Publication detail

Parametric Reduction of Jacobian Matrix for Fault Analysis

KOLKA, Z. KINCL, Z. BIOLEK, D. BIOLKOVÁ, V.

Original Title

Parametric Reduction of Jacobian Matrix for Fault Analysis

English Title

Parametric Reduction of Jacobian Matrix for Fault Analysis

Type

conference paper

Language

en

Original Abstract

The paper deals with a fast numerical method for generating the Jacobian matrix of network function for large analog linear circuits. It is based on the use of cofactor matrices. The computational cost of the Jacobian is comparable to the cost of computing a single network function on selected frequencies. In addition, the method allows performing the reduction of Jacobian matrix based on large-change sensitivities in order to decrease the computational complexity of subsequent testability analysis. An example analysis of a frequency filter is presented.

English abstract

The paper deals with a fast numerical method for generating the Jacobian matrix of network function for large analog linear circuits. It is based on the use of cofactor matrices. The computational cost of the Jacobian is comparable to the cost of computing a single network function on selected frequencies. In addition, the method allows performing the reduction of Jacobian matrix based on large-change sensitivities in order to decrease the computational complexity of subsequent testability analysis. An example analysis of a frequency filter is presented.

Keywords

Analog fault diagnosis, circuit reduction, fault location, numerical methods.

RIV year

2010

Released

19.12.2010

Publisher

IEEE

Location

Cairo

ISBN

978-1-61284-151-9

Book

Proc. of the 22nd IEEE International Conference on Microelectronics (ICM 2010)

Pages from

503

Pages to

506

Pages count

4

BibTex


@inproceedings{BUT34795,
  author="Zdeněk {Kolka} and Zdeněk {Kincl} and Dalibor {Biolek} and Viera {Biolková}",
  title="Parametric Reduction of Jacobian Matrix for Fault Analysis",
  annote="The paper deals with a fast numerical method for generating the Jacobian matrix of network function for large analog linear circuits. It is based on the use of cofactor matrices. The computational cost of the Jacobian is comparable to the cost of computing a single network function on selected frequencies. In addition, the method allows performing the reduction of Jacobian matrix based on large-change sensitivities in order to decrease the computational complexity of subsequent testability analysis. An example analysis of a frequency filter is presented.",
  address="IEEE",
  booktitle="Proc. of the 22nd IEEE International Conference on Microelectronics (ICM 2010)",
  chapter="34795",
  doi="10.1109/ICM.2010.5696200",
  howpublished="electronic, physical medium",
  institution="IEEE",
  year="2010",
  month="december",
  pages="503--506",
  publisher="IEEE",
  type="conference paper"
}