A linear enhancement GaN Doherty power amplifier

By using GaN Doherty power amplifiers with non-uniform power distribution and independent bias, the efficiency and linearity issues of traditional Doherty power amplifiers under peak-to-average power ratio (PAPR) signals are solved, achieving a wider power back-off range and higher linearity.

CN123122472APending Publication Date: 2026-11-17UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611324692.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-28
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Traditional Doherty power amplifiers have limited efficiency fallback range under peak-to-average power ratio (PAPR) signals and lack effective compensation for the nonlinear characteristics of GaN HEMT devices, resulting in reduced linearity and limited operating bandwidth.

Method used

By employing a non-uniform power distribution network and independently gate-biased GaN power transistor branches, combined with carrier and peak output matching units, dynamic load modulation is achieved by adjusting impedance and phase transmission characteristics to compensate for signal phase shift.

Benefits of technology

The power back-off range has been expanded, improving the linearity and efficiency of the power amplifier and adapting to the requirements of 5G peak-to-average power ratio signals.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123122472A_ABST
    Figure CN123122472A_ABST
Patent Text Reader

Abstract

This invention discloses a linearly enhanced GaN Doherty power amplifier, relating to the field of radio frequency power amplification technology. It solves the technical problem of reduced linearity in existing GaN Doherty power amplifiers caused by the interaction between active load modulation and the nonlinear characteristics of GaN power transistors under different power states. The device includes an input power distribution network, a carrier power amplification branch, a peak power amplification branch, and an output combining and matching network. The input power distribution network performs non-uniform power distribution based on the power capacity difference between the two power amplification branches. The output combining and matching network adjusts the impedance transformation characteristics and phase transmission characteristics through carrier output matching units and peak output matching units, forming a reverse phase response based on the admittance change during active load modulation. This invention is used to reduce nonlinear distortion during power amplification and improve the linearity of the power amplifier.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of radio frequency communication technology, and in particular to a linearly enhanced GaN Doherty power amplifier. Background Technology

[0002] Radio frequency (RF) power amplifiers are core modules in wireless transmission systems, responsible for amplifying modulated RF signals to sufficient power before sending them to the antenna for radiation. As 5G communication systems evolve towards higher spectral efficiency, wider bandwidth, and multiple antennas, base station RF front-ends place comprehensive demands on power amplifiers, requiring high efficiency, high linearity, wide dynamic range, and broadband performance. GaN HEMT devices, with their high breakdown voltage, high power density, and high efficiency, are well-suited for use in 5G base station RF power amplifiers.

[0003] In practical communication signals, OFDM, multi-carrier, and high-order modulated signals typically have a high peak-to-average power ratio (PAPR). Power amplifiers cannot operate at saturation for extended periods, as this can easily lead to clipping and nonlinear distortion. Therefore, they need to maintain high efficiency over a large output power back-off region. However, simply implementing power back-off would significantly reduce the efficiency of traditional power amplifiers. Doherty power amplifiers, through active load modulation of the carrier power amplifier and the peak power amplifier, increase the equivalent load impedance of the carrier power amplifier in the back-off region, representing a crucial technical approach for high-efficiency power amplifiers in current base stations.

[0004] A traditional Doherty power amplifier typically consists of a power divider, a carrier amplifier, a peak amplifier, a quarter-wavelength impedance transformation line, and a power combining point. Figure 1 As shown. At low input power, only the carrier power amplifier works. As the input power increases, the peak power amplifier gradually turns on and injects current into the combining node, which modulates the load impedance of the carrier power amplifier, thereby improving the efficiency of the power back-off region.

[0005] In the process of realizing this invention, the inventors discovered at least the following problems in the prior art: 1. The high-efficiency backoff point of the traditional symmetric Doherty architecture is usually limited to about 6 dB, which is difficult to cover the output power backoff range of about 9 dB or even greater required by the peak average signal of 5G. 2. Traditional output synthesis matching networks typically only perform impedance transformation and power synthesis, lacking the ability to actively compensate for the inherent AM-AM and AM-PM nonlinearities of carrier GaNHEMT devices. Partial linearization schemes suppress AM-PM distortion through two-way phase mismatch, but phase mismatch will disrupt the two-way power synthesis conditions, leading to a decrease in saturated output power and back-off efficiency. 3. Traditional quarter-wavelength impedance transformation structures are frequency-sensitive, and when the impedance transformation ratio is too large, the equivalent Q value of the matching network increases, limiting the operating bandwidth. The carrier branch is prone to premature gain compression under large back-off and asymmetric bias conditions, thus deteriorating the overall AM-AM linearity. Summary of the Invention

[0006] The purpose of this invention is to provide a linearly enhanced GaN Doherty power amplifier to solve the technical problem of reduced linearity during power amplification caused by the mismatch between active load modulation and the nonlinear characteristics of the power transistor in existing Doherty power amplifiers under different output power states. The various technical effects of the preferred solutions among the many technical solutions provided by this invention are detailed below.

[0007] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a linear enhancement GaN Doherty power amplifier, including an input power distribution network for receiving an input radio frequency signal and distributing the input radio frequency signal in a non-uniform manner according to the power capacity difference between the carrier power amplification branch and the peak power amplification branch, and outputting a carrier input signal and a peak input signal; The carrier power amplification branch has a first GaN power transistor inside, which is used to amplify the power of the carrier input signal and output the carrier power amplification signal. The peak power amplification branch has a second GaN power transistor inside, which is used to amplify the peak input signal and output the peak power amplification signal. An output synthesis matching network is used to perform impedance transformation and power synthesis processing on the carrier power amplified signal and the peak power amplified signal, and output an RF output signal. The peak power amplification branch and the carrier power amplification branch are configured with different power capacities so that the peak power amplification branch and the carrier power amplification branch participate in power output in different output power ranges; The output synthesis matching network includes a carrier output matching unit and a peak output matching unit. The carrier output matching unit and the peak output matching unit are respectively connected to the carrier power amplification branch and the peak power amplification branch, and are used to perform impedance transformation on the carrier power amplification signal and the peak power amplification signal, and jointly adjust the load impedance in the power synthesis process. The carrier output matching unit and the peak output matching unit adjust the impedance transformation characteristics and phase transmission characteristics so that the output synthesis matching network forms a reverse phase response based on the admittance change of the power synthesis point during active load modulation, thereby compensating for the signal phase offset of the first GaN power transistor and the second GaN power transistor.

[0008] Preferably, the carrier power amplification branch and the peak power amplification branch employ independent gate bias and drain bias to adjust the DC operating state of the first GaN power transistor and the second GaN power transistor, respectively.

[0009] Preferably, the carrier power amplifier branch is set to a Class-AB or Class-B conduction state so that the carrier power amplifier branch remains conduction in the power back-off region or when the input power is lower than a set threshold; The peak power amplification branch is set to a Class-C bias state so that the peak power amplification branch is turned off when the input power is lower than a set threshold, and turned on after the input power rises to the set threshold, injecting current into the power combining point of the output combining matching network to change the equivalent load impedance of the carrier power amplification branch at the power combining point, thereby making the carrier power amplification branch correspond to different load impedances in the off state and the on state of the peak power amplification branch.

[0010] Preferably, the range of the power back-off region of the linearly enhanced GaN Doherty power amplifier is expressed as follows: OBO(dB) = 10lg[(1+α)(1+αβ)] Where OBO(dB) represents the range of the power back-off region, α is the ratio of the saturated fundamental current of the peak power amplification branch to that of the carrier power amplification branch, and β is the ratio of the drain bias voltage of the peak power amplification branch to that of the carrier power amplification branch.

[0011] Preferably, the carrier output matching unit is disposed between the drain output terminal of the first GaN power transistor and the power combining point, and is used to match the carrier power amplification branch to the optimal load admittance at the power saturation point and form a first target electrical length at the center frequency, so as to adjust the amplitude response and phase response of the carrier power amplification branch.

[0012] Preferably, the peak output matching unit is disposed between the drain output terminal of the second GaN power transistor and the power combining point, and is used to adjust the equivalent output admittance presented by the peak power amplification branch to the power combining point in the power back-off region, and to adjust the current phase injected into the power combining point after the peak power amplification branch is turned on.

[0013] Preferably, the peak output matching unit forms a second target electrical length at the center frequency. The second target electrical length is used to adjust the equivalent output admittance of the peak power amplification branch and the current phase injected into the power combining point, so that the output combining matching network forms a phase response opposite to the phase shift direction of the first GaN power transistor and the second GaN power transistor as the input power changes.

[0014] Preferably, the input power distribution network performs a 1:2 power distribution on the input radio frequency signal to obtain a one-part input power signal and a two-part input power signal; The input power signal is divided into two parts, and the input power signal is divided into two parts. .... The input power signal is divided into two parts. The input

[0015] Preferably, the output synthesis matching network forms an amplitude compensation amount based on the change in the transmission coefficient from the power back-off point to the power saturation point, so as to mitigate the gain compression of the first GaN power transistor as the input power increases.

[0016] Preferably, the normalized load admittance of the carrier power amplifier branch at the power back-off point is expressed as: Y C,OBO =g OBO (1+jb OBO )Y CO Among them, Y C,OBO g represents the load admittance of the carrier power amplifier branch at the power back-off point. OBO and b OBO Let these represent the normalized real part and the normalized imaginary part of the load admittance, respectively. The output synthesis matching network adjusts the transmission coefficient change from the power backoff point to the power saturation point based on the changes in the normalized real part and the normalized imaginary part, thereby adjusting the amplitude compensation amount.

[0017] Implementing one of the above-described technical solutions of the present invention has the following advantages or beneficial effects: 1. This invention uses an input power distribution network to perform non-proportional power distribution on the input RF signal based on the power capacity difference between the carrier power amplification branch and the peak power amplification branch. By configuring different power capacities, the two power amplification branches can participate in power output in different output power ranges. This solves the problem of insufficient adaptability of the power configuration of the two power amplification branches to power change states in traditional Doherty power amplifiers, and achieves the effect of coordinated power amplification in different power ranges. 2. This invention uses a carrier output matching unit and a peak output matching unit to jointly adjust the load impedance of the carrier power amplification branch and the peak power amplification branch, which solves the problem of power amplification characteristic change caused by power change in the power combining point admittance during active load modulation, and achieves the effect of dynamic adjustment of load impedance during power combining. 3. This invention adjusts the impedance transformation characteristics and phase transmission characteristics by using a carrier output matching unit and a peak output matching unit. This enables the output synthesis matching network to form a phase response opposite to the phase shift direction of the GaN power transistor signal based on the admittance change during the active load modulation process. This solves the problem of power amplifier signal phase shift caused by the nonlinear amplitude and phase characteristics of GaN power transistors, and achieves the effect of compensating for signal phase shift and improving the linearity of the power amplifier. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1 This is a schematic diagram of the structure of a traditional Doherty power amplifier; Figure 2 This is a schematic diagram of the structure of a linear enhancement GaN Doherty power amplifier according to an embodiment of the present invention; Figure 3 These are simulation results of AM-AM characteristics with a 1:2 power allocation according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the output synthesis matching network according to an embodiment of the present invention; Figure 5 This is a graph showing the relationship between the drain efficiency and output power back-off of the Doherty power amplifier according to an embodiment of the present invention. Figure 6 The Doherty power amplifier in this embodiment of the invention is in AM-AM OCN,S Simulation results under 0dB conditions; Among them, (a) is the AM-PM simulation result diagram, and (b) is the AM-AM simulation result diagram; Figure 7 The Doherty power amplifier in this embodiment of the invention is in AM-AM OCN,S Simulation results under the condition of 0.5dB; Among them, (a) is the AM-PM simulation result diagram, and (b) is the AM-AM simulation result diagram; Figure 8The Doherty power amplifier in this embodiment of the invention is in AM-AM OCN,S = Simulation results under 1dB condition; Among them, (a) is the AM-PM simulation result diagram, and (b) is the AM-AM simulation result diagram; Figure 9 These are the simulation results of the AM-PM characteristics of the Doherty power amplifier in an embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the present invention clearer, various exemplary embodiments described below will be referenced to the accompanying drawings, which form part of the exemplary embodiments, illustrating various exemplary embodiments that may be used to implement the present invention. Unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. It should be understood that they are merely examples of processes, methods, and apparatuses consistent with some aspects of the present invention disclosed as detailed in the appended claims, and other embodiments may be used, or structural and functional modifications may be made to the embodiments listed herein without departing from the scope and spirit of the present invention.

[0020] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," etc., indicate the orientation or positional relationship based on the accompanying drawings, and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the referred element must have a specific orientation, or be constructed and operated in a specific orientation. The terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. The term "a plurality of" means two or more. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0021] To illustrate the technical solution described in this invention, specific embodiments are described below, showing only the parts related to the embodiments of this invention.

[0022] Example 1: like Figure 2 As shown, the present invention provides a linear enhancement GaN Doherty power amplifier, comprising: An input power distribution network is used to receive input radio frequency signals and, based on the power capacity difference between the carrier power amplification branch and the peak power amplification branch, perform non-uniform power distribution on the input radio frequency signals, and output carrier input signals and peak input signals. The carrier power amplification branch has a first GaN power transistor inside, which is used to amplify the power of the carrier input signal and output the carrier power amplification signal. The peak power amplification branch has a second GaN power transistor inside, which is used to amplify the peak input signal and output the peak power amplification signal. An output synthesis matching network is used to perform impedance transformation and power synthesis processing on the carrier power amplified signal and the peak power amplified signal, and output an RF output signal. The peak power amplification branch and the carrier power amplification branch are configured with different power capacities so that the peak power amplification branch and the carrier power amplification branch participate in power output in different output power ranges; The output synthesis matching network includes a carrier output matching unit and a peak output matching unit. The carrier output matching unit and the peak output matching unit are respectively connected to the carrier power amplification branch and the peak power amplification branch, and are used to perform impedance transformation on the carrier power amplification signal and the peak power amplification signal, and jointly adjust the load impedance in the power synthesis process. The carrier output matching unit and the peak output matching unit adjust the impedance transformation characteristics and phase transmission characteristics so that the output synthesis matching network forms a reverse phase response based on the admittance change of the power synthesis point during active load modulation, thereby compensating for the signal phase offset of the first GaN power transistor and the second GaN power transistor.

[0023] In the description of the embodiments of this application, the power combining point refers to a specific node or equivalent reference surface in the linear enhancement GaN Doherty power amplifier where the carrier power amplified signal and the peak power amplified signal undergo energy superposition and impedance transformation in the output combining matching network. The key feature of the power combining point is that its admittance state dynamically changes with the operating state of the two transistors, i.e., the conduction degree of the carrier power amplification branch and the peak power amplification branch, so that the output combining matching network can generate a corresponding reverse phase response in response to the admittance change, thereby compensating for the signal phase shift of each branch transistor. However, it does not necessarily mean that the power combining point has the same impedance characteristics at the fundamental frequency and each harmonic of the operating frequency, nor does it specifically indicate that there is a strict linear correspondence between the admittance change and the phase shift compensation at this point.

[0024] In this embodiment, the impedance transformation characteristic is used to characterize the ability of the carrier output matching unit and the peak output matching unit to transform the impedance or admittance of the power synthesis point admittance change. The change of the impedance transformation characteristic can change the transmission coefficient amplitude of the output synthesis matching network from the power back-off point to the power saturation point, thereby forming an amplitude compensation response, which is used to mitigate the gain compression of GaN power transistors as the input power increases, corresponding to the compensation for AM-AM distortion.

[0025] The phase transmission characteristic is used to characterize the phase transmission and phase adjustment characteristics of the carrier output matching unit and the peak output matching unit for the power amplified signal. The change in the phase transmission characteristic enables the output synthesis matching network to generate a phase response opposite to the phase shift direction of the GaN power transistor as the input power changes, thereby compensating for AM-PM distortion.

[0026] Therefore, the impedance transformation characteristic and the phase transmission characteristic correspond to the amplitude response adjustment and phase response adjustment of the output synthesis matching network, respectively, and together achieve compensation for the nonlinear amplitude and phase distortion of the power amplifier.

[0027] As an optional implementation, the carrier power amplification branch and the peak power amplification branch adopt independent gate bias and drain bias to adjust the DC operating state of the first GaN power transistor and the second GaN power transistor respectively.

[0028] In terms of bias control, as an optional implementation, the carrier power amplifier branch is set to a Class-AB or Class-B conduction state so that the carrier power amplifier branch remains conduction in the power back-off region or when the input power is lower than a set threshold. The peak power amplification branch is set to a Class-C bias state so that the peak power amplification branch is turned off when the input power is lower than a set threshold, and turned on after the input power rises to the set threshold, injecting current into the power combining point of the output combining matching network to change the equivalent load impedance of the carrier power amplification branch at the power combining point, thereby making the carrier power amplification branch correspond to different load impedances in the off state and the on state of the peak power amplification branch.

[0029] In the description of the embodiments of this application, the power back-off region refers to a specific input power range where the input signal power is lower than a preset peak amplification branch conduction threshold, so that the peak power amplification branch is in a cut-off state and only the carrier power amplification branch remains on and outputs power.

[0030] Specifically, the input power being below a set threshold is the conduction condition for the peak power amplification branch, and the power back-off region is the power operating range of the entire power amplifier.

[0031] In a preferred embodiment, the first GaN power transistor can be a CGH40010F, and the second GaN power transistor can be a CGH40025F. The bias voltage of the first GaN power transistor can be set to 20 V, and the voltage of the second GaN power transistor can be set to 32 V, so that the ratio β of the drain bias voltage of the peak power amplification branch to the carrier power amplification branch is approximately 1.6. The gate bias of the carrier power amplifier can be set to Class-AB, and the gate bias of the peak power amplifier can be set to Class-C, to form an asymmetric Doherty operating mode. The ratio α of the saturation fundamental current of the peak power amplification branch to the carrier power amplification branch can be set to approximately 1.4. Through the asymmetric carrier power amplification branch and peak power amplification branch configuration, the high-efficiency region of the Doherty power amplifier is extended from the traditional approximately 6 dB backoff to approximately 9 dB backoff, which is more suitable for 5G peak-to-average power ratio signals. Simulation results are as follows: Figure 3 As shown, the amplifier maintains high efficiency from an output power of 45dBm down to 36dBm.

[0032] Specifically, let the saturation fundamental currents of the carrier power amplifier branch and the peak power amplifier branch be I... cs and I ps Both satisfy I ps =αI cs The drain bias voltages of the carrier power amplifier branch and the peak power amplifier branch are V, respectively. dc and V dp Both satisfy V dp =βV dc .

[0033] The saturation power of the carrier power amplifier branch can be expressed as P cs =V dc ·I cs / 2, the saturation power of the peak power amplification branch can be expressed as P ps =V dp ·I ps / 2=αβP cs Based on the asymmetric Doherty active load modulation relationship, the output power back-off range can be expressed as: OBO(dB) = 10lg[(1+α)(1+αβ)] Wherein, OBO(dB) represents the range of the power back-off region. When α is 1.4 and β is 1.6, the value of OBO(dB) is approximately 8.9 dB, which can approximately achieve a 9 dB output power back-off range.

[0034] Specifically, α and β are coupled: when the target power back-off region is determined, increasing β reduces the α required to achieve the same back-off range, thereby decreasing the impedance conversion ratio between the power back-off point and the power saturation point, which is beneficial for broadband matching; however, β should not be too large, otherwise the drain voltage of the carrier power amplifier will be too low, which can easily lead to premature gain compression of the carrier branch and deteriorate AM-AM linearity. Therefore, in this embodiment, α is preferably 1.4 and β is 1.6, which is a trade-off between the 9dB power back-off range, the impedance conversion ratio, and the linearity of the carrier branch.

[0035] The power back-off point refers to the system operating point corresponding to the critical input power at which the peak power amplification branch just enters the conduction state from the cutoff state during the process of increasing the input power from a small value. This operating point is also the upper boundary point of the power back-off region, and its output power value is equal to the power value corresponding to the preset amount of saturated output power back-off.

[0036] The power saturation point refers to the critical operating point where the peak power amplification branch is fully turned on and the output power no longer increases significantly with the increase of input power. This operating point corresponds to the maximum output power capability of the power amplifier. Further increases in input power thereafter will cause the amplifier to enter a deep compression region, and the output power will approach a constant value.

[0037] As an optional implementation, the Output Matching Network–Carrier (OMNC) is disposed between the drain output terminal of the first GaN power transistor and the power combining point, and is used to match the carrier power amplification branch to the optimal load admittance at the power saturation point and form a first target electrical length at the center frequency, for adjusting the amplitude response and phase response of the carrier power amplification branch.

[0038] As an optional implementation, the Output Matching Network-Peaking (OMNP) is disposed between the drain output terminal of the second GaN power transistor and the power combining point, and is used to adjust the equivalent output admittance presented by the peak power amplification branch to the power combining point in the power back-off region, and to adjust the current phase injected into the power combining point after the peak power amplification branch is turned on.

[0039] The peak output matching unit forms a second target electrical length at the center frequency. The second target electrical length is used to adjust the equivalent output admittance of the peak power amplification branch and the current phase injected into the power combining point, so that the output combining matching network forms a phase response opposite to the phase shift direction of the first GaN power transistor and the second GaN power transistor as the input power changes.

[0040] In addition, α and β will also interact with the first electric length θ C Second electric length θ P Indirect coupling is formed, and α determines the equivalent load admittance Y at the power combining point. L =(1+α)Y CO And determine the characteristic admittance αY of OMNP. CO Therefore, after α is changed, the backoff point load modulation trajectory, the required electrical lengths for OMNC and OMNP, and the AM-AM / AM-PM compensation all need to be re-optimized. The first electrical length θ C Second electric length θ P These are not isolated constants, but rather output synthetic matching network parameters obtained by combining α, β, the target power backoff range, and the target linear compensation amount.

[0041] The embodiment is merely a specific example and does not indicate that this is the only way to implement the present invention.

[0042] Example 2: The difference between this second embodiment and the first embodiment is that the second embodiment uses a 1:2 unequal power distribution network to distribute the input radio frequency signal in a 1:2 ratio, resulting in a one-part input power signal and a two-part input power signal. The input power signal is divided into two parts, and the input power signal is divided into two parts. .... The input power signal is divided into two parts. The input

[0043] The reason for choosing a 1:2 power split is that the carrier power amplifier branch uses a lower drain bias and bears the main output in the power back-off region. If a 1:1 power divider is used, the input drive of the carrier branch is relatively large, which may cause it to enter the gain compression region prematurely before the peak power amplifier branch is fully turned on, resulting in increased AM-AM distortion. With a 1:2 split, the carrier power amplifier branch receives about 1 / 3 of the input power, and the peak branch receives about 2 / 3 of the input power. Compared with a 1:1 split, the input drive of the carrier power amplifier branch is reduced by about 1.8~2 dB, and the peak power amplifier branch receives more sufficient drive, which is conducive to its timely turn-on in the high-power region and participation in load modulation.

[0044] Using a 1:1 power allocation and retaining only OMNC and OMNP, the output synthesized matching network can still generate some AM-PM compensation, but the carrier power amplification branch is more easily compressed due to the large input power, resulting in insufficient AM-AM improvement; therefore, the 1:2 power allocation and OMNC / OMNP are complementary.

[0045] By reducing the input drive of the carrier branch through a 1:2 unequal power allocation, premature gain compression of the carrier power amplifier is suppressed, further reducing AM-AM distortion. The results are as follows: Figure 4 As shown, the AM-AM distortion of the power amplifier is within 1.5dB across the entire operating frequency band, exhibiting excellent linearity performance.

[0046] Example 3: The difference between this third embodiment and embodiments one and two is that, because the load admittance of the carrier power amplifier changes dynamically with the active load modulation process, the transmission coefficient of the output synthetic matching network will exhibit amplitude and phase changes with the input power. For example... Figure 5 As shown, in this embodiment, the controllable nonlinearity of the output synthesis matching network is used to reverse compensate for the inherent AM-AM distortion and AM-PM distortion of the GaN power transistor, thereby reducing the overall AM-AM and AM-PM distortion of the power amplifier.

[0047] exist Figure 5 Middle, Y CO Y is the characteristic admittance of the carrier power amplifier branch. c V is the output admittance of the carrier power amplifier branch. c Y is the equivalent output voltage of the carrier power amplifier branch. c1 V is the admittance at the output side of the carrier output matching unit, i.e., the admittance from the output of the carrier output matching unit to the power combining point. Ic is the current at the output side of the carrier output matching unit. p Y is the equivalent output voltage of the peak power amplifier branch. p Y is the output admittance of the peak power amplifier branch. p1 V is the admittance at the output side of the peak output matching unit. L This represents the load voltage at the power combining point.

[0048] In a preferred embodiment, the characteristic admittance of the OMNC can be set to Y. CO First electric length θ C The characteristic admittance of OMNP can be set to 71.7°, and can be set to αY. CO The second electric length θ P It can be set to 155°. (Using θ) C and θ PThe combined configuration allows the output synthetic matching network to generate a compressed phase response that is opposite to the inherent phase spread trend of the carrier power amplifier branch, thereby compensating for AM-PM distortion in the carrier power amplifier branch. Simultaneously, this compensation method does not rely on phase mismatch between the two outputs, avoiding the decrease in saturated output power and back-off efficiency caused by phase mismatch.

[0049] θ C and θ P The determination process may include the following steps: First, determine α, β, and the combined node load admittance based on the target power back-off region; Secondly, the desired output amplitude compensation amount AM-AM for the synthetic matching network is set. OCN,S and scan θ P ; The real part g of the back-off point normalized admittance is solved using the transmission line ABCD matrix and the output synthesized matched network transmission coefficient GT. OBO and the imaginary part b OBO ; Solve for θ based on the reflection coefficient relationship. C The AM-AM and AM-PM compensation effects were verified through large-signal simulation.

[0050] In this embodiment, α is set to 1.4, and AM-AM OCN,S Candidate values ​​such as 0 dB, 0.5 dB, and 1 dB, and θ P Scanning is performed within a certain range. For example... Figure 6 , Figure 7 and Figure 8 As shown, simulation results indicate that AM-AM OCN,S A value of approximately 0.5 dB strikes a good balance between main transistor gain compression compensation and overall DPA linearity, corresponding to θ. C 71.7°, θ P The value is -25°. Since it's inconvenient to achieve a negative charge length in practical microstrip lines, θ can be... P The 180° increase is equivalent to 155°, and the two paths are kept in phase at the power junction point through the phase compensation line.

[0051] in, Figure 6 , Figure 7 and Figure 8 The DPA mentioned in the text refers to the Doherty Power Amplifier, which is the Doherty power amplifier described in this invention, and the main transistor refers to the first GaN power transistor.

[0052] In particular, θ C With θ P There are still constraints. θ PThis will affect the output admittance Y of the peak power amplifier branch in the power back-off region. outp This, in turn, changes the admittance Y at the power junction. C1,OBO and carrier-side load admittance Y C,OBO ;θ C This determines the conversion relationship of the carrier power amplifier branch output synthesis matching network to the dynamic admittance. Both need to be solved jointly under given α, target AM-AM compensation, and target AM-PM compensation.

[0053] For the embodiment with 1.7~2.1 GHz and α≈1.4, θ C The preferred range is approximately 68° to 74°, θ P The preferred equivalent range can be approximately 149° to 159°; under different devices, frequency bands, or target compensation amounts, θ C Scalable design at approximately 60°~85°, θ P The scalable design is within the range of approximately 120° to 180°, and is further optimized and determined through electromagnetic simulation and large-signal simulation.

[0054] The output synthesis matching network generates an amplitude compensation amount based on the change in the transmission coefficient from the power back-off point to the power saturation point, so as to mitigate the gain compression of the first GaN power transistor as the input power increases.

[0055] The AM-AM compensation of the output synthetic matching network is mainly determined by the ratio α of the real part, imaginary part, and asymmetric current of the normalized admittance at the back-off point. AM-PM compensation can be further improved by adjusting the peak branch output network electrical length θ. P The adjustment enables the present invention to achieve coordinated compensation of amplitude distortion and phase distortion at the circuit level.

[0056] The transmission coefficient of the output synthetic matching network can be defined as: GT=V L / I C , where V L I is the voltage at the power junction. C This represents the current in the carrier power amplifier branch.

[0057] The normalized load admittance of the carrier power amplifier branch at the power back-off point is expressed as follows: Y C,OBO =g OBO (1+jb OBO )Y CO Among them, Y C,OBO g represents the load admittance of the carrier power amplifier branch at the power back-off point. OBO and bOBOLet represent the normalized real part and the normalized imaginary part of the load admittance, respectively. The output synthetic matching network adjusts the transmission coefficient change from the power back-off point to the power saturation point based on the changes in the normalized real part and the normalized imaginary part, thereby adjusting the amplitude compensation amount. Then, the gain change of the output synthetic matching network from the power back-off point to the power saturation point can be expressed as: ΔGT=G T,SAT / G T,OBO Wherein, ΔGT is the gain change of the output synthetic matching network from the power back-off point to the power saturation point, and ΔGT corresponds to the amplitude compensation introduced by the output synthetic matching network.

[0058] The AM-AM compensation amount can be expressed as: AM-AM OCN,S =10lg(|ΔGT|²) In admittance form, through the derivation of the transmission coefficients at the power back-off point and the power saturation point, we can obtain |ΔGT|² and g OBO b OBO Related to α, that is: |ΔGT|²=g OBO (1+b OBO ²)(1+α) Therefore, the asymmetric current ratio α not only determines the power back-off range, but also affects the amplitude compensation capability that the output synthesis matching network can generate.

[0059] Its physical function is as follows: GaN power transistors usually experience gain compression in the large signal region, resulting in insufficient increase in output amplitude as the input power increases. This invention sets the dynamic admittance transition trajectory of OMNC and OMNP so that the output synthesis matching network generates an appropriate amount of gain expansion or reduces gain compression during the process from the power back-off point to the power saturation point, thereby offsetting the inherent AM-AM distortion of the carrier device.

[0060] By having a designable θ C and θ P The output synthesis matching network is configured to generate controllable AM-AM and AM-PM responses, compensating for the inherent nonlinearity of the carrier power amplification branch and thus improving overall linearity. Simulation results are as follows: Figure 9 As shown, the AM-PM distortion of the power amplifier is within 6.3° throughout the entire operating frequency band, exhibiting excellent linearity performance.

[0061] The above description is merely a preferred embodiment of the present invention. Those skilled in the art will understand that various changes or equivalent substitutions can be made to these features and embodiments without departing from the spirit and scope of the present invention. Furthermore, under the teachings of the present invention, these features and embodiments can be modified to adapt to specific situations and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the protection scope of the present invention.

Claims

1. A linear enhancement GaN Doherty power amplifier, characterized in that, include: An input power distribution network is used to receive input radio frequency signals and, based on the power capacity difference between the carrier power amplification branch and the peak power amplification branch, perform non-uniform power distribution on the input radio frequency signals, and output carrier input signals and peak input signals. The carrier power amplification branch has a first GaN power transistor inside, which is used to amplify the power of the carrier input signal and output the carrier power amplification signal. The peak power amplification branch has a second GaN power transistor inside, which is used to amplify the peak input signal and output the peak power amplification signal. An output synthesis matching network is used to perform impedance transformation and power synthesis processing on the carrier power amplified signal and the peak power amplified signal, and output an RF output signal. The peak power amplification branch and the carrier power amplification branch are configured with different power capacities so that the peak power amplification branch and the carrier power amplification branch participate in power output in different output power ranges; The output synthesis matching network includes a carrier output matching unit and a peak output matching unit. The carrier output matching unit and the peak output matching unit are respectively connected to the carrier power amplification branch and the peak power amplification branch, and are used to perform impedance transformation on the carrier power amplification signal and the peak power amplification signal, and jointly adjust the load impedance in the power synthesis process. The carrier output matching unit and the peak output matching unit adjust the impedance transformation characteristics and phase transmission characteristics so that the output synthesis matching network forms a reverse phase response based on the admittance change of the power synthesis point during active load modulation, thereby compensating for the signal phase offset of the first GaN power transistor and the second GaN power transistor.

2. The linear enhancement GaN Doherty power amplifier according to claim 1, characterized in that, The carrier power amplification branch and the peak power amplification branch employ independent gate bias and drain bias to adjust the DC operating state of the first GaN power transistor and the second GaN power transistor, respectively.

3. The linear enhancement GaN Doherty power amplifier according to claim 2, characterized in that, The carrier power amplifier branch is set to Class-AB or Class-B conduction state so that the carrier power amplifier branch remains conduction in the power back-off region or when the input power is lower than a set threshold; The peak power amplification branch is set to a Class-C bias state so that the peak power amplification branch is turned off when the input power is lower than a set threshold, and turned on after the input power rises to the set threshold, injecting current into the power combining point of the output combining matching network to change the equivalent load impedance of the carrier power amplification branch at the power combining point, thereby making the carrier power amplification branch correspond to different load impedances in the off state and the on state of the peak power amplification branch.

4. The linear enhancement GaN Doherty power amplifier according to claim 1, characterized in that, The range of the power back-off region of the linearly enhanced GaN Doherty power amplifier is expressed as follows: OBO(dB) = 10lg[(1+α)(1+αβ)] Where OBO(dB) represents the range of the power back-off region, α is the ratio of the saturated fundamental current of the peak power amplification branch to that of the carrier power amplification branch, and β is the ratio of the drain bias voltage of the peak power amplification branch to that of the carrier power amplification branch.

5. The linear enhancement GaN Doherty power amplifier according to claim 1, characterized in that, The carrier output matching unit is disposed between the drain output terminal of the first GaN power transistor and the power combining point, and is used to match the carrier power amplification branch to the optimal load admittance at the power saturation point and form a first target electrical length at the center frequency, for adjusting the amplitude response and phase response of the carrier power amplification branch.

6. The linear enhancement GaN Doherty power amplifier according to claim 1, characterized in that, The peak output matching unit is disposed between the drain output terminal of the second GaN power transistor and the power combining point, and is used to adjust the equivalent output admittance presented by the peak power amplification branch to the power combining point in the power back-off region, and to adjust the current phase injected into the power combining point after the peak power amplification branch is turned on.

7. The linear enhancement GaN Doherty power amplifier according to claim 6, characterized in that, The peak output matching unit forms a second target electrical length at the center frequency. The second target electrical length is used to adjust the equivalent output admittance of the peak power amplification branch and the current phase injected into the power combining point, so that the output combining matching network forms a phase response opposite to the phase shift direction of the first GaN power transistor and the second GaN power transistor as the input power changes.

8. The linear enhancement GaN Doherty power amplifier according to claim 1, characterized in that, The input power distribution network performs a 1:2 power distribution on the input radio frequency signal to obtain a one-part input power signal and a two-part input power signal. The input power signal is divided into two parts, and the input power signal is divided into two parts. .... The input power signal is divided into two parts. The input 9. The linear enhancement GaN Doherty power amplifier according to claim 1, characterized in that, The output synthesis matching network generates an amplitude compensation amount based on the change in the transmission coefficient from the power back-off point to the power saturation point, so as to mitigate the gain compression of the first GaN power transistor as the input power increases.

10. The linearly enhanced GaN Doherty power amplifier according to claim 9, characterized in that, The normalized load admittance of the carrier power amplifier branch at the power back-off point is expressed as follows: Y C,OBO =g OBO (1+jb OBO )Y CO Among them, Y C,OBO g represents the load admittance of the carrier power amplifier branch at the power back-off point. OBO and b OBO Let these represent the normalized real part and the normalized imaginary part of the load admittance, respectively. The output synthesis matching network adjusts the transmission coefficient change from the power backoff point to the power saturation point based on the changes in the normalized real part and the normalized imaginary part, thereby adjusting the amplitude compensation amount.