Millimeter wave CMOS power amplifier with bendable layout

Abstract

The invention relates to the field of integrated circuits, and discloses a millimeter wave CMOS power amplifier with a bendable layout, which comprises a CMOS power amplifier body, and the CMOS power amplifier body comprises an input matching network, a driving amplification stage, an inter-stage matching network, a power output stage and an output matching network which are connected in sequence, each of the input matching network, the inter-stage matching network and the output matching network comprises primary-secondary coupling coils which are arranged along the same horizontal direction and are partially overlapped; any one of the primary and secondary coupling coils of the input matching network or the inter-stage matching network of the CMOS power amplifier body arranged in the outer channel can rotate to adjust the output direction of the input matching network or the inter-stage matching network. The method has the advantages that extra link loss cannot be generated, extra chip area is not occupied, transmission amplitude / phase consistency before and after layout bending is good, secondary redesign of the circuit is avoided, and the method is efficient and convenient and has high engineering practical value.

Description

Technical Field

[0001] The present invention relates to the field of integrated circuits, and in particular to a millimeter wave CMOS power amplifier with a bendable layout. Background Art

[0002] For silicon-based millimeter-wave multi-channel whole chips in phased array system applications, using the same set of schematics for all channels (outer channels and inner channels) can not only save design costs, but also maintain the phase / amplitude consistency between channels to the greatest extent. Figure 1 As shown in the figure, the output pads of the power amplifiers of the outer and inner channels are generally set on the left and right sides of the chip. However, in order to match the layout of the actual application (PCB board, package, etc.), it is necessary to flexibly adjust the circuits in some channels of the whole chip to avoid the coupling effect caused by the output pads of adjacent channels after the chip is mounted on the board or packaged. Figure 2 As shown, the position of the output pad of the final power amplifier of the outer channel can be adjusted so that it is not on the same side of the chip as the output pad of the final power amplifier of the internal middle channel, that is, the output direction of the final power amplifier of the outer channel is not in the same direction as the output direction of the final power amplifier of the internal channel, and the two are orthogonal to each other, thereby eliminating the coupling effect to the greatest extent.

[0003] To achieve the above effect, the traditional design method usually rotates the final power amplifier module of the outer channel as a whole, and uses a transmission line to transition between the output of the previous module and the input of the final power amplifier (at this time, the unrotated internal channel also needs to be introduced) to ensure the consistency of the output signal between channels. However, since the power amplifier usually adopts a 2-3-stage structure and is relatively long in size, the overall rotation will seriously increase the area of ​​the entire chip. The introduction of the transition transmission line will also bring additional link insertion loss, and its loss will increase with the increase of the operating frequency. Therefore, it is hoped that the internal circuit of the final power amplifier of the outer channel can be rotated to achieve the corner output. However, the change of the internal characteristics of the circuit often causes mismatch between channels, especially when the frequency rises to the millimeter wave band, which will seriously affect the performance of the system-level chip. At the same time, considering the convenience of implementation and avoiding a complete redesign of the bending part, the transmission characteristics before and after the bending also naturally need to have a certain consistency. Summary of the invention

[0004] In order to overcome the problem of introducing transmission lines to perform or change the internal circuit of a power amplifier to achieve rotational bending in the above-mentioned background technology, the present invention provides a millimeter-wave CMOS power amplifier with a bendable layout. By directly rotating and bending the input matching network or the inter-stage matching network in the CMOS power amplifier on the outer channel, no additional link loss is generated and no additional chip area is occupied. The transmission amplitude / phase consistency before and after the layout is bent is good, and secondary redesign of the circuit is avoided. It is efficient and convenient, and has the beneficial effect of high engineering practical value.

[0005] The technical solution of the present invention is as follows:

[0006] A millimeter-wave CMOS power amplifier with a bendable layout comprises a CMOS power amplifier body, wherein the CMOS power amplifier body comprises an input matching network, a driving amplifier stage, an interstage matching network, a power output stage and an output matching network connected in sequence, wherein the input matching network, the interstage matching network and the output matching network all comprise primary-secondary coupling coils arranged in the same horizontal direction and partially overlapping, and any coupling coil in the primary-secondary coupling coils of the input matching network or the interstage matching network of the CMOS power amplifier body arranged in an outer channel can be rotated to adjust its output direction; a radio frequency input signal is converted to the driving amplifier stage for amplification via the input matching network, then converted to the power output stage for amplification via the interstage matching network, and then converted and output via the output matching network.

[0007] Preferably, the coupling coil with a larger radius in the rotating primary-secondary coupling coils in the input matching network and the inter-stage matching network is a regular polygon, and the rotating coupling coil rotates 90° around the center of the coupling coil which is a regular polygon.

[0008] Further preferably, the coupling coil with a larger radius in the primary-secondary coupling coil is a square or a regular octagon.

[0009] Further preferably, the primary-secondary coupling coils of the same matching network are on the same metal layer or on upper and lower metal layers respectively. When on the same metal layer, the jumper of the primary-secondary coupling coils is located on the coupling coil with a smaller radius.

[0010] Preferably, the coupling mode of the rotating primary-secondary coupling coils in the input matching network and the inter-stage matching network is magnetic coupling, and the electrical coupling is parasitic coupling.

[0011] Preferably, the minimum electrical coupling of the primary-secondary coupling coils that rotate in the input matching network and the inter-stage matching network is obtained through their size, spacing, shape and coupling coefficient, and the corresponding primary-secondary coupling coils under the minimum electrical coupling condition must meet the impedance matching requirements of the front and rear stages and the network amplitude / phase of the corresponding primary-secondary coupling coils before and after rotation must meet the consistency requirements.

[0012] Preferably, the turns ratio of the primary-secondary coupling coils of the input matching network, the inter-stage matching network and the output matching network is determined according to the impedance transformation ratio of the previous and next stages and the two-port impedance matching conversion theory.

[0013] Preferably, the input end of the input matching network is connected to the output end of the previous module, and also includes an input capacitor C 11 , secondary power supply center tap CT1 and output capacitor C 12 , the primary-secondary coupling coil includes a primary coupling coil L 11 and the secondary coupling coil L 12 , the input capacitor C 11 The anode of the primary coupling coil L 11 The positive terminal of the input matching network is connected as the single-ended input terminal, and the input capacitor C 11 The cathode of the primary coupling coil L is grounded. 11 The negative terminal of the output capacitor C 12 The anode of the secondary coupling coil L 12 The positive terminal of the output capacitor C is connected as the differential output positive terminal of the input matching network. 12 The cathode of the secondary coupling coil L 12 The negative terminal of the output matching network is connected as the negative terminal of the differential output, and the center tap CT1 is connected from the secondary coil L 12 The inter-stage matching network also includes input capacitor C 21 , the primary power supply center tap CT2 and the output capacitor C 22 , the primary-secondary coupling coil includes a primary coupling coil L 21 and the secondary coupling coil L 22 , where the input capacitor C 21 The anode of the primary coupling coil L 21 The positive electrode is connected as the differential input positive terminal of the interstage matching network, and the input capacitor C 21 The cathode of the primary coupling coil L 21 The negative terminal of the output capacitor C is connected as the negative terminal of the differential input of the interstage matching network. 22 The anode of the secondary coupling coil L 22 The positive stage is connected as the differential output positive terminal of the inter-stage matching network, and the output capacitor C 22 The cathode of the secondary coupling coil L22 The negative stage is connected as the negative terminal of the differential output of the interstage matching network, and the center tap CT2 is connected from the primary coil L 21 The input capacitor and the output capacitor of the input matching network and the inter-stage matching network are both implemented by MOM capacitors; the input matching network is a single-ended-to-differential passive structure, and the inter-stage matching network is a differential-to-differential passive structure, and the passive structures of the two are both implemented by a coupled resonant cavity based on the transformer principle.

[0014] Preferably, the driving amplifier stage includes an input matching resistor R1, a transistor M 11 , transistor M 12 , and the capacitor C C11 And the neutralizing capacitor C C12 , the positive electrode of the input matching resistor R1 is connected to the transistor M 11 The gate and neutralizing capacitor C C12 The cathode connection of transistor M 11 The source of transistor M is grounded. 11 The drain terminal is connected to the neutralizing capacitor C C11 anode of transistor M 11 The gate of transistor M is used as the positive input of the driving amplifier stage. 11 The drain of the drive amplifier stage is used as the negative output, and the negative electrode of the input matching resistor R1 is connected to the transistor M 12 The gate and neutralizing capacitor C C11 The cathode of transistor M 12 The source of transistor M is grounded. 12 The drain is connected to the neutralizing capacitor C C12 anode of transistor M 12 The gate of the transistor M is used as the negative input of the driving amplifier stage. 12 The drain of the power amplifier stage is used as the positive output of the driving amplifier stage; the power output stage includes a bias resistor R b1 , bias resistor R b2 , transistor M 21 , transistor M 22 , and the capacitor C C21 And the neutralizing capacitor C C22 , bias resistor R in the power output stage b1 The positive electrode is connected to transistor M 21 The gate and neutralizing capacitor C C22 The cathode of transistor M 21 The source of transistor M is grounded. 21 The drain terminal is connected to the neutralizing capacitor C C21 anode of transistor M 21 The gate of transistor M is used as the positive input of the power output stage. 21The drain is used as the negative output of the power output stage, and the bias resistor R b2 The negative terminal is connected to transistor M 22 The gate and neutralizing capacitor C C21 The cathode of transistor M 22 The source of transistor M is grounded. 22 The drain terminal is connected to the neutralizing capacitor C C22 anode of transistor M 22 The gate of transistor M is used as the negative input of the power output stage. 22 The drain of the power output stage is used as the positive output, and the bias resistor R b1 The negative terminal of the bias resistor R b2 The positive electrode is used as the gate bias point of the power output stage; the driving amplifier stage and the power output stage both adopt a differential common source transistor amplification structure and a neutralizing capacitor technology, and the output power of the driving amplifier stage is less than the output power of the power output stage.

[0015] Preferably, the output matching network is a differential-to-single-ended passive structure, including an input capacitor C 31 , primary coupling coil L 31 , secondary coupling coil L 32 , the primary power supply center tap CT3 and the output capacitor C 32 , where the input capacitor C 31 Anode and primary coupling coil L 31 The positive terminal is connected as the differential input positive terminal of the output matching network, and the input capacitor C 31 Cathode and primary coupling coil L 31 The negative terminal is connected as the negative terminal of the differential input of the output matching network, and the output capacitor C 32 Anode and secondary coupling coil L 32 The positive terminal is connected as the single-ended output terminal of the output matching network, and the output capacitor C 32 The cathode is grounded, and the secondary coupling coil L 32 The negative terminal is grounded, and the center tap CT3 is connected from the primary coil L 31 Lead out from the internal physical center.

[0016] Compared with the prior art, the above technical solution has the following beneficial effects:

[0017] (1) The CMOS power amplifier provided by the present application can directly rotate and bend the inner part of the layout of the CMOS power amplifier on the outer channel for the outer channel and the inner channel in the multi-channel whole chip, and the rotation and bending can be implemented in any coupling coil of the primary-secondary coupling coil of the input matching network or the inter-stage matching network, so as to obtain two different link sizes and ensure that the overall efficiency of the power amplifier is not affected;

[0018] (2) By designing a CMOS power amplifier that only relies on the internal matching network itself for rotation and bending, no additional link loss is generated and no additional chip area is occupied. This avoids the problem of rotating the final power amplifier module as a whole in the background technology, which will seriously increase the area of ​​the entire chip and the transmission line introduced in the transition will cause additional link insertion loss. This flexibly solves the problem of signal output layout of each channel in a multi-channel chip and reduces the coupling effect caused by the signal pads between channels after the chip is mounted on the board or packaged.

[0019] (3) The RF input signal is amplified and output through a multi-stage structure, which can amplify the signal power and improve the transmission distance and efficiency of the signal;

[0020] (4) By designing the coupling coil with a larger radius in the primary-secondary coupling coil of the matching network that rotates in the input matching network and the inter-stage matching network to be a regular polygon, it is possible to ensure that the magnetic coupling area of ​​the primary-secondary coupling coil is consistent before and after the rotation, thereby avoiding the problem of impedance mismatch between the front and rear stages;

[0021] (5) By designing the coupling mode of the primary-secondary coupling coil to be magnetic coupling and the electrical coupling to be parasitic coupling, the phase change generated before and after the rotation of the primary-secondary coupling coil can be minimized;

[0022] (6) Since the minimum electrical coupling of the primary-secondary coupling coils that rotate in the input matching network and the inter-stage matching network is obtained through their size, spacing, shape and coupling coefficient, and the corresponding primary-secondary coupling coils under the minimum electrical coupling condition must meet the impedance matching requirements of the front and rear stages and the network amplitude / phase of the corresponding primary-secondary coupling coils before and after rotation must meet the consistency requirements, this design can ensure that the transmission amplitude / phase of the determined primary-secondary coils remains consistent before and after rotation, avoiding secondary redesign, which is efficient and convenient and has high engineering practical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The present invention will be described with reference to the accompanying drawings, in which:

[0024] Figure 1 Schematic diagram of the traditional multi-channel whole chip layout (i.e., all channel output pads are located on the left and right sides of the chip);

[0025] Figure 2 It is a schematic diagram of a multi-channel whole chip structure in which a transmission line is introduced to adjust the position of an output pad of an outer channel in the prior art;

[0026] Figure 3 It is a schematic diagram of the planar structure of the primary-secondary coupling coil direct version of the input matching network;

[0027] Figure 4A schematic diagram of a planar structure in which the primary coupling coil of the primary-secondary coupling coil of the input matching network rotates;

[0028] Figure 5 It is a schematic diagram of the planar structure of the primary-secondary coupling coil through version of the inter-stage matching network;

[0029] Figure 6 A schematic diagram of a circuit plan structure in which a secondary coupling coil of a primary-secondary coupling coil of an inter-stage matching network rotates;

[0030] Figure 7 A schematic diagram of a multi-channel whole chip structure using a kinked power amplifier input matching network to adjust the position of the outer channel output pad;

[0031] Figure 8 A schematic diagram of a multi-channel whole chip structure using an inter-stage matching network of a bent power amplifier to adjust the position of an outer channel output pad;

[0032] Fig. 9 This is a schematic diagram of the circuit structure of the bendable millimeter-wave CMOS power amplifier;

[0033] Fig.10 Comparison of transmission amplitude and phase before and after bending the input matching network. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0035] Example 1: Figures 3 to 10 A millimeter-wave CMOS power amplifier with a bendable layout is shown. The CMOS power amplifier body includes an input matching network, a driving amplifier stage, an interstage matching network, a power output stage and an output matching network connected in sequence. The input matching network, the interstage matching network and the output matching network all include primary-secondary coupling coils arranged in the same horizontal direction and partially overlapping. The primary-secondary coupling coils include a primary coupling coil and a secondary coupling coil. Any coupling coil in the input matching network or the primary-secondary coupling coil of the interstage matching network of the CMOS power amplifier body arranged in the outer channel of the multi-channel whole chip can be rotated to adjust its output direction, that is, when adjusting the output direction of the CMOS power amplifier arranged in the outer channel, only the primary-secondary coupling coils in the input matching network and the interstage matching network participate in the bending rotation, so that it can ensure that the overall efficiency of the power amplifier is not affected, the input matching network and the interstage matching network do not participate in the bending rotation at the same time, and only one of the primary-secondary coupling coils rotates.

[0036] The two specific rotation methods are as follows: the first method is based on the rotation of the primary coupling coil of the input matching network of the CMOS power amplifier body. Figure 3 This is a schematic diagram of the planar structure of the input matching network with the primary-secondary coupling coil not rotated. At this time, the primary and secondary coupling coils are arranged in the same horizontal direction, and the radius of the secondary coupling coil is larger than the primary coupling coil. When the input matching network needs to be rotated and bent, the primary coupling coil is rotated 90° counterclockwise around the center of the secondary coupling coil, as shown in Figure 1. Figure 4 As shown in FIG. 1 , the output direction of the secondary coupling coil is not changed. The multi-channel whole chip structure after adjusting the position of the outer channel output pad in this way is shown in FIG. Figure 7 shown.

[0037] The second type is based on the rotation of the secondary coupling coil of the inter-stage matching network of the CMOS power amplifier body. Figure 5 This is a schematic diagram of the planar structure of the primary-secondary coupling coil of the interstage matching network without rotation. At this time, the primary and secondary coupling coils are arranged in the same horizontal direction, and the radius of the primary coupling coil is larger than that of the secondary coupling coil. When the interstage matching network needs to be rotated and bent, the secondary coupling coil is rotated 90° clockwise around the center of the primary coupling coil, as shown in Figure 1. Figure 6 As shown in FIG. 1 , the input direction of the primary coupling coil is not changed. The multi-channel chip structure after adjusting the position of the outer channel output pad in this way is shown in FIG. Figure 8 shown.

[0038] The output directions of the outer channel and the inner channel of the multi-channel chip made by the above two rotation methods are both located in different directions of the chip, which reduces the coupling effect caused by the signal pads between the channels after the chip is mounted on the board or packaged, and does not need to introduce transmission lines, does not generate additional link loss, and does not occupy additional chip area. Two different link sizes can be obtained by using a bent input matching network and a bent inter-stage matching network. It should be noted that the rotational bending methods inside the CMOS power amplifier body layout include but are not limited to the above two methods.

[0039] The RF input signal is converted to the driving amplifier stage for amplification by the input matching network, and then converted to the power output stage for amplification by the interstage matching network, and then converted and output by the output matching network. Specifically, the single-ended input end of the input matching network is connected to the output of the channel link pre-stage module, the differential output positive end of the input matching network is connected to the positive input of the driving amplifier stage, the differential output negative end of the input matching network is connected to the negative input of the driving amplifier stage, the negative output of the driving amplifier stage is connected to the differential input positive end of the interstage matching network, the positive output of the driving amplifier stage is connected to the differential input negative end of the interstage matching network, the differential output positive end of the interstage matching network is connected to the positive input of the power output stage, the differential output negative end of the interstage matching network is connected to the negative input of the power output stage, the negative output of the power output stage is connected to the differential input positive end of the output matching network, the positive output of the power output stage is connected to the differential input negative end of the output matching network, and the single-ended output end of the output matching network is connected to the output pad.

[0040] The CMOS power amplifier provided in this embodiment can be directly rotated and bent inside the layout of the CMOS power amplifier on the outer channel for the outer channel and the inner channel, and the rotation and bending can be implemented in any coupling coil of the primary-secondary coupling coil of the input matching network or the inter-stage matching network, so as to obtain two different link sizes and ensure that the overall efficiency of the power amplifier is not affected; by designing a CMOS power amplifier that only relies on the internal matching network itself for rotation and bending, no additional link loss is generated and no additional chip area is occupied, thereby avoiding the problem of rotating the final-stage power amplifier module as a whole in the background technology, which will seriously increase the area of ​​the entire chip and the transmission line introduced in the transition will bring additional link insertion loss, and flexibly solves the problem of signal output layout of each channel in a multi-channel whole chip, reducing the coupling effect caused by the signal pad between channels after the chip is mounted on the board or packaged; the RF input signal is amplified and output through a multi-stage structure, which can amplify the signal power and improve the transmission distance and efficiency of the signal.

[0041] Embodiment 2: Based on Embodiment 1, the shape of the primary-secondary coupling coil is optimally designed. The coupling coil with a larger radius in the primary-secondary coupling coil of the matching network that rotates in the input matching network and the inter-stage matching network is a regular polygon, and the rotating coupling coil rotates 90° around the center of the coupling coil that is a regular polygon, that is, the shape of the primary-secondary coupling coil of the input matching network and the inter-stage matching network is determined according to whether the corresponding primary-secondary coupling coil is rotated and bent. There is no specific requirement for the shape style of the matching network that does not bend and rotate. The coupling coil with a larger radius in the primary-secondary coupling coil of the matching network that needs to be bent and rotated must be a regular polygon structure, and the corresponding coupling coil of the other stage may not be a regular polygon structure. This setting can ensure that the magnetic coupling area of ​​the primary-secondary coupling coil is consistent before and after rotation, avoiding the problem of impedance mismatch between the front and rear stages, and in actual chip processing and manufacturing, the inner angle of the coupling coil can only be an integer multiple of 45°, so the coupling coil with a larger radius in the primary-secondary coupling coil is a square or a regular octagon. By setting the coupling coil to rotate 90 degrees, the output direction of the outer channel can be made perpendicular to the output direction of the inner channel, which is more conducive to the matching connection between its output pad and the PCB board, package, etc. in actual application.

[0042] Furthermore, the primary-secondary coupling coils of the same matching network are on the same metal layer or on two upper and lower metal layers respectively. When the primary-secondary coupling coils are arranged on the same metal layer, their cross-interface is located on the coupling coil with a smaller radius, and the cross-interface is connected by other metal traces to form a cross-connection, forming a complete current loop to avoid short circuit; when the primary-secondary coupling coils are arranged on different metal layers, the primary and secondary coils have overlapping parts in the vertical direction.

[0043] Embodiment 3: Based on Embodiment 1, the coupling mode of the primary-secondary coupling coil is optimally designed, and the coupling mode of the primary-secondary coupling coil that rotates in the input matching network and the inter-stage matching network is magnetic coupling, and the electric coupling is parasitic coupling. Since when the coupling coil rotates, the shape of the coil can be designed so that the magnetic coupling area of ​​the primary-secondary coupling coil before and after the rotation remains unchanged, the coupling mode of the primary-secondary coupling coil is designed to be magnetic coupling, and the electric coupling is parasitic coupling, so that the phase change generated before and after the rotation of the primary-secondary coupling coil can be minimized.

[0044] Embodiment 4: Based on Embodiment 1, the minimum electrical coupling of the primary-secondary coupling coil is optimally designed. The minimum electrical coupling of the primary-secondary coupling coil that rotates in the input matching network and the inter-stage matching network is determined by its determining parameters, which include size, spacing, shape and coupling coefficient. The primary-secondary coupling coil corresponding to the determined minimum electrical coupling must meet the impedance matching requirements of the front and rear stages.

[0045] Specifically, the initial styles of the direct versions of the primary-secondary coupling coils of the input matching network and the inter-stage matching network are first modeled in the electromagnetic field simulation tool, and then the size, spacing, shape and coupling coefficient of the primary-secondary coupling coils are adjusted to match. Among them, the size of the primary-secondary coupling coil refers to the line width, thickness, number of turns and radius of the primary coupling coil and the secondary coupling coil, the spacing refers to the spacing distance between the primary coupling coil and the secondary coupling coil and the distance between the two coils when they are multi-turn coils, the shape refers to the appearance of the primary coupling coil and the secondary coupling coil, and the coupling coefficient refers to the ratio of the mutual inductance (absolute value) of the primary coupling coil and the secondary coupling coil to its maximum limit value. When one of the determining parameters is adjusted, the other part or all of the determining parameters also need to be adjusted accordingly to ensure that the electrical coupling index between the primary and secondary coils changes while other performance indicators remain unchanged. For example, when the size of any coupling coil in the primary and secondary coupling coils changes, the spacing distance between the primary and secondary coupling coils will also change. At this time, the spacing distance needs to be adjusted to keep the previous spacing distance to ensure the only variable. The adjusted primary and secondary coupling coils satisfy the minimum projection on the XY plane at the crossover point and the minimum side wall electrical coupling caused by the thickness of the metal layer of the coupling coil itself, so as to determine the minimum electrical coupling and the primary and secondary coupling coil corresponding to the minimum electrical coupling; Then observe whether the primary-secondary coupling coefficient and the inductance value corresponding to the coil size meet the impedance matching requirements of the front and rear stages. If not, readjust the determination parameters of the primary-secondary coupling coil until the requirements are met; finally, rotate and bend the primary-secondary coil 90° and perform simulation, and compare the amplitude / phase consistency of the matching network corresponding to the primary-secondary coupling coil before and after rotation, that is, the amplitude / phase consistency requirements set by the system. If the consistency requirements cannot be met, readjust the determination parameters of the primary-secondary coupling coil again until the consistency requirements are met. If the consistency requirements can be met, end the simulation, and the size, spacing, shape and coupling coefficient corresponding to the rotated primary-secondary coil are obtained.

[0046] The primary-secondary coil determined by this design is applied to the input matching network. The transmission amplitude and phase comparison before and after bending in the input matching network is as follows: Fig.10 As shown, the amplitude error does not exceed 0.24dB, and the phase error does not exceed 2.1°. The primary-secondary coils determined by this design ensure that the transmission amplitude / phase remains consistent before and after rotation, avoiding secondary redesign, which is efficient and convenient and has high engineering practical value.

[0047] Example 5: Based on Example 1, the turns ratio of the primary-secondary coupling coil is optimally designed, and the turns ratio of the primary-secondary coupling coil of the input matching network, the inter-stage matching network and the output matching network is determined according to the impedance transformation ratio of the front and rear stages and the two-port impedance matching conversion theory.

[0048] Embodiment 6: Based on Embodiment 1, the circuit structure of the input matching network and the inter-stage matching network is optimally designed, the input end of the input matching network is connected to the output end of the previous module, and also includes an input capacitor C 11 , secondary power supply center tap CT1 and output capacitor C 12 , the primary-secondary coupling coil includes a primary coupling coil L 11 and the secondary coupling coil L 12 , input capacitance C 11 The anode of the primary coupling coil L 11 The positive terminal of the input matching network is connected as the single-ended input terminal, and the input capacitor C 11 The cathode of the primary coupling coil L is grounded. 11 The negative terminal of the output capacitor C 12 The anode of the secondary coupling coil L 12 The positive terminal of the output capacitor C is connected as the differential output positive terminal of the input matching network. 12 The cathode of the secondary coupling coil L 12 The negative terminal of the output matching network is connected as the negative terminal of the differential output, and the center tap CT1 is connected from the secondary coil L 12 The inter-stage matching network also includes input capacitor C 21 , the primary power supply center tap CT2 and the output capacitor C 22 , the primary-secondary coupling coil includes a primary coupling coil L 21 and the secondary coupling coil L 22 , where the input capacitor C 21 The anode of the primary coupling coil L 21 The positive electrode is connected as the differential input positive terminal of the interstage matching network, and the input capacitor C 21 The cathode of the primary coupling coil L 21 The negative terminal of the output capacitor C is connected as the negative terminal of the differential input of the interstage matching network. 22 The anode of the secondary coupling coil L 22 The positive stage is connected as the differential output positive terminal of the inter-stage matching network, and the output capacitor C 22 The cathode of the secondary coupling coil L 22 The negative stage is connected as the negative terminal of the differential output of the interstage matching network, and the center tap CT2 is connected from the primary coil L 21Lead out from the internal physical center. The input capacitors and output capacitors of the input matching network and the inter-stage matching network are all implemented with MOM capacitors. The input matching network is a single-ended-to-differential passive structure, and the inter-stage matching network is a differential-to-differential passive structure. The passive structures of both are based on the coupled resonant cavity of the transformer principle. Energy transmission is only carried out by coupling between the primary and secondary coupling coils, including magnetic coupling and electrical coupling, and there is no hard connection physically.

[0049] Embodiment 7: Based on Embodiment 1, the circuit structure of the driving amplifier stage and the power output stage is optimally designed. The driving amplifier stage includes an input matching resistor R1, a transistor M 11 , transistor M 12 , and the capacitor C C11 And the neutralizing capacitor C C12 , the positive electrode of the input matching resistor R1 is connected to the transistor M 11 The gate and neutralizing capacitor C C12 The cathode connection of transistor M 11 The source of transistor M is grounded. 11 The drain terminal is connected to the neutralizing capacitor C C11 anode of transistor M 11 The gate of transistor M is used as the positive input of the driving amplifier stage. 11 The drain of the drive amplifier stage is used as the negative output, and the negative electrode of the input matching resistor R1 is connected to the transistor M 12 The gate and neutralizing capacitor C C11 The cathode of transistor M 12 The source of transistor M is grounded. 12 The drain terminal is connected to the neutralizing capacitor C C12 anode of transistor M 12 The gate of the transistor M is used as the negative input of the driving amplifier stage. 12 The drain of the amplifier is used as the positive output of the driving amplifier stage; the power output stage includes a bias resistor R b1 , bias resistor R b2 , transistor M 21 , transistor M 22 , and the capacitor C C21 And the neutralizing capacitor C C22 , bias resistor R in the power output stage b1 The positive electrode is connected to transistor M 21 The gate and neutralizing capacitor C C22 The cathode of transistor M 21 The source of transistor M is grounded. 21 The drain terminal is connected to the neutralizing capacitor C C21 anode of transistor M 21 The gate of transistor M is used as the positive input of the power output stage. 21The drain is used as the negative output of the power output stage, and the bias resistor R b2 The negative terminal is connected to transistor M 22 The gate and neutralizing capacitor C C21 The cathode of transistor M 22 The source of transistor M is grounded. 22 The drain terminal is connected to the neutralizing capacitor C C22 anode of transistor M 22 The gate of transistor M is used as the negative input of the power output stage. 22 The drain of the power output stage is used as the positive output, and the bias resistor R b1 The negative terminal of the bias resistor R b2 The positive pole is used as the gate bias point of the power output stage. Both the drive amplifier stage and the power output stage adopt differential common source transistor amplification structure and neutralization capacitor technology, which can improve gain and stability. The output power of the drive amplifier stage is less than that of the power output stage. The power output stage gives priority to power output and high efficiency. The gain may decrease with the increase of frequency within the frequency band. There is no need to completely level it when designing the power stage. It can be leveled when designing the drive amplifier stage.

[0050] Embodiment 8: Based on Embodiment 1, the output matching network is optimally designed. The output matching network is a differential-to-single-ended passive structure, including an input capacitor C 31 , primary coupling coil L 31 , secondary coupling coil L 32 , the primary power supply center tap CT3 and the output capacitor C 32 , where the input capacitor C 31 Anode and primary coupling coil L 31 The positive terminal is connected as the differential input positive terminal of the output matching network, and the input capacitor C 31 Cathode and primary coupling coil L 31 The negative terminal is connected as the negative terminal of the differential input of the output matching network, and the output capacitor C 32 Anode and secondary coupling coil L 32 The positive terminal is connected as the single-ended output terminal of the output matching network, and the output capacitor C 32 The cathode is grounded, and the secondary coupling coil L 32 The negative terminal is grounded, and the center tap CT3 is connected from the primary coil L 31 The single-ended output of the output matching network is connected to the output pad.

[0051] The above embodiments only express the specific implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the protection scope of the present application. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the technical solution concept of the present application, and these all belong to the protection scope of the present application.

Claims

1. A millimeter-wave CMOS power amplifier with a flexible layout, comprising a CMOS power amplifier body, wherein the CMOS power amplifier body comprises an input matching network, a driving amplifier stage, an inter-stage matching network, a power output stage and an output matching network connected in sequence, wherein the input matching network, the inter-stage matching network and the output matching network all comprise primary-secondary coupling coils arranged in the same horizontal direction and partially overlapping with each other, characterized in that: Any coupling coil in the input matching network of the CMOS power amplifier body arranged in the outer channel or the primary-secondary coupling coil of the interstage matching network can be rotated to adjust its output direction; the RF input signal is converted to the driving amplifier stage for amplification through the input matching network, and then converted to the power output stage for amplification through the interstage matching network and then converted and output through the output matching network.

2. The millimeter-wave CMOS power amplifier with a flexible layout according to claim 1, characterized in that: The coupling coil with a larger radius in the rotating primary-secondary coupling coils in the input matching network and the inter-stage matching network is a regular polygon, and the rotating coupling coil rotates 90° around the center of the coupling coil which is a regular polygon.

3. The millimeter-wave CMOS power amplifier with a flexible layout according to claim 2, characterized in that: The coupling coil with a larger radius in the primary-secondary coupling coil is a square or a regular octagon.

4. The millimeter-wave CMOS power amplifier with a flexible layout according to claim 2, characterized in that: The primary and secondary coupling coils of the same matching network are on the same metal layer or on upper and lower metal layers respectively. When on the same metal layer, the crossover point of the primary and secondary coupling coils is located on the coupling coil with a smaller radius.

5. The millimeter-wave CMOS power amplifier with a flexible layout according to claim 1, characterized in that: The coupling mode of the primary-secondary coupling coils rotating in the input matching network and the inter-stage matching network is magnetic coupling, and the electrical coupling is parasitic coupling.

6. The millimeter-wave CMOS power amplifier with a flexible layout according to claim 1, characterized in that: The minimum electrical coupling of the primary-secondary coupling coils that rotate in the input matching network and the inter-stage matching network is obtained through their size, spacing, shape and coupling coefficient, and the corresponding primary-secondary coupling coils under the minimum electrical coupling condition must meet the impedance matching requirements of the front and rear stages and the network amplitude / phase of the corresponding primary-secondary coupling coils before and after rotation must meet the consistency requirements.

7. The millimeter-wave CMOS power amplifier with a flexible layout according to claim 1, characterized in that: The turns ratio of the primary-secondary coupling coils of the input matching network, the inter-stage matching network and the output matching network is determined according to the impedance transformation ratio of the front and rear stages and the two-port impedance matching conversion theory.

8. The millimeter-wave CMOS power amplifier with a flexible layout according to claim 1, characterized in that: The input end of the input matching network is connected to the output end of the previous module, and also includes an input capacitor C 11 , secondary power supply center tap CT1 and output capacitor C 12 , the primary-secondary coupling coil includes a primary coupling coil L 11 and the secondary coupling coil L 12 , the input capacitor C 11 The anode of the primary coupling coil L 11 The positive terminal of the input matching network is connected as the single-ended input terminal, and the input capacitor C 11 The cathode of the primary coupling coil L is grounded. 11 The negative terminal of the output capacitor C 12 The anode of the secondary coupling coil L 12 The positive terminal of the output capacitor C is connected as the differential output positive terminal of the input matching network. 12 The cathode of the secondary coupling coil L 12 The negative terminal of the output matching network is connected as the negative terminal of the differential output, and the center tap CT1 is connected from the secondary coil L 12 The inter-stage matching network also includes input capacitor C 21 , the primary power supply center tap CT2 and the output capacitor C 22 , the primary-secondary coupling coil includes a primary coupling coil L 21 and the secondary coupling coil L 22 , where the input capacitor C 21 The anode of the primary coupling coil L 21 The positive electrode is connected as the differential input positive terminal of the interstage matching network, and the input capacitor C 21 The cathode of the primary coupling coil L 21 The negative terminal of the output capacitor C is connected as the negative terminal of the differential input of the interstage matching network. 22 The anode of the secondary coupling coil L 22 The positive stage is connected as the differential output positive terminal of the inter-stage matching network, and the output capacitor C 22 The cathode of the secondary coupling coil L 22 The negative stage is connected as the negative terminal of the differential output of the interstage matching network, and the center tap CT2 is connected from the primary coil L 21 The input capacitor and the output capacitor of the input matching network and the inter-stage matching network are both implemented by MOM capacitors; the input matching network is a single-ended-to-differential passive structure, and the inter-stage matching network is a differential-to-differential passive structure, and the passive structures of the two are both implemented by a coupled resonant cavity based on the transformer principle.

9. The millimeter-wave CMOS power amplifier with a flexible layout according to claim 1, characterized in that: The driving amplifier stage includes an input matching resistor R1, a transistor M 11 , transistor M 12 , and the capacitor C C11 And the neutralizing capacitor C C12 , the positive electrode of the input matching resistor R1 is connected to the transistor M 11 The gate and neutralizing capacitor C C12 The cathode connection of transistor M 11 The source of transistor M is grounded. 11 The drain is connected to the neutralizing capacitor C C11 anode of transistor M 11 The gate of the transistor M is used as the positive input of the driving amplifier stage. 11 The drain of the drive amplifier stage is used as the negative output, and the negative electrode of the input matching resistor R1 is connected to the transistor M 12 The gate and neutralizing capacitor C C11 The cathode of transistor M 12 The source of transistor M is grounded. 12 The drain terminal is connected to the neutralizing capacitor C C12 anode of transistor M 12 The gate of the transistor M is used as the negative input of the driving amplifier stage. 12 The drain of the power amplifier stage is used as the positive output of the driving amplifier stage; the power output stage includes a bias resistor R b1 , bias resistor R b2 , transistor M 21 , transistor M 22 , and the capacitor C C21 And the neutralizing capacitor C C22 , bias resistor R in the power output stage b1 The positive electrode is connected to transistor M 21 The gate and neutralizing capacitor C C22 The cathode of transistor M 21 The source of transistor M is grounded. 21 The drain terminal is connected to the neutralizing capacitor C C21 anode of transistor M 21 The gate of transistor M is used as the positive input of the power output stage. 21 The drain is used as the negative output of the power output stage, and the bias resistor R b2 The negative terminal is connected to transistor M 22 The gate and neutralizing capacitor C C21 The cathode of transistor M 22 The source of transistor M is grounded. 22 The drain terminal is connected to the neutralizing capacitor C C22 anode of transistor M 22 The gate of transistor M is used as the negative input of the power output stage. 22 The drain of the power output stage is used as the positive output, and the bias resistor R b1 The negative terminal of the bias resistor R b2 The positive electrode is used as the gate bias point of the power output stage; the driving amplifier stage and the power output stage both adopt a differential common source transistor amplification structure and a neutralizing capacitor technology, and the output power of the driving amplifier stage is less than the output power of the power output stage.

10. The millimeter-wave CMOS power amplifier with a flexible layout according to claim 1, characterized in that: The output matching network is a differential-to-single-ended passive structure, including an input capacitor C 31 , primary coupling coil L 31 , secondary coupling coil L 32 , the primary power supply center tap CT3 and the output capacitor C 32 , where the input capacitor C 31 Anode and primary coupling coil L 31 The positive terminal is connected as the differential input positive terminal of the output matching network, and the input capacitor C 31 Cathode and primary coupling coil L 31 The negative terminal is connected as the negative terminal of the differential input of the output matching network, and the output capacitor C 32 Anode and secondary coupling coil L 32 The positive terminal is connected as the single-ended output terminal of the output matching network, and the output capacitor C 32 The cathode is grounded, and the secondary coupling coil L 32 The negative terminal is grounded, and the center tap CT3 is connected from the primary coil L 31 Lead out from the internal physical center.

Images