Seung-Jun Bae

Seung-Jun Bae

Auto-generated profile, based on publicly available information

Seung-Jun Bae

DRAM Design Team Memory Division Samsung Electronics Hwasung Republic of Korea

Research Focus & Metrics

Academic Career Length

20Years (2007-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 May 2024 in IEEE Journal of Solid-State Circuits
Verified
Journal Article
Closed access
HAHyun-A Ahn
Hyun-A Ahn
YSYoo-Chang Sung
Yoo-Chang Sung
JKJanghoo Kim
Janghoo Kim
KKKihan Kim
Kihan Kim
DLDong-Hun Lee
Dong-Hun Lee
TOTae-Young Oh
Tae-Young Oh
For low-power 8.5-Gbps operation, 4th-generation 10-nm 16-Gb LPDDR5x DRAM I/O circuits and control methods are proposed in this article. The proposed I/O improves signal integrity (SI) and power integrity (PI) by using a self-pre-emphasized stacked driver in transmitter (Tx), a supply voltage insensitive data receiver (Rx), and an optimized clock tree in write clock (WCK) clock paths. The measured eye widths of Tx and Rx, which are the indicators of the SI/PI, are 0.66 unit interval (UI) and 0.57 UI at 8.5 Gbps, respectively. Also, the measured power consumption is reduced by 20.0% compared to the previous LPDDR5 SDRAM product.

1 February 2007
Verified
Conference Paper
Closed access
JIJeong-Don Ihm
Jeong-Don Ihm
SBSeung-Jun Bae
Seung-Jun Bae
KPKwang Il Park
Kwang Il Park
HSHo-Young Song
Ho-Young Song
WLWoo Jin Lee
Woo Jin Lee
SCSoo-In Cho
Soo-In Cho
A 4Gb/s/pin 32b parallel 512Mb GDDR4 SDRAM is implemented in an 80nm DRAM process. It employs a data-bus inversion coding scheme with an analog majority voter insensitive to mismatch, which reduces peak-to-peak jitter by 21 ps and voltage fluctuation by 68mV. A dual duty-cycle corrector is proposed to average duty error, and tuning is added to the auto-calibration of driver and termination impedance.

1 February 2009
Verified
Conference Paper
Closed access
KHKyung-Soo Ha
Kyung-Soo Ha
LKLee-Sup Kim
Lee-Sup Kim
SBSeung-Jun Bae
Seung-Jun Bae
KPKwang Il Park
Kwang Il Park
JCJoo Sun Choi
Joo Sun Choi
KKKinam Kim
Kinam Kim
This paper describes pseudo-differential signaling schemes for DRAM interfaces that minimize the skew between data and suppress common-mode noise. The chip is implemented in a 0.13 mum process and occupies 2.7times2.3 mm2 and the active area is 1.0times0.3 mm2.

15 January 2019 in IEEE Journal of Solid-State Circuits
Verified
Journal Article
Closed access
YFYohan Frans
Yohan Frans
WDWim Dehaene
Wim Dehaene
MMMasato Motomura
Masato Motomura
SBSeung-Jun Bae
Seung-Jun Bae
This Special Issue of the IEEE Journal of Solid-State Circuits is dedicated to a set of the best papers selected from the 2018 IEEE International Solid-State Circuits Conference (ISSCC) that took place in San Francisco, CA, USA, on February 11–15, 2018. It covers papers from the Wireline (WLN), Digital Circuits (DCT), Digital Architectures and Systems (DAS), and Memory (MEM) committees.

6 March 2017
Verified
Conference Paper
Closed access
YFYohan Frans
Yohan Frans
IFIchiro Fujimori
Ichiro Fujimori
SBSeung-Jun Bae
Seung-Jun Bae
SPSam Palermo
Sam Palermo
HNHideyuki Nosaka
Hideyuki Nosaka
SESimone Erba
Simone Erba
With the explosive data growth due to HD video content, IoT, and the rapid shift from enterprise to cloud computing, Mega Data Centers have become an essential part of the global network infrastructure. Despite their massive scale, Data Center architectures require software-defined network capability and scalability, with lower latency. Interconnects will need to support diversified media-types, higher data-rates (>50Gb/s per lane), and longer distance (up to 2km) to achieve this and still make economic sense. Once a niche for mainly VCSEL links in HPC, “Optical Ethernet” is now becoming mainstream as interconnects for both MMF and SMF in various fiber configurations are being defined within the standards. Also, copper links including backplane continue to be the interconnect with highest volume. The Forum will cover various wireline transceivers enabling Mega Data Centers such as ADC-based transceivers and Silicon Photonics, and challenges moving forward.

23 June 2025 in arXiv
Verified
Preprint
Open access
PDF file
MUMuhammad Usama
Muhammad Usama
HJHee-Deok Jang
Hee-Deok Jang
SSSoham Shanbhag
Soham Shanbhag
YSYoo-Chang Sung
Yoo-Chang Sung
SBSeung-Jun Bae
Seung-Jun Bae
DCDong Eui Chang
Dong Eui Chang
This paper addresses the dual challenge of improving anomaly detection and signal integrity in high-speed dynamic random access memory signals. To achieve this, we propose a joint training framework that integrates an autoencoder with a classifier to learn more distinctive latent representations by focusing on valid data features. Our approach is evaluated across three anomaly detection algorithms and consistently outperforms two baseline methods. Detailed ablation studies further support these findings. Furthermore, we introduce a signal integrity enhancement algorithm that improves signal integrity by an average of 11.3%. The source code and data used in this study are available at https://github.com/Usama1002/learning-latent-representations.