Chuxiong Lin
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Chuxiong Lin has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Weifeng He
Shanghai Jiao Tong University
9 shared publications - Yuxuan Qin
Shanghai Jiao Tong University
5 shared publications - Mingoo Seok
Columbia University
4 shared publications - Jieyu Li
Department of Micro-Nano Electronics, Shanghai Jiao Tong University, Shanghai, China
3 shared publications - Yanan Sun
Shanghai Jiao Tong University
3 shared publications - Bo Zhang
Electrical Engineering, Columbia University, New York, New York, USA
2 shared publications
Latest Publications
- Open Access
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
DRCA: Reliable Bitwise Logic in DRAM via Dual-Rail Compute and Access
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessDRCA: Reliable Bitwise Logic in DRAM via Dual-Rail Compute and Access
- 3 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 75 (1) , 350-364
TPDA-DRAM: A Variation-Aware DRAM Improving System Performance via In-Situ Timing Margin Detection and Adaptive Mitigation
DRAM latency remains a critical bottleneck in the performance of modern computing systems. However, the latency is excessively conservative due to the timing margins imposed by DRAM vendors to accommodate rare worst-case scenarios, such as weak cells and high temperatures. In this study, we introduce a temperature- and process-variation-aware timing detection and adaptation DRAM (TPDA-DRAM) architecture that dynamically mitigates timing margins at runtime. TPDA-DRAM leverages innovative in-situ cross-coupled detectors to monitor voltage differences between bitline pairs inside DRAM arrays, ensuring precise detection of timing margins. Additionally, the proposed detector inherently accelerates the precharge operation of DRAM, thereby reducing the precharge latency by up to 62.5%. Building upon this architecture, we propose two variation-aware timing adaptation schemes: 1) a process-variation-aware adaptation (PVA) scheme that accelerates access to weak cells, mitigating process-induced timing margins, and 2) a temperature-variation-aware adaptation (TVA) scheme that leverages temperature information and the restoration truncation technique to reduce DRAM latency, mitigating temperature-induced timing margins. Evaluations on an eight-core computing system show that TPDA-DRAM improves average performance by 21.8% and energy efficiency by 18.2%.
- 3 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 75 (1) , 350-364
TPDA-DRAM: A Variation-Aware DRAM Improving System Performance via In-Situ Timing Margin Detection and Adaptive Mitigation
DRAM latency remains a critical bottleneck in the performance of modern computing systems. However, the latency is excessively conservative due to the timing margins imposed by DRAM vendors to accommodate rare worst-case scenarios, such as weak cells and high temperatures. In this study, we introduce a temperature- and process-variation-aware timing detection and adaptation DRAM (TPDA-DRAM) architecture that dynamically mitigates timing margins at runtime. TPDA-DRAM leverages innovative in-situ cross-coupled detectors to monitor voltage differences between bitline pairs inside DRAM arrays, ensuring precise detection of timing margins. Additionally, the proposed detector inherently accelerates the precharge operation of DRAM, thereby reducing the precharge latency by up to 62.5%. Building upon this architecture, we propose two variation-aware timing adaptation schemes: 1) a process-variation-aware adaptation (PVA) scheme that accelerates access to weak cells, mitigating process-induced timing margins, and 2) a temperature-variation-aware adaptation (TVA) scheme that leverages temperature information and the restoration truncation technique to reduce DRAM latency, mitigating temperature-induced timing margins. Evaluations on an eight-core computing system show that TPDA-DRAM improves average performance by 21.8% and energy efficiency by 18.2%.
- Open Access
- 29 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 292-297
ASF-CRB: an Energy-Efficient Activity-Balanced Charge-Recycling Bus Architecture Based on Alternate Signal Flipping
- 29 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 292-297
Open AccessASF-CRB: an Energy-Efficient Activity-Balanced Charge-Recycling Bus Architecture Based on Alternate Signal Flipping
- 9 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 44 (12) , 4740-4751
An Ultraspeed Middle Voltage and Timing Analyzer With Near-SPICE Accuracy for Charge-Recycling Buses
By stacking two data channels between$V_{DD}$ and$V_{SS}$ , charge recycling buses (CRBs) halve the voltage swing on interconnects, achieving significant power savings for energy-efficient on-chip data transmission. However, the middle voltage ($V_{MID}$ ) between the two channels may fluctuate dynamically due to the diversity of input data, which significantly impacts data propagation delay and reliability. Unfortunately, existing SPICE-based simulators do not run fast enough to identify the worst-case$V_{MID}$ fluctuation and propagation delay, posing great challenges in CRB design. In this article, we present a dedicated CRB Simulator for fast and accurate$V_{MID}$ and timing analysis. A highly condensed$V_{MID}$ fluctuation model, which integrates each cycle’s$V_{MID}$ changes into a single closed-form formula, is embedded in the simulator to predict$V_{MID}$ values at clock edges. In addition, a speed-monitoring algorithm is developed to track the continuously changing signal propagation speed under intracycle$V_{MID}$ fluctuations for accurate delay estimation. Both the$V_{MID}$ fluctuation model and the delay estimation algorithm involve only a small number of arithmetic operations, thus featuring remarkably low computational complexity. Compared with HSPICE, our CRB Simulator runs >$1.0\times 10^{5} $ times faster on average across various CRB circuits, with a$V_{MID}$ prediction error of only 1.3 mV and a delay estimation error as low as 0.6%. The significant speed improvement combined with high accuracy makes our CRB Simulator an efficient and reliable solution for$V_{MID}$ and timing analysis in CRB circuit design.- 9 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 44 (12) , 4740-4751
An Ultraspeed Middle Voltage and Timing Analyzer With Near-SPICE Accuracy for Charge-Recycling Buses
By stacking two data channels between$V_{DD}$ and$V_{SS}$ , charge recycling buses (CRBs) halve the voltage swing on interconnects, achieving significant power savings for energy-efficient on-chip data transmission. However, the middle voltage ($V_{MID}$ ) between the two channels may fluctuate dynamically due to the diversity of input data, which significantly impacts data propagation delay and reliability. Unfortunately, existing SPICE-based simulators do not run fast enough to identify the worst-case$V_{MID}$ fluctuation and propagation delay, posing great challenges in CRB design. In this article, we present a dedicated CRB Simulator for fast and accurate$V_{MID}$ and timing analysis. A highly condensed$V_{MID}$ fluctuation model, which integrates each cycle’s$V_{MID}$ changes into a single closed-form formula, is embedded in the simulator to predict$V_{MID}$ values at clock edges. In addition, a speed-monitoring algorithm is developed to track the continuously changing signal propagation speed under intracycle$V_{MID}$ fluctuations for accurate delay estimation. Both the$V_{MID}$ fluctuation model and the delay estimation algorithm involve only a small number of arithmetic operations, thus featuring remarkably low computational complexity. Compared with HSPICE, our CRB Simulator runs >$1.0\times 10^{5} $ times faster on average across various CRB circuits, with a$V_{MID}$ prediction error of only 1.3 mV and a delay estimation error as low as 0.6%. The significant speed improvement combined with high accuracy makes our CRB Simulator an efficient and reliable solution for$V_{MID}$ and timing analysis in CRB circuit design.- 2 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 60 (5) , 1805-1817
A 394-TOPS/W Matched Filter With Charge-Domain Computing for GPS Signal Acquisition
As the most compute-intensive signal processing task in the global positioning system (GPS) receiver, signal acquisition accounts for most of the power budget, which is a key bottleneck in extending the battery life of edge devices. To improve the energy efficiency of GPS signal acquisition, we present charge-domain computing for GPS signal acquisition (CCSA), a matched filter with CCSA. It employs a $64 {\times } 64$ mixed-signal processing element (PE) array, a 2.8-b flash analog-to-digital converter (ADC), and a digital controller. Each PE handles the element-wise multiplication with a mixed-signal multiplier (MSMUL), which can perform the multiplication and digital-to-charge conversion simultaneously with a two-phase operation scheme. We develop the two-phase digital-to-charge mapping method for the MSMUL, where the digital multiplication result is converted into the charge difference stored in a capacitor between the charge reset phase and the charge evaluation phase. The charge-domain outputs of all MSMULs are dumped and redistributed on a global output wire for the accumulation operation. Compared to prior charge-domain works, the proposed design reduces the number of capacitors or additional voltage levels by half, thereby leading to better energy efficiency and lower circuit complexity. The 28-nm test chip achieves 114–394-TOPS/W energy efficiency across 0.34–0.9-V supply voltage, which is $8.2{\times }$ higher than state-of-the-art at the same throughput.
- 2 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 60 (5) , 1805-1817
A 394-TOPS/W Matched Filter With Charge-Domain Computing for GPS Signal Acquisition
As the most compute-intensive signal processing task in the global positioning system (GPS) receiver, signal acquisition accounts for most of the power budget, which is a key bottleneck in extending the battery life of edge devices. To improve the energy efficiency of GPS signal acquisition, we present charge-domain computing for GPS signal acquisition (CCSA), a matched filter with CCSA. It employs a $64 {\times } 64$ mixed-signal processing element (PE) array, a 2.8-b flash analog-to-digital converter (ADC), and a digital controller. Each PE handles the element-wise multiplication with a mixed-signal multiplier (MSMUL), which can perform the multiplication and digital-to-charge conversion simultaneously with a two-phase operation scheme. We develop the two-phase digital-to-charge mapping method for the MSMUL, where the digital multiplication result is converted into the charge difference stored in a capacitor between the charge reset phase and the charge evaluation phase. The charge-domain outputs of all MSMULs are dumped and redistributed on a global output wire for the accumulation operation. Compared to prior charge-domain works, the proposed design reduces the number of capacitors or additional voltage levels by half, thereby leading to better energy efficiency and lower circuit complexity. The 28-nm test chip achieves 114–394-TOPS/W energy efficiency across 0.34–0.9-V supply voltage, which is $8.2{\times }$ higher than state-of-the-art at the same throughput.
- Open Access
- 23 June 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Reducing DRAM Latency via In-situ Temperature- and Process-Variation-Aware Timing Detection and Adaption
Long DRAM access latency has a significant impact on modern system performance. However, the improvement of DRAM access latency is limited, as the DRAM vendors reserve considerable timing margins against seldom worst-case conditions. To mitigate such pessimistic timing margins, we propose a temperature- and process-variation-aware timing detection and adaption DRAM (TPDA-DRAM) architecture. It equips in-situ cross-coupled detectors to monitor the voltage difference between bitline pairs, enabling estimation of timing margins caused by process and temperature variations. Moreover, TPDA-DRAM incorporates two collaborative timing adaption schemes: 1) a process-variation-aware timing adaption scheme (PVA) that selectively accelerates the access to weak cells, and 2) a temperature-variation-aware timing adaption scheme (TVA) that precisely adjusts timing parameters by adopting temperature information. Compared to prior art, the proposed detector reduces detection deviation by 54.8% and area overhead by 88.1%. The system-level evaluation in an eight-core system shows that TPDA-DRAM improves the average performance and energy efficiency by 20.5% and 15.0%, respectively.
- 23 June 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Open AccessReducing DRAM Latency via In-situ Temperature- and Process-Variation-Aware Timing Detection and Adaption
Long DRAM access latency has a significant impact on modern system performance. However, the improvement of DRAM access latency is limited, as the DRAM vendors reserve considerable timing margins against seldom worst-case conditions. To mitigate such pessimistic timing margins, we propose a temperature- and process-variation-aware timing detection and adaption DRAM (TPDA-DRAM) architecture. It equips in-situ cross-coupled detectors to monitor the voltage difference between bitline pairs, enabling estimation of timing margins caused by process and temperature variations. Moreover, TPDA-DRAM incorporates two collaborative timing adaption schemes: 1) a process-variation-aware timing adaption scheme (PVA) that selectively accelerates the access to weak cells, and 2) a temperature-variation-aware timing adaption scheme (TVA) that precisely adjusts timing parameters by adopting temperature information. Compared to prior art, the proposed detector reduces detection deviation by 54.8% and area overhead by 88.1%. The system-level evaluation in an eight-core system shows that TPDA-DRAM improves the average performance and energy efficiency by 20.5% and 15.0%, respectively.
- 23 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 59 (2) , 616-625
A Metastability Risk Prediction and Mitigation Technique for Clock-Domain Crossing With Single-Stage Synchronizer in Near-Threshold-Voltage Multivoltage/ Frequency-Domain Network-on-Chip
For a network-on-chip (NoC) with multiple voltage/frequency domains, metastability hurts the reliability during the clock-domain crossing, especially in the near-threshold-voltage (NTV) region. Conventional multistage synchronizers reduce the probability of metastability but have a high latency penalty. This article presents a technique titled metastability risk prediction and mitigation (MPAM) that predicts the near-future metastability risks by a triple-phase clock monitoring circuitry and mitigates them by a metastability-free clock scheme. Therefore, the MPAM enables only one flip-flop for data synchronization without degrading the reliability against metastability, thus improving the latency and throughput of NoC. We prototyped a 2-by-2 NoC test chip with four independent voltage/frequency domains in a 40-nm low-power (LP) process, featuring the MPAM technique. The measurement shows that the MPAM significantly reduces the metastability condition rate by $10^{10}$ times under different clock frequency ratios. Moreover, by enabling only one flip-flop for synchronization, the MPAM-based NoC achieves packet latency reduction, throughput improvement, and energy efficiency gain by 58%, 13.4%, and 8.6%, respectively.
- 23 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 59 (2) , 616-625
A Metastability Risk Prediction and Mitigation Technique for Clock-Domain Crossing With Single-Stage Synchronizer in Near-Threshold-Voltage Multivoltage/ Frequency-Domain Network-on-Chip
For a network-on-chip (NoC) with multiple voltage/frequency domains, metastability hurts the reliability during the clock-domain crossing, especially in the near-threshold-voltage (NTV) region. Conventional multistage synchronizers reduce the probability of metastability but have a high latency penalty. This article presents a technique titled metastability risk prediction and mitigation (MPAM) that predicts the near-future metastability risks by a triple-phase clock monitoring circuitry and mitigates them by a metastability-free clock scheme. Therefore, the MPAM enables only one flip-flop for data synchronization without degrading the reliability against metastability, thus improving the latency and throughput of NoC. We prototyped a 2-by-2 NoC test chip with four independent voltage/frequency domains in a 40-nm low-power (LP) process, featuring the MPAM technique. The measurement shows that the MPAM significantly reduces the metastability condition rate by $10^{10}$ times under different clock frequency ratios. Moreover, by enabling only one flip-flop for synchronization, the MPAM-based NoC achieves packet latency reduction, throughput improvement, and energy efficiency gain by 58%, 13.4%, and 8.6%, respectively.
- 19 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 42 (9) , 3078-3091
CDAR-DRAM: Enabling Runtime DRAM Performance and Energy Optimization via In-Situ Charge Detection and Adaptive Data Restoration
With the increasing of DRAM’s capacity, the refresh operation rapidly becomes a major concern to the performance of the current computational system. Moreover, conservative timing parameters adopted for access operations make an increasing amount of negative impact on system performance and energy efficiency. In this paper, we propose an in-situ charge detection and adaptive data restoration DRAM (CDAR-DRAM) architecture, which can dynamically adjust the refresh rate and relax the constraints on access timing by removing pessimistic timing margins for PVT variations. CDAR-DRAM employs a lowcost skewed-inverter-based detector to monitor the bitline voltage in runtime and estimate real-time timing parameters of cells. Based on the detector, an adaptive refresh and restore scheme (CDAR-ref) is presented, which progressively reduces the refresh rate and partially restores cells’ voltage just enough for cells with sufficient charge, thereby optimizing both refresh and restoration operations. Moreover, a supplementary adaptive access scheme (CDAR-acc) is presented, which detects the runtime charge level of recently-accessed rows and reduces access latency aggressively, benefitting workloads in a single-core system and memory nonintensive workloads in a multi-core system. CDAR’s flexibility allows the two schemes to be combined. The evaluation shows that in an 8-core system, the combined scheme improves performance and energy efficiency by 15.2% and 22.6%, respectively.
- 19 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 42 (9) , 3078-3091
CDAR-DRAM: Enabling Runtime DRAM Performance and Energy Optimization via In-Situ Charge Detection and Adaptive Data Restoration
With the increasing of DRAM’s capacity, the refresh operation rapidly becomes a major concern to the performance of the current computational system. Moreover, conservative timing parameters adopted for access operations make an increasing amount of negative impact on system performance and energy efficiency. In this paper, we propose an in-situ charge detection and adaptive data restoration DRAM (CDAR-DRAM) architecture, which can dynamically adjust the refresh rate and relax the constraints on access timing by removing pessimistic timing margins for PVT variations. CDAR-DRAM employs a lowcost skewed-inverter-based detector to monitor the bitline voltage in runtime and estimate real-time timing parameters of cells. Based on the detector, an adaptive refresh and restore scheme (CDAR-ref) is presented, which progressively reduces the refresh rate and partially restores cells’ voltage just enough for cells with sufficient charge, thereby optimizing both refresh and restoration operations. Moreover, a supplementary adaptive access scheme (CDAR-acc) is presented, which detects the runtime charge level of recently-accessed rows and reduces access latency aggressively, benefitting workloads in a single-core system and memory nonintensive workloads in a multi-core system. CDAR’s flexibility allows the two schemes to be combined. The evaluation shows that in an 8-core system, the combined scheme improves performance and energy efficiency by 15.2% and 22.6%, respectively.
- 20 November 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 56 (7) , 2270-2280
MEDAC: A Metastability Condition Detection and Correction Technique for a Near-Threshold-Voltage Multi-Voltage-/Frequency-Domain Network-on-Chip
This article presents a technique titled metastability condition detection and correction (MEDAC) that is featured in a near-threshold voltage (NTV) network-on-chip (NoC) hardware. The proposed technique introduces a double-sampling-based circuitry to detect whether the input data arrives too close to the receiver's clock edge, which is defined as a metastability condition. The MEDAC also contains a metastability condition mitigation scheme to monitor if the occurrence of such metastability conditions is frequent, and adaptively tune the receiver's clock phase to increase the mean time between metastability conditions. We prototype a test chip of the NoC in a 65-nm low-power process, which supports four independent voltage/frequency (V/F) domains in globally asynchronous locally synchronous (GALS) systems. Measurements across 0.4-1-V supply voltages with 13 different frequency ratios from 0.25 to 4 show that the MEDAC can well detect the metastability conditions and reduce the probability of entering such conditions by one to three orders of magnitude. This reduction, in turn, reduces the potential incorrect packets, thereby improving the data rate and energy efficiency by up to 21.1% and 27.2%, respectively.
- 20 November 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 56 (7) , 2270-2280
MEDAC: A Metastability Condition Detection and Correction Technique for a Near-Threshold-Voltage Multi-Voltage-/Frequency-Domain Network-on-Chip
This article presents a technique titled metastability condition detection and correction (MEDAC) that is featured in a near-threshold voltage (NTV) network-on-chip (NoC) hardware. The proposed technique introduces a double-sampling-based circuitry to detect whether the input data arrives too close to the receiver's clock edge, which is defined as a metastability condition. The MEDAC also contains a metastability condition mitigation scheme to monitor if the occurrence of such metastability conditions is frequent, and adaptively tune the receiver's clock phase to increase the mean time between metastability conditions. We prototype a test chip of the NoC in a 65-nm low-power process, which supports four independent voltage/frequency (V/F) domains in globally asynchronous locally synchronous (GALS) systems. Measurements across 0.4-1-V supply voltages with 13 different frequency ratios from 0.25 to 4 show that the MEDAC can well detect the metastability conditions and reduce the probability of entering such conditions by one to three orders of magnitude. This reduction, in turn, reduces the potential incorrect packets, thereby improving the data rate and energy efficiency by up to 21.1% and 27.2%, respectively.