RX equalization for a high-speed channel based on bayesian active learning using dropout

Xianbo Yang; Junyan Tang; Hakki M. Torun; Wiren D. Becker; Jose A. Hejase; Madhavan Swaminathan

doi:10.1109/EPEPS48591.2020.9231471

RX equalization for a high-speed channel based on bayesian active learning using dropout

Xianbo Yang, Junyan Tang, Hakki M. Torun, Wiren D. Becker, Jose A. Hejase, Madhavan Swaminathan

Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

6 Scopus citations

Abstract

Determining optimal equalization settings in highspeed bus design simulations is becoming more important due to increased complexity and data rates of current server systems, but it is also time and resource consuming. In this paper, a probabilistic machine learning technique, Bayesian Active Learning using Dropout (BAL-DO), is utilized to perform RX equalization and optimization to address this issue. Largest HEYE opening and corresponding equalization settings are obtained with high prediction accuracy without performing extensive time-domain analysis, thereby significantly reducing the cost of engineering time and computational resources.

Original language	English (US)
Title of host publication	EPEPS 2020 - IEEE 29th Conference on Electrical Performance of Electronic Packaging and Systems
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781728161617
DOIs	https://doi.org/10.1109/EPEPS48591.2020.9231471
State	Published - Oct 2020
Event	29th IEEE Conference on Electrical Performance of Electronic Packaging and Systems, EPEPS 2020 - San Jose, United States Duration: Oct 5 2020 → Oct 7 2020

Publication series

Name	EPEPS 2020 - IEEE 29th Conference on Electrical Performance of Electronic Packaging and Systems

Conference

Conference	29th IEEE Conference on Electrical Performance of Electronic Packaging and Systems, EPEPS 2020
Country/Territory	United States
City	San Jose
Period	10/5/20 → 10/7/20

All Science Journal Classification (ASJC) codes

Electrical and Electronic Engineering
Electronic, Optical and Magnetic Materials
Energy Engineering and Power Technology

Access to Document

10.1109/EPEPS48591.2020.9231471

Cite this

Yang, X., Tang, J., Torun, H. M., Becker, W. D., Hejase, J. A., & Swaminathan, M. (2020). RX equalization for a high-speed channel based on bayesian active learning using dropout. In EPEPS 2020 - IEEE 29th Conference on Electrical Performance of Electronic Packaging and Systems Article 9231471 (EPEPS 2020 - IEEE 29th Conference on Electrical Performance of Electronic Packaging and Systems). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/EPEPS48591.2020.9231471

Yang, Xianbo ; Tang, Junyan ; Torun, Hakki M. et al. / RX equalization for a high-speed channel based on bayesian active learning using dropout. EPEPS 2020 - IEEE 29th Conference on Electrical Performance of Electronic Packaging and Systems. Institute of Electrical and Electronics Engineers Inc., 2020. (EPEPS 2020 - IEEE 29th Conference on Electrical Performance of Electronic Packaging and Systems).

@inproceedings{2e01ebfe4bdc440ebddc84b58ca24e0d,

title = "RX equalization for a high-speed channel based on bayesian active learning using dropout",

abstract = "Determining optimal equalization settings in highspeed bus design simulations is becoming more important due to increased complexity and data rates of current server systems, but it is also time and resource consuming. In this paper, a probabilistic machine learning technique, Bayesian Active Learning using Dropout (BAL-DO), is utilized to perform RX equalization and optimization to address this issue. Largest HEYE opening and corresponding equalization settings are obtained with high prediction accuracy without performing extensive time-domain analysis, thereby significantly reducing the cost of engineering time and computational resources.",

author = "Xianbo Yang and Junyan Tang and Torun, {Hakki M.} and Becker, {Wiren D.} and Hejase, {Jose A.} and Madhavan Swaminathan",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 29th IEEE Conference on Electrical Performance of Electronic Packaging and Systems, EPEPS 2020 ; Conference date: 05-10-2020 Through 07-10-2020",

year = "2020",

month = oct,

doi = "10.1109/EPEPS48591.2020.9231471",

language = "English (US)",

series = "EPEPS 2020 - IEEE 29th Conference on Electrical Performance of Electronic Packaging and Systems",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "EPEPS 2020 - IEEE 29th Conference on Electrical Performance of Electronic Packaging and Systems",

address = "United States",

}

Yang, X, Tang, J, Torun, HM, Becker, WD, Hejase, JA & Swaminathan, M 2020, RX equalization for a high-speed channel based on bayesian active learning using dropout. in EPEPS 2020 - IEEE 29th Conference on Electrical Performance of Electronic Packaging and Systems., 9231471, EPEPS 2020 - IEEE 29th Conference on Electrical Performance of Electronic Packaging and Systems, Institute of Electrical and Electronics Engineers Inc., 29th IEEE Conference on Electrical Performance of Electronic Packaging and Systems, EPEPS 2020, San Jose, United States, 10/5/20. https://doi.org/10.1109/EPEPS48591.2020.9231471

RX equalization for a high-speed channel based on bayesian active learning using dropout. / Yang, Xianbo; Tang, Junyan; Torun, Hakki M. et al.
EPEPS 2020 - IEEE 29th Conference on Electrical Performance of Electronic Packaging and Systems. Institute of Electrical and Electronics Engineers Inc., 2020. 9231471 (EPEPS 2020 - IEEE 29th Conference on Electrical Performance of Electronic Packaging and Systems).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - RX equalization for a high-speed channel based on bayesian active learning using dropout

AU - Yang, Xianbo

AU - Tang, Junyan

AU - Torun, Hakki M.

AU - Becker, Wiren D.

AU - Hejase, Jose A.

AU - Swaminathan, Madhavan

PY - 2020/10

Y1 - 2020/10

N2 - Determining optimal equalization settings in highspeed bus design simulations is becoming more important due to increased complexity and data rates of current server systems, but it is also time and resource consuming. In this paper, a probabilistic machine learning technique, Bayesian Active Learning using Dropout (BAL-DO), is utilized to perform RX equalization and optimization to address this issue. Largest HEYE opening and corresponding equalization settings are obtained with high prediction accuracy without performing extensive time-domain analysis, thereby significantly reducing the cost of engineering time and computational resources.

AB - Determining optimal equalization settings in highspeed bus design simulations is becoming more important due to increased complexity and data rates of current server systems, but it is also time and resource consuming. In this paper, a probabilistic machine learning technique, Bayesian Active Learning using Dropout (BAL-DO), is utilized to perform RX equalization and optimization to address this issue. Largest HEYE opening and corresponding equalization settings are obtained with high prediction accuracy without performing extensive time-domain analysis, thereby significantly reducing the cost of engineering time and computational resources.

UR - http://www.scopus.com/inward/record.url?scp=85096972338&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85096972338&partnerID=8YFLogxK

U2 - 10.1109/EPEPS48591.2020.9231471

DO - 10.1109/EPEPS48591.2020.9231471

M3 - Conference contribution

AN - SCOPUS:85096972338

T3 - EPEPS 2020 - IEEE 29th Conference on Electrical Performance of Electronic Packaging and Systems

BT - EPEPS 2020 - IEEE 29th Conference on Electrical Performance of Electronic Packaging and Systems

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 29th IEEE Conference on Electrical Performance of Electronic Packaging and Systems, EPEPS 2020

Y2 - 5 October 2020 through 7 October 2020

ER -

Yang X, Tang J, Torun HM, Becker WD, Hejase JA, Swaminathan M. RX equalization for a high-speed channel based on bayesian active learning using dropout. In EPEPS 2020 - IEEE 29th Conference on Electrical Performance of Electronic Packaging and Systems. Institute of Electrical and Electronics Engineers Inc. 2020. 9231471. (EPEPS 2020 - IEEE 29th Conference on Electrical Performance of Electronic Packaging and Systems). doi: 10.1109/EPEPS48591.2020.9231471

RX equalization for a high-speed channel based on bayesian active learning using dropout

Abstract

Publication series

Conference

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this