Publications

Publications in 2013 of type Article, Conference Proceedings and Edited Conference Proceedings

Filter by Year: All, 2018, 2017, 2016, 2015, 2014, 2013, 2012, 2011, 2010, 2009, 2008,
Filter by Type: All, Book, Article, Conference Proceedings, Edited Conference Proceedings, Master Theses, Bachelor Theses, Technical Reports, Miscellaneous,
Filter by Language: All, English, German,

    2013

    • Philipp Meyer, Till Steinbach, Franz Korf, and Thomas C. Schmidt. Extending IEEE 802.1 AVB with Time-triggered Scheduling: A Simulation Study of the Coexistence of Synchronous and Asynchronous Traffic. In: 2013 IEEE Vehicular Networking Conference (VNC). Pages 47—54, Piscataway, NJ, USA, Dec. 2013, IEEE Press,
      [Abstract], [Fulltext Document (pdf)], [DOI], [IEEE Xplore], [Bibtex]

      In-car networks based on Ethernet are expected to be the first choice for future applications in the domain of info- and entertainment. However, the full benefit of a technologically integrated in-car network will only become rewarding with an Ethernet-based backbone, unifying several automotive domains in a single infrastructure. Today, there is remarkable interest in the IEEE 802.1 Audio/Video Bridging (AVB) protocol suite, that provides end-to-end performance guarantees in Ethernet networks. But for the strict timing requirements of automotive control-traffic, these guarantees are too weak. An extension of Ethernet AVB with synchronous time-triggered traffic can overcome these limitations. In this paper, we investigate the coexistence of synchronous and asynchronous traffic by experimentally adding time-triggered messages to the credit-based shaper of AVB in a straightforward way. Based on simulations and analytical evaluations, we quantify the impact of such integration concepts for a reasonable design range. Our results demonstrate the feasibility of a shaping strategy with concurrent AVB and time-triggered message, but show a significant impact of the schedule design on the asynchronous AVB streams. Based on our findings, we provide recommendations for configurations that can improve end-to-end network performance for in-car applications by over 100 percent

      @InProceedings{   msks-eatts-13,
        author        = {Philipp Meyer AND Till Steinbach AND Franz Korf AND Thomas
                        C. Schmidt},
        title         = {{Extending IEEE 802.1 AVB with Time-triggered Scheduling:
                        A Simulation Study of the Coexistence of Synchronous and
                        Asynchronous Traffic}},
        booktitle     = {2013 IEEE Vehicular Networking Conference (VNC)},
        location      = {Boston, Massachusetts},
        month         = dec,
        year          = 2013,
        pages         = {47--54},
        publisher     = {IEEE Press},
        address       = {Piscataway, NJ, USA},
        isbn          = {978-1-4799-2686-2},
        issn          = {2157-9865},
        doi           = {10.1109/VNC.2013.6737589},
        eprinttype    = {ieeexplore},
        eprint        = {6737589},
        abstract      = {In-car networks based on Ethernet are expected to be the
                        first choice for future applications in the domain of info-
                        and entertainment. However, the full benefit of a
                        technologically integrated in-car network will only become
                        rewarding with an Ethernet-based backbone, unifying several
                        automotive domains in a single infrastructure. Today, there
                        is remarkable interest in the IEEE 802.1 Audio/Video
                        Bridging (AVB) protocol suite, that provides end-to-end
                        performance guarantees in Ethernet networks. But for the
                        strict timing requirements of automotive control-traffic,
                        these guarantees are too weak. An extension of Ethernet AVB
                        with synchronous time-triggered traffic can overcome these
                        limitations. In this paper, we investigate the coexistence
                        of synchronous and asynchronous traffic by experimentally
                        adding time-triggered messages to the credit-based shaper
                        of AVB in a straightforward way. Based on simulations and
                        analytical evaluations, we quantify the impact of such
                        integration concepts for a reasonable design range. Our
                        results demonstrate the feasibility of a shaping strategy
                        with concurrent AVB and time-triggered message, but show a
                        significant impact of the schedule design on the
                        asynchronous AVB streams. Based on our findings, we provide
                        recommendations for configurations that can improve
                        end-to-end network performance for in-car applications by
                        over 100 percent},
        langid        = {english}
      }
    • Stefan Buschmann, Till Steinbach, Franz Korf, and Thomas C. Schmidt. Simulation based Timing Analysis of FlexRay Communication at System Level. In: Proceedings of the 6th International ICST Conference on Simulation Tools and Techniques. Pages 285—290, New York, Mar. 2013, ACM-DL,
      [Abstract], [Fulltext Document (pdf)], [Slides (pdf)], [ACM DL], [Bibtex]

      In modern cars the communication infrastructure consists of different application specific bus systems that are interconnected with each other. Due to the growing complexity of the communication infrastructure, the corresponding timing analysis at system level is currently a hot topic in the automotive industry. FlexRay is a state-of-the-art fieldbus for cars. While FlexRay simulations below system level are already established in automotive tool chains, FlexRay system level simulation is not yet common. This paper focuses on simulation-based timing analyses for FlexRay communication. Based on different scenarios the practical relevance is shown. The performance results promise that our simulation approach is a good building block for the simulation of heterogeneous communication consisting of several bus systems and technologies. An evaluation of the simulation results in comparison with the CANoe network simulator proves the conformance of the implementation with the FlexRay specification.

      @InProceedings{   bsks-stafc-13,
        author        = {Stefan Buschmann AND Till Steinbach AND Franz Korf AND
                        Thomas C. Schmidt},
        title         = {{Simulation based Timing Analysis of FlexRay Communication
                        at System Level}},
        booktitle     = {Proceedings of the 6th International ICST Conference on
                        Simulation Tools and Techniques},
        location      = {Cannes, France},
        month         = mar,
        year          = 2013,
        pages         = {285--290},
        publisher     = {ACM-DL},
        address       = {New York},
        isbn          = {978-1-4503-2464-9},
        eprinttype    = {acmdl},
        eprint        = {2512775},
        abstract      = {In modern cars the communication infrastructure consists
                        of different application specific bus systems that are
                        interconnected with each other. Due to the growing
                        complexity of the communication infrastructure, the
                        corresponding timing analysis at system level is currently
                        a hot topic in the automotive industry. FlexRay is a
                        state-of-the-art fieldbus for cars. While FlexRay
                        simulations below system level are already established in
                        automotive tool chains, FlexRay system level simulation is
                        not yet common. This paper focuses on simulation-based
                        timing analyses for FlexRay communication. Based on
                        different scenarios the practical relevance is shown. The
                        performance results promise that our simulation approach is
                        a good building block for the simulation of heterogeneous
                        communication consisting of several bus systems and
                        technologies. An evaluation of the simulation results in
                        comparison with the CANoe network simulator proves the
                        conformance of the implementation with the FlexRay
                        specification.},
        langid        = {english}
      }
    • Oleg Karfich, Florian Bartols, Till Steinbach, Franz Korf, and Thomas C. Schmidt. A Hardware/Software Platform for Real-time Ethernet Cluster Simulation in OMNeT++. In: Proceedings of the 6th International ICST Conference on Simulation Tools and Techniques. Pages 334—337, New York, Mar. 2013, ACM-DL,
      [Abstract], [Fulltext Document (pdf)], [Poster (pdf)], [Slides (pdf)], [ACM DL], [Bibtex]

      Cluster simulation is a popular method for supporting system integration in various distributed applications by simulating the environment of a subsystem under test. Particularly in real-time systems, the timing requirements of transmission and reception must be fulfilled, which is not easy to achieve. In this paper, we contribute a scheme for cluster simulation of real-time Ethernet (RTEthernet) based distributed systems. It relies on the discrete event-based simulation framework OMNeT++, interconnected with an ARM-based co-processor. Our approach allows coupling a real-world RTEthernet subsystem with virtual components running in the discrete simulation, that realise the required behaviour for the subsystem. We have evaluated the performance limits of our approach regarding latency and jitter, when running the simulation on a Linux system with the real-time Kernel patch. The results show that the timing requirements for the cluster simulation of small RTEthernet networks can be achieved.

      @InProceedings{   ksbks-hspre-13,
        ids           = {ksbks-eifre-13},
        author        = {Oleg Karfich AND Florian Bartols AND Till Steinbach AND
                        Franz Korf AND Thomas C. Schmidt},
        title         = {{A Hardware/Software Platform for Real-time Ethernet
                        Cluster Simulation in OMNeT++}},
        booktitle     = {Proceedings of the 6th International ICST Conference on
                        Simulation Tools and Techniques},
        location      = {Cannes, France},
        month         = mar,
        year          = 2013,
        pages         = {334--337},
        publisher     = {ACM-DL},
        address       = {New York},
        isbn          = {978-1-4503-2464-9},
        eprinttype    = {acmdl},
        eprint        = {2512782},
        abstract      = {Cluster simulation is a popular method for supporting
                        system integration in various distributed applications by
                        simulating the environment of a subsystem under test.
                        Particularly in real-time systems, the timing requirements
                        of transmission and reception must be fulfilled, which is
                        not easy to achieve. In this paper, we contribute a scheme
                        for cluster simulation of real-time Ethernet (RTEthernet)
                        based distributed systems. It relies on the discrete
                        event-based simulation framework OMNeT++, interconnected
                        with an ARM-based co-processor. Our approach allows
                        coupling a real-world RTEthernet subsystem with virtual
                        components running in the discrete simulation, that realise
                        the required behaviour for the subsystem. We have evaluated
                        the performance limits of our approach regarding latency
                        and jitter, when running the simulation on a Linux system
                        with the real-time Kernel patch. The results show that the
                        timing requirements for the cluster simulation of small
                        RTEthernet networks can be achieved.},
        langid        = {english}
      }
    • Lazar T. Todorov, Till Steinbach, Franz Korf, and Thomas C. Schmidt. Evaluating Requirements of High Precision Time Synchronisation Protocols using Simulation. In: Proceedings of the 6th International ICST Conference on Simulation Tools and Techniques. Pages 307—313, New York, Mar. 2013, ACM-DL,
      [Abstract], [Fulltext Document (pdf)], [Slides (pdf)], [ACM DL], [Bibtex]

      High precision time synchronisation protocols are used in distributed real-time systems such as trains, planes, cars or industrial installations. In time-triggered systems, with a coordinated time division multiple access media allocation strategy, the achievable precision of time synchronisation among sending participants determines the quality of communication and the available bandwidth. The simulation of time synchronisation protocols allows to find problems at the earliest time — in general, during the design and configuration — of a synchronised distributed system. In this work we show a concept for the simulation of distributed real-time synchronisation protocols that uses discrete event-based simulation. Our model for the OMNeT++ Framework is adaptable and thus allows for providing highly accurate results or fast simulations. The precise simulation of a real-time synchronisation protocol usually consumes considerable simulation time. This paper presents an approach to speed up accurate simulation, based on recordings of previous runs. We evaluate typical real-world use cases for the introduced concept by simulating the AS6802 standard for time synchronisation. Our results show that the simulation can help to reduce the effort of determining configuration parameters for clock synchronisation protocols. We further quantify the performance increase of our evolutionary approach.

      @InProceedings{   tsks-erhpt-13,
        author        = {Lazar T. Todorov AND Till Steinbach AND Franz Korf AND
                        Thomas C. Schmidt},
        title         = {{Evaluating Requirements of High Precision Time
                        Synchronisation Protocols using Simulation}},
        booktitle     = {Proceedings of the 6th International ICST Conference on
                        Simulation Tools and Techniques},
        location      = {Cannes, France},
        month         = mar,
        year          = 2013,
        pages         = {307--313},
        publisher     = {ACM-DL},
        address       = {New York},
        isbn          = {978-1-4503-2464-9},
        eprinttype    = {acmdl},
        eprint        = {2512778},
        abstract      = {High precision time synchronisation protocols are used in
                        distributed real-time systems such as trains, planes, cars
                        or industrial installations. In time-triggered systems,
                        with a coordinated time division multiple access media
                        allocation strategy, the achievable precision of time
                        synchronisation among sending participants determines the
                        quality of communication and the available bandwidth. The
                        simulation of time synchronisation protocols allows to find
                        problems at the earliest time -- in general, during the
                        design and configuration -- of a synchronised distributed
                        system. In this work we show a concept for the simulation
                        of distributed real-time synchronisation protocols that
                        uses discrete event-based simulation. Our model for the
                        OMNeT++ Framework is adaptable and thus allows for
                        providing highly accurate results or fast simulations. The
                        precise simulation of a real-time synchronisation protocol
                        usually consumes considerable simulation time. This paper
                        presents an approach to speed up accurate simulation, based
                        on recordings of previous runs. We evaluate typical
                        real-world use cases for the introduced concept by
                        simulating the AS6802 standard for time synchronisation.
                        Our results show that the simulation can help to reduce the
                        effort of determining configuration parameters for clock
                        synchronisation protocols. We further quantify the
                        performance increase of our evolutionary approach.},
        langid        = {english}
      }