A Doherty-like power amplifier with improved back-off efficiency

Abstract

The present invention discloses a Doherty-like power amplifier with improved back-off efficiency, which relates to an amplifier, including: an input power distribution network for isolating a driver amplifier unit (DR) stage and an adaptive bias circuit; a driver amplifier unit (DR) stage, whose output end is connected to the input end of a power output unit (PA) stage; an adaptive bias circuit for detecting changes in the envelope of the power amplifier input signal in real time based on the radio frequency signal transmitted by the input power distribution network, and outputting a voltage signal; a power output unit (PA) stage, which is provided with an input end for receiving a voltage signal, the voltage signal being used to control the bias tail pipe current of the power output unit (PA) stage; and an output matching balun network, whose input end is connected to the output end of the power output unit (PA) stage. The present invention achieves better back-off efficiency than a traditional Class AB power amplifier, and only requires one power output unit (PA) stage, without considering the load modulation effect, and can realize broadband applications.

Description

Technical Field

[0001] The present invention relates to an amplifier, and more particularly, to a Doherty-like power amplifier with improved back-off efficiency. Background Art

[0002] As the largest energy consumer in a transmission system, the power added efficiency (PAE) of the power amplifier (PA) is crucial, representing its ability to convert DC power into AC power. Inadequate amplifier efficiency negatively impacts overall system energy efficiency. Due to amplifier nonlinearity, PAs in high-order communications (such as 16QAM and 64QAM) typically operate within the saturation region at a back-off point of 6-8dBm. Operating at this back-off point significantly reduces the efficiency of traditional Class AB PAs, impacting overall system efficiency. To improve back-off efficiency, researchers have proposed a number of specialized PA structures, including the Doherty PA. The traditional Doherty PA architecture combines the outputs of two PAs. Active load modulation allows the two PAs to operate at different inputs, one biased in Class AB and the other in Class C. At low inputs, the main PA is enabled while the auxiliary PA is disabled. When the input increases, the auxiliary PA is enabled, maintaining high overall circuit efficiency. This allows the circuit to operate at high efficiency even at the back-off point.

[0003] However, the following problems exist when using traditional Doherty amplifiers:

[0004] 1. Two PAs are required, which also include bias adjustment, passive components and other structures, which will increase the chip area.

[0005] 2. In the millimeter wave frequency band, strong parasitic effects will affect the circuit function, resulting in limited output power and efficiency of the Doherty power amplifier operating in this frequency band, and the back-off efficiency curve is not obvious.

[0006] 3. Traditional Doherty PAs rely on quarter-wavelength lines and impedance transformation networks (such as Zopt→2Zopt transformation), which makes it difficult to achieve broadband matching and therefore limits bandwidth. Summary of the Invention

[0007] The technical problem to be solved by the present invention is to address the shortcomings of the existing technology and provide a Doherty-like power amplifier with improved back-off efficiency. By controlling the current of the bias tube of the power output unit PA stage through an adaptive bias circuit based on envelope tracking, the output current of the power output unit PA stage is increased to achieve a higher output power, thereby further improving efficiency, thereby achieving better back-off efficiency than traditional Class AB power amplifiers. In addition, the present invention only requires one power output unit PA stage, without considering load modulation effects, and can achieve broadband applications.

[0008] The present invention provides a Doherty-like power amplifier with improved fallback efficiency, comprising:

[0009] An input power distribution network outputs two RF signals, one of which is input to the driver amplifier unit DR stage, and the other is input to the adaptive bias circuit, so as to isolate the driver amplifier unit DR stage from the adaptive bias circuit;

[0010] The output terminal of the driving amplifier unit DR stage is connected to the input terminal of the power output unit PA stage;

[0011] an adaptive bias circuit for detecting changes in the envelope of the power amplifier input signal in real time based on the radio frequency signal transmitted from the input power distribution network, and outputting a voltage signal;

[0012] A power output unit PA stage, which is provided with an input terminal for receiving the voltage signal, wherein the voltage signal is used to control the bias tail pipe current of the power output unit PA stage to adjust the output power and conversion efficiency of the power output unit PA stage;

[0013] The output matches the balun network, whose input is connected to the output of the power output unit PA stage.

[0014] At high input signal power, an adaptive bias circuit based on envelope tracking controls the current of the power output unit PA-level bias tube, increasing the output current of the power output unit PA-level to achieve a higher output power. This further increases efficiency, thereby achieving better back-off efficiency than traditional Class AB power amplifiers. Furthermore, the present invention only requires one power output unit PA-level, eliminating the need to consider load modulation effects and enabling broadband applications. This invention has achieved good output power and back-off effects using a 140GHz broadband PA based on TSMC's 65nm process.

[0015] Preferably, the adaptive bias circuit includes a detector and a buffer stage; the input end of the detector inputs a radio frequency signal, the output end of the detector is connected to the input end of the buffer stage, and the input end of the buffer stage outputs the voltage signal.

[0016] Preferably, the detector includes a self-mixing differential pair consisting of a first transistor M1 and a second transistor M2, and a load network; the gates of the first transistor M1 and the second transistor M2 serve as input terminals, respectively, to input the INP signal and the INN signal of the RF signal; the sources of the first transistor M1 and the second transistor M2 are connected to each other and to a common terminal; the drains of the first transistor M1 and the second transistor M2 are connected to each other and to the load network, and the connection terminal of the load network and the drains of the first transistor M1 and the second transistor M2 are connected to the buffer stage as the output terminal of the detector.

[0017] Preferably, the load network is a resistor or a PMOS transistor.

[0018] Preferably, the output end of the detector is grounded via a first capacitor C1.

[0019] Preferably, the buffer stage includes a fourth transistor M4, a first resistor R1, and a filter network; the gate of the fourth transistor M4 is connected to the output end of the detector, the source of the fourth transistor M4 is grounded, the drain of the fourth transistor M4 is connected to one end of the first resistor R1 and the input end of the filter network, the other end of the first resistor R1 is connected to the power supply end, and the output end of the filter network outputs an output voltage.

[0020] Preferably, the power output unit PA stage includes an amplifier structure with a cross-neutralizing capacitor and a power tail tube consisting of a fifth transistor M5 and a sixth transistor M6; the gate of the fifth transistor M5 inputs a voltage signal, the gate of the sixth transistor M6 inputs a fixed bias signal, the sources of the fifth transistor M5 and the sixth transistor M6 are both grounded, and the drains of the fifth transistor M5 and the sixth transistor M6 are commonly connected to the amplifier structure with the cross-neutralizing capacitor.

[0021] Preferably, the amplifier structure with cross-neutralization capacitors is a common-source amplifier structure with cross-neutralization capacitors or a common-source and common-gate amplifier structure with cross-neutralization capacitors.

[0022] Preferably, the driving amplifier unit DR stage includes a common source amplifier structure with cross-neutralizing capacitors.

[0023] Preferably, the driving amplifier unit DR stage includes an amplifier structure with a cross-neutralizing capacitor and a power tail pipe.

[0024] Beneficial effects

[0025] The advantages of the present invention are:

[0026] 1. The compensation tube within the Doherty-like power amplifier of the present invention is set with a certain bias voltage, so that the power output unit PA stage maintains high efficiency at low input power. As the input power increases, the output power increases accordingly until it reaches the power corresponding to the set output current. However, at this time, the power output unit PA stage has not entered the saturation region, and the efficiency of the power output unit PA stage continues to increase with the increase of the input signal. Under high input signal power, the current of the power output unit PA stage bias tube is controlled by an adaptive bias circuit based on envelope tracking, so that the output current of the power output unit PA stage is increased to achieve a higher output power. Therefore, the efficiency continues to increase, thereby achieving better back-off efficiency than traditional Class AB power amplifiers. In addition, the present invention only requires one power output unit PA stage, without considering the load modulation effect, and can achieve broadband applications. The present invention has achieved good output power and back-off effects using a 140GHz broadband PA based on TSMC 65nm process.

[0027] 2. High back-off efficiency: Traditional Doherty-type amplifiers require multi-channel power combining solutions, which are difficult to strike a balance between broadband matching and low insertion loss. This results in significant losses, resulting in reduced efficiency and back-off efficiency. This is more pronounced at high millimeter-wave frequencies due to strong parasitic effects. The present invention only requires a single-channel structure, which has lower losses and area than the traditional two-channel structure, making it more suitable for high-frequency design. It can avoid the losses caused by power combining and achieve efficiency comparable to that of a single-channel power amplifier.

[0028] 3. Suitable for high-frequency designs and high-order modulation communication systems: The present invention features a single-channel structure, resulting in a smaller footprint and suitable for environments with strong parasitics at high millimeter-wave frequencies. The adaptive bias circuit employed in the present invention can meet the high-speed communication requirements of high-order modulation signals, which contributes to its superior communication capabilities.

[0029] 4. Excellent broadband design performance: The slot-line-coupled balun-based output-end structure is a distributed matching structure. Compared with the traditional transformer-based lumped matching structure, it has better broadband characteristics and is more suitable for broadband design. Furthermore, the present invention is a Doherty-like power amplifier based on current source control. Compared with traditional Doherty power amplifiers, it can achieve no load modulation effect, making it suitable for broadband design. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 Schematic diagram of the modular structure of the Doherty-like power amplifier of the present invention;

[0031] Figure 2a Schematic diagram of the asymmetric Wilkinson power splitter circuit structure of the present invention;

[0032] Figure 2bSchematic diagram of the circuit structure of the three-coupling transformer of the present invention;

[0033] Figure 3 Schematic diagram of the structure of the triple-coupling transformer of the present invention;

[0034] Figure 4 This is a schematic diagram of the output matching balun network structure of the present invention;

[0035] Figure 5a Schematic diagram of a first embodiment of the circuit structure of the driving amplifier unit DR of the present invention;

[0036] Figure 5b This is a schematic diagram of a second embodiment of the circuit structure of the driving amplifier unit DR stage of the present invention;

[0037] Figure 6a This is a schematic diagram of a first embodiment of the PA-level circuit structure of the power output unit of the present invention;

[0038] Figure 6b Schematic diagram of a second embodiment of the PA-level circuit structure of the power output unit of the present invention;

[0039] Figure 7 This is a graph showing the effect of the bias voltage of the tail pipe of the compensation power supply on the back-off point power of the present invention;

[0040] Figure 8 This is a fallback curve diagram when the bias voltage of the tail tube of the compensation power supply of the present invention changes;

[0041] Figure 9 A graph showing the effect of the bias tube of the present invention on saturation power and back-off amount;

[0042] Figure 10 Schematic diagram of the circuit structure of the adaptive bias circuit of the present invention;

[0043] Figure 11a Schematic diagram of a first embodiment of the detector circuit structure of the present invention;

[0044] Figure 11b Schematic diagram of a second embodiment of the detector circuit structure of the present invention;

[0045] Figure 12a Schematic diagram of a first embodiment of the detector load network circuit structure of the present invention;

[0046] Figure 12b Schematic diagram of a second embodiment of the detector load network circuit structure of the present invention;

[0047] Figure 13 is a graph showing how the output voltage of the adaptive bias circuit of the present invention changes with input power;

[0048] Figure 14 The communication performance waveform diagram of the adaptive bias circuit of the present invention;

[0049] Figure 15 This is a small signal simulation diagram of the Doherty-like power amplifier of the present invention;

[0050] Figure 16 This is a large signal simulation diagram of the Doherty-like power amplifier of the present invention;

[0051] Figure 17 These are large signal simulation diagrams of the Doherty-like power amplifier of the present invention at different frequencies. DETAILED DESCRIPTION

[0052] The present invention will be further described below in conjunction with the embodiments, but this does not constitute any limitation to the present invention. Any limited number of modifications made by anyone within the scope of the claims of the present invention are still within the scope of the claims of the present invention.

[0053] like Figure 1 As shown, a Doherty-like power amplifier with improved back-off efficiency of the present invention includes an input power distribution network 100, an output matching balun network 400, an adaptive bias circuit 500, a driving amplifier unit DR stage 200, and a power output unit PA stage 300.

[0054] The input power distribution network 100 is an asymmetric power distribution network that outputs two RF signals, one of which is input to the driver amplifier unit DR stage 200, and the other is input to the adaptive bias circuit 500. The purpose is to match the input of the driver amplifier unit DR stage 200 and maintain the isolation between the adaptive bias circuit and the driver amplifier unit DR stage 200, so that the input power enters the RF link as much as possible. Common structures of asymmetric power distribution networks include asymmetric Wilkinson power splitters, three-coupled transformers, etc., and their circuit structures are as follows: Figure 2a and Figure 2b As shown in Figure 2. Among them, the asymmetric Wilkinson power divider is difficult to use as a differential output, so a three-coupled transformer is often used as an asymmetric power divider. Its three-dimensional structure is shown in Figure 2. Figure 3 As shown in the figure, the two transformers connected to the output end of the three-coupled transformer are the high-layer metal and low-layer metal of the input end respectively. This can reduce the mutual coupling between the two output ends and increase the isolation, while ensuring that the RF input matching has a small insertion loss.

[0055] The output matching balun network 400 is used for power matching of the output stage. Common structures include transformer and slot balun. Compared with the transformer network, the slot balun has lower loss and better broadband characteristics at high frequencies. Its structure is as follows: Figure 4As shown, since it is the existing technology, no further explanation will be given.

[0056] like Figure 5a As shown, Figure 5a An embodiment of the driver amplifier unit DR stage 200 is shown, which is a common-source amplifier structure with cross-neutralizing capacitors, providing a larger output for the subsequent amplifier. Figure 5b This is another embodiment of a driver amplifier unit DR stage 200, comprising a common-source amplifier structure with cross-neutralizing capacitors and a power tail tube. The power tail tube includes a compensation tube and a bias tube. Since this common-source amplifier structure with cross-neutralizing capacitors is already used in the DR stage of traditional Doherty-like power amplifiers and is not improved upon in the present invention, it will not be described in detail.

[0057] like Figure 6a and Figure 6b As shown, the power output unit PA stage 300 includes an amplifier structure with cross-neutralizing capacitors and a power tail composed of a fifth transistor M5 and a sixth transistor M6. The gate of the fifth transistor M5 inputs a voltage signal output by the adaptive bias circuit 500, which is an adaptive bias signal. The gate of the sixth transistor M6 inputs a fixed bias signal. The sources of the fifth transistor M5 and the sixth transistor M6 are both grounded, and the drains of the fifth transistor M5 and the sixth transistor M6 are commonly connected to the amplifier structure with cross-neutralizing capacitors.

[0058] in, Figure 6a An embodiment of the power output unit PA stage 300 is shown. In this embodiment, the power output unit PA stage 300 is composed of a common source amplifier structure with a cross-neutralizing capacitor and a power tail pipe. It should be noted that in a Doherty-like power amplifier, the driver amplifier unit DR stage 200 and the power output unit PA stage 300 can be composed of multiple stages to provide greater gain; and according to Figure 5b and Figure 6a It can be seen from the circuit structure that the driving amplifier unit DR stage 200 can adopt the same circuit structure as the power output unit PA stage 300, thereby improving the control capability of the circuit back-off efficiency.

[0059] Figure 6b As another embodiment of the power output unit PA stage 300, in this embodiment, the power output unit PA stage 300 adopts a common source and common gate amplifier structure with cross neutralization capacitors combined with a power tail pipe, thereby improving the saturated output power and efficiency.

[0060] In the power output unit PA stage 300, the power tail tube is divided into a compensation tube and a bias tube. Specifically, the fifth transistor M5 serves as a bias tube, and the sixth transistor M6 serves as a compensation tube. The compensation tube is given a fixed bias to provide an initial amplification state for the circuit. Therefore, the size of the compensation tube and its bias voltage determine the circuit's fallback power and efficiency. Figure 7 The figure shows the curve of the circuit output power change when the compensation tube bias voltage changes. Figure 8 This is the circuit back-off curve under the change of compensation tube bias. It can be seen that when the bias tube is determined, the circuit saturation output power is fixed, and the back-off point power and efficiency are affected by the compensation tube bias. Another power tail tube is the bias tube, and the bias voltage of this tube is given by the adaptive bias circuit 500. When a smaller RF signal is input, the adaptive bias circuit 500 outputs a smaller voltage of about 0.2V. At this time, the bias tube provides almost no current, and the output current of the power output unit PA level 300 is determined by the compensation tube. This is the initial state of the circuit. When the input RF signal reaches a certain level, the adaptive bias circuit 500 outputs a high voltage to the bias tube. At this time, the circuit output current is jointly determined by the bias tube and the compensation tube, and the circuit outputs greater power.

[0061] The gate width and bias voltage of a transistor control its current output capability. In high-order communications, power amplifiers are typically required to operate at a power back-off of 3-6 dBm. This means the bias transistor needs to provide 1-2 times the current of the compensation transistor. Therefore, the bias transistor size is set to approximately 1-2 times the size of the compensation transistor. Figure 9 This figure shows the change in the circuit back-off curve when the gate width of the bias transistor in the power output unit PA stage 300 changes from 20µm to 160µm. It can be seen that as the gate width of the bias transistor increases, both the saturation power and the maximum PAE increase, while the back-off amount remains almost unchanged. Therefore, to ensure sufficient back-off, the bias transistor in the power output unit PA stage 300 should be as large as possible, and can be set to 1-4 times the size of the compensation transistor, depending on requirements.

[0062] The schematic diagram of the adaptive bias circuit 500 is shown in FIG. Figure 10 As shown in Figure 1, it consists of a detector and a buffer stage. The detector structure is as follows: Figure 11a and Figure 11b As shown, it can be seen that in the two embodiments of the detector, it is composed of a self-mixing differential pair and a load network. Among them, the self-mixing differential pair is composed of a first transistor M1 and a second transistor M2, and these two transistors can be PMOS or NMOS tubes. Specifically, the gates of the first transistor M1 and the second transistor M2 serve as input terminals, respectively, to input the INP signal and the INN signal of the RF signal. The source of the first transistor M1 and the second transistor M2 are connected and connected to the common terminal. Among them, as shown in FIG. Figure 11aAs shown, in this embodiment, the common terminal is the ground terminal, and the load network is connected to the power supply terminal. Figure 11b As shown, in this embodiment, the common terminal is the power supply terminal, and the load network is grounded. The drains of the first transistor M1 and the second transistor M2 are connected, and are also connected to the load network. The connection between the load network and the drains of the first transistor M1 and the second transistor M2 serves as the output terminal of the detector and is connected to the buffer stage. The output terminal of the detector is grounded through the first capacitor C1. In other words, the bias of the self-mixing differential pair is provided by the input power distribution network 100 of the previous stage. This bias voltage is controlled to be slightly less than the threshold voltage, allowing the adaptive bias circuit 500 to operate normally after reaching a certain input power, and maintain a low input before this input.

[0063] like Figure 12a and Figure 12b As shown, the load network of the present invention can be replaced by a resistor or a PMOS tube. Specifically, based on the implementation method of the common terminal as the power supply terminal, as shown in FIG. Figure 12a The circuit diagram shown is a PMOS tube load network; Figure 12b The circuit diagram shows a load network with resistors. By changing the resistor size or the PMOS transistor bias voltage, the equivalent resistance can be changed, thereby controlling the bandwidth of the envelope detector to meet higher communication requirements.

[0064] The buffer stage includes a fourth transistor M4, a first resistor R1, and a filter network. The gate of the fourth transistor M4 is connected to the output of the detector, the source of the fourth transistor M4 is grounded, the drain of the fourth transistor M4 is connected to one end of the first resistor R1 and the input of the filter network, the other end of the first resistor R1 is connected to the power supply, and the output of the filter network provides an output voltage. The buffer stage, based on the above circuit structure, can increase the envelope amplitude of the detection output and provide a certain degree of reverse isolation. The filter network in the buffer stage is used to prevent the RF signal from the tailpipe from leaking into the adaptive bias circuit.

[0065] Figure 13 The output voltage of the adaptive bias circuit 500 is shown for different input powers. It can be seen that the output voltage of the adaptive bias circuit 500 remains low when the input power is below a certain value. After the input power exceeds a certain threshold, the output voltage rapidly increases to close to the power supply voltage of 1.2V. Furthermore, the detector's differential tube bias voltage can adjust the detector's turn-on power and turn-on speed. As shown in the figure, when the detector's differential tube bias voltage changes from 0.33 to 0.38V, the adaptive bias circuit 500 turns on with lower power and slower turn-on speed. Furthermore, the detector's load network also affects the turn-on power of the adaptive bias circuit 500. Furthermore, the load network affects the detector's amplification and filtering capabilities, thereby affecting its communication rate.

[0066] Based on the aforementioned power amplifier circuit structure, the high-backoff-efficiency Doherty-like power amplifier circuit proposed in the present invention, using envelope tracking technology, can achieve high-order communications. The circuit's communication performance depends on the communication rate of the adaptive bias. Specifically, the speed at which the adaptive bias circuit 500 tracks the modulated signal envelope determines the signal rate the circuit can handle. Figure 14 The output curve of adaptive bias circuit 500 is shown for a given rate-modulated signal. As can be seen, even with a 16QAM modulated signal at a 10 Gps bit rate, the output of adaptive bias circuit 500 still tracks the envelope well. Therefore, the circuit exhibits excellent communication capabilities.

[0067] like Figure 15 As shown in Figure 2, the Doherty-like power amplifier of the present invention finally achieves a bandwidth design of 132-152 GHz, with a maximum gain of 25.6 dB, and the range of input and output matching less than -10 dB is greater than 20 GHz. The large signal simulation data of the power amplifier at 140 GHz with input power is shown in Figure 2. Figure 16 shown. Figure 17 It shows that within the entire observed bandwidth, an output power greater than 10dBm is achieved, with a maximum of 14.1dBm and an optimal energy efficiency greater than 12.3%.

[0068] The above is only a preferred embodiment of the present invention. It should be pointed out that for those skilled in the art, several modifications and improvements can be made without departing from the structure of the present invention. These modifications and improvements will not affect the effect of the implementation of the present invention and the practicality of the patent.

Claims

1. A Doherty-like power amplifier with improved fallback efficiency, characterized in that: include: An input power distribution network (100) outputs two radio frequency signals, wherein one of the radio frequency signals is input into the driver amplifier unit DR stage (200), and the other of the radio frequency signals is input into the adaptive bias circuit (500), so as to isolate the driver amplifier unit DR stage (200) and the adaptive bias circuit (500); The output end of the driving amplifier unit DR stage (200) is connected to the input end of the power output unit PA stage (300); An adaptive bias circuit (500) is used to detect changes in the envelope of the power amplifier input signal in real time based on the radio frequency signal transmitted by the input power distribution network (100), and output a voltage signal; A power output unit PA stage (300) is provided with an input terminal for receiving the voltage signal, wherein the voltage signal is used to control the bias tail pipe current of the power output unit PA stage (300) to adjust the output power and conversion efficiency of the power output unit PA stage (300); An output matching balun network (400), the input end of which is connected to the output end of the power output unit PA stage (300); The adaptive bias circuit (500) comprises a detector and a buffer stage; the input end of the detector inputs a radio frequency signal, the output end of the detector is connected to the input end of the buffer stage, and the input end of the buffer stage outputs the voltage signal; The detector comprises a self-mixing differential pair consisting of a first transistor (M1) and a second transistor (M2), and a load network; the gates of the first transistor (M1) and the second transistor (M2) serve as input terminals, respectively, to input an INP signal and an INN signal of a radio frequency signal; the sources of the first transistor (M1) and the second transistor (M2) are connected and connected to a common terminal; the drains of the first transistor (M1) and the second transistor (M2) are connected and connected to the load network, and the connection terminal of the load network and the drains of the first transistor (M1) and the second transistor (M2) are connected to a buffer stage as the output terminal of the detector.

2. The Doherty-like power amplifier with improved back-off efficiency according to claim 1, characterized in that: The load network is a resistor or a PMOS transistor.

3. The Doherty-like power amplifier with improved back-off efficiency according to claim 1, characterized in that: The output end of the detector is grounded via a first capacitor (C1).

4. A Doherty-like power amplifier with improved back-off efficiency according to any one of claims 1 to 3, characterized in that: The buffer stage comprises a fourth transistor (M4), a first resistor (R1) and a filter network; the gate of the fourth transistor (M4) is connected to the output end of the detector, the source of the fourth transistor (M4) is grounded, the drain of the fourth transistor (M4) is simultaneously connected to one end of the first resistor (R1) and the input end of the filter network, the other end of the first resistor (R1) is connected to the power supply end, and the output end of the filter network outputs an output voltage.

5. The Doherty-like power amplifier with improved back-off efficiency according to claim 1, characterized in that: The power output unit PA stage (300) comprises an amplifier structure with a cross-neutralizing capacitor and a power tail tube composed of a fifth transistor (M5) and a sixth transistor (M6); a gate of the fifth transistor (M5) inputs a voltage signal, a gate of the sixth transistor (M6) inputs a fixed bias signal, sources of the fifth transistor (M5) and the sixth transistor (M6) are both grounded, and drains of the fifth transistor (M5) and the sixth transistor (M6) are commonly connected to the amplifier structure with the cross-neutralizing capacitor.

6. The Doherty-like power amplifier with improved back-off efficiency according to claim 5, characterized in that: The amplifier structure with cross-neutralizing capacitors is a common-source amplifier structure with cross-neutralizing capacitors or a common-source and common-gate amplifier structure with cross-neutralizing capacitors.

7. The Doherty-like power amplifier with improved back-off efficiency according to claim 5, characterized in that: The driver amplifier unit DR stage (200) comprises a common source amplifier structure with cross-neutralizing capacitors.

8. The Doherty-like power amplifier with improved back-off efficiency according to claim 5, characterized in that: The driver amplifier unit DR stage (200) comprises an amplifier structure with a cross-neutralizing capacitor and a power tail pipe.

Images