• DocumentCode
    271871
  • Title

    Schedule integration framework for time-triggered automotive architectures

  • Author

    Sagstetter, Florian ; Andalam, Sidharta ; Waszecki, Peter ; Lukasiewycz, Martin ; Stähle, Hauke ; Chakraborty, Shiladri ; Knoll, Aaron

  • Author_Institution
    TUM CREATE, Singapore, Singapore
  • fYear
    2014
  • fDate
    1-5 June 2014
  • Firstpage
    1
  • Lastpage
    6
  • Abstract
    Automotive Electrical/Electronic (E/E)-architectures consist of various components which are generally developed independently. Due to the increasing size and complexity, component integration is highly challenging and already slight modifications to components or subsystems often require expensive re-testing and re-validation. As a remedy, we propose a framework for modular architectures based on a data-centric description and a fully time-triggered scheduling. This modular design approach is enabled by a novel methodology for schedule integration where local schedules are defined independently for subsystems before being integrated into a global schedule. This divide-and-conquer approach significantly reduces the integration complexity while the system becomes highly composable. Our experimental results give evidence of the efficiency and versatility of the proposed approach, using networks based on a time-triggered automotive Ethernet.
  • Keywords
    automotive electronics; computational complexity; divide and conquer methods; local area networks; scheduling; automotive electrical-electronic architecture; component integration; data-centric description; divide-and-conquer approach; integration complexity; modular architecture; modular design approach; schedule integration framework; time-triggered automotive Ethernet; time-triggered automotive architectures; time-triggered scheduling; Automotive engineering; Delays; Hardware; Optimal scheduling; Runtime; Schedules; Software;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Design Automation Conference (DAC), 2014 51st ACM/EDAC/IEEE
  • Conference_Location
    San Francisco, CA
  • Type

    conf

  • DOI
    10.1145/2593069.2593211
  • Filename
    6881347