Jae-Jin Kim
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Jae-Jin Kim has a total of 44 co-authors. Here are the top co-authors by number of shared publications:
- Gunmo Koo
School of Electronic and Electrical Engineering, Kyungpook National University, Daegu, South Korea
4 shared publications - Kunhee Cho
Sungkyunkwan University
4 shared publications - Joo-Young Kim
KAIST
2 shared publications Seongmin Lee
School of Electronic and Electrical Engineering, Kyungpook National University, Daegu, South Korea
2 shared publications- Chanheum Han
Kwangwoon University
1 shared publication - Di Chen
Tsinghua University
1 shared publication
Latest Publications
- Open Access
- 30 September 2026
- Published by Korean Institute of Electromagnetic Engineering and Science in The Journal of Korean Institute of Electromagnetic Engineering and Science
- Vol. 37 (9) , 917-924
A Broadband Four-Element Transceiver IC for Single-Beam Phased Arrays in 65-nm CMOS
한국전자파학회논문지 (The Journal of Korean Institute of Electromagnetic Engineering and Science)
- 30 September 2026
- Published by Korean Institute of Electromagnetic Engineering and Science in The Journal of Korean Institute of Electromagnetic Engineering and Science
- Vol. 37 (9) , 917-924
Open AccessA Broadband Four-Element Transceiver IC for Single-Beam Phased Arrays in 65-nm CMOS
한국전자파학회논문지 (The Journal of Korean Institute of Electromagnetic Engineering and Science)
- 1 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 61 (5) , 2349-2364
DPIM: A 2T1C eDRAM Transformer-in-Memory Chip With Sparsity-Aware Quantization and Heterogeneous Dense–Sparse Core
Transformer models have revolutionized artificial intelligence (AI) applications across various domains, but their increasing complexity poses significant challenges in terms of computational and memory demands. While processing-in-memory (PIM) paradigms have been adopted to address these limitations, existing PIM-based transformer accelerators still face hurdles such as: 1) focusing solely on optimizing attention layers; 2) lack of sparsity exploitation for transformers; and 3) limited PIM macro capacity and low cell density, which degrades on-chip data reuse and increases external memory access (EMA). This article presents DPIM, a novel 2T1C eDRAM-based transformer-in-memory chip that addresses these challenges through three key innovations: 1) a sparsity-aware quantization (SAQ) scheme that significantly increases bit-slice sparsity in both activation and weight data, achieving ratios of 83.3% and 88.4%, respectively, with minimal accuracy loss; 2) a heterogeneous PIM core capable of efficiently handling both sparse and dense matrix multiplications (MMs); and 3) a high-density 2T1C eDRAM cell with a density of 1.38 Mb/mm2, enabling large-capacity PIM macros. By integrating these features, DPIM achieves improved computational efficiency and reduced EMA with enhanced on-chip data reuse. The DPIM chip, fabricated using 28-nm CMOS technology, achieves a throughput of 3.03–12.12 TOPS and an energy efficiency of 4.84–19.36 TOPS/W, all measured across INT8 and INT4 operations, respectively. It achieves a throughput density of 0.55 TOPS/mm2 with INT8 operation. With a total macro size of 4608 kb, the chip occupies a die area of 20.25 mm2 and operates at frequencies from 50 to 285 MHz with a supply voltage of 0.85–1.0 V. The DPIM successfully executes BERT-Large on the general language understanding evaluation (GLUE) dataset. Its macro density is 1413 kb/mm2, and the resulting density figure-of-merit (FoM) (macro density$\times $ throughput density) is$1.6\times $ –$115.8\times $ higher than previous works, representing a significant advancement in hardware design for efficient transformer processing.- 1 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 61 (5) , 2349-2364
DPIM: A 2T1C eDRAM Transformer-in-Memory Chip With Sparsity-Aware Quantization and Heterogeneous Dense–Sparse Core
Transformer models have revolutionized artificial intelligence (AI) applications across various domains, but their increasing complexity poses significant challenges in terms of computational and memory demands. While processing-in-memory (PIM) paradigms have been adopted to address these limitations, existing PIM-based transformer accelerators still face hurdles such as: 1) focusing solely on optimizing attention layers; 2) lack of sparsity exploitation for transformers; and 3) limited PIM macro capacity and low cell density, which degrades on-chip data reuse and increases external memory access (EMA). This article presents DPIM, a novel 2T1C eDRAM-based transformer-in-memory chip that addresses these challenges through three key innovations: 1) a sparsity-aware quantization (SAQ) scheme that significantly increases bit-slice sparsity in both activation and weight data, achieving ratios of 83.3% and 88.4%, respectively, with minimal accuracy loss; 2) a heterogeneous PIM core capable of efficiently handling both sparse and dense matrix multiplications (MMs); and 3) a high-density 2T1C eDRAM cell with a density of 1.38 Mb/mm2, enabling large-capacity PIM macros. By integrating these features, DPIM achieves improved computational efficiency and reduced EMA with enhanced on-chip data reuse. The DPIM chip, fabricated using 28-nm CMOS technology, achieves a throughput of 3.03–12.12 TOPS and an energy efficiency of 4.84–19.36 TOPS/W, all measured across INT8 and INT4 operations, respectively. It achieves a throughput density of 0.55 TOPS/mm2 with INT8 operation. With a total macro size of 4608 kb, the chip occupies a die area of 20.25 mm2 and operates at frequencies from 50 to 285 MHz with a supply voltage of 0.85–1.0 V. The DPIM successfully executes BERT-Large on the general language understanding evaluation (GLUE) dataset. Its macro density is 1413 kb/mm2, and the resulting density figure-of-merit (FoM) (macro density$\times $ throughput density) is$1.6\times $ –$115.8\times $ higher than previous works, representing a significant advancement in hardware design for efficient transformer processing.- 17 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. 71 (11) , 5081-5090
A DVS-Enabled Distributed Digital LDO Providing Rapid Uniform Power Grid and Ripple Reduction Achieving 20.1-ps FOM in 28 nm CMOS
A dynamic voltage scaling (DVS) enabled distributed digital low-dropout voltage regulator (LDO) is described. The proposed distributed LDO utilizes a multi-point average sensing to enable rapid and uniform output voltage regulation across a large-scale power grid, even during unbalanced load transients. A 16-bit thermometer-code flash analog-to-digital converter (FADC) combined with unary passgate configurations and an adaptive on-resistance (R $_{\mathrm {ON}}$ ) modulation is employed to ensure a small output voltage ripple during DVS operation, using only a small 13.4nF output capacitor. The proposed distributed LDO has been implemented in 28nm CMOS, achieves a 10.4A/mm2 current density, 99.96% current efficiency, and a 20.1ps FOM. It has also been tested under various unbalanced load transient conditions and can rapidly regulate the output voltage back to the target level.
- 17 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. 71 (11) , 5081-5090
A DVS-Enabled Distributed Digital LDO Providing Rapid Uniform Power Grid and Ripple Reduction Achieving 20.1-ps FOM in 28 nm CMOS
A dynamic voltage scaling (DVS) enabled distributed digital low-dropout voltage regulator (LDO) is described. The proposed distributed LDO utilizes a multi-point average sensing to enable rapid and uniform output voltage regulation across a large-scale power grid, even during unbalanced load transients. A 16-bit thermometer-code flash analog-to-digital converter (FADC) combined with unary passgate configurations and an adaptive on-resistance (R $_{\mathrm {ON}}$ ) modulation is employed to ensure a small output voltage ripple during DVS operation, using only a small 13.4nF output capacitor. The proposed distributed LDO has been implemented in 28nm CMOS, achieves a 10.4A/mm2 current density, 99.96% current efficiency, and a 20.1ps FOM. It has also been tested under various unbalanced load transient conditions and can rapidly regulate the output voltage back to the target level.
- 27 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Solid-State Circuits Letters
- Vol. 7, 199-202
An Ultrawide Load-Range Fast-Transient Output Capacitor-Less Digital LDO With Adaptive Gate Modulation and Droop Detection
An ultra-wide load-range output capacitor-less digital LDO (DLDO) with an adaptive gate modulation scheme is described. The proposed DLDO is primarily regulated by digital codes with a synchronous clock signal while the gate driving level is dynamically adjusted according to the load current level. The proposed gate modulation scheme can significantly widen the dynamic range of load current and reduce the output voltage ripple. In addition, an asynchronous droop detection circuit, coupled with adaptive gate modulation, is added to improve the voltage droop and ensure fast recovery from load transients. The proposed DLDO was fabricated in 28 nm CMOS process. The dynamic load range of 57,143 × (1.4μ A-80 mA) is achieved and the output voltage ripple of under 17 mV is measured across the entire load current range. A response time of less than 10 ns and a recovery time of less than 30 ns are measured in various load transient conditions.
- 27 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Solid-State Circuits Letters
- Vol. 7, 199-202
An Ultrawide Load-Range Fast-Transient Output Capacitor-Less Digital LDO With Adaptive Gate Modulation and Droop Detection
An ultra-wide load-range output capacitor-less digital LDO (DLDO) with an adaptive gate modulation scheme is described. The proposed DLDO is primarily regulated by digital codes with a synchronous clock signal while the gate driving level is dynamically adjusted according to the load current level. The proposed gate modulation scheme can significantly widen the dynamic range of load current and reduce the output voltage ripple. In addition, an asynchronous droop detection circuit, coupled with adaptive gate modulation, is added to improve the voltage droop and ensure fast recovery from load transients. The proposed DLDO was fabricated in 28 nm CMOS process. The dynamic load range of 57,143 × (1.4μ A-80 mA) is achieved and the output voltage ripple of under 17 mV is measured across the entire load current range. A response time of less than 10 ns and a recovery time of less than 30 ns are measured in various load transient conditions.
- 20 July 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. 70 (12) , 4975-4985
An Output-Capacitor-Free NMOS Digital LDO Using Gate Driving Strength Modulation and Droop Detector
An output-capacitor-free NMOS digital LDO (DLDO) using gate driving strength modulation (GDSM) is described. The proposed DLDO is mainly based on the time-driven topology while the GDSM changes the gate driving level adaptively according to the load current condition. The proposed GDSM lowers the gate driving level in the light load condition which improves the load transient response, widens the load current dynamic range, and reduces the output voltage ripple. A droop detector combined with the modulation scheme is also proposed to further improve the undershoot and recovery time. The proposed DLDO has been implemented in 28 nm CMOS and the 20,000 $\times$ load range is obtained even with 256 unity power switch arrays. When the load current changes from 5 mA to 85 mA, the $V_{OUT}$ droop and recovery time are 130 mV and 2.5 $\mu $ s, which are 4.92 $\times$ and 70 $\times$ improvements compared to the baseline time-driven DLDO, respectively.
- 20 July 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. 70 (12) , 4975-4985
An Output-Capacitor-Free NMOS Digital LDO Using Gate Driving Strength Modulation and Droop Detector
An output-capacitor-free NMOS digital LDO (DLDO) using gate driving strength modulation (GDSM) is described. The proposed DLDO is mainly based on the time-driven topology while the GDSM changes the gate driving level adaptively according to the load current condition. The proposed GDSM lowers the gate driving level in the light load condition which improves the load transient response, widens the load current dynamic range, and reduces the output voltage ripple. A droop detector combined with the modulation scheme is also proposed to further improve the undershoot and recovery time. The proposed DLDO has been implemented in 28 nm CMOS and the 20,000 $\times$ load range is obtained even with 256 unity power switch arrays. When the load current changes from 5 mA to 85 mA, the $V_{OUT}$ droop and recovery time are 130 mV and 2.5 $\mu $ s, which are 4.92 $\times$ and 70 $\times$ improvements compared to the baseline time-driven DLDO, respectively.
- 12 April 2022
- Published by Felipe Herrera Library (Inter-American Development Bank)
A Roadmap for Digitalization of Tax Systems: Lessons from Korea
This publication reviews the history of digitalization of tax administration in Korea dating back to the 1990s and shares the countrys experience and know-how in building an efficient e-taxation architecture. Its main emphasis is on how the Korean government managed to make the best use of a wide range of taxpayer information efficiently and securely. It highlights information security and presents three case studies of an institutional framework for using third-party data: tax schemes for credit card usage, a cash receipt system, and e-invoicing. It then lays out a range of policy implications for consideration by tax authorities in the Latin American and Caribbean region.
- 12 April 2022
- Published by Felipe Herrera Library (Inter-American Development Bank)
A Roadmap for Digitalization of Tax Systems: Lessons from Korea
This publication reviews the history of digitalization of tax administration in Korea dating back to the 1990s and shares the countrys experience and know-how in building an efficient e-taxation architecture. Its main emphasis is on how the Korean government managed to make the best use of a wide range of taxpayer information efficiently and securely. It highlights information security and presents three case studies of an institutional framework for using third-party data: tax schemes for credit card usage, a cash receipt system, and e-invoicing. It then lays out a range of policy implications for consideration by tax authorities in the Latin American and Caribbean region.
- 6 April 2020
- Published by Springer Nature in Archives of Civil and Mechanical Engineering
- Vol. 20 (2) , 1-18
Spacing and bundling effects on rate-dependent pullout behavior of various steel fibers embedded in ultra-high-performance concrete
- 6 April 2020
- Published by Springer Nature in Archives of Civil and Mechanical Engineering
- Vol. 20 (2) , 1-18
Spacing and bundling effects on rate-dependent pullout behavior of various steel fibers embedded in ultra-high-performance concrete
- 25 April 2011
- Published by The Electrochemical Society in ECS Transactions
- Vol. 33 (40) , 51-57
(High Temperature Materials Division Outstanding Achievement Award) Measurement and Modeling of Electrical, Mechanical, and Chemical Properties of a Model Mixed Ionic Electronic Conductor: Pr Doped Ceria
Praseodymium doped cerium oxide (PCO) shows mixed ionic and electronic conducting (MIEC) characteristics at relatively high pO2 (e.g. air) and enhanced oxygen storage capacity (OSC) resulting from the ready reduction of Pr to its trivalent state. These features are important in applications such as solid oxide fuel cell (SOFC) cathodes and permeation membranes, oxygen buffers in three-way catalysts, and gas sensors. In this proceedings article, we present an overview of our current and past efforts towards the electrical, chemical, mechanical, and optical characterization of PCO.
- 25 April 2011
- Published by The Electrochemical Society in ECS Transactions
- Vol. 33 (40) , 51-57
(High Temperature Materials Division Outstanding Achievement Award) Measurement and Modeling of Electrical, Mechanical, and Chemical Properties of a Model Mixed Ionic Electronic Conductor: Pr Doped Ceria
Praseodymium doped cerium oxide (PCO) shows mixed ionic and electronic conducting (MIEC) characteristics at relatively high pO2 (e.g. air) and enhanced oxygen storage capacity (OSC) resulting from the ready reduction of Pr to its trivalent state. These features are important in applications such as solid oxide fuel cell (SOFC) cathodes and permeation membranes, oxygen buffers in three-way catalysts, and gas sensors. In this proceedings article, we present an overview of our current and past efforts towards the electrical, chemical, mechanical, and optical characterization of PCO.
- 1 June 2009
- Published by Taylor & Francis in International Journal of Urban Sciences
- Vol. 13 (1) , 63-80
On-line Estimation of Departure Time-Based Link Travel Times from Spatial Detection System
- 1 June 2009
- Published by Taylor & Francis in International Journal of Urban Sciences
- Vol. 13 (1) , 63-80
On-line Estimation of Departure Time-Based Link Travel Times from Spatial Detection System
- 1 November 2003
- Published by Springer Nature in Korean Journal of Chemical Engineering
- Vol. 20 (6) , 1158-1165
Surface characterization of biocompatible polysulfone membranes modified with poly(ethylene glycol) derivatives
- 1 November 2003
- Published by Springer Nature in Korean Journal of Chemical Engineering
- Vol. 20 (6) , 1158-1165
Surface characterization of biocompatible polysulfone membranes modified with poly(ethylene glycol) derivatives