A crosstalk sensitivity analysis on bundles of twisted wire pairs

Jesper Hermanus Gerhardus Johannes Lansink Rotgerink, Frank Leferink

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    3 Citations (Scopus)
    163 Downloads (Pure)

    Abstract

    Uncertainties in the geometry of complex cable bundles highly complicate crosstalk predictions. A change in for instance the position or the twist rate of individual cables in a bundle might have an impact on crosstalk levels. Application of sensitivity analyses can indicate which model parameters are most sensitive, and in which cabling configurations. In this paper, the efficient Stochastic Reduced Order Models (SROM) method is used to perform such a sensitivity analysis. It is applied to two cable bundles with two and seven twisted wire pairs, respectively. Monte Carlo simulations are performed to determine the accuracy of the SROM method. The sensitivity of parameters like inter-pair and intra-pair separation distance and the twist rate is determined in two different cases. Moreover, the effects of bundle twist and cable meandering to parameter sensitivities is investigated.
    Original languageEnglish
    Title of host publicationProceedings in 2018 International Symposium on Electromagnetic Compatibility
    PublisherIEEE
    Pages11-16
    Number of pages6
    ISBN (Electronic)978-1-4673-9698-1
    ISBN (Print)978-1-4673-9699-8
    DOIs
    Publication statusPublished - 8 Oct 2018
    EventEMC EUROPE 2018: International Symposium on Electromagnetic Compatibility - Beurs van Berlage, Amsterdam, Netherlands
    Duration: 27 Aug 201830 Aug 2018
    https://www.emceurope2018.org/

    Conference

    ConferenceEMC EUROPE 2018
    Abbreviated titleEMC EUROPE
    Country/TerritoryNetherlands
    CityAmsterdam
    Period27/08/1830/08/18
    Internet address

    Fingerprint

    Dive into the research topics of 'A crosstalk sensitivity analysis on bundles of twisted wire pairs'. Together they form a unique fingerprint.

    Cite this