High-frequency amplifier
The high-frequency amplifier addresses the challenge of external control requirements by using in-phase transistor groups and matching circuits for self-compensation, reducing distortion and power consumption, making it suitable for satellite communications.
Patent Information
- Application Number
- PCT/JP2023/020579
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-06-02
- Publication Date
- 2024-12-05
AI Technical Summary
Conventional high-frequency amplifiers require external device control for nonlinear distortion compensation, leading to increased size, weight, and power consumption, which is unsustainable for satellite communications due to limited solar panel and battery capacities.
A high-frequency amplifier design featuring multiple transistor groups with in-phase signal distribution and matching circuits, including tip short-circuit stubs that allow impedance matching within the operating frequency band, enabling self-compensation for nonlinear distortion without external control.
This design reduces intermodulation distortion, lowers power consumption, and enhances efficiency, making it suitable for satellite communications by eliminating the need for external control and minimizing size and weight.
Smart Images

Figure JP2023020579_05122024_PF_FP_ABST
Abstract
Description
High-frequency amplifier
[0001] The present disclosure relates to high frequency amplifiers.
[0002] High-frequency amplifiers are widely used as amplifiers for communications in satellites, etc. For example, Patent Document 1 describes an amplifier device with a distortion control function that compensates for nonlinear distortion of a Doherty amplifier by combining a Doherty amplifier equipped with a carrier amplifier circuit and a peak amplifier circuit with a predistorter. This amplifier device can achieve both high amplification efficiency and reduced distortion during communications.
[0003] Japanese Patent Application Laid-Open No. 2008-22513
[0004] Conventional high-frequency amplifiers have the problem of requiring control by an external device such as a computer to compensate for nonlinear distortion. For example, a predistortion distortion compensation circuit for compensating for nonlinear distortion is large in circuit scale and requires a computer to control it. This increases the volume, weight, and power consumption of the entire high-frequency amplifier. In particular, in satellite communications, there are restrictions on the performance of the computers that can be used, and power consumption is high in proportion to the computer's performance. Furthermore, solar panels and their battery capacities are limited, so high-frequency amplifiers with high power consumption cannot be used in satellite communications.
[0005] The present disclosure is intended to solve the above-mentioned problems, and has an object to provide a high-frequency amplifier that can compensate for nonlinear distortion without being controlled by an external device.
[0006] A high-frequency amplifier according to the present disclosure comprises a plurality of transistor groups, each composed of one or a plurality of transistors, to which in-phase signals are distributed; a plurality of mutually different input-side matching circuits connected respectively to the plurality of transistor groups; a plurality of parallel circuits connected respectively to the plurality of input-side matching circuits and short-circuiting at harmonic frequencies of signals output from the plurality of transistor groups; and a plurality of mutually different output-side matching circuits, one end connected to a connection point between the input-side matching circuit and the parallel circuit, and the other end connected to the plurality of output-side matching circuits, and circuit constants are set so that the impedance looking into the transistor groups from the connection point between the output-side matching circuits is equal in the operating frequency band.
[0007] According to the present disclosure, a high-frequency amplifier according to the present disclosure includes a plurality of transistor groups, a plurality of mutually different input matching circuits connected to the plurality of transistor groups, each of the plurality of parallel circuits connected to the plurality of input matching circuits and shorted at harmonic frequencies of signals output from the plurality of transistor groups, and a plurality of mutually different output matching circuits, one end of which is connected to a junction point between the input matching circuit and the parallel circuit, and the other end of which is connected to the plurality of output matching circuits. The plurality of output matching circuits are set with circuit constants such that the impedance looking into the transistor groups from the junction point between the output matching circuits is equal in the operating frequency band. Because intermodulation distortion is reduced at the junction point between the output matching circuits, the high-frequency amplifier according to the present disclosure can compensate for nonlinear distortion without control by an external device.
[0008] 1 is a circuit diagram showing the configuration of a high-frequency amplifier according to a first embodiment. 2 is a Smith chart showing the impedance characteristics of a short-circuited stub. 3 is a circuit diagram showing an equivalent circuit in the operating frequency band of the high-frequency amplifier according to the first embodiment. 4 is a circuit diagram showing an equivalent circuit in a frequency band twice the operating frequency of the high-frequency amplifier according to the first embodiment. 5 is a block diagram showing the configuration of the high-frequency amplifier according to the first embodiment, including a matching circuit made of a transmission line. 6 is a diagram showing an example of the characteristic impedance and electrical length of the transmission line in the high-frequency amplifier of FIG. 5. 7 is a Smith chart showing the impedance characteristics of the high-frequency amplifier of FIG. 5. 8A, 8B, 8C, and 8D are contour diagrams showing the results of load-pull calculations of the output power and the phase of the low-frequency side frequency of IMD3 with respect to impedance. 9A and 9B are diagrams showing an example of vector synthesis of an IMD3 signal. 10 are Smith charts showing the results of normalizing the LP measurement results of the load impedance at the frequency of IMD3 by the load impedance at the operating frequency. 11A and 11B are characteristic diagrams showing the relationship between the output power, IMD3 signal, and output efficiency of a high-frequency amplifier. It is a circuit diagram showing a tip short-circuit stub.It is a circuit diagram showing the configuration of a modified example of the high-frequency amplifier according to the first embodiment.It is a circuit diagram showing the configuration of the high-frequency amplifier according to the second embodiment.
[0009] Embodiment 1. Fig. 1 is a circuit diagram showing the configuration of a high-frequency amplifier 1 according to embodiment 1. In Fig. 1, high-frequency amplifier 1 is an amplifier having a function of compensating for nonlinear distortion, and includes a transistor group 2, matching circuits 3A, 3B, 3C, 3D, a transistor group 4, and short-circuited stubs 5A and 5B. Transistor group 2 and transistor group 4 are each composed of one or a plurality of transistors for amplifying an input high-frequency signal.
[0010] 1 shows a case where each of the transistor groups 2 and 4 is composed of one transistor. The transistor groups 2 and 4 may each be composed of a plurality of transistors connected in series or parallel, or a combination of series and parallel. When the transistor groups 2 and 4 are each composed of a plurality of transistors, there is one input terminal to which a high-frequency signal is input, and one or more output terminals from which a high-frequency signal is output. When the transistor group has multiple output terminals, the multiple output terminals are connected to a downstream input-side matching circuit.
[0011] The control terminals of the transistors constituting the transistor groups 2 and 4 are connected as input terminals to which a high-frequency signal is input. For example, as shown in Figure 1, since the transistor groups 2 and 4 are each composed of one field-effect transistor (hereinafter referred to as FET), the high-frequency signal is input to the gate terminal of the FET.
[0012] The FETs constituting transistor group 2 have source terminals connected to ground potential and drain terminals connected to matching circuit 3 A. The FETs constituting transistor group 4 have source terminals connected to ground potential and drain terminals connected to matching circuit 3 C. A radio frequency signal input to the input terminal of radio frequency amplifier 1 is split into two in-phase signals by an input matching circuit not shown in FIG. 1 , with one in-phase signal being input to the gate terminal of the FETs constituting transistor group 2 and the other in-phase signal being input to the gate terminal of the FETs constituting transistor group 4.
[0013] The conventional Doherty amplifier described in Patent Document 1 also includes transistors corresponding to transistor groups 2 and 4 for amplifying high-frequency signals, similar to high-frequency amplifier 1. However, in the conventional Doherty amplifier, there is a phase difference of about 90 degrees between the high-frequency signals input to each of these transistor groups, and no in-phase signals are input. Thus, one of the features of high-frequency amplifier 1 that distinguishes it from conventional amplifiers is that the high-frequency signals input to each of the multiple transistor groups are in-phase.
[0014] Matching circuits 3A, 3B, 3C, and 3D are denoted as MN in FIG. 1 . Matching circuit 3A is an input-side matching circuit connected to the output terminal of transistor group 2. A short-circuited stub 5A is connected to the output terminal of matching circuit 3A. Impedance Z1 indicated by an arrow in FIG. 1 is the impedance seen from the output terminal of transistor group 2 (e.g., the drain terminal of an FET) toward the circuit side. Matching circuit 3C is an input-side matching circuit connected to the output terminal of transistor group 4. A short-circuited stub 5B is connected to the output terminal of matching circuit 3C. Impedance Z2 indicated by an arrow in FIG. 1 is the impedance seen from the output terminal of transistor group 4 (e.g., the drain terminal of an FET) toward the circuit side. Matching circuits 3A and 3B have circuit constants set to have different electrical lengths in the operating frequency band of high-frequency amplifier 1.
[0015] The short-circuited stub 5A has one end connected to the output side of the matching circuit 3A and the other end grounded, and is a parallel circuit that is shorted at the harmonic frequency of the signal output from the transistor group 2. Furthermore, the short-circuited stub 5B has one end connected to the output side of the matching circuit 3C and the other end grounded, and is a parallel circuit that is shorted at the harmonic frequency of the signal output from the transistor group 4. Note that the impedance Z3 indicated by the arrow in Figure 1 is the impedance created by the short-circuited stubs 5A and 5B. The length of each of the short-circuited stubs 5A and 5B is approximately ¼ wavelength near the center of the operating frequency of the high-frequency amplifier 1.
[0016] Matching circuit 3B is an output-side matching circuit having one end connected to the connection point between matching circuit 3A and short-circuited stub 5A and the other end connected to the output side of matching circuit 3D. Matching circuit 3D is an output-side matching circuit having one end connected to the connection point between matching circuit 3C and short-circuited stub 5B and the other end connected to the output side of matching circuit 3B. Matching circuits 3B and 3D have circuit constants set to have different electrical lengths in the operating frequency band. An output terminal (not shown in FIG. 1 ) is connected to combination point S, which is the connection point between the output end of matching circuit 3B and the output end of matching circuit 3D.
[0017] Impedance Z4 indicated by an arrow in Fig. 1 is the impedance that is measured from the combining point S of matching circuits 3B and 3D to transistor group 2 that is connected to matching circuit 3B via matching circuit 3A. Impedance Z5 indicated by an arrow in Fig. 1 is the impedance that is measured from the combining point S to transistor group 4 that is connected to matching circuit 3D via matching circuit 3C. In high-frequency amplifier 1, matching circuits 3B and 3D have circuit constants that make impedance Z4 and impedance Z5 equal in the operating frequency band.
[0018] 2 is a Smith chart showing the impedance characteristics of the shorted stubs 5A and 5B. The Smith chart on the left side of FIG. 2 shows the characteristics of the impedance Z3 created by the shorted stubs 5A and 5B in the operating frequency band of the high-frequency amplifier 1 (9 GHz to 11 GHz in FIG. 2). The triangular plot represents the impedance Z3. By sweeping the frequency from 9 GHz (Min) to 11 GHz (Max), the impedance Z3 changes within the range indicated by the black line. As the Smith chart shows, the impedance Z3 in the operating frequency band is high.
[0019] The Smith chart on the right side of Figure 2 shows the characteristics of the impedance Z3 created by the short-circuited stubs 5A and 5B in a frequency band twice the operating frequency of the high-frequency amplifier 1 (18 GHz to 22 GHz in Figure 2). The triangular plot represents the impedance Z3. By sweeping the frequency from 18 GHz (Min) to 22 GHz (Max), the impedance Z3 changes within the range indicated by the black line. As the Smith chart shows, the impedance Z3 in the frequency band twice the operating frequency becomes low.
[0020] 2, the short-circuited stubs 5A and 5B do not affect the circuit operation of the high-frequency amplifier 1 in the operating frequency band, and have the function of reflecting second harmonics output from the transistor groups 2 and 4 back to the transistor groups 2 and 4 in a frequency band twice the operating frequency. Note that in the high-frequency amplifier 1, the short-circuited stubs 5A and 5B need only be circuits that have the above function, and are not required to be short-circuited stubs. For example, a parallel circuit such as a parallel resonator may be provided instead of the short-circuited stubs 5A and 5B.
[0021] Next, the effect of the short-circuited stubs 5A and 5B will be described with reference to Figures 3 and 4. Figure 3 is a circuit diagram showing an equivalent circuit in the operating frequency band of the high-frequency amplifier 1. As explained with reference to Figure 2, the short-circuited stubs 5A and 5B do not affect the circuit operation of the high-frequency amplifier 1 in the operating frequency band. For this reason, as shown in Figure 3, the short-circuited stubs 5A and 5B are omitted from the equivalent circuit in the operating frequency band.
[0022] FIG. 4 is a circuit diagram showing an equivalent circuit in a frequency band twice the operating frequency of the high-frequency amplifier 1. As described with reference to FIG. 2, in the frequency band twice the operating frequency, the short-circuited stubs 5A and 5B reflect the second harmonic output from the transistor groups 2 and 4 back to the transistor groups 2 and 4. That is, in the frequency band twice the operating frequency, the impedance Z3 of each of the short-circuited stubs 5A and 5B becomes very low and is equivalent to ground. Therefore, as shown in FIG. 4, in the equivalent circuit in the frequency band twice the operating frequency, the output terminal of matching circuit 3A is essentially grounded, and the output terminal of matching circuit 3C is grounded, so that matching circuits 3B and 3D are omitted. Furthermore, in the high-frequency amplifier 1, matching circuits 3A and 3C have circuit constants set to have different electrical lengths in the frequency band twice the operating frequency.
[0023] Next, a specific configuration of the matching circuits 3A to 3D will be described. FIG. 5 is a block diagram showing the configuration of a high-frequency amplifier 1 equipped with matching circuits 3A to 3D, each consisting of transmission lines A to D. In FIG. 5, transmission line A is matching circuit 3A, transmission line B is matching circuit 3B, transmission line C is matching circuit 3C, and transmission line D is matching circuit 3D. Furthermore, one of the two transmission lines E is a short-circuited stub 5A, and the other is a short-circuited stub 5B. Furthermore, transistor 6 is an FET that constitutes transistor group 2, and transistor 7 is an FET that constitutes transistor group 4. Furthermore, transmission lines A to E are denoted as TL in FIG. 5.
[0024] In Fig. 5, the output load impedance (=optimum load impedance) of transistor 6 is assumed to be 50 Ω. Furthermore, the connection point (combination point S) between transmission line B and transmission line D is the output terminal of high-frequency amplifier 1 and is connected to a 50 Ω load. Fig. 6 is a diagram showing an example of the characteristic impedance and electrical length of transmission lines A to E in the high-frequency amplifier 1 shown in Fig. 5. In Fig. 6, the characteristic impedance of all transmission lines A to E is 70.7 Ω. Furthermore, by setting the electrical length of transmission line E to 90 degrees, the impedance becomes high in the operating frequency band and low in a frequency band twice the operating frequency.
[0025] Also, by setting the sum of the electrical lengths of transmission lines A and B to 90 degrees and the sum of the electrical lengths of transmission lines C and D to 90 degrees, the impedance Z1 looking from the output end of transistor 6 toward the circuit side and the impedance Z2 looking from the output end of transistor 7 toward the circuit side are both set to 50 Ω at the operating frequency. Furthermore, because the electrical lengths of transmission lines A and C are different from each other, the impedance Z1 looking from the output end of transistor 6 toward the circuit side and the impedance Z2 looking from the output end of transistor 7 toward the circuit side are different at a frequency twice the operating frequency.
[0026] Fig. 7 is a Smith chart showing the impedance characteristics of the high-frequency amplifier 1 of Fig. 5. The Smith chart on the left side of Fig. 7 shows the characteristics of impedance Z1 and impedance Z2 at the operating frequency (1 GHz in Fig. 7). The triangular plots show the value of impedance Z1, and the square plots show the value of impedance Z2, which are normalized to 50 Ω. As is clear from the Smith chart, perfect impedance matching is achieved between impedance Z1 and impedance Z2 at the operating frequency.
[0027] On the other hand, the Smith chart on the right side of FIG. 7 shows the characteristics of impedances Z1 and Z2 at a frequency twice the operating frequency (2 GHz in FIG. 7). The triangular plots show the value of impedance Z1, and the square plots show the value of impedance Z2, which are normalized to 50 Ω. As is clear from the Smith chart, at a frequency twice the operating frequency, impedances Z1 and Z2 have different values, and impedance matching is not achieved. Furthermore, as is clear from the Smith charts on the left and right sides of FIG. 7, the values of impedances Z1 and Z2 at the operating frequency are different from the values of impedances Z1 and Z2 at a frequency twice the operating frequency.
[0028] Next, we will explain why the high-frequency amplifier 1 is able to compensate for nonlinear distortion without control by an external device. In wireless communication systems that use high-frequency amplifiers, when communication signal distortion is analyzed in frequency space, a characteristic is observed in which the noise floor is elevated on the low-frequency and high-frequency sides of the used frequency band in the frequency spectrum. What appears to be an elevated noise floor is actually a superposition of intermodulation distortions such as IMD3 (third-order intermodulation distortion), IMD5, IMD7, and so on.
[0029] Since IMD3 has the highest signal level among all intermodulation distortions, the following describes how distortion can be reduced by reducing IMD3 in the high-frequency amplifier 1. One method for observing IMD3 is to apply two high-frequency signals (hereinafter referred to as two-tone signals) to the input terminal of the high-frequency amplifier 1. In this method, when the frequency on the lower frequency side of the two-tone signal is designated as FL and the frequency on the higher frequency side is designated as FH, the frequencies that cause IMD3 (hereinafter referred to as IMD3 frequencies) are FL-(FH-FL) and FH+(FH-FL).
[0030] 8A, 8B, 8C, and 8D are contour diagrams showing the results of load-pull calculations of the output power Pout (left side of each diagram) and the phase IMD3L Phase of the low-frequency IMD3 frequency (right side of each diagram) versus impedance in the high-frequency amplifier 1. These results are obtained when a two-tone signal is input to the input terminal of the high-frequency amplifier 1, and the load reflection coefficient ∠Γ(2f 0 ) is 1 and the phase angles are -90 degrees, 0 degrees, 90 degrees, and 180 degrees, and load pull calculations are performed so that the load impedance is the same at the two operating frequencies of the two tones and at the four frequencies on the low-frequency side and high-frequency side of the IMD3 frequency.
[0031] As shown in Figures 8A, 8B, 8C, and 8D, markers M1 and M2 are set at the impedance where the output is maximum. The impedance where markers M1 and M2 are set is the impedance where the load reflection coefficient ∠Γ(2f 0) has the same impedance regardless of the phase angle. On the other hand, as is clear from the calculation results for the IMD3 frequency phase shown in the right diagrams of Figures 8A, 8B, 8C, and 8D, even though the load impedance at the operating frequency and the IMD3 frequency is the same, the phase of the IMD3 signal (hereinafter referred to as the IMD3 signal) generated by transistor groups 2 and 4 varies depending on the load phase of the double frequency. In other words, the phase angle dependency of the reflection coefficient of the double frequency of the IMD3 signal generated by the transistors also occurs in the high-frequency amplifier 1. Specifically, the phases of the IMD3 signals generated by transistor groups 2 and 4 are no longer in phase. In this case, in the high-frequency amplifier 1, IMD3 signals with different phases are vector-combined at the combination point S.
[0032] Next, vector synthesis of IMD3 signals with different phases will be described. Fig. 9A is a diagram showing an example (1) of vector synthesis of IMD3 signals. Synthesis example (1) shows a case where vector a, which represents the IMD3 signal generated in transistor group 2, and vector b, which represents the IMD3 signal generated in transistor group 4, have approximately the same phase angle. When the phase angles are approximately the same, as shown in Fig. 9A, the synthesized vector of vector a and vector b is larger than vector a and vector b.
[0033] 9B is a diagram showing an example (2) of vector synthesis of IMD3 signals. Synthesis example (2) shows a case where the phase angle is significantly different between vector a, which represents the IMD3 signal generated in transistor group 2, and vector b, which represents the IMD3 signal generated in transistor group 4. When the phase angles are significantly different, as shown in FIG. 9B , the synthesized vector of vector a and vector b is smaller than vector a and vector b.
[0034] By comparing the composite vectors in composite example (1) and composite example (2), it can be seen that the amplitude of the signal corresponding to the composite vector decreases when the difference in phase angle is large. At composite point S, the phase angles of vector a corresponding to the IMD3 signal generated in transistor group 2 and vector b corresponding to the IMD3 signal generated in transistor group 4 differ, and vector a and vector b are vector-combined as shown in composite example (2).
[0035] For example, in the conventional amplifier described in Patent Document 1, the phases of the IMD3 signals are in-phase with each other, similar to the phases of the operating frequency signals. In contrast, in the high-frequency amplifier 1, the phases of the IMD3 signals are different from each other, as described above. For this reason, it can be said that the amplitude of the IMD3 signal in the high-frequency amplifier 1 is smaller than that in the conventional amplifier.
[0036] Furthermore, from the perspective of active impedance, when traveling waves of different phases are combined, an impedance mismatch occurs. When an impedance mismatch occurs, some components of the IMD3 signal are returned to the transistor that generated the IMD3 signal, and the remaining components are applied to the other transistor. In this case, the active impedance at the current source end of the transistor at the IMD3 frequency varies from the passive load impedance generated by the matching circuit. In other words, the load impedance at the IMD3 frequency seen from transistor groups 2 and 4 differs from the load impedance at the operating frequency.
[0037] FIG. 10 is a Smith chart plotting the LP measurement results of the load impedance at the IMD3 frequency normalized by the load impedance at the operating frequency. As shown in FIG. 10, it can be seen that the greater the difference between the load impedance at the IMD3 frequency and the load impedance at the operating frequency, the smaller the amplitude of the IMD3 signal output from the transistor. This verification result is described, for example, in the following reference: (Reference) "Effects of Load Impedances at Third Order Intermodulation Tones", EUMIC 2020.
[0038] Next, the reason why the amplitude of the IMD3 signal is reduced will be explained. FIG. 11A is a characteristic diagram showing the relationship between the output power and the amplitude of the IMD3 signal for a conventional amplifier and the high-frequency amplifier 1. The conventional amplifier is described in Patent Document 1, and the phases of the IMD3 signals are in-phase. In FIG. 11A, characteristic F is the characteristic of the conventional amplifier, showing the relationship between the output power (dBm) and the amplitude (dBc) of the IMD3 signal. Furthermore, characteristic G is the characteristic of the high-frequency amplifier 1, showing the relationship between the output power (dBm) and the amplitude (dBc) of the IMD3 signal. FIG. 11A compares characteristic F of the conventional amplifier with characteristic G of the high-frequency amplifier 1 at the same output power. As indicated by the arrows in FIG. 11A, at the same output power, the amplitude H2 of the IMD3 signal for the high-frequency amplifier 1 is smaller than the amplitude H1 of the IMD3 signal for the conventional amplifier, thereby achieving low distortion.
[0039] FIG. 11B is a characteristic diagram showing the relationship between output power, IMD3 signal amplitude, and amplifier efficiency for a conventional amplifier and high-frequency amplifier 1. In the conventional amplifier, the phases of the IMD3 signals are in-phase. In FIG. 11B, characteristic F shows the relationship between the output power (dBm) and the IMD3 signal amplitude (dBc) of the conventional amplifier shown in FIG. 11A. Characteristic G shows the relationship between the output power (dBm) and the IMD3 signal amplitude (dBc) of high-frequency amplifier 1 shown in FIG. 11A. Characteristic I shows the relationship between output power (dBm) and efficiency. In FIG. 11B, characteristic F of the conventional amplifier and characteristic G of high-frequency amplifier 1 are compared at the same IMD3 signal amplitude. As indicated by the arrows in FIG. 11B, at the same IMD3 signal amplitude, high-frequency amplifier 1 achieves higher output power and higher efficiency than the conventional amplifier.
[0040] Next, the structure of the short-circuited stubs 5A and 5B will be described. FIG. 12 is a circuit diagram showing the short-circuited stubs 5A and 5B. As shown in FIG. 12, the short-circuited stubs 5A and 5B may be configured with a matching circuit 3A1 and a capacitor 8. The matching circuit 3A1 is an intra-stub matching circuit having one end connected to the input-side matching circuit and the other end connected to the capacitor 8. The capacitor 8 is a capacitor that cuts the DC component in the signal (hereinafter referred to as DC cut), and has one end connected to the matching circuit 3A1 and the other end grounded. The matching circuit 3A1 is denoted as MN in FIG. 12.
[0041] 12 , there is no need to provide a dedicated DC blocking circuit between the transistor group 2 and the short-circuited stub 5A, and there is no need to provide a dedicated DC blocking circuit between the transistor group 4 and the short-circuited stub 5B. This reduces circuit loss caused by the dedicated DC blocking circuit, achieving high efficiency of the high-frequency amplifier 1. Furthermore, it is also possible to supply power to the transistor groups 2 and 4 from between the matching circuit 3A1 and the capacitor 8.
[0042] The high-frequency amplifier according to the first embodiment may include two of the high-frequency amplifiers 1 shown in FIG. 1. FIG. 13 is a circuit diagram showing the configuration of a modified high-frequency amplifier 1A. In FIG. 13, the high-frequency amplifier 1A is an amplifier having the function of compensating for nonlinear distortion, and includes two of the high-frequency amplifiers 1 shown in FIG. 1, and a matching circuit 9. The matching circuit 9 has the output terminals of the two high-frequency amplifiers 1 connected to its input side, and has a single output terminal 10 on its output side. The high-frequency amplifier 1A configured in this manner can also achieve the same effects as described above. The matching circuit 9 is denoted as MN in FIG. 13.
[0043] As described above, the high-frequency amplifier 1 according to the first embodiment includes transistor groups 2 and 4, mutually different matching circuits 3A and 3C connected to the transistor groups 2 and 4, short-circuited stubs 5A and 5B connected to the matching circuits 3A and 3C and shorted at the harmonic frequencies of the signal, and mutually different matching circuits 3B and 3D each having one end connected to a connection point between the matching circuits 3A and 3C and the short-circuited stubs 5A and 5B and the other end connected at a combining point S. The matching circuits 3B and 3D have circuit constants set such that an impedance Z4 looking into the transistor group 2 from the combining point S and an impedance Z5 looking into the transistor group 4 from the combining point S are equal in the operating frequency band. This reduces intermodulation distortion at the combining point S, allowing the high-frequency amplifier 1 to compensate for nonlinear distortion without control by an external device, thereby achieving low distortion and high efficiency due to the compensation for nonlinear distortion. Because compensation control by an external device, including a dedicated external control circuit, is not required, the high-frequency amplifier 1 can be made small in volume, lightweight, and low in power consumption.
[0044] In the high-frequency amplifier 1 according to the first embodiment, the short-circuited stub 5A includes a matching circuit 3A1, one end of which is connected to the matching circuit 3A, and a capacitor 8, one end of which is connected to the matching circuit 3A1 and the other end of which is grounded, for cutting DC components contained in the signal. This reduces the circuit loss caused by the dedicated DC-cut circuit in the high-frequency amplifier 1, enabling high efficiency. Furthermore, power can be supplied to the transistor groups 2 and 4 from between the matching circuit 3A1 and the capacitor 8.
[0045] Second Embodiment Fig. 14 is a block diagram showing the configuration of a high-frequency amplifier 1A according to a second embodiment. In Fig. 14, a high-frequency amplifier 1B is an amplifier having a function of compensating for nonlinear distortion, and includes N high-frequency amplifiers 1 and a matching circuit 11. The matching circuit 11 has one end connected to the output terminals of the N high-frequency amplifiers 1 arranged in parallel, and the other end having a single output terminal 12, where N is an integer equal to or greater than 2. The matching circuit 11 is represented as MN in Fig. 14.
[0046] In one high-frequency amplifier 1, transistor group 2 has a wiring that connects it to matching circuit 3A, and transistor group 4 has a wiring that connects its output terminal to matching circuit 3C. That is, transistor groups 2 and 4 each have two wirings. If one transistor group is configured with a large number of transistors to further increase the output power of high-frequency amplifier 1, a large number of transistors will be connected to the wiring that connects it to the input-side matching circuit. When the number of transistors becomes so large, the combined impedance becomes too small, and it may become impossible to create a matching circuit.
[0047] Therefore, in the high-frequency amplifier 1B, N high-frequency amplifiers 1 are arranged in parallel and each connected to a matching circuit 11. This allows the number of wires for connecting the N high-frequency amplifiers 1 to the matching circuit 11 to be increased to N without increasing the number of transistors in each high-frequency amplifier 1. As a result, the matching circuit 11 is capable of harmonic matching of signals from the N high-frequency amplifiers 1.
[0048] Furthermore, when transistor groups 2 and 4 are configured with a large number of transistors, parasitic oscillation may occur, particularly in the high frequency range. In contrast, high frequency amplifier 1B prevents parasitic oscillation by connecting N high frequency amplifiers 1 to a single matching circuit 11 without increasing the number of transistors in each high frequency amplifier 1.
[0049] Various lossless filter technologies are known for broadening the bandwidth of matching circuits. For example, a typical lossless filter used in a wideband amplifier has a matching circuit configured with transmission lines connected in multiple stages. In the high-frequency amplifier 1B, a single high-frequency amplifier 1 has a first-stage matching circuit, and matching circuit 11 is the second-stage matching circuit. Therefore, it is possible to configure a lossless filter using these matching circuits and achieve a wide bandwidth.
[0050] As described above, high-frequency amplifier 1B according to the second embodiment includes N high-frequency amplifiers 1 and matching circuit 11, one end of which is connected to the output terminals of the N parallel-arranged high-frequency amplifiers 1, and the other end of which has single output terminal 12. Because amplification processing can be performed by N high-frequency amplifiers 1, high-frequency amplifier 1B can achieve higher output, higher frequency, and wider bandwidth in addition to the effects described in the first embodiment.
[0051] It is possible to combine the embodiments, modify any of the components of the embodiments, or omit any of the components of the embodiments.
[0052] A high-frequency amplifier according to the present disclosure can be used as a communications amplifier for, for example, an artificial satellite.
[0053] 1, 1A, 1B: high frequency amplifier; 2: transistor group; 3A, 3A1, 3B, 3C, 3D; 9, 11: matching circuit; 4: transistor group; 5A, 5B: tip short-circuit stub; 6, 7: transistor; 8: capacitor; 10, 12: output terminal.
Claims
1. A high-frequency amplifier comprising: a plurality of transistor groups, each composed of one or a plurality of transistors, to which in-phase signals are distributed; a plurality of different input matching circuits connected respectively to the plurality of transistor groups; a plurality of parallel circuits connected respectively to the plurality of input matching circuits, and shorted at harmonic frequencies of signals output from the plurality of transistor groups; and a plurality of different output matching circuits, one end of which is connected to a connection point between the input matching circuit and the parallel circuit and the other ends of which are connected to each other, wherein the plurality of output matching circuits have circuit constants set so that the impedance looking into the transistor groups from the connection point between the output matching circuits is equal in the operating frequency band.
2. The high frequency amplifier according to claim 1, wherein the parallel circuit is a plurality of short-circuited stubs that are shorted at the double wave of the signal.
3. The high-frequency amplifier according to claim 2, characterized in that the short-circuited stub comprises: an in-stub matching circuit, one end of which is connected to the input-side matching circuit; and a capacitor, one end of which is connected to the other end of the in-stub matching circuit and the other end of which is grounded, for cutting out DC components contained in the signal.
4. A high frequency amplifier comprising: a plurality of high frequency amplifiers according to any one of claims 1 to 3; and a matching circuit having one end connected to the output ends of the plurality of high frequency amplifiers arranged in parallel, and having a single output terminal at the other end.
Citation Information
Patent Citations
Power amplifier
JP1988153904A
High frequency amplifier and radio transmitter using the same
JP2002368553A
High frequency amplifier
JP2006311300A
High frequency amplifier
JP2011030069A
