Fa Foster Dai
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Fa Foster Dai has a total of 40 co-authors. Here are the top co-authors by number of shared publications:
- Feng Zhao6 shared publications
- Dongyi Liao
Department of Electrical and Computer Engineering, Auburn University, Auburn, AL 36849 USA
3 shared publications - Hechen Wang
Intel Labs, Intel Corporation, Hillsboro, OR, USA
3 shared publications - Hua Wang
ETH Zurich
2 shared publications Yanjie Wang
2 shared publicationsBradley F. Dutton
Semiconductor company, Chandler
1 shared publication
Latest Publications
- 20 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Microwave and Wireless Technology Letters
- Vol. 34 (11) , 1275-1277
A 2.4-GHz Multiphase Inductorless PLL With Coupled-Ring Oscillators and Time-Amplifying Phase-Frequency Detector for Low Phase Noise and Robust Locking Performances
This letter presents an inductorless compact 12-phase integer-N phase-locked-loop (PLL) with time-amplifying phase–frequency detector (TAPFD) to achieve low in-band phase noise. A time amplifier with automatic gain control is proposed to provide a wide detectable range during acquisition and a high gain at lock condition. The charge-pump gain is also adaptively tuned to ensure that the overall loop gain is constant for robust operation. The PLL includes a 12-phase inverter-based coupled ring oscillator. Instead of placing all the delay cells in one ring, the multiple phase outputs are achieved by capacitive coupling of two identical ring oscillators. The proposed double-ring coupled ring oscillator provides additional output phases without scarifying the output frequency and phase noise. The inductorless PLL is implemented in 22-nm FDX CMOS technology with a core area of 0.0253 mm $^{2}$ and achieves up to 23-dB in-band phase noise improvement to $-$ 113.1 dBc/Hz and a measured integrated jitter of 1.274 ps at 2.4 GHz. The 12-phase ring PLL consumes 8.18 mW from a 0.8-V power supply and achieves a measured PLL FoM of 229 dB.
- 20 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Microwave and Wireless Technology Letters
- Vol. 34 (11) , 1275-1277
A 2.4-GHz Multiphase Inductorless PLL With Coupled-Ring Oscillators and Time-Amplifying Phase-Frequency Detector for Low Phase Noise and Robust Locking Performances
This letter presents an inductorless compact 12-phase integer-N phase-locked-loop (PLL) with time-amplifying phase–frequency detector (TAPFD) to achieve low in-band phase noise. A time amplifier with automatic gain control is proposed to provide a wide detectable range during acquisition and a high gain at lock condition. The charge-pump gain is also adaptively tuned to ensure that the overall loop gain is constant for robust operation. The PLL includes a 12-phase inverter-based coupled ring oscillator. Instead of placing all the delay cells in one ring, the multiple phase outputs are achieved by capacitive coupling of two identical ring oscillators. The proposed double-ring coupled ring oscillator provides additional output phases without scarifying the output frequency and phase noise. The inductorless PLL is implemented in 22-nm FDX CMOS technology with a core area of 0.0253 mm $^{2}$ and achieves up to 23-dB in-band phase noise improvement to $-$ 113.1 dBc/Hz and a measured integrated jitter of 1.274 ps at 2.4 GHz. The 12-phase ring PLL consumes 8.18 mW from a 0.8-V power supply and achieves a measured PLL FoM of 229 dB.
- 14 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 70 (8) , 4028-4035
Nonconducting RF and DC Hot Carrier Stresses in 14/16-nm FinFETs for RF Power Amplifiers
Hot carrier reliability under nonconducting (NC) RF and dc stresses is measured and modeled on 14/16-nm FinFETs used for RF PAs. The impact of stress on $\textbf{I}-\textbf{V}$ and RF parameters is examined. RF parameter stress response suggests that degradations are located near the drain end of the channel within the pinch-off region. For classic $\textit{V}_{\text{gs}}=\text{0}$ V OFF-state RF stress, quasi-static-approximation (QSA) significantly underestimates degradation, necessitating measurement-based lifetime modeling. At near-threshold $\textit{V}_{\text{gs}}$ , a condition of interest for our PAs, the degradation shows significant die-to-die variations dominated by variations of the subthreshold channel current that initiates the hot carriers. Modeling accounting for the subthreshold channel current variations shows that the near-threshold RF stress is approximately quasi-static. The results show that these FinFETs provide enough margins against NC RF stress for the intended PA applications.
- 14 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 70 (8) , 4028-4035
Nonconducting RF and DC Hot Carrier Stresses in 14/16-nm FinFETs for RF Power Amplifiers
Hot carrier reliability under nonconducting (NC) RF and dc stresses is measured and modeled on 14/16-nm FinFETs used for RF PAs. The impact of stress on $\textbf{I}-\textbf{V}$ and RF parameters is examined. RF parameter stress response suggests that degradations are located near the drain end of the channel within the pinch-off region. For classic $\textit{V}_{\text{gs}}=\text{0}$ V OFF-state RF stress, quasi-static-approximation (QSA) significantly underestimates degradation, necessitating measurement-based lifetime modeling. At near-threshold $\textit{V}_{\text{gs}}$ , a condition of interest for our PAs, the degradation shows significant die-to-die variations dominated by variations of the subthreshold channel current that initiates the hot carriers. Modeling accounting for the subthreshold channel current variations shows that the near-threshold RF stress is approximately quasi-static. The results show that these FinFETs provide enough margins against NC RF stress for the intended PA applications.
- 19 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 58 (10) , 2883-2896
A 12-Bit 260-MS/s Pipelined-SAR ADC With Ring-TDC-Based Fine Quantizer for Automatic Cross-Domain Scale Alignment
This article presents a power efficient and process, voltage, and temperature (PVT) robust pipelined successive approximation register (SAR) analog-to-digital converter (ADC) that quantizes signals in both voltage and time domains. In this work, a low-power SAR ADC is adopted as the coarse quantizer, while a ring-configured time-to-digital converter (TDC) is utilized in the fine quantizer to improve the linearity and power efficiency. In addition, the ring TDC also participates in the voltage-to-time conversion to guarantee that one-lap delay in the ring TDC is aligned with the least quantization step in the voltage domain. As a result, an auto-scale alignment between voltage and time domains is promised regardless of PVT variations. The ADC prototype IC was fabricated in a 22-nm CMOS technology. When measured at 260 MS/s, the ADC achieves 60.5-dB signal-to-noise and distortion ratio (SNDR) and 77-dB spurious-free dynamic range (SFDR) with a Nyquist input, while consuming 0.97 mW from a 0.8-V power supply. The calculated Walden and Schreier figures-of-merit (FoMs) are 4.27 fJ/conversion step and 171.8 dB, respectively.
- 19 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 58 (10) , 2883-2896
A 12-Bit 260-MS/s Pipelined-SAR ADC With Ring-TDC-Based Fine Quantizer for Automatic Cross-Domain Scale Alignment
This article presents a power efficient and process, voltage, and temperature (PVT) robust pipelined successive approximation register (SAR) analog-to-digital converter (ADC) that quantizes signals in both voltage and time domains. In this work, a low-power SAR ADC is adopted as the coarse quantizer, while a ring-configured time-to-digital converter (TDC) is utilized in the fine quantizer to improve the linearity and power efficiency. In addition, the ring TDC also participates in the voltage-to-time conversion to guarantee that one-lap delay in the ring TDC is aligned with the least quantization step in the voltage domain. As a result, an auto-scale alignment between voltage and time domains is promised regardless of PVT variations. The ADC prototype IC was fabricated in a 22-nm CMOS technology. When measured at 260 MS/s, the ADC achieves 60.5-dB signal-to-noise and distortion ratio (SNDR) and 77-dB spurious-free dynamic range (SFDR) with a Nyquist input, while consuming 0.97 mW from a 0.8-V power supply. The calculated Walden and Schreier figures-of-merit (FoMs) are 4.27 fJ/conversion step and 171.8 dB, respectively.
- 3 May 2021
- Published by Springer Nature in Analog Integrated Circuits and Signal Processing
- Vol. 108 (2) , 229-239
A dual-residue pipelined SAR ADC using only zero-crossing signals
- 3 May 2021
- Published by Springer Nature in Analog Integrated Circuits and Signal Processing
- Vol. 108 (2) , 229-239
A dual-residue pipelined SAR ADC using only zero-crossing signals
- 6 November 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 56 (3) , 694-704
A Fractional-N Reference Sampling PLL With Linear Sampler and CDAC Based Fractional Spur Cancellation
In this article, a fractional-N reference sampling phase-locked loop (PLL) (RSPLL) is presented. A capacitor-based digital-to-analog converter (CDAC) is implemented at the output of the reference sampling phase detector (RSPD) to cancel the divider quantization error in fractional mode. The RSPD is linearized by maintaining a constant discharge current from the sampling capacitor. Although the discharging current source may induce some noise compared to a sampling PD. However, its noise can be effectively suppressed with the high-gain setting for the RSPD when the loop is in lock. The linear range of RSPD can be programmed to cover the quantization error in a frac-N mode without degrading the PLL in-band phase noise. To mitigate the nonlinearity of RSPD, the CDAC is implemented with a high-order cancellation scheme using only one capacitor array but multiple reference voltages. The prototype chip was fabricated in a 45-nm partially depleted silicon-on-insulator (PDSOI) CMOS process. The measurement showed an output frequency range covering 7.7-9.1 GHz with an integrated jitter (10 kHz-50 MHz) of 135 fs and an in-band fractional spur level of -55 dBc at an offset frequency of 50 kHz. The entire PLL consumes 4.5 mW and achieves an figure of merit (FoM) of -250.8 dB.
- 6 November 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 56 (3) , 694-704
A Fractional-N Reference Sampling PLL With Linear Sampler and CDAC Based Fractional Spur Cancellation
In this article, a fractional-N reference sampling phase-locked loop (PLL) (RSPLL) is presented. A capacitor-based digital-to-analog converter (CDAC) is implemented at the output of the reference sampling phase detector (RSPD) to cancel the divider quantization error in fractional mode. The RSPD is linearized by maintaining a constant discharge current from the sampling capacitor. Although the discharging current source may induce some noise compared to a sampling PD. However, its noise can be effectively suppressed with the high-gain setting for the RSPD when the loop is in lock. The linear range of RSPD can be programmed to cover the quantization error in a frac-N mode without degrading the PLL in-band phase noise. To mitigate the nonlinearity of RSPD, the CDAC is implemented with a high-order cancellation scheme using only one capacitor array but multiple reference voltages. The prototype chip was fabricated in a 45-nm partially depleted silicon-on-insulator (PDSOI) CMOS process. The measurement showed an output frequency range covering 7.7-9.1 GHz with an integrated jitter (10 kHz-50 MHz) of 135 fs and an in-band fractional spur level of -55 dBc at an offset frequency of 50 kHz. The entire PLL consumes 4.5 mW and achieves an figure of merit (FoM) of -250.8 dB.
- 1 April 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 55 (4) , 1064-1076
Sub-Sampling Direct RF-to-Digital Converter With 1024-APSK Modulation for High Throughput Polar Receiver
- 1 April 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 55 (4) , 1064-1076
Sub-Sampling Direct RF-to-Digital Converter With 1024-APSK Modulation for High Throughput Polar Receiver
- 10 January 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 55 (3) , 536-546
An mm-Wave Synthesizer With Robust Locking Reference-Sampling PLL and Wide-Range Injection-Locked VCO
In this article, a two-stage millimeter (mm)-wave frequency synthesizer with low in-band noise and robust locking reference-sampling techniques is presented. Using a two-stage scheme allows separately dealing with the low phase noise (PN) frequency synthesis in the first stage and the mm-wave frequency multiplication in the second stage, achieving the best overall power efficiency. In the first stage, a voltage domain reference-sampling phase detector (RSPD)-locked loop (RSPLL) is adopted to achieve both low PN and robust locking without additional frequency locking loop. A reference reshaping buffer is implemented to improve the phase detector gain and in-band PN. The reference rising/falling time is programmable to achieve optimal RSPLL performance even under external disturbances. The second stage employs an injection-locked voltage-controlled oscillator (ILVCO) for 4x frequency multiplication. A low-power digital frequency tracking loop (FTL) detecting actual frequency errors is implemented in order to achieve wide operation range for the ILVCO while using a high Q tank with low power. The prototype synthesizer was fabricated in a 45-nm partially depleted silicon on insulator (PDSOI) CMOS technology. The first stage 9-GHz RSPLL achieves 144-fs integrated jitter with 7.2-mW power consumption, achieving a figure of merit (FoM) of -248 dB and the overall mm-wave synthesizer achieves 251-fs integrated jitter with 20.6-mW power consumption at 35.84 GHz, achieving an FoM of -238.9 dB.
- 10 January 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 55 (3) , 536-546
An mm-Wave Synthesizer With Robust Locking Reference-Sampling PLL and Wide-Range Injection-Locked VCO
In this article, a two-stage millimeter (mm)-wave frequency synthesizer with low in-band noise and robust locking reference-sampling techniques is presented. Using a two-stage scheme allows separately dealing with the low phase noise (PN) frequency synthesis in the first stage and the mm-wave frequency multiplication in the second stage, achieving the best overall power efficiency. In the first stage, a voltage domain reference-sampling phase detector (RSPD)-locked loop (RSPLL) is adopted to achieve both low PN and robust locking without additional frequency locking loop. A reference reshaping buffer is implemented to improve the phase detector gain and in-band PN. The reference rising/falling time is programmable to achieve optimal RSPLL performance even under external disturbances. The second stage employs an injection-locked voltage-controlled oscillator (ILVCO) for 4x frequency multiplication. A low-power digital frequency tracking loop (FTL) detecting actual frequency errors is implemented in order to achieve wide operation range for the ILVCO while using a high Q tank with low power. The prototype synthesizer was fabricated in a 45-nm partially depleted silicon on insulator (PDSOI) CMOS technology. The first stage 9-GHz RSPLL achieves 144-fs integrated jitter with 7.2-mW power consumption, achieving a figure of merit (FoM) of -248 dB and the overall mm-wave synthesizer achieves 251-fs integrated jitter with 20.6-mW power consumption at 35.84 GHz, achieving an FoM of -238.9 dB.
- 8 May 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 53 (9) , 2454-2464
A Multi-Phase Coupled Oscillator Using Inductive Resonant Coupling and Modified Dual-Tank Techniques
This paper presents a novel coupled oscillator RFIC for multi-phase clock generation. The design achieves low phase noise while maintaining strong coupling among oscillator cores. The proposed transformer-based dual-tank topology forms a loop of coupling path for enhanced multi-phase coupling. To facilitate strong magnetic coupling among oscillator cores, the proposed transformers utilize resonant inductive coupling to enhance the voltage swing at the coupled wing. The phase noise optimization is accomplished by leveraging the adaptive biasing feedback and the enhanced dual-tank technique. The behavioral model based on Alder's equation as well as the analysis of the tank response is given. Full electromagnetic (EM) modeling involving all transformers and the passive interconnecting routes has been performed using the EMX software in order to ensure that simulated performance reflects measurement environment. The prototype RFIC of the proposed circuit with eight phases is implemented in a 65-nm CMOS radio frequency (RF) silicon on insulator technology. The measured phase noise is -124.3 dBc/Hz at 1-MHz offset from 2.41 GHz with 7-mW power consumption per core, and the operating frequency can be digitally tuned from 1.81 to 2.41 GHz. The phase noise can be further improved to -128.2 dBc/Hz at 1-MHz offset from 2.41 GHz by increasing the power consumption to 15 mW per core.
- 8 May 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 53 (9) , 2454-2464
A Multi-Phase Coupled Oscillator Using Inductive Resonant Coupling and Modified Dual-Tank Techniques
This paper presents a novel coupled oscillator RFIC for multi-phase clock generation. The design achieves low phase noise while maintaining strong coupling among oscillator cores. The proposed transformer-based dual-tank topology forms a loop of coupling path for enhanced multi-phase coupling. To facilitate strong magnetic coupling among oscillator cores, the proposed transformers utilize resonant inductive coupling to enhance the voltage swing at the coupled wing. The phase noise optimization is accomplished by leveraging the adaptive biasing feedback and the enhanced dual-tank technique. The behavioral model based on Alder's equation as well as the analysis of the tank response is given. Full electromagnetic (EM) modeling involving all transformers and the passive interconnecting routes has been performed using the EMX software in order to ensure that simulated performance reflects measurement environment. The prototype RFIC of the proposed circuit with eight phases is implemented in a 65-nm CMOS radio frequency (RF) silicon on insulator technology. The measured phase noise is -124.3 dBc/Hz at 1-MHz offset from 2.41 GHz with 7-mW power consumption per core, and the operating frequency can be digitally tuned from 1.81 to 2.41 GHz. The phase noise can be further improved to -128.2 dBc/Hz at 1-MHz offset from 2.41 GHz by increasing the power consumption to 15 mW per core.
- 1 April 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
A bidirectional lens-free digital-bits-in/-out 0.57mm2 terahertz nano-radio in CMOS with 49.3mW Peak power consumption supporting 50cm Internet-of-Things communication
The following topics are dealt with: CMOS integrated circuits; low-power electronics; analogue-digital conversion; calibration; voltage-controlled oscillators; integrated circuit design; delta-sigma modulation; phase noise; oscillators; radio transceivers.
- 1 April 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
A bidirectional lens-free digital-bits-in/-out 0.57mm2 terahertz nano-radio in CMOS with 49.3mW Peak power consumption supporting 50cm Internet-of-Things communication
The following topics are dealt with: CMOS integrated circuits; low-power electronics; analogue-digital conversion; calibration; voltage-controlled oscillators; integrated circuit design; delta-sigma modulation; phase noise; oscillators; radio transceivers.
- 1 March 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 53 (3) , 715-727
A 2.4-GHz 16-Phase Sub-Sampling Fractional-N PLL With Robust Soft Loop Switching
A 16-modulo fractional-N sub-sampling phase-locked loop (SSPLL) with a quadrature voltage-controlled oscillator (VCO) interpolating 16 output phases is presented in this paper. Automatic soft switching between the sub-sampling phase control loop and the frequency control loop is proposed to improve loop robustness against perturbations and interferences, achieving more stable loop dynamics for a larger range of phase errors compared with prior art SSPLL designs. A capacitive phase interpolation network is implemented for 16-phase clock generation starting from quadrature phases. The 16 phases are further utilized to achieve fractional-N operation with a sub-sampling phase detector. This passive phase interpolation at the VCO frequency introduces no extra noise or power and avoids in-band phase noise degradation for fractional-N mode. Implemented in a 130-nm CMOS technology, the SSPLL chip achieves a measured in-band phase noise of -120 dBc/Hz and a measured integrated jitter of 158 fs at 2.4 GHz, while consuming 21 mW with 16 output phases. The measured reference spur and fractional spur levels are -72 and -52 dBc, respectively.
- 1 March 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 53 (3) , 715-727
A 2.4-GHz 16-Phase Sub-Sampling Fractional-N PLL With Robust Soft Loop Switching
A 16-modulo fractional-N sub-sampling phase-locked loop (SSPLL) with a quadrature voltage-controlled oscillator (VCO) interpolating 16 output phases is presented in this paper. Automatic soft switching between the sub-sampling phase control loop and the frequency control loop is proposed to improve loop robustness against perturbations and interferences, achieving more stable loop dynamics for a larger range of phase errors compared with prior art SSPLL designs. A capacitive phase interpolation network is implemented for 16-phase clock generation starting from quadrature phases. The 16 phases are further utilized to achieve fractional-N operation with a sub-sampling phase detector. This passive phase interpolation at the VCO frequency introduces no extra noise or power and avoids in-band phase noise degradation for fractional-N mode. Implemented in a 130-nm CMOS technology, the SSPLL chip achieves a measured in-band phase noise of -120 dBc/Hz and a measured integrated jitter of 158 fs at 2.4 GHz, while consuming 21 mW with 16 output phases. The measured reference spur and fractional spur levels are -72 and -52 dBc, respectively.