Yuhang Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yuhang Zhang has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Wenhua Chen
Tsinghua University
6 shared publications - Zhenghe Feng
Department of Electronic Engineering, Tsinghua University, Beijing, China
4 shared publications - Shun Wan
Department of Electronic Engineering, Tsinghua University, Beijing, China
3 shared publications - Shuyang Li
Department of Electronic Engineering, Tsinghua University, Beijing, China
2 shared publications - Xiaofan Chen
Beijing National Research Center for Information Science and Technology (BNRist), Tsinghua University, Beijing, China
2 shared publications - Donghao Li
Department of Electronic Engineering, Tsinghua University, Beijing, China
1 shared publication
Latest Publications
- 6 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Microwave Theory and Techniques
- Vol. 74 (4) , 3538-3550
A 2.2–5 GHz Differential AC-Stacked GaAs HBT Power Amplifier Using Compact Lumped Hybrid Three-Conductor Balun for Low-Voltage Handset Applications
This article presents a broadband power amplifier (PA) with a novel lumped hybrid three-conductor (LHTC) balun. The proposed LHTC balun employs an asymmetric lumped compensation network, which enhances amplitude and phase balance across a wide bandwidth. This design relaxes the stringent shielding requirements of conventional TC baluns and provides an inherent dc-blocking function. Furthermore, the adoption of lumped components enables a more compact layout suitable for sub-6 GHz applications, reducing the overall footprint by nearly 50%. These attributes collectively offer exceptional design flexibility and establish a new methodology for improving balance performance in TC baluns. Several important issues, including the impact of lumped components, the loss of practical inductors, size compression, and layout considerations, are also discussed. As a proof of concept, by combining the LHTC balun with an AC-stacked configuration, a differential PA is implemented in a GaAs HBT process for handset applications. This design overcomes the bandwidth limitations imposed by high impedance transformation ratios (ITRs) at low supply voltages. The fabricated PA achieves 33.8–35.2 dBm saturated output power with 42.4%–51.9% peak power added efficiency (PAE) from 2.2 to 5GHz with a chip size of $1.1\times 1.2$ mm. Under a 100 MHz 5G NR modulated signal with a 6-dB peak-to-average power ratio (PAPR), the PA achieves an adjacent channel power ratio (ACPR) of −30 dBc at average output powers of 26.2–28.8 dBm.
- 6 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Microwave Theory and Techniques
- Vol. 74 (4) , 3538-3550
A 2.2–5 GHz Differential AC-Stacked GaAs HBT Power Amplifier Using Compact Lumped Hybrid Three-Conductor Balun for Low-Voltage Handset Applications
This article presents a broadband power amplifier (PA) with a novel lumped hybrid three-conductor (LHTC) balun. The proposed LHTC balun employs an asymmetric lumped compensation network, which enhances amplitude and phase balance across a wide bandwidth. This design relaxes the stringent shielding requirements of conventional TC baluns and provides an inherent dc-blocking function. Furthermore, the adoption of lumped components enables a more compact layout suitable for sub-6 GHz applications, reducing the overall footprint by nearly 50%. These attributes collectively offer exceptional design flexibility and establish a new methodology for improving balance performance in TC baluns. Several important issues, including the impact of lumped components, the loss of practical inductors, size compression, and layout considerations, are also discussed. As a proof of concept, by combining the LHTC balun with an AC-stacked configuration, a differential PA is implemented in a GaAs HBT process for handset applications. This design overcomes the bandwidth limitations imposed by high impedance transformation ratios (ITRs) at low supply voltages. The fabricated PA achieves 33.8–35.2 dBm saturated output power with 42.4%–51.9% peak power added efficiency (PAE) from 2.2 to 5GHz with a chip size of $1.1\times 1.2$ mm. Under a 100 MHz 5G NR modulated signal with a 6-dB peak-to-average power ratio (PAPR), the PA achieves an adjacent channel power ratio (ACPR) of −30 dBc at average output powers of 26.2–28.8 dBm.
- 17 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Microwave Theory and Techniques
- Vol. 74 (3) , 2526-2539
A Power-Efficient 308-to-331-GHz 4 × 2 Radiator Array Achieving 13.6-dBm Chip-Only EIRP in 65-nm CMOS
In this article, a coherent power-efficient terahertz (THz) radiator array with eight coupled oscillator elements is presented. First, an efficiency-enhanced resonant tank based on the low- $k$ transformer is proposed, which can drive fundamental oscillation into the deep triode region while effectively reducing harmonic leakage between the gate and the drain. Depending on this resonant tank, a 2nd-harmonic oscillator is designed to generate high output power. In addition, a broadband dual-mode on-chip patch antenna array with U-shaped slots is proposed and optimized for chip-only high effective isotropic radiated power (EIRP). The harmonic oscillators finally use a ring-scalable coupling network to realize a 2-D array configuration for spatial power combining. The chip is fabricated by the TSMC 65-nm CMOS process, and the core area of the chip is $1.8\times 0.82$ mm. The measured tuning range of the proposed radiator array is 7.2% from 308.2 to 331.4 GHz. The maximum chip-only EIRP achieves 13.6 dBm with a peak radiated power ( ${P} _{\mathbf {rad}}$ ) of 1 dBm. The peak dc-to-RF efficiency achieves 0.493%.
- 17 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Microwave Theory and Techniques
- Vol. 74 (3) , 2526-2539
A Power-Efficient 308-to-331-GHz 4 × 2 Radiator Array Achieving 13.6-dBm Chip-Only EIRP in 65-nm CMOS
In this article, a coherent power-efficient terahertz (THz) radiator array with eight coupled oscillator elements is presented. First, an efficiency-enhanced resonant tank based on the low- $k$ transformer is proposed, which can drive fundamental oscillation into the deep triode region while effectively reducing harmonic leakage between the gate and the drain. Depending on this resonant tank, a 2nd-harmonic oscillator is designed to generate high output power. In addition, a broadband dual-mode on-chip patch antenna array with U-shaped slots is proposed and optimized for chip-only high effective isotropic radiated power (EIRP). The harmonic oscillators finally use a ring-scalable coupling network to realize a 2-D array configuration for spatial power combining. The chip is fabricated by the TSMC 65-nm CMOS process, and the core area of the chip is $1.8\times 0.82$ mm. The measured tuning range of the proposed radiator array is 7.2% from 308.2 to 331.4 GHz. The maximum chip-only EIRP achieves 13.6 dBm with a peak radiated power ( ${P} _{\mathbf {rad}}$ ) of 1 dBm. The peak dc-to-RF efficiency achieves 0.493%.
- 28 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Microwave and Wireless Technology Letters
- Vol. 35 (10) , 1554-1557
A Wideband Vector-Modulated Phase Shifter With Complementary Gain Control Topology
This letter presents the design and analysis of a wideband active vector-sum phase shifter in a 22-nm fully depleted silicon-on-insulator (FD-SOI) CMOS process. The phase shifter consists of an inductive fine-tuning polyphase filter (PPF) for wideband quadrature signal generation and an impedance-invariant variable gain amplifier (IIVGA) for vector modulation. By exploiting the complementary characteristics of transistors’ intrinsic parameters, the input and output impedances of the VGA can be kept almost constant during gain control. Meanwhile, the cross-coupled structure effectively neutralizes the Miller capacitance of the MOSFETs, thereby minimizing the phase variation. The measurement results of the prototype circuit reveal that the proposed phase shifter achieves an average gain of −1.3 dB, an rms gain error (GE) of less than 0.67 dB, and an rms phase error (PE) of 1°–2.5° over 24–38 GHz with single-frequency calibration.
- 28 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Microwave and Wireless Technology Letters
- Vol. 35 (10) , 1554-1557
A Wideband Vector-Modulated Phase Shifter With Complementary Gain Control Topology
This letter presents the design and analysis of a wideband active vector-sum phase shifter in a 22-nm fully depleted silicon-on-insulator (FD-SOI) CMOS process. The phase shifter consists of an inductive fine-tuning polyphase filter (PPF) for wideband quadrature signal generation and an impedance-invariant variable gain amplifier (IIVGA) for vector modulation. By exploiting the complementary characteristics of transistors’ intrinsic parameters, the input and output impedances of the VGA can be kept almost constant during gain control. Meanwhile, the cross-coupled structure effectively neutralizes the Miller capacitance of the MOSFETs, thereby minimizing the phase variation. The measurement results of the prototype circuit reveal that the proposed phase shifter achieves an average gain of −1.3 dB, an rms gain error (GE) of less than 0.67 dB, and an rms phase error (PE) of 1°–2.5° over 24–38 GHz with single-frequency calibration.
- 27 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Microwave Theory and Techniques
- Vol. 73 (8) , 4894-4904
A 6–18 GHz 13 W GaN Push–Pull Power Amplifier MMIC With Better Than −23 dBc Second Harmonic Suppression
Harmonics in broadband gallium nitride (GaN) power amplifiers (PAs) are usually quite strong, so additional filtering modules are required in practical applications, which degrade the system’s performance. Therefore, designing PAs with excellent harmonic suppression performance is of great significance. The push–pull architecture is a widely utilized technique for harmonic suppression. However, the design of a 6–18 GHz push–pull PA monolithic microwave integrated circuit (MMIC) poses significant challenges. This is attributed to the fact that on-chip broadband baluns typically exhibit high losses, which significantly degrade the PA’s performance. This article proposes a design method suitable for the 6–18 GHz high-performance push–pull PAs. Based on a 250-nm GaN HEMT technology, we design a three-line balun with a dissipation loss (DL) of less than 0.5 dB and utilize it in a 6–18 GHz 13 W push–pull PA MMIC. Under pulse-CW measurement conditions, the MMIC achieves a small gain of 19.7–23 dB, a saturated output power of 41.1–42.3 dBm, a saturated PAE of 18.8%–25.9% and a second harmonic suppression (HS2) of better than -23 dBc. To the best of the authors’ knowledge, this work is the first 6–18 GHz 10 W fully integrated GaN push–pull PA with good large-signal performance.
- 27 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Microwave Theory and Techniques
- Vol. 73 (8) , 4894-4904
A 6–18 GHz 13 W GaN Push–Pull Power Amplifier MMIC With Better Than −23 dBc Second Harmonic Suppression
Harmonics in broadband gallium nitride (GaN) power amplifiers (PAs) are usually quite strong, so additional filtering modules are required in practical applications, which degrade the system’s performance. Therefore, designing PAs with excellent harmonic suppression performance is of great significance. The push–pull architecture is a widely utilized technique for harmonic suppression. However, the design of a 6–18 GHz push–pull PA monolithic microwave integrated circuit (MMIC) poses significant challenges. This is attributed to the fact that on-chip broadband baluns typically exhibit high losses, which significantly degrade the PA’s performance. This article proposes a design method suitable for the 6–18 GHz high-performance push–pull PAs. Based on a 250-nm GaN HEMT technology, we design a three-line balun with a dissipation loss (DL) of less than 0.5 dB and utilize it in a 6–18 GHz 13 W push–pull PA MMIC. Under pulse-CW measurement conditions, the MMIC achieves a small gain of 19.7–23 dB, a saturated output power of 41.1–42.3 dBm, a saturated PAE of 18.8%–25.9% and a second harmonic suppression (HS2) of better than -23 dBc. To the best of the authors’ knowledge, this work is the first 6–18 GHz 10 W fully integrated GaN push–pull PA with good large-signal performance.
- 25 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Microwave and Wireless Technology Letters
- Vol. 35 (2) , 201-204
A 1.7–6.5-GHz 35-W Reactively Matched GaN Power Amplifier Using a Balance-Enhanced Power Distribution Network
In this letter, a 1.7–6.5-GHz reactively matched power amplifier (RMPA) is designed using a 0.2- $\mu$ m gallium nitride (GaN) high-electron mobility transistor (HEMT) process. To achieve an output power of more than 30 W, eight transistors are required in the power stage. The coupling effects and the asymmetry of the layout result in a significant imbalance among these transistors, which could deteriorate the performance of the power amplifier (PA). We propose a novel power distribution network (PDN) based on tightly coupled lines to mitigate the imbalance introduced by the compact interstage matching network (ISMN). The fabricated RMPA achieves an output power of 35 W and a PAE of 38%–50% under continuous-wave (CW) test.
- 25 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Microwave and Wireless Technology Letters
- Vol. 35 (2) , 201-204
A 1.7–6.5-GHz 35-W Reactively Matched GaN Power Amplifier Using a Balance-Enhanced Power Distribution Network
In this letter, a 1.7–6.5-GHz reactively matched power amplifier (RMPA) is designed using a 0.2- $\mu$ m gallium nitride (GaN) high-electron mobility transistor (HEMT) process. To achieve an output power of more than 30 W, eight transistors are required in the power stage. The coupling effects and the asymmetry of the layout result in a significant imbalance among these transistors, which could deteriorate the performance of the power amplifier (PA). We propose a novel power distribution network (PDN) based on tightly coupled lines to mitigate the imbalance introduced by the compact interstage matching network (ISMN). The fabricated RMPA achieves an output power of 35 W and a PAE of 38%–50% under continuous-wave (CW) test.
- 1 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Solid-State Circuits Letters
- Vol. 7, 307-310
Broadband GaN MMIC Doherty Power Amplifier Using Compact Short-Circuited Coupler
In this letter, a broadband GaN MMIC Doherty power amplifier using a compact short-circuited coupler (CSC) is presented. To enhance the bandwidth and reduce the size of integrated DPA, the conventional λ/2 transmission line in the peaking output matching network is replaced by the CSC structure. Detailed theoretical analysis and design procedures are provided. Based on the proposed solution, a 5.1 GHz-7.2 GHz DPA is designed using a 0.12-μm GaN HEMT process. The fractional bandwidth (FBW) is 34.1%. The measurement results show a saturated output power of 37.2-39 dBm and a 6 dB back-off DE of 38.4-50.5% across the design bands with a chip size of 2.6 × 2.6 mm. The adjacent channel power ratio (ACPR) under 100MHz single-carrier 64 QAM modulation signal with a 6dB peak-to-average power ratio (PAPR) excitation is better than -45dBc with digital predistortion (DPD).
- 1 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Solid-State Circuits Letters
- Vol. 7, 307-310
Broadband GaN MMIC Doherty Power Amplifier Using Compact Short-Circuited Coupler
In this letter, a broadband GaN MMIC Doherty power amplifier using a compact short-circuited coupler (CSC) is presented. To enhance the bandwidth and reduce the size of integrated DPA, the conventional λ/2 transmission line in the peaking output matching network is replaced by the CSC structure. Detailed theoretical analysis and design procedures are provided. Based on the proposed solution, a 5.1 GHz-7.2 GHz DPA is designed using a 0.12-μm GaN HEMT process. The fractional bandwidth (FBW) is 34.1%. The measurement results show a saturated output power of 37.2-39 dBm and a 6 dB back-off DE of 38.4-50.5% across the design bands with a chip size of 2.6 × 2.6 mm. The adjacent channel power ratio (ACPR) under 100MHz single-carrier 64 QAM modulation signal with a 6dB peak-to-average power ratio (PAPR) excitation is better than -45dBc with digital predistortion (DPD).