Huahuang Luo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Huahuang Luo has a total of 75 co-authors. Here are the top co-authors by number of shared publications:
- Yi-Kuen Lee
The Hong Kong University of Science and Technology
11 shared publications - Yuan Wang
University of Macau
8 shared publications Qingqing Ke
Sun Yat-sen University
5 shared publications- Wei Xu
State Key Laboratory of Radio Frequency Heterogeneous Integration, Shenzhen University, Shenzhen, China
5 shared publications - Michael Kraft
KU Leuven
4 shared publications - Mingzheng Duan
The Hong Kong University of Science and Technology
4 shared publications
Latest Publications
- Open Access
- 30 September 2026
- Published by Springer Nature in Research Square
Enhancing Performance of a Novel Piezoelectric MEMS Microphone Through Single Diaphragm Mode Localization via Piezoelectric Layer Segmentation
This study introduces a novel Aluminum Nitride (AlN) piezoelectric MEMS microphone featuring a piezoelectric layer segmentation comprising a circular, fully clamped membrane and multiple circular patterned electrodes. The innovative design achieves simultaneous multi-readout metrics without complex structures, size factors, or challenging manufacturing processes. By leveraging the unique single diaphragm mode-localization-like behavior of the architecture, the performance of the proposed subject is significantly enhanced. The proposed microphone was systematically characterized through electrical measurements and acoustic validation, demonstrating a distinct improvement in acoustic sensitivity compared to conventional counterparts. Baseline experiments over an acoustic pressure range of 0.16 Pa to 1.42 Pa show that the sensitivities measured by the outer electrodes of the proposed subject outperform those of the reference microphone, with the highest sensitivity reaching 66.74 ± 3.18 μV/Pa. Additionally, the sensitivities of the three inner electrodes were consistent across both devices, with the discrepancy being only 9.39 μV/Pa. The electrical characterization results reveal a weak coupling phenomenon in the first modal of the proposed device due to the engineered piezoelectric layer segmentation, facilitating a foundation for the mode-localization-like dynamics. As such, simultaneous multi-readout schemes can be garnered with respect to various acoustic disturbances, particularly encompassing infrasound regime (<1 Hz), i.e., amplitude modulation (AM), frequency modulation (FM), magnitude ratio (MR), frequency ratio (FR) and cross ratio (CR), in terms of both relative (Rel.) sensitivities and absolute (Abs.) sensitivities. Remarkably, a resolution of 16.24 nPa @ 1 Hz is achieved using the FR readout method, demonstrating a six-order-of-magnitude improvement over the AM metric. These findings underscore the effectiveness of piezoelectric layer segmentation in enhancing the performance of piezoelectric MEMS microphones without necessitating structural modifications or extra interface circuit and providing a pathway for design criteria of next-generation high-performance MEMS acoustic devices.
- 30 September 2026
- Published by Springer Nature in Research Square
Open AccessEnhancing Performance of a Novel Piezoelectric MEMS Microphone Through Single Diaphragm Mode Localization via Piezoelectric Layer Segmentation
This study introduces a novel Aluminum Nitride (AlN) piezoelectric MEMS microphone featuring a piezoelectric layer segmentation comprising a circular, fully clamped membrane and multiple circular patterned electrodes. The innovative design achieves simultaneous multi-readout metrics without complex structures, size factors, or challenging manufacturing processes. By leveraging the unique single diaphragm mode-localization-like behavior of the architecture, the performance of the proposed subject is significantly enhanced. The proposed microphone was systematically characterized through electrical measurements and acoustic validation, demonstrating a distinct improvement in acoustic sensitivity compared to conventional counterparts. Baseline experiments over an acoustic pressure range of 0.16 Pa to 1.42 Pa show that the sensitivities measured by the outer electrodes of the proposed subject outperform those of the reference microphone, with the highest sensitivity reaching 66.74 ± 3.18 μV/Pa. Additionally, the sensitivities of the three inner electrodes were consistent across both devices, with the discrepancy being only 9.39 μV/Pa. The electrical characterization results reveal a weak coupling phenomenon in the first modal of the proposed device due to the engineered piezoelectric layer segmentation, facilitating a foundation for the mode-localization-like dynamics. As such, simultaneous multi-readout schemes can be garnered with respect to various acoustic disturbances, particularly encompassing infrasound regime (<1 Hz), i.e., amplitude modulation (AM), frequency modulation (FM), magnitude ratio (MR), frequency ratio (FR) and cross ratio (CR), in terms of both relative (Rel.) sensitivities and absolute (Abs.) sensitivities. Remarkably, a resolution of 16.24 nPa @ 1 Hz is achieved using the FR readout method, demonstrating a six-order-of-magnitude improvement over the AM metric. These findings underscore the effectiveness of piezoelectric layer segmentation in enhancing the performance of piezoelectric MEMS microphones without necessitating structural modifications or extra interface circuit and providing a pathway for design criteria of next-generation high-performance MEMS acoustic devices.
- 16 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Microelectromechanical Systems
- Vol. 35 (3) , 421-428
Two-Dimensional Modeling for Parametric Optimization and Crosstalk Analysis of Dual-Axis Micro Thermal Convective Tilt Sensors
This paper presents a novel two-dimensional (2D) model based on partial differential equations (PDEs) to characterize the input-output response of the dual-axis micro thermal convective tilt ( $\mu $ TCT) sensors. The proposed model enables highly efficient performance analysis of $\mu $ TCT sensors with a computational time of 3 minutes. Based on the theoretical framework, the key parameters of the $\mu $ TCT sensor, including the heater-detector distance, sensor length, film thickness, and cavity depth, are optimized to achieve high sensitivity and low power consumption. Furthermore, the model provides a reliable method for quantifying cross-axis crosstalk relative to the sensitivity of the sensitive axis. Accordingly, a $\mu $ TCT sensor has been designed and fabricated, demonstrating bidirectional inclination detection over ±180° range with a power consumption of 3.29mW. Measured sensitivities along x-axis and y-axis were measured as 37.53 mV/° and 43.03 mV/°, respectively. Good agreement between experimental results and theoretical predictions validates the model’s accuracy and feasibility. With a normalized sensitivity of 13.07 mV/°/mW, the $\mu $ TCT sensor surpasses the previous micro tilt sensors by >3 times, while maintaining the cross-axis crosstalk error ( $\delta $ ) below 5%. This study establishes the developed 2D model as an indispensable tool for the design and optimization of high-performance dual-axis $\mu $ TCT sensors. The significant performance improvements highlight the model’s critical role in advancing micro tilt sensor technology for demanding applications such as high-precision attitude detection in industrial IoT systems. [2025-0218]
- 16 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Microelectromechanical Systems
- Vol. 35 (3) , 421-428
Two-Dimensional Modeling for Parametric Optimization and Crosstalk Analysis of Dual-Axis Micro Thermal Convective Tilt Sensors
This paper presents a novel two-dimensional (2D) model based on partial differential equations (PDEs) to characterize the input-output response of the dual-axis micro thermal convective tilt ( $\mu $ TCT) sensors. The proposed model enables highly efficient performance analysis of $\mu $ TCT sensors with a computational time of 3 minutes. Based on the theoretical framework, the key parameters of the $\mu $ TCT sensor, including the heater-detector distance, sensor length, film thickness, and cavity depth, are optimized to achieve high sensitivity and low power consumption. Furthermore, the model provides a reliable method for quantifying cross-axis crosstalk relative to the sensitivity of the sensitive axis. Accordingly, a $\mu $ TCT sensor has been designed and fabricated, demonstrating bidirectional inclination detection over ±180° range with a power consumption of 3.29mW. Measured sensitivities along x-axis and y-axis were measured as 37.53 mV/° and 43.03 mV/°, respectively. Good agreement between experimental results and theoretical predictions validates the model’s accuracy and feasibility. With a normalized sensitivity of 13.07 mV/°/mW, the $\mu $ TCT sensor surpasses the previous micro tilt sensors by >3 times, while maintaining the cross-axis crosstalk error ( $\delta $ ) below 5%. This study establishes the developed 2D model as an indispensable tool for the design and optimization of high-performance dual-axis $\mu $ TCT sensors. The significant performance improvements highlight the model’s critical role in advancing micro tilt sensor technology for demanding applications such as high-precision attitude detection in industrial IoT systems. [2025-0218]
- 12 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75 (00189456) , 1-8
A Single-Pulse Actuated Micro Flow Sensor With Thermal Feedback and Digitally Controlled Power Distribution
In this work, we present a single-pulse-driven micro flow sensor that features a wide measurable range, extended linearity, and fast response, which was realized by the synergistic integration of thermal feedback and digitally controlled power distribution across dual microheaters. Employing a comprehensive computational fluid dynamics (CFD) modeling approach, we have developed and validated a sensor architecture capable of measuring nitrogen flow velocities across an exceptionally wide range (>26.6m/s). The CMOS-MEMS fabricated device utilizes poly-silicon thermosensitive elements with an optimized heater-detector spacing of 3 μm. Experimental results demonstrate a high sensitivity of 1.9 %/(m/s) across flow rates ranging from 0 to 26.66 m/s, with excellent linearity (R2 = 0.99, obtained from least-squares regression) maintained within 0~4.44 m/s. Besides, the proposed system achieves a breakthrough thermal response time of 0.36 s, which is a 22× improvement over the previous dual-pulse TF-based designs. These advancements establish a new paradigm for high-performance flow sensing in dynamic environments, combining wide measurement range, excellent linearity, and rapid response in a single integrated platform.
- 12 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75 (00189456) , 1-8
A Single-Pulse Actuated Micro Flow Sensor With Thermal Feedback and Digitally Controlled Power Distribution
In this work, we present a single-pulse-driven micro flow sensor that features a wide measurable range, extended linearity, and fast response, which was realized by the synergistic integration of thermal feedback and digitally controlled power distribution across dual microheaters. Employing a comprehensive computational fluid dynamics (CFD) modeling approach, we have developed and validated a sensor architecture capable of measuring nitrogen flow velocities across an exceptionally wide range (>26.6m/s). The CMOS-MEMS fabricated device utilizes poly-silicon thermosensitive elements with an optimized heater-detector spacing of 3 μm. Experimental results demonstrate a high sensitivity of 1.9 %/(m/s) across flow rates ranging from 0 to 26.66 m/s, with excellent linearity (R2 = 0.99, obtained from least-squares regression) maintained within 0~4.44 m/s. Besides, the proposed system achieves a breakthrough thermal response time of 0.36 s, which is a 22× improvement over the previous dual-pulse TF-based designs. These advancements establish a new paradigm for high-performance flow sensing in dynamic environments, combining wide measurement range, excellent linearity, and rapid response in a single integrated platform.
- Open Access
- 9 February 2026
- Published by Springer Nature in Research Square
Real-time 3D imaging with 11×11 rigid PMUT array on conventional silicon wafer
Piezoelectric micromachined ultrasonic transducers (PMUTs) can push medical imaging forward by enabling smaller pitch sizes and reduced sidelobes for sharper resolution. However, earlier research on ultrasound imaging with PMUTs primarily used linear phased arrays, which are limited to 2D imaging, or row-column (RC) arrays, which present challenges for 3D imaging due to circuit complexity. Existing PMUT arrays for 3D imaging suffer from a complex fabrication process, raising yield and reliability concerns. Here, we demonstrate real-time 3D ultrasound imaging using an 11×11 PMUT array fabricated on a low-cost conventional silicon wafer with a polyimide membrane, a first in the literature. Operating at 2 MHz in tissue-mimicking media, an array element achieves 50 Pa/V output pressure at 2.5 cm and 5 µV/V round-trip sensitivity with 5 V DC bias. Up to 4 cm deep imaging results are recorded with a voltage as low as 10 VAC+10 VDC. Unlike prior SOI-based PMUTs, this approach drastically reduces fabrication expenses while maintaining performance for low-frequency imaging. The novel interconnect scheme enables simplified fabrication and packaging processes, opening a pathway for on-probe electronics. This technology offers a cost-effective solution for low-frequency real-time 3D ultrasound.
- 9 February 2026
- Published by Springer Nature in Research Square
Open AccessReal-time 3D imaging with 11×11 rigid PMUT array on conventional silicon wafer
Piezoelectric micromachined ultrasonic transducers (PMUTs) can push medical imaging forward by enabling smaller pitch sizes and reduced sidelobes for sharper resolution. However, earlier research on ultrasound imaging with PMUTs primarily used linear phased arrays, which are limited to 2D imaging, or row-column (RC) arrays, which present challenges for 3D imaging due to circuit complexity. Existing PMUT arrays for 3D imaging suffer from a complex fabrication process, raising yield and reliability concerns. Here, we demonstrate real-time 3D ultrasound imaging using an 11×11 PMUT array fabricated on a low-cost conventional silicon wafer with a polyimide membrane, a first in the literature. Operating at 2 MHz in tissue-mimicking media, an array element achieves 50 Pa/V output pressure at 2.5 cm and 5 µV/V round-trip sensitivity with 5 V DC bias. Up to 4 cm deep imaging results are recorded with a voltage as low as 10 VAC+10 VDC. Unlike prior SOI-based PMUTs, this approach drastically reduces fabrication expenses while maintaining performance for low-frequency imaging. The novel interconnect scheme enables simplified fabrication and packaging processes, opening a pathway for on-probe electronics. This technology offers a cost-effective solution for low-frequency real-time 3D ultrasound.
- 25 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-8
Substantial Performance Optimization of Thermal Expansion-Based Angular Motion Sensors With Parallel Heater-Temperature Detector Configuration
This paper presents a novel parallel heater-detector configuration (PaH) for thermal expansion-based angular motion (TEAM) sensors to achieve significant performance enhancement without increasing chip area or power consumption. Through the systematic analysis based on the extended 2D PDE model, we demonstrate that the novel PaH structure improves thermal working efficiency (η) by over 4X (26.9% vs. 5.3%), relative to the conventional perpendicular heater-temperature detector design (PeH), primarily by minimizing heat loss. The improved η directly contributes to a 68% sensitivity increase (0.4859 mV/°/s vs. 0.2893 mV/°/s). Theoretical and experimental analyses elucidate the underlying thermofluidic mechanisms, revealing accelerated dynamic response (thermal time constant τ* reduced by 24% to 13.6 ms) and elevated thermal resonance frequency (fr = 37 Hz vs. 28 Hz) for the PaH sensor. Consequently, the extended operable frequency range maximizes sensitivity enhancement through amplified periodic thermal flow generation. Furthermore, we establish a universal design criterion via Strouhal number normalization (St* = 0.5) for optimal driving frequency prediction (fPWM_Optimal = St*/τ*). Experimental validation under ±3600 °/s confirms the PaH sensor’s superiority, particularly with SF6 encapsulation where sensitivity reaches 3.2761 mV/°/s – 75% higher than the PeH counterpart. Thus, this work pioneers a high-performance paradigm for miniaturized thermal inertial sensors, enabling next-generation industrial IoT applications in precision robotics and autonomous navigation systems.
- 25 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-8
Substantial Performance Optimization of Thermal Expansion-Based Angular Motion Sensors With Parallel Heater-Temperature Detector Configuration
This paper presents a novel parallel heater-detector configuration (PaH) for thermal expansion-based angular motion (TEAM) sensors to achieve significant performance enhancement without increasing chip area or power consumption. Through the systematic analysis based on the extended 2D PDE model, we demonstrate that the novel PaH structure improves thermal working efficiency (η) by over 4X (26.9% vs. 5.3%), relative to the conventional perpendicular heater-temperature detector design (PeH), primarily by minimizing heat loss. The improved η directly contributes to a 68% sensitivity increase (0.4859 mV/°/s vs. 0.2893 mV/°/s). Theoretical and experimental analyses elucidate the underlying thermofluidic mechanisms, revealing accelerated dynamic response (thermal time constant τ* reduced by 24% to 13.6 ms) and elevated thermal resonance frequency (fr = 37 Hz vs. 28 Hz) for the PaH sensor. Consequently, the extended operable frequency range maximizes sensitivity enhancement through amplified periodic thermal flow generation. Furthermore, we establish a universal design criterion via Strouhal number normalization (St* = 0.5) for optimal driving frequency prediction (fPWM_Optimal = St*/τ*). Experimental validation under ±3600 °/s confirms the PaH sensor’s superiority, particularly with SF6 encapsulation where sensitivity reaches 3.2761 mV/°/s – 75% higher than the PeH counterpart. Thus, this work pioneers a high-performance paradigm for miniaturized thermal inertial sensors, enabling next-generation industrial IoT applications in precision robotics and autonomous navigation systems.
- 3 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Microelectromechanical Systems
- Vol. 35 (1) , 29-39
Decoupling of Temperature-Strain Signals in Surface Acoustic Wave Sensors Using Machine Learning Algorithms
Surface acoustic wave (SAW) sensors exhibit temperature-strain cross-sensitivity, which poses significant challenges for precise temperature and strain detection in complex environments. However, currently reported decoupling methods generally increase assembling complexity of SAW sensors, therefore failing to achieve ideal decoupling performance. In this work, we proposed a scheme of decoupling the temperature-strain signal of the SAW sensor using a machine learning model, which was verified by the measurement from our developed SAW sensor based on 128° Y-cut lithium niobate. This sensor simultaneously excited three SAW modes (first-order Rayleigh wave, leaky longitudinal wave, and second-order Rayleigh wave), demonstrating excellent temperature/strain sensing characteristics under 25- $120~^{\circ }$ C and 0- $800\mu \varepsilon $ loading conditions. Using temperature and strain as output variables, we extracted key features from the frequency response signals of the three acoustic wave modes and constructed six machine learning models for prediction. Results show that the polynomial regression (PR) model achieved optimal overall performance across all three modes. Specifically, for strain prediction using the first-order Rayleigh wave and leaky longitudinal wave, the model demonstrates normalized root mean square errors (NRMSE) of 0.0232 and 0.0113, mean absolute errors (MAE) of $12.29~\mu \varepsilon $ and 7.64 $\mu \varepsilon $ , and coefficients of determination (R2 ) of 99.63% and 99.91%, respectively. For temperature prediction, it shows NRMSE values of 0.0027 and 0.0024, MAE was $0.21~^{\circ }$ C for both cases, and R2 reached 99.99% and 99.9953% respectively, achieving ideal decoupling performance. This work provides a highly efficient and easily implementable solution for temperature-strain signal decoupling of the SAW sensors.[2025-0128]
- 3 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Microelectromechanical Systems
- Vol. 35 (1) , 29-39
Decoupling of Temperature-Strain Signals in Surface Acoustic Wave Sensors Using Machine Learning Algorithms
Surface acoustic wave (SAW) sensors exhibit temperature-strain cross-sensitivity, which poses significant challenges for precise temperature and strain detection in complex environments. However, currently reported decoupling methods generally increase assembling complexity of SAW sensors, therefore failing to achieve ideal decoupling performance. In this work, we proposed a scheme of decoupling the temperature-strain signal of the SAW sensor using a machine learning model, which was verified by the measurement from our developed SAW sensor based on 128° Y-cut lithium niobate. This sensor simultaneously excited three SAW modes (first-order Rayleigh wave, leaky longitudinal wave, and second-order Rayleigh wave), demonstrating excellent temperature/strain sensing characteristics under 25- $120~^{\circ }$ C and 0- $800\mu \varepsilon $ loading conditions. Using temperature and strain as output variables, we extracted key features from the frequency response signals of the three acoustic wave modes and constructed six machine learning models for prediction. Results show that the polynomial regression (PR) model achieved optimal overall performance across all three modes. Specifically, for strain prediction using the first-order Rayleigh wave and leaky longitudinal wave, the model demonstrates normalized root mean square errors (NRMSE) of 0.0232 and 0.0113, mean absolute errors (MAE) of $12.29~\mu \varepsilon $ and 7.64 $\mu \varepsilon $ , and coefficients of determination (R2 ) of 99.63% and 99.91%, respectively. For temperature prediction, it shows NRMSE values of 0.0027 and 0.0024, MAE was $0.21~^{\circ }$ C for both cases, and R2 reached 99.99% and 99.9953% respectively, achieving ideal decoupling performance. This work provides a highly efficient and easily implementable solution for temperature-strain signal decoupling of the SAW sensors.[2025-0128]