Hongzhi Liang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Hongzhi Liang has a total of 51 co-authors. Here are the top co-authors by number of shared publications:
- Shubin Liu
Xidian University
18 shared publications - Zhangming Zhu
Xidian University
18 shared publications - Ruixue Ding
Xidian University
10 shared publications - Dengquan Li
Xidian University
9 shared publications - Yi Shen
Xidian University
9 shared publications - Chenxi Han
Xidian University
5 shared publications
Latest Publications
- 1 October 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 34 (10) , 3137-3147
A 0.054-mm 2 Clock Generator With 26-GHz Double-Edge Sampling PLL and 7-bit Multifrequency PI Scoring 42.5-fs rms Jitter and −251.0-dB FoM J
- 1 October 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 34 (10) , 3137-3147
A 0.054-mm 2 Clock Generator With 26-GHz Double-Edge Sampling PLL and 7-bit Multifrequency PI Scoring 42.5-fs rms Jitter and −251.0-dB FoM J
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 73 (9) , 913-917
A 112 GS/s 7 b 64× Time-Interleaved Hybrid ADC Achieving 43.1-dB SFDR at 44.1-GHz Input in 28-nm CMOS
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 73 (9) , 913-917
A 112 GS/s 7 b 64× Time-Interleaved Hybrid ADC Achieving 43.1-dB SFDR at 44.1-GHz Input in 28-nm CMOS
- 17 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. PP (00189200) , 1-11
A 16× Interleaved 32-GS/s 8b Hybrid ADC With Self-Tracking Inter-Stage Gain Achieving 44.3-dB SFDR at 20.9-GHz Input
This article presents a 32-GS/s 16-channel hierarchical time-interleaved (TI) hybrid analog-to-digital converter (ADC). The prototype utilizes the intrinsic high-speed quantization of time-domain (TD) ADC to reduce the interleaving factor. By incorporating hierarchical sampling and a cascode sampler, the compact TI-ADC achieves 44.3-dB spurious-free dynamic range (SFDR) at 20.9-GHz input. The sub-ADC adopts a two-stage hybrid-domain architecture that combines a 3-bit successive approximation register (SAR) ADC with a 6-bit inverter-based opposite phase propagation (OPP) time-to-digital converter (TDC). Aprototype chip is fabricated in a 28-nm CMOS process, occupying a core area of 0.141mm2. At 32GS/s, it achieves a signal-to-noise and distortion ratio (SNDR) of 36dB with a Nyquist input while consuming 142.3mW, corresponding to a Walden figure-of-merit of 86.3fJ/conv-step. Meanwhile, a delay-locked loop (DLL)-based calibration loop is proposed to calibrate the full-scale quantization range variation introduced by PVT variations, together with an analog self-tracking inter-stage gain calibration loop. The measured SNDR varies by 1.84dB over a temperature range of $0~{^{\circ } }$ C to $100~{^{\circ } }$ C and fluctuates by 2.25dB with a ±10% supply variation.
- 17 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. PP (00189200) , 1-11
A 16× Interleaved 32-GS/s 8b Hybrid ADC With Self-Tracking Inter-Stage Gain Achieving 44.3-dB SFDR at 20.9-GHz Input
This article presents a 32-GS/s 16-channel hierarchical time-interleaved (TI) hybrid analog-to-digital converter (ADC). The prototype utilizes the intrinsic high-speed quantization of time-domain (TD) ADC to reduce the interleaving factor. By incorporating hierarchical sampling and a cascode sampler, the compact TI-ADC achieves 44.3-dB spurious-free dynamic range (SFDR) at 20.9-GHz input. The sub-ADC adopts a two-stage hybrid-domain architecture that combines a 3-bit successive approximation register (SAR) ADC with a 6-bit inverter-based opposite phase propagation (OPP) time-to-digital converter (TDC). Aprototype chip is fabricated in a 28-nm CMOS process, occupying a core area of 0.141mm2. At 32GS/s, it achieves a signal-to-noise and distortion ratio (SNDR) of 36dB with a Nyquist input while consuming 142.3mW, corresponding to a Walden figure-of-merit of 86.3fJ/conv-step. Meanwhile, a delay-locked loop (DLL)-based calibration loop is proposed to calibrate the full-scale quantization range variation introduced by PVT variations, together with an analog self-tracking inter-stage gain calibration loop. The measured SNDR varies by 1.84dB over a temperature range of $0~{^{\circ } }$ C to $100~{^{\circ } }$ C and fluctuates by 2.25dB with a ±10% supply variation.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- p. 1-14
A Reference-Less CDR Using SAR-Based Frequency-Acquisition Technique Achieving Constant Band-Searching Time and up to 63.64 Gb/s/ μ s Acquisition Speed
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- p. 1-14
A Reference-Less CDR Using SAR-Based Frequency-Acquisition Technique Achieving Constant Band-Searching Time and up to 63.64 Gb/s/ μ s Acquisition Speed
- 17 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. PP (15498328) , 1-10
A 65.5-dB SNDR 100-MS/s Pipelined-SAR ADC With Nested Negative-C and Dynamic Auto-Zeroing Residue Amplifier Achieving $\le$2.2-dB ΔSNDR Over −20 °C to 125 °C
This paper demonstrates a 12-bit 100-MS/s PVT-stabilized pipelined-SAR ADC with fully integrated on-chip calibration. Two key innovations facilitate over 65-dB SNDR with less than 2.2-dB variation from -20 °C to 125 °C: First, a PVT-stabilized nested negative-C topology boosts open-loop gain ( $G_{ol}$ ) of the residue amplifier (RA) by 28 dB, enabling one-time foreground calibration that avoids persistent inter-stage gain error tracking. Second, a dynamic auto-zeroing technique addresses offsets from both the RA and first-stage comparator through a switched-capacitor filter that extracts offsets based on the stochastic characteristics of residue signals. Dynamically level-shifted capacitors (DLSC) are embedded at the differential inputs to neutralize aliased noise and thermal noise, so that the offset drifts under high-temperature environment are effectively cancelled. Fabricated in 40-nm CMOS, the prototype occupies 0.22 mm2 and achieves 65.5-dB SNDR and 83-dB SFDR at Nyquist input while consuming 3.7 mW. Across six tested chips, the prototype design exhibits merely 2.2-dB SNDR degradation over -20 °C ~ 125 °C and 1.3-dB variation under ± 5 % supply voltage (1.1 V) fluctuations.
- 17 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. PP (15498328) , 1-10
A 65.5-dB SNDR 100-MS/s Pipelined-SAR ADC With Nested Negative-C and Dynamic Auto-Zeroing Residue Amplifier Achieving $\le$2.2-dB ΔSNDR Over −20 °C to 125 °C
This paper demonstrates a 12-bit 100-MS/s PVT-stabilized pipelined-SAR ADC with fully integrated on-chip calibration. Two key innovations facilitate over 65-dB SNDR with less than 2.2-dB variation from -20 °C to 125 °C: First, a PVT-stabilized nested negative-C topology boosts open-loop gain ( $G_{ol}$ ) of the residue amplifier (RA) by 28 dB, enabling one-time foreground calibration that avoids persistent inter-stage gain error tracking. Second, a dynamic auto-zeroing technique addresses offsets from both the RA and first-stage comparator through a switched-capacitor filter that extracts offsets based on the stochastic characteristics of residue signals. Dynamically level-shifted capacitors (DLSC) are embedded at the differential inputs to neutralize aliased noise and thermal noise, so that the offset drifts under high-temperature environment are effectively cancelled. Fabricated in 40-nm CMOS, the prototype occupies 0.22 mm2 and achieves 65.5-dB SNDR and 83-dB SFDR at Nyquist input while consuming 3.7 mW. Across six tested chips, the prototype design exhibits merely 2.2-dB SNDR degradation over -20 °C ~ 125 °C and 1.3-dB variation under ± 5 % supply voltage (1.1 V) fluctuations.
- 10 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 72 (12) , 1957-1961
FR-CNN: Frequency Recognition Convolutional Neural Network-Based Calibration for Timing Skew in TI-ADCs
This brief proposes a convolutional neural network (CNN)-based calibration method for timing skew in time-interleaved (TI) ADCs. Compared to previous neural network-based methods, this work employs a lightweight network for frequency recognition and then dynamically configures the structure of a subsequent CNN based on the recognition result. By adapting the calibration scheme to the input signal characteristics, the proposed method achieves high accuracy, reduces overfitting at low frequencies, and saves hardware cost. A 12-bit 3.6-GS/s TI-ADC test board with programmable timing skew is used to validate the method. Measurement results show that at Nyquist input, SNDR and SFDR are improved from 34.23 and 35.77 dB to 51.85 and 62.67 dB after calibration, respectively. Implemented on FPGA, the power consumption of the calibrator is 40 mW, where the frequency recognition network contributes 12.5%. The proposed method is applicable to various types of input signals, including single-tone, multi-tone, and modulated signals.
- 10 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 72 (12) , 1957-1961
FR-CNN: Frequency Recognition Convolutional Neural Network-Based Calibration for Timing Skew in TI-ADCs
This brief proposes a convolutional neural network (CNN)-based calibration method for timing skew in time-interleaved (TI) ADCs. Compared to previous neural network-based methods, this work employs a lightweight network for frequency recognition and then dynamically configures the structure of a subsequent CNN based on the recognition result. By adapting the calibration scheme to the input signal characteristics, the proposed method achieves high accuracy, reduces overfitting at low frequencies, and saves hardware cost. A 12-bit 3.6-GS/s TI-ADC test board with programmable timing skew is used to validate the method. Measurement results show that at Nyquist input, SNDR and SFDR are improved from 34.23 and 35.77 dB to 51.85 and 62.67 dB after calibration, respectively. Implemented on FPGA, the power consumption of the calibrator is 40 mW, where the frequency recognition network contributes 12.5%. The proposed method is applicable to various types of input signals, including single-tone, multi-tone, and modulated signals.
- 22 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Microwave Theory and Techniques
- Vol. 73 (4) , 2036-2049
A Configurable Multimode Nonuniform Eye-Opening Monitor for High-Speed Wireline Communication Achieving 3-μs EW/EH Evaluation and 0.98-R² Accuracy
This article presents a configurable multimode eye-opening monitor (EOM) with nonuniform sampling and quantization for an on-chip high-speed link built-in self-test (BIST). The EOM can operate in three modes, enabling it to not only capture the color-graded eye diagram but also rapidly outline its contour, eye height (EH), and eye width (EW). In the nonuniform scanning mode (NSM), the EH and EW are swiftly measured using an optimization algorithm to adjust the control codes of a phase interpolator (PI) and a digital-to-analog converter (DAC). In the fast-multisampling mode (FMSM), the reduced number of samples in each pixel facilitates a prompt generation of the eye contour. The EOM can also reconstruct the eye diagram in the multisampling mode (MSM) by analyzing the probability density function (PDF) of a 128 × 63 pixel array. Fabricated in a 28-nm CMOS technology, the EOM can operate up to 26 Gb/s within 0.005-mm2 area. At 26 Gb/s, it consumes 14.55 mW with more than 0.98-R2 accuracy and takes up 933 μs in the MSM, 597 μs in the FMSM, and ~3 μs in the NSM, respectively.
- 22 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Microwave Theory and Techniques
- Vol. 73 (4) , 2036-2049
A Configurable Multimode Nonuniform Eye-Opening Monitor for High-Speed Wireline Communication Achieving 3-μs EW/EH Evaluation and 0.98-R² Accuracy
This article presents a configurable multimode eye-opening monitor (EOM) with nonuniform sampling and quantization for an on-chip high-speed link built-in self-test (BIST). The EOM can operate in three modes, enabling it to not only capture the color-graded eye diagram but also rapidly outline its contour, eye height (EH), and eye width (EW). In the nonuniform scanning mode (NSM), the EH and EW are swiftly measured using an optimization algorithm to adjust the control codes of a phase interpolator (PI) and a digital-to-analog converter (DAC). In the fast-multisampling mode (FMSM), the reduced number of samples in each pixel facilitates a prompt generation of the eye contour. The EOM can also reconstruct the eye diagram in the multisampling mode (MSM) by analyzing the probability density function (PDF) of a 128 × 63 pixel array. Fabricated in a 28-nm CMOS technology, the EOM can operate up to 26 Gb/s within 0.005-mm2 area. At 26 Gb/s, it consumes 14.55 mW with more than 0.98-R2 accuracy and takes up 933 μs in the MSM, 597 μs in the FMSM, and ~3 μs in the NSM, respectively.