Wang Kang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Wang Kang has a total of 277 co-authors. Here are the top co-authors by number of shared publications:
- Weisheng Zhao
National Key Laboratory of Spintronics, Hangzhou International Innovation Institute, Beihang University 1 , Hangzhou 311115
42 shared publications - He Zhang
School of Integrated Circuit Science and Engineering, Beihang University, Beijing, China
24 shared publications Youguang Zhang
School of Electronic and Information Engineering, Beihang University, Beijing, China
23 shared publications- Xichao Zhang
Waseda University
10 shared publications Xing Chen
School of Microelectronics, Fert Beijing Research Institute, BDBC, Beihang University, Beijing 100191, China
9 shared publications- Yan Zhou
The Chinese University of Hong Kong
9 shared publications
Latest Publications
- 1 October 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 75 (10) , 3713-3727
Continuous Bayesian Machine Learning for High-Dimensional Analog and Mixed-Signal Circuit Yield Analysis and Optimization
- 1 October 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 75 (10) , 3713-3727
Continuous Bayesian Machine Learning for High-Dimensional Analog and Mixed-Signal Circuit Yield Analysis and Optimization
- 23 September 2026
- Published by Elsevier in Measurement
- Vol. 292
A 60 fps, 0.69nJ/frame∙pixel low-power CMOS image sensor with multi-phase gray code generation, error calibration, and split-readout strategy
- 23 September 2026
- Published by Elsevier in Measurement
- Vol. 292
A 60 fps, 0.69nJ/frame∙pixel low-power CMOS image sensor with multi-phase gray code generation, error calibration, and split-readout strategy
- Open Access
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
IGNITE: An Inquisitive and Graph-Aware Neural Circuit Simulator with Two-Stage Enhanced In-Context Learning
Neural circuit simulators promise fast alternatives to SPICE but remain impractical due to three fundamental limitations: per-circuit hyperparameter tuning, cold-start problems from lacking cross-task knowledge sharing, and prohibitive online retraining costs that negate computational advantages. We introduce IGNITE, a pretrained neural simulator that addresses these challenges through in-context learning. Built on a self-supervised topology encoder and Prior-data Fitted Network decoder, IGNITE adapts to new circuits via a single forward pass conditioned on a small context set, eliminating online training overhead. An information-theoretic active learning strategy enables intelligent sample selection, maximizing data efficiency. IGNITE transfers knowledge across topologies and technology nodes, achieving R2 > 0.9 with only 95 samples in average—a 15.8× improvement over SOTA. Integration with optimization frameworks yields 2–30× speedup for transistor sizing and more than 5× speedup for yield optimization, demonstrating practical viability for industrial analog circuit design.
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessIGNITE: An Inquisitive and Graph-Aware Neural Circuit Simulator with Two-Stage Enhanced In-Context Learning
Neural circuit simulators promise fast alternatives to SPICE but remain impractical due to three fundamental limitations: per-circuit hyperparameter tuning, cold-start problems from lacking cross-task knowledge sharing, and prohibitive online retraining costs that negate computational advantages. We introduce IGNITE, a pretrained neural simulator that addresses these challenges through in-context learning. Built on a self-supervised topology encoder and Prior-data Fitted Network decoder, IGNITE adapts to new circuits via a single forward pass conditioned on a small context set, eliminating online training overhead. An information-theoretic active learning strategy enables intelligent sample selection, maximizing data efficiency. IGNITE transfers knowledge across topologies and technology nodes, achieving R2 > 0.9 with only 95 samples in average—a 15.8× improvement over SOTA. Integration with optimization frameworks yields 2–30× speedup for transistor sizing and more than 5× speedup for yield optimization, demonstrating practical viability for industrial analog circuit design.
- Open Access
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Thermal Stability Strategy for a 12nm/28nm Hybrid-Bonding Wafer-on-Wafer 3D-Integrated eFlash In-Memory Computing SoC
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessThermal Stability Strategy for a 12nm/28nm Hybrid-Bonding Wafer-on-Wafer 3D-Integrated eFlash In-Memory Computing SoC
- 11 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. PP (15498328) , 1-14
HE-FR: A High-Efficiency Fault-Resilient Framework for Stuck-at-Faults Mitigation in ReRAM-Based In-Memory Computing
ReRAM-based in-memory computing (IMC) architectures enable efficient neural-network inference on edge devices, due to their high density and non-volatility. However, ReRAM is prone to the stuck-at faults (SAFs) which would distort weight mappings and reduce model accuracy substantially. To tackle this problem, a high-efficiency fault-resilient (HE-FR) framework is introduced in the paper, which couples layered-precision quantization based on the generalized gaussian distribution cumulative distribution function (GGD-CDF), preserving weight distributions with the minimal information loss. Further, a fault-aware weight re-decomposition method is proposed. Specifically, a Q-agent-based method is proposed for sparse SAFs, which employs offline reinforcement learning to construct a globally optimized hash table with reduced compilation complexity, while incorporating both weight matching and cell-state stability into the reward function to suppress electro-stress. For dense SAFs, we exploit a greedy search over the remaining cells to provide rapid and accurate mappings. Experimental results show that GGD-CDF quantization improves accuracy by 2.18% over Float32. Under fabrication faults, HE-FR surpasses a fault-free method by 12.97% and accelerates compilation by $197\times $ with only 0.024% picojoule-level energy overhead. In electro-stress evaluations, HE-FR reduces performance degradation by 46.27%. It demonstrates that the proposed framework has advantages in robustness, efficiency, and reliability.
- 11 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. PP (15498328) , 1-14
HE-FR: A High-Efficiency Fault-Resilient Framework for Stuck-at-Faults Mitigation in ReRAM-Based In-Memory Computing
ReRAM-based in-memory computing (IMC) architectures enable efficient neural-network inference on edge devices, due to their high density and non-volatility. However, ReRAM is prone to the stuck-at faults (SAFs) which would distort weight mappings and reduce model accuracy substantially. To tackle this problem, a high-efficiency fault-resilient (HE-FR) framework is introduced in the paper, which couples layered-precision quantization based on the generalized gaussian distribution cumulative distribution function (GGD-CDF), preserving weight distributions with the minimal information loss. Further, a fault-aware weight re-decomposition method is proposed. Specifically, a Q-agent-based method is proposed for sparse SAFs, which employs offline reinforcement learning to construct a globally optimized hash table with reduced compilation complexity, while incorporating both weight matching and cell-state stability into the reward function to suppress electro-stress. For dense SAFs, we exploit a greedy search over the remaining cells to provide rapid and accurate mappings. Experimental results show that GGD-CDF quantization improves accuracy by 2.18% over Float32. Under fabrication faults, HE-FR surpasses a fault-free method by 12.97% and accelerates compilation by $197\times $ with only 0.024% picojoule-level energy overhead. In electro-stress evaluations, HE-FR reduces performance degradation by 46.27%. It demonstrates that the proposed framework has advantages in robustness, efficiency, and reliability.
- 30 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Emerging and Selected Topics in Circuits and Systems
- Vol. 16 (2) , 338-347
HyIMC: A 16 TOPS/W Hybrid Analog-Digital In-Memory Computing SoC With Model Compression and Recurrent Optimization for Deep Learning-Based Speech Enhancement
In-memory computing (IMC) presents a promising avenue for deep learning-based speech enhancement (DL-SE). However, the current state-of-the-art IMC architectures are hindered by their inability to simultaneously achieve high precision and energy efficiency, coupled with a lack of parallelism necessary for DL-SE’s distinctive temporal dependency. This paper introduces the HyIMC, which features: 1) an analog-digital hybrid IMC architecture that facilitates DL-SE algorithms; 2) a schedule controller designed to optimize recurrent dataflow in skip connections; and 3) non-key dimension shrinkage (NKDS), a model compression technique that retains superior accuracy while compressing DL-SE models. Our approach achieves 16 TOPS/W energy efficiency at system-level, implemented on a 40nm eFlash-based IMC SoC prototype. The HyIMC with NKDS compresses DL-SE model size by ∼83.3%, enhances the figure of merit by ∼1200%, and improves perceptual evaluation of speech quality score by ∼120%.
- 30 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Emerging and Selected Topics in Circuits and Systems
- Vol. 16 (2) , 338-347
HyIMC: A 16 TOPS/W Hybrid Analog-Digital In-Memory Computing SoC With Model Compression and Recurrent Optimization for Deep Learning-Based Speech Enhancement
In-memory computing (IMC) presents a promising avenue for deep learning-based speech enhancement (DL-SE). However, the current state-of-the-art IMC architectures are hindered by their inability to simultaneously achieve high precision and energy efficiency, coupled with a lack of parallelism necessary for DL-SE’s distinctive temporal dependency. This paper introduces the HyIMC, which features: 1) an analog-digital hybrid IMC architecture that facilitates DL-SE algorithms; 2) a schedule controller designed to optimize recurrent dataflow in skip connections; and 3) non-key dimension shrinkage (NKDS), a model compression technique that retains superior accuracy while compressing DL-SE models. Our approach achieves 16 TOPS/W energy efficiency at system-level, implemented on a 40nm eFlash-based IMC SoC prototype. The HyIMC with NKDS compresses DL-SE model size by ∼83.3%, enhances the figure of merit by ∼1200%, and improves perceptual evaluation of speech quality score by ∼120%.
- 2 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 34 (3) , 891-903
A 40 nm Buffer-Free 7T-SRAM Analog Charge-Domain CIM Macro With Merging Timing Based On Time-Row Division Strategy
Computing-in-memory (CIM) macros based on static random access memory (SRAM) are meant to increase capacity while improving energy efficiency and reducing computing latency. However, traditional analog designs still face several key challenges, including long computing latency from separated computing phases, negative voltage fluctuations from massive parallel computing, and low bitcell density from additional transistors and capacitors for multiplication. On the other hand, only time-aligned inputs are supported in the works. To overcome the above challenges, this work proposes a buffer-free 7T-SRAM charge-domain CIM macro. It has four key features: 1) a compact 7T SRAM bitcell structure for high-energy efficiency; 2) a configurable input unit to support different sizes of input activations; 3) a time-row division (RD) strategy to support real-time processing and alleviate negative voltage fluctuations; and 4) a merging timing to conceal the input phase for high throughput. The fabricated 512-Kb SRAM-CIM macro in 40 nm achieves 79.3–290.4 Tops/W at 4-bit precision.
- 2 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 34 (3) , 891-903
A 40 nm Buffer-Free 7T-SRAM Analog Charge-Domain CIM Macro With Merging Timing Based On Time-Row Division Strategy
Computing-in-memory (CIM) macros based on static random access memory (SRAM) are meant to increase capacity while improving energy efficiency and reducing computing latency. However, traditional analog designs still face several key challenges, including long computing latency from separated computing phases, negative voltage fluctuations from massive parallel computing, and low bitcell density from additional transistors and capacitors for multiplication. On the other hand, only time-aligned inputs are supported in the works. To overcome the above challenges, this work proposes a buffer-free 7T-SRAM charge-domain CIM macro. It has four key features: 1) a compact 7T SRAM bitcell structure for high-energy efficiency; 2) a configurable input unit to support different sizes of input activations; 3) a time-row division (RD) strategy to support real-time processing and alleviate negative voltage fluctuations; and 4) a merging timing to conceal the input phase for high throughput. The fabricated 512-Kb SRAM-CIM macro in 40 nm achieves 79.3–290.4 Tops/W at 4-bit precision.
- 13 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 34 (3) , 1067-1071
Self-Calibrating Analog Circuitry for Softmax-Scaled Function With Analog Computing-In-Memory
Analog computing-in-memory (ACIM) has garnered widespread attention due to its advantage of high energy efficiency. However, it faces large power and hardware costs to handle sophisticated nonlinear functions, such as the softmax, due to costly exponentiation and division. Existing digital-domain approaches often rely on dedicated modules to carry out these operations, leading to a cost expensive area and high-power consumption. To address the issues, we propose a self-calibrating analog circuitry for a softmax-scaled function with ACIM. By exploiting transistor subthreshold properties, the work eliminates expensive digital operations while mapping exponentiation and division to successive analog circuits. A self-calibration module further mitigates partial mismatch-induced deviations by dynamically tuning bias voltages, improving overall fitting accuracy and system robustness. The proposed softmax-enabled ACIM work achieves energy efficiency of 55.06–60.08TOPS/W and 684.15 GOPS/mm2 at 4-bit precision. In comparison with the state-of-the-art ACIMs with softmax implications, our proposed work shows higher energy efficiency and area efficiency.
- 13 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 34 (3) , 1067-1071
Self-Calibrating Analog Circuitry for Softmax-Scaled Function With Analog Computing-In-Memory
Analog computing-in-memory (ACIM) has garnered widespread attention due to its advantage of high energy efficiency. However, it faces large power and hardware costs to handle sophisticated nonlinear functions, such as the softmax, due to costly exponentiation and division. Existing digital-domain approaches often rely on dedicated modules to carry out these operations, leading to a cost expensive area and high-power consumption. To address the issues, we propose a self-calibrating analog circuitry for a softmax-scaled function with ACIM. By exploiting transistor subthreshold properties, the work eliminates expensive digital operations while mapping exponentiation and division to successive analog circuits. A self-calibration module further mitigates partial mismatch-induced deviations by dynamically tuning bias voltages, improving overall fitting accuracy and system robustness. The proposed softmax-enabled ACIM work achieves energy efficiency of 55.06–60.08TOPS/W and 684.15 GOPS/mm2 at 4-bit precision. In comparison with the state-of-the-art ACIMs with softmax implications, our proposed work shows higher energy efficiency and area efficiency.