal is required, and the automatic phase control function is performed as an automatic phase control circuit and a delay line, which will be described later. That is, the phase delay time of the distortion signal amplifier is compensated so that the phase of the output signal of the subtraction circuit is changed within the operating region of the automatic phase control circuit, and the length of the delay line is calculated and applied to the circuit. The length of DL3, the delay line used for the feedforward linearizer for phase control, measures the delay time from the output point E1 of the subtraction circuit to the G-block input point E3 through the distortion signal amplifier. The delay time up to E2 can be applied. In order to calculate the length of DL1, the delay time from the input terminal A to the C1 point passing through the large power amplifier is measured and the delay time from the A point to the C2 point is applied in the same way as this value. As for the length of DL2, the delay time from the point B to the point D2 via the subtraction circuit and the distortion signal amplifier is measured, and the delay time from the point B to the point D1 is equally applied to this value.

Second, the present invention minimizes the variation of input signal power and distortion of various signals without using a pilot tone. The pilot tone method is to determine and adjust the degree of removal of the distortion signal component based on the artificially applied pilot tone. Since the pilot tone has a larger power than the intermodulation signal, the degree of adjustment of the equalizer is determined. This is relatively easy. When the phases of each pilot tone combined at the final output coupler are 180 ° out of phase and equal in magnitude, the intermodulation signal generated by the large power amplifier is automatically minimized. However, in order to maintain the broadband characteristics, the use of a large number of pilot tones is inevitable, and the circuit of the equalizer and monitor circuits is also complicated by several blocks covering a specific frequency band. In addition, the pilot tones pass through the large power amplifier, causing unwanted signal distortion, and it takes considerable time to initialize the signal.

In addition, since the active circuit is not used in the section used as a reference value in the linearizer, the circuit stability is improved by preventing signal distortion and oscillation. This method has a disadvantage in that the attenuator and phase shifter are placed in front of the large power amplifier to cause attenuation of the output signal due to insertion loss of these circuits. In the present invention, an insertion loss of about 1 to 2 dB occurred. Considering the circuit design considerations, the two signals input to the magnitude control circuit must be the same magnitude. The two signals input to the phase control circuit must be mutually in phase at the mixer input point in the phase control circuit. Considering the abnormal operation of the subtraction circuit, if the magnitude of the main signal is larger than the third-order intermodulation signal, it is preferable to apply a kind of limiter circuit that limits the size of the input signal so as not to generate a distortion signal. . RF circuit and digital circuit of linearizer should be designed and arranged separately to prevent signal interference between each other. The main path through the high power amplifier passes through the 35W high power signal, so the circuits (directional coupler in this paper) must be separated to minimize the effects of signal interference on the surrounding circuits. Can be.

Looking at the design and fabrication features of the linearizer of the present invention, the structure of the A block and the E block has the attenuator and the phase converter as the main circuit to facilitate the fabrication and characterization as the same structure. In the B and G blocks, the main circuits are the size control circuit and the phase control circuit, which seeks mutual identity, which provides much convenience in measuring and fabricating the circuit. In addition, when designing the internal circuit, the signal interference is minimized by separating the digital circuit part and the RF circuit part, and the circuit integration degree is reflected to the maximum.

Commercial linearizers are generally developed not to be larger than large power amplifiers, and the structure in which the large power amplifier and the linearizer face each other is common. In addition, integration of pure linearization circuits is an important part of the technology because circuits such as heat dissipation, alarm and shock protection are added. The linearizer integration of the present invention needs to increase the degree of directness more to be commercial level compared to the size of a large power amplifier, and the problem of mutual signal interference should also be improved. The H block, which is a synchronous signal generator, consists of pure digital circuits to minimize the mutual signal interference problem.

The C block, which is a subtraction circuit, is preferably made to block alone so that the fabrication and measurement of the input signal can be easily measured and the offset signal of the output signal can be easily measured. It is preferable to block the F block connected to the final output stage to facilitate accurate measurement and adjustment of the degree of linearization, and to add a large-capacity isolator to improve the reflection characteristic of the distortion signal amplifier output signal. It is preferable to block D block, which is a limiter circuit, so that the block can be easily adjusted according to the offset of the signal adjusting and subtracting circuit. In order to measure the linearizer, first, the delay time of the circuits should be measured for each section to obtain the required length of the delay line. As a result, phase compensation delay line of signal loop was 23.42nSec., Insertion loss value -2.9dB, and error compensation phase delay delay line was 26.83nSec., Insertion The loss value was -4.2dB, and the delay line for phase control located at the minus circuit output was 6.19nSec. And the insertion loss value was -0.74dB.

In the following order, as to automate the size and phase control function of the B block, this will be described with reference to FIG. First adjust the phase control after automation of the size control.

The magnitude control of the B block of FIG. 10 is sufficient to match the magnitude of the main signal at the point F of FIG. For this purpose, the DC signal, which is a magnitude control signal, is input to the attenuator through point A to adjust the nonlinear high power amplifier output signal variation due to the change of the input signal. At this time, the signal at the point F becomes a reference signal because it linearly responds to the change of the input signal. In designing the signal size control circuit, the applied reference signal was 12dBm in size and the attenuation of the attenuator was adjusted to be about 3dB. At this time, the attenuator control signal was about 5.7 [V]. The phase control of the B block may also be operated by the control signal at the B point so that both signals are in phase at the E and F points. To do this, first, the input phase of the phase control circuit part in the B block must be matched. That is, since the signal input to the D point acts as the LO signal of the mixer as a reference signal, the phase value at the mixer input point of this signal and the phase value of the signal input to the C point acting as the RF signal of the mixer are Match at the RF signal input point. That is, since the RF signal is input to the mixer via the directional coupler and the RF switch, its phase delay time must be compensated in the LO signal input section. Since the mixer operates normally when the LO signal is larger than the RF signal, the mixer needs to be adjusted in consideration of this. The output signal of the mixer is divided into the I signal and the Q signal by the DC switch. Since the magnitude components of the signals are the same, they are canceled with each other by the division circuit and only the pure phase components remain. For ALC input, 0 [V] was applied as the reference value and 0 [V] was applied as the initial value of the divider circuit. That is, the denominator I signal must be greater than 0 [V], and the Q signal of the numerator can be adjusted to 0 [V]. Although the function of the subtraction circuit cancels the main signal, in order for the linearizer to operate stably, there should be no oscillation of the canceled main signal. That is, even if the main signal size of the output signal is slightly higher than the IM signal size, it is not a problem because it is a negligible size compared to the output signal of the large power amplifier. The phase change amount of the large power amplifier was about 18 ° in the input signal change range of about 15dB, and the phase shifter control signal was adjusted to input about 5 [V].

The automation of the magnitude and phase control of the G-block is to adjust the non-linear magnitude and phase change by the distortion signal amplifier to have a fixed value. That is, the attenuator may be controlled through the H point, which is the output point of the magnitude control signal, so that the magnitudes of the two parts coincide at the G and K points, which are the input points of the G-block. The phase control may also be operated by the control signal at the point J so that both signals remain in phase at the points G and K.

The automation method is the same as the B block. The D block is a circuit that performs a kind of limiter function to prevent a large signal from being input into the distortion signal amplifier when the main signal canceling function of the subtraction circuit is not normal. For the successful execution of the above process, the operating range of the size control circuit and the phase control circuit should be measured in advance and reflected in the system design. In addition, the input signal must have a positive component for the normal operation of the division circuit, and the mutual interference of the RF signals should be considered. In the final stage, the IM signal should have a 180 ° phase difference, so adjust the phase at the L point. In other words, the IM signals cancel each other and only the main signal components are outputted, thereby successfully performing the function of the linearizer. The linearizer studied in this paper is a linearizer for PCS base station. The center frequency is 1855 MHz, the input signal is 2-Tone, and the tone interval is 0.6MHz. The large power amplifier, which is to be linearized, applied a commercial product of 35MW capacity of 'KMW'. The C / I improvement by the linearizer is about 18 ~ 24dB and 17 ~ 25dB, which resulted in the required level of design and is commercially available.

An example of the intensity control circuit will now be described with reference to FIG. 13A. In FIG. 13A, if the input signal A is a reference signal and B is a pressure signal, the RF switch 201 outputs the A and B signals at a 10 kHz period by the synchronization signal (the synchronization signal is a 1 MHz signal generated by the oscillator 202). The signal is output as a 10 kHz signal by the frequency divider 203), and it is converted into a DC voltage by the detector 204 and sent to the DC switch 205. If the DC switch is that because synchronization signal is operating only in "high" state, when the input it to the inverting and non-inverting state, A, a B signal and alternately outputs the output signals of the A signal V _ref, the B signal V _i , V _i by matching the signal strength on the basis of the signal strength V _ref in the ALC (206) by using the attenuator 207 of the same size as the value of V _ref, and an automatic intensity control circuit.

In operation principle, if the output power of the amplifier is too large, V _i is larger than V _ref , and V _out is increased to the time constant τ = RC, and the attenuator of the reflector increases the attenuation value.

If the attenuation value becomes excessively large, V _i becomes smaller than V _ref and this time V _out decreases, thus reducing the attenuation of the attenuator. When the RF output is detected as a DC voltage, V _ref increases and decreases due to the center of gravity. It can be seen that the operation time of the circuit is fixed because it is much smaller than the time constant. If V _out is fixed in this way, it can produce constant output no matter what input is applied. Therefore, the level of the output signal is determined constant according to the value of V _ref . At this time, the output Vout, which is the DC voltage, approaches the value that adjusts the attenuation amount of the attenuator so that the Vi value and the Vref value coincide, and outputs this value constantly. In addition, the resistance value in the ALC can be determined in consideration of the DC voltage characteristics of the attenuator. When the signal detected with a value smaller than the reference voltage of the ALC is input, the input power cannot be increased, so that the ALC is not ALC.

13B shows an example of the above intensity control circuit.

Next, an example of the phase control circuit will be described with reference to FIG. 14A. In Fig. 14A, B∠φ, which is an output signal of the large power amplifier, is input to the 90 ° 3dB coupler 702 as the input signal of the phase control circuit through the power divider 701, and the magnitudes of the signals are respectively. It generates two signals, cosφ and sinφ, which have a 90 ° phase difference.

This is converted into a tanφ form so that the magnitude of the phase can be controlled.

These signals are alternately output by the synchronous signal from the RF switch 703 in 10 kHz periods (the synchronous signal is output as the 1 kHz signal generated by the oscillator 704 as the 10 kHz signal in the divider 705), and a frequency mixer ( 706 is inputted as an RF signal.

The reference signal A 신호 θ is input to the LO signal of the frequency mixer 706 through the power divider 707. The frequency mixer has a magnitude multiplied by the ratio of the conversion loss K, and the frequency phase of the output IF signal is determined by the frequency and phase of the RF and LO signals.

When the RF signal is equal to the frequency of the LO signal, the IF output frequency of the frequency mixer generates a DC component and other frequency components. At this time, the magnitude of the DC component is determined by the phase relationship between the RF signal and the LO signal.

Expanding the equation by Equation 1, the input signal strength of the frequency mixer equally , Was applied.

When the harmonic component is removed by the low pass filter of the IF output signal, the IF output signal is expressed by Equation 2 below.

Applying the same method to IF2, Equation 3 is obtained.

When the frequency mixer 706 sends these IF1 and IF2 signals converted to DC voltage to the DC switch 708, the DC switch alternately outputs the IF1 and IF2 signals at a 10 kHz period.

Since these output signals have a 90 ° phase difference, the IF1 signal can be applied as an I signal (In-phase) and the IF2 signal as a Q signal (Quadrature-phase). When the I signal is used as the denominator value and the Q signal is input as the numerator value, the (4 / K) AB magnitude component disappears as the elapse of the Q / I division circuit, and tan (φ-θ) Only pure phase component values can be extracted.

The reference value of the ALC 710 is set to 0 V, and the phase value of tan (?-?) Which is a DC component is input to the ALC as an input value.

The operation principle of the ALC is the same as that of the intensity control, where 0 V is selected as the reference value because the tan value is set to 0 with the phase value of tan (φ−θ) as φ = θ as the initial input value.

That is, when the phase of the large power amplifier is changed and the value of φ is changed, the V _out value is sent to the phase converter 711 so as to satisfy the condition of φ = θ in the ALC to control the phase of the large power amplifier in the phase converter.

13B shows an example of the above phase control circuit.

Fig. 11 shows the measuring points of the linearizer according to the present invention. For example, as a linearizer for a PCS base station, the center frequency is 1855 MHz, the input signal is applied 2 Tone, and the tone interval is 0.6 MHz. The large power amplifier for linearization applied KMW's 35W commercial product. The measuring method is composed of the whole system first and then under normal operating conditions.

'Measurement 1' measures the MAU (6) output signal. The through port of the 30dB directional coupler (7) is shorted by 50 ohms and the output signal of the coupled port is connected to the spectrum analyzer. do.

'Measure 2' measures the output signal of the subtraction circuit 10, and is performed by connecting the signal of the subtraction circuit output terminal to the spectrum analyzer.

'Measurement 3' measures the output signal when the linearization function is applied by connecting between the CAU ((16) output terminal and the 10dB directional coupler 17), and outputs the output signal when the linearization function is not applied by shorting the same point. The measurement was performed at 1 W, 2 W, 10 W, 20 W, 25 W and 30 W, respectively, with and without linearization at the measurement points.

IM improvement by the linearizer was about 18-24 dBm and 17-25 dBc, showing a fairly good effect.

Table 1 shows the output signal of the MAU output signal and the subtraction circuit, and Tables 2 and 3 show the output characteristics of the large power amplifier without the linearizer and the final stage when the feedforward linearizer designed and manufactured in the present invention is applied. The output signal is shown, and the improvement amount of 3rd-IM and C / I by the linearizer is expressed in dBc.

Output power (dBm / tone) MAU output power Subtraction Circuit Output Power f-3 f + 3 f-3 f + 3 30 W (41.7) 0.62 -2.40 -38.80 -42.45 25 W (41.0) -1.25 -3.60 -38.17 -40.15 20 W (40.0) -1.00 -3.12 -37.77 -39.32 10 W (37.0) -2.52 -3.65 -38.97 -39.77 2 W (30.0) -9.75 -9.70 -49.40 -49.40 1W (27.0) -19.10 -16.40 -57.65 -57.65

Output power (dBm / tone) No feedforward Use feedforward Improvement amount (dB) f-3 f + 3 f-3 f + 3 f-3 f + 3 30 W (41.7) -5.17 -7.30 -26.62 -26.05 21.45 18.75 25 W (41.0) -4.92 -7.52 -26.67 -25.65 21.75 18.13 20 W (40.0) -2.92 -4.25 -27.30 -27.32 24.38 23.07 10 W (37.0) -4.68 -5.89 -23.71 -27.54 19.03 21.65 2 W (30.0) -13.62 -13.15 -32.87 -32.20 19.25 19.05 1W (27.0) -20.62 -20.72 -37.37 -37.30 16.75 16.58

Output power (dBm / tone) No feedforward Enable feedforward Improvement amount (dB) 30 W (41.7) 41.47 62.94 21.47 25 W (41.0) 41.29 63.17 21.88 20 W (40.0) 38.17 62.75 24.58 10 W (37.0) 36.68 60.18 23.50 2 W (30.0) 39.19 58.49 19.30 1W (27.0) 42.94 59.84 16.90

The present invention has been described above with reference to one embodiment shown in the accompanying drawings, but the present invention is not limited thereto, and various modifications are possible within a range easily conceived by those skilled in the art. Therefore, the limitation of the present invention should be limited only by the following claims.

12A and 12B show the results of measuring the output spectrum when the linearizer is not applied when the output power of the large power amplifier is 30W and when the linearizer is applied. While there is almost no difference in the signal levels of the carriers (No. 1, No. 2), it can be seen that the intermodulation signal levels (No. 3 to No. 6) are significantly reduced as a result of using the linearizer according to the present invention.

As described above, in the present invention, a novel feedforward linearizer using a delay amplifier for correction amplifier control has been proposed, and has exhibited 58.5 dBc to 63.2 dBc IMD characteristics in an operating range of about 15 dB. A C / I improvement of about 16.9 dB to 24.6 dB was achieved. As a result, it becomes possible to obtain an effective linearizer.

Claims (3)

A first power divider (2) for distributing the carrier signal input through the carrier input terminal (1) to the main path and the auxiliary path; A first variable attenuator (4) and a first variable phase converter (5) for changing the strength and phase of the signal in the main path; A large power amplifier (6) connected to the rear end of the first variable attenuator (4) and the first variable phase converter (5) and amplifying to a desired output level; A subtraction circuit 10 outputting only intermodulation distortion signals among the carrier wave and intermodulation distortion signals generated in the main path; A first directional coupler (7) which partially extracts carrier and intermodulation distortion signal components from the output of the large power amplifier (6) and applies them as a comparison signal to the subtraction circuit (10); A first delay line (3) for delaying the carrier signal in the auxiliary path and applying it as a reference signal to the subtraction circuit (10); A second variable attenuator 14 and a second variable phase converter 15 for changing the intensity and phase of the intermodulation distortion signal components obtained by the subtraction circuit 10; An auxiliary amplifier (16) connected to the second variable attenuator (14) and the second variable phase converter (15), for amplifying to a desired output level; A carrier wave by combining the intermodulation distortion signal components on the auxiliary path amplified to an appropriate size by the auxiliary amplifier 16 and a sum signal of the carrier wave and the intermodulation distortion signals passing through the second delay line 13 on the main path; A second directional coupler 18 for extracting and outputting only the signal; Including but not limited to: The carrier signal component extracts the comparison signal information and applies it to the first magnitude control circuit (AGC1) 24 and the first phase control circuit (APC1) 25 so that the spectrum of the reference signal and the comparison signal coincide. And extract the reference signal information to be applied to the first magnitude control circuit (AGC1) 24 and the first phase control circuit (APC1) 25 so that the carrier signal components in the main path and the auxiliary path The first variable amplitude control circuit (AGC1) 24 and the first phase control circuit (APC1) 25 allow the first variable attenuator 4 and the first variable phase converter 5 to have the same spectrum. A B block 200 for controlling to appropriately change the magnitude and phase of the comparison signal; And When the output signal of the subtraction circuit is coupled to the main path again, the output signal information of the subtraction circuit is extracted to have the same magnitude as that of the intermodulation distortion component on the main path and become in the reverse phase (180 °). The second magnitude control circuit (AGC2) 34 to be applied to the control circuit (AGC2) 34 and the second phase control circuit (APC2) 35, and to extract the output signal information of the auxiliary amplifier (16) And the second phase control circuit (AGC2) 34 and the second phase control circuit (APC2) 35 are applied to the second phase control circuit (APC2) 35 so that the main path and the auxiliary path. A G block for controlling the second variable attenuator 14 and the second variable phase converter 15 to appropriately change the magnitude and phase of the output signal of the subtraction circuit so that the intermodulation distortion signal components are in opposite spectra. 700; Wherein the final output is such that intermodulation signal components are removed and only pure carriers remain. The method of claim 1, The third directional coupler 8 is coupled to the input signal of the subtraction circuit 10, and the extracted signal is passed through the second power divider 9 to the first magnitude control circuit AGC1 24. A fourth directional coupler 28 is applied to a signal on the auxiliary path of the first power divider 2 and is applied to the first phase control circuit APC1 25, and the extracted signal is a third power. The first magnitude control circuit AGC1 24 and the first phase control circuit APC1 25 are applied to the first magnitude control circuit AGC1 24 and the first phase control circuit AGC1 24 and the first phase through the distributor 29. A linear power amplifier having a linearizer characterized in that the control circuit (APC1) 25 controls the first variable attenuator 4 and the first variable phase converter 5 to appropriately change the magnitude and phase of the input signal. . The method according to claim 1 or 2, The fifth directional coupler 18 is coupled to the amplified signal of the auxiliary amplifier 16, and the extracted signal is passed through the fourth power divider 19 to the second magnitude control circuit AGC2 34. And a second phase control circuit (APC2) 35, and a sixth directional coupler 38 is coupled to the output signal of the subtraction circuit 10, and the extracted signal is a third delay line 33. And the second magnitude control circuit (AGC2) 34 and the second phase control circuit (APC2) 35 through the fifth power divider 39, thereby providing the second magnitude control circuit AGC2 (34). And the second variable attenuator 14 and the second variable phase converter 15 are controlled by the second phase control circuit (APC2) 35 so as to appropriately change the magnitude and phase of the distortion signal. Linear power amplifier with group.