A machine learning methodology for inferring network S-parameters in the presence of variability

Xiao Ma; Maxim Raginsky; Andreas C. Cangellaris

doi:10.1109/SaPIW.2018.8401643

A machine learning methodology for inferring network S-parameters in the presence of variability

Xiao Ma, Maxim Raginsky, Andreas C. Cangellaris

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

Abstract

This paper proposes the use of Variational Autoencoders, a generative modeling technique, for the problem of inferring S-parameters of linear multiport networks in the presence of manufacturing variability. The Variational Autoencoder learns the underlying data generation process and yields a generative network that can approximately mimic the probability distribution of the training data. The generated samples can be used for subsequent statistical simulations. A post-processing step, applying Vector Fitting to the predicted S-parameters, constrains the model to a finite-order rational function form and enforces appropriate physical constraints. The method is validated through its application to a coupled micro strip transmission line.

Original language	English (US)
Title of host publication	2018 IEEE 22nd Workshop on Signal and Power Integrity, SPI 2018 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1-4
Number of pages	4
ISBN (Electronic)	9781538622995
DOIs	https://doi.org/10.1109/SaPIW.2018.8401643
State	Published - Jun 29 2018
Event	22nd IEEE Workshop on Signal and Power Integrity, SPI 2018 - Brest, France Duration: May 22 2018 → May 25 2018

Publication series

Name	2018 IEEE 22nd Workshop on Signal and Power Integrity, SPI 2018 - Proceedings

Other

Other	22nd IEEE Workshop on Signal and Power Integrity, SPI 2018
Country	France
City	Brest
Period	5/22/18 → 5/25/18

Keywords

Bayes methods
Inference algorithms
Integrated circuit interconnections
Statistical analysis
Unsupervised learning

ASJC Scopus subject areas

Signal Processing
Energy Engineering and Power Technology
Electrical and Electronic Engineering
Safety, Risk, Reliability and Quality

Access to Document

10.1109/SaPIW.2018.8401643

Fingerprint Dive into the research topics of 'A machine learning methodology for inferring network S-parameters in the presence of variability'. Together they form a unique fingerprint.

Cite this

Ma, X., Raginsky, M., & Cangellaris, A. C. (2018). A machine learning methodology for inferring network S-parameters in the presence of variability. In 2018 IEEE 22nd Workshop on Signal and Power Integrity, SPI 2018 - Proceedings (pp. 1-4). (2018 IEEE 22nd Workshop on Signal and Power Integrity, SPI 2018 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SaPIW.2018.8401643

A machine learning methodology for inferring network S-parameters in the presence of variability. / Ma, Xiao; Raginsky, Maxim ; Cangellaris, Andreas C.

2018 IEEE 22nd Workshop on Signal and Power Integrity, SPI 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. p. 1-4 (2018 IEEE 22nd Workshop on Signal and Power Integrity, SPI 2018 - Proceedings).

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

Ma, X, Raginsky, M & Cangellaris, AC 2018, A machine learning methodology for inferring network S-parameters in the presence of variability. in 2018 IEEE 22nd Workshop on Signal and Power Integrity, SPI 2018 - Proceedings. 2018 IEEE 22nd Workshop on Signal and Power Integrity, SPI 2018 - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 1-4, 22nd IEEE Workshop on Signal and Power Integrity, SPI 2018, Brest, France, 5/22/18. https://doi.org/10.1109/SaPIW.2018.8401643

Ma X, Raginsky M , Cangellaris AC. A machine learning methodology for inferring network S-parameters in the presence of variability. In 2018 IEEE 22nd Workshop on Signal and Power Integrity, SPI 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2018. p. 1-4. (2018 IEEE 22nd Workshop on Signal and Power Integrity, SPI 2018 - Proceedings). https://doi.org/10.1109/SaPIW.2018.8401643

Ma, Xiao ; Raginsky, Maxim ; Cangellaris, Andreas C. / A machine learning methodology for inferring network S-parameters in the presence of variability. 2018 IEEE 22nd Workshop on Signal and Power Integrity, SPI 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 1-4 (2018 IEEE 22nd Workshop on Signal and Power Integrity, SPI 2018 - Proceedings).

@inproceedings{5315efc06d144fe1b5cf9d8094caa163,

title = "A machine learning methodology for inferring network S-parameters in the presence of variability",

abstract = "This paper proposes the use of Variational Autoencoders, a generative modeling technique, for the problem of inferring S-parameters of linear multiport networks in the presence of manufacturing variability. The Variational Autoencoder learns the underlying data generation process and yields a generative network that can approximately mimic the probability distribution of the training data. The generated samples can be used for subsequent statistical simulations. A post-processing step, applying Vector Fitting to the predicted S-parameters, constrains the model to a finite-order rational function form and enforces appropriate physical constraints. The method is validated through its application to a coupled micro strip transmission line.",

keywords = "Bayes methods, Inference algorithms, Integrated circuit interconnections, Statistical analysis, Unsupervised learning",

author = "Xiao Ma and Maxim Raginsky and Cangellaris, {Andreas C.}",

year = "2018",

month = jun,

day = "29",

doi = "10.1109/SaPIW.2018.8401643",

language = "English (US)",

series = "2018 IEEE 22nd Workshop on Signal and Power Integrity, SPI 2018 - Proceedings",

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

pages = "1--4",

booktitle = "2018 IEEE 22nd Workshop on Signal and Power Integrity, SPI 2018 - Proceedings",

address = "United States",

note = "22nd IEEE Workshop on Signal and Power Integrity, SPI 2018 ; Conference date: 22-05-2018 Through 25-05-2018",

}

TY - GEN

T1 - A machine learning methodology for inferring network S-parameters in the presence of variability

AU - Ma, Xiao

AU - Raginsky, Maxim

AU - Cangellaris, Andreas C.

PY - 2018/6/29

Y1 - 2018/6/29

N2 - This paper proposes the use of Variational Autoencoders, a generative modeling technique, for the problem of inferring S-parameters of linear multiport networks in the presence of manufacturing variability. The Variational Autoencoder learns the underlying data generation process and yields a generative network that can approximately mimic the probability distribution of the training data. The generated samples can be used for subsequent statistical simulations. A post-processing step, applying Vector Fitting to the predicted S-parameters, constrains the model to a finite-order rational function form and enforces appropriate physical constraints. The method is validated through its application to a coupled micro strip transmission line.

AB - This paper proposes the use of Variational Autoencoders, a generative modeling technique, for the problem of inferring S-parameters of linear multiport networks in the presence of manufacturing variability. The Variational Autoencoder learns the underlying data generation process and yields a generative network that can approximately mimic the probability distribution of the training data. The generated samples can be used for subsequent statistical simulations. A post-processing step, applying Vector Fitting to the predicted S-parameters, constrains the model to a finite-order rational function form and enforces appropriate physical constraints. The method is validated through its application to a coupled micro strip transmission line.

KW - Bayes methods

KW - Inference algorithms

KW - Integrated circuit interconnections

KW - Statistical analysis

KW - Unsupervised learning

UR - http://www.scopus.com/inward/record.url?scp=85050505565&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85050505565&partnerID=8YFLogxK

U2 - 10.1109/SaPIW.2018.8401643

DO - 10.1109/SaPIW.2018.8401643

M3 - Conference contribution

AN - SCOPUS:85050505565

T3 - 2018 IEEE 22nd Workshop on Signal and Power Integrity, SPI 2018 - Proceedings

SP - 1

EP - 4

BT - 2018 IEEE 22nd Workshop on Signal and Power Integrity, SPI 2018 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 22nd IEEE Workshop on Signal and Power Integrity, SPI 2018

Y2 - 22 May 2018 through 25 May 2018

ER -