Advanced search
2 files | 21.74 MB

A generative modeling framework for statistical link analysis based on sparse data

Simon De Ridder (UGent) , Paolo Manfredi (UGent) , Jan De Geest, Dirk Deschrijver (UGent) , Daniël De Zutter (UGent) , Tom Dhaene (UGent) and Dries Vande Ginste (UGent)
Author
Organization
Abstract
This paper proposes a novel strategy for creating generative models of stochastic link responses starting from limited available data. Whereas state-of-the-art techniques, e.g., based on generalized polynomial chaos expansions, require a considerable amount of (expensive) input data, here we start from a small set of "training" responses. These responses are obtained either from simulations or measurements to construct a comprehensive stochastic model. Using this model, new response samples can be generated with a distribution as similar as possible to the real data distribution, for use in Monte Carlo-like analyses. The methodology first uses the standard Vector Fitting algorithm to fit the S-parameter data with rational functions having common poles. Then, a generative model for the residues is created by means of principal component analysis and kernel density estimation. An a posteriori selection of passive samples is performed on the generated data to ensure the new samples are physically consistent. The proposed modeling approach is applied to a commercial connector and to a set of differential striplines. Both are concatenated to produce the stochastic analysis of a complete link. Comparisons on the prediction of time-domain responses are also provided.
Keywords
IBCN, PRINCIPAL COMPONENT ANALYSIS, POLYNOMIAL CHAOS, UNCERTAINTY, QUANTIFICATION, PASSIVITY ENFORCEMENT, NONLINEAR CIRCUITS, FREQUENCY-DOMAIN, DENSITY-FUNCTION, MACROMODELS, PARAMETERS, LOSSY, High-speed connectors and links, kernel density estimation (KDE), principal component analysis (PCA), statistical link analysis, stochastic modeling

Downloads

  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 9.65 MB
  • 7096 i.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 12.10 MB

Citation

Please use this url to cite or link to this publication:

Chicago
De Ridder, Simon, Paolo Manfredi, Jan De Geest, Dirk Deschrijver, Daniël De Zutter, Tom Dhaene, and Dries Vande Ginste. 2018. “A Generative Modeling Framework for Statistical Link Analysis Based on Sparse Data.” Ieee Transactions on Components Packaging and Manufacturing Technology 8 (1): 21–31.
APA
De Ridder, Simon, Manfredi, P., De Geest, J., Deschrijver, D., De Zutter, D., Dhaene, T., & Vande Ginste, D. (2018). A generative modeling framework for statistical link analysis based on sparse data. IEEE TRANSACTIONS ON COMPONENTS PACKAGING AND MANUFACTURING TECHNOLOGY, 8(1), 21–31.
Vancouver
1.
De Ridder S, Manfredi P, De Geest J, Deschrijver D, De Zutter D, Dhaene T, et al. A generative modeling framework for statistical link analysis based on sparse data. IEEE TRANSACTIONS ON COMPONENTS PACKAGING AND MANUFACTURING TECHNOLOGY. Piscataway: Ieee-inst Electrical Electronics Engineers Inc; 2018;8(1):21–31.
MLA
De Ridder, Simon, Paolo Manfredi, Jan De Geest, et al. “A Generative Modeling Framework for Statistical Link Analysis Based on Sparse Data.” IEEE TRANSACTIONS ON COMPONENTS PACKAGING AND MANUFACTURING TECHNOLOGY 8.1 (2018): 21–31. Print.
@article{8547195,
  abstract     = {This paper proposes a novel strategy for creating generative models of stochastic link responses starting from limited available data. Whereas state-of-the-art techniques, e.g., based on generalized polynomial chaos expansions, require a considerable amount of (expensive) input data, here we start from a small set of {\textacutedbl}training{\textacutedbl} responses. These responses are obtained either from simulations or measurements to construct a comprehensive stochastic model. Using this model, new response samples can be generated with a distribution as similar as possible to the real data distribution, for use in Monte Carlo-like analyses. The methodology first uses the standard Vector Fitting algorithm to fit the S-parameter data with rational functions having common poles. Then, a generative model for the residues is created by means of principal component analysis and kernel density estimation. An a posteriori selection of passive samples is performed on the generated data to ensure the new samples are physically consistent. The proposed modeling approach is applied to a commercial connector and to a set of differential striplines. Both are concatenated to produce the stochastic analysis of a complete link. Comparisons on the prediction of time-domain responses are also provided.},
  author       = {De Ridder, Simon and Manfredi, Paolo and De Geest, Jan and Deschrijver, Dirk and De Zutter, Dani{\"e}l and Dhaene, Tom and Vande Ginste, Dries},
  issn         = {2156-3950},
  journal      = {IEEE TRANSACTIONS ON COMPONENTS PACKAGING AND MANUFACTURING TECHNOLOGY},
  keyword      = {IBCN,PRINCIPAL COMPONENT ANALYSIS,POLYNOMIAL CHAOS,UNCERTAINTY,QUANTIFICATION,PASSIVITY ENFORCEMENT,NONLINEAR CIRCUITS,FREQUENCY-DOMAIN,DENSITY-FUNCTION,MACROMODELS,PARAMETERS,LOSSY,High-speed connectors and links,kernel density estimation (KDE),principal component analysis (PCA),statistical link analysis,stochastic modeling},
  language     = {eng},
  number       = {1},
  pages        = {21--31},
  publisher    = {Ieee-inst Electrical Electronics Engineers Inc},
  title        = {A generative modeling framework for statistical link analysis based on sparse data},
  url          = {http://dx.doi.org/10.1109/TCPMT.2017.2761907},
  volume       = {8},
  year         = {2018},
}

Altmetric
View in Altmetric
Web of Science
Times cited: