Adaptive Gramian-Angular-Field Segmentation Integration Based Generative Adversarial Network (AGSI-GAN) for Eye Diagram Estimation of High Bandwidth Memory (HBM) Interposer

被引：3

作者：

Lee, Junghyun ^{[1
]}

Choi, Seonguk ^{[1
]}

Son, Keeyoung ^{[1
]}

Park, Joonsang ^{[1
]}

Kim, Hyunwoo ^{[1
]}

Kim, Keunwoo ^{[1
]}

Shin, Taein ^{[1
]}

Sim, Boogyo ^{[1
]}

Hong, Jonghyun ^{[1
]}

Yoon, Jiwon ^{[1
]}

Kim, Juneyoung ^{[1
]}

Kim, Joungho ^{[1
]}

机构：

[1] Korea Adv Inst Scic & Technol KAIST, Sch Elect Engn, Daejeon, South Korea

来源：

2023 IEEE 32ND CONFERENCE ON ELECTRICAL PERFORMANCE OF ELECTRONIC PACKAGING AND SYSTEMS, EPEPS | 2023年

基金：

新加坡国家研究基金会;

关键词：

Adaptive Gramian-Angular-Field segmentation integration; eye diagram estimation; generative adversarial networks; high bandwidth memory;

D O I：

10.1109/EPEPS58208.2023.10314929

中图分类号：

TP301 [理论、方法];

学科分类号：

081202 ;

摘要：

In this paper, we propose an adaptive Gramian-Angular-Field segmentation integration (AGSI) based generative adversarial networks (GANs) for eye diagram estimation, which is a novel approach in high bandwidth memory (HBM) interposer design. The proposed AGSI framework effectively integrated relevant signal integrity (SI) components, namely single bit response (SBR) and far-end crosstalk (FEXT). AGSI allows for a customized and accurate representation of various design scenarios. By employing GAN-based estimator with AGSI as input data, the model significantly improved the time efficiency and accuracy of eye diagram estimation. This novel approach has potential applications in other domains, including signal integrity/power integrity (SI/PI) co-simulation, and is expected to enhance the design process in terms of time cost for HBM interposers and beyond.

引用

页数：3

共 7 条

[1] An accurate and efficient analysis method for multi-Gb/s chip-to-chip signaling schemes [J].

Casper, BK ;

Haycock, M ;

Mooney, R .

2002 SYMPOSIUM ON VLSI CIRCUITS, DIGEST OF TECHNICAL PAPERS, 2002, :54-57

[2]

Choi S, 2021, 2021 IEEE 30 C ELECT, P1

[3] Image-to-Image Translation with Conditional Adversarial Networks [J].

Isola, Phillip ;

Zhu, Jun-Yan ;

Zhou, Tinghui ;

Efros, Alexei A. .

30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, :5967-5976

[4] High Speed Receiver Modeling Using Generative Adversarial Networks [J].

Kashyap, Priyank ;

Pitts, W. Shepherd ;

Baron, Dror ;

Wong, Chau-Wai ;

Wu, Tianfu ;

Franzon, Paul D. .

IEEE 30TH CONFERENCE ON ELECTRICAL PERFORMANCE OF ELECTRONIC PACKAGING AND SYSTEMS (EPEPS 2021), 2021,

[5]

Lho D, 2018, IEEE C ELECTR PERFOR, P209, DOI 10.1109/EPEPS.2018.8534277

[6]

Standard JEDEC., 2021, High Bandwidth Memory DRAM (HBM1, HBM2), JESD235D

[7]

Wang ZG, 2015, PROCEEDINGS OF THE TWENTY-FOURTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE (IJCAI), P3939

← 1 →