Fine-grained Simulation in the Design of Automotive Communication Systems

Early in the design cycle, the two main approaches for verifying timing constraints and di- mensioning automotive embedded networks are worst-case schedulability analysis and simulation. The first aim of the paper is to demonstrate that both provide complementary results and that, most often, none of them alone is sufficient. In this paper, we present a simulation approach accounting for the clock drifts that occur on the network nodes at run- time and evaluate the extent to which the results ob- tained with this approach are relevant for the design- ers in order to validate the performances of a CAN- based communication system. One of the practical outcome of this study is to show that initial phasings between nodes, as well as the values of the clock drifts, do not significantly impact the frame response time distributions that can be observed on the long run.

[1]  Alan Burns,et al.  Calculating controller area network (can) message response times , 1994 .

[2]  Jörn Migge,et al.  Frame latency evaluation: when simulation and analysis alone are not enough , 2010, WFCS 2010.

[3]  Alan Burns,et al.  Controller Area Network (CAN) schedulability analysis: Refuted, revisited and revised , 2007, Real-Time Systems.

[4]  Jörn Migge,et al.  Aperiodic traffic in response time analyses with adjustable safety level , 2009, 2009 IEEE Conference on Emerging Technologies & Factory Automation.

[5]  Alf J. Isaksson,et al.  Simulation of Process Control with WirelessHART Networks Subject to Clock Drift , 2008, 2008 32nd Annual IEEE International Computer Software and Applications Conference.

[6]  Reinder J. Bril,et al.  Integrating hardware limitations in CAN schedulability analysis , 2010, 2010 IEEE International Workshop on Factory Communication Systems Proceedings.

[7]  Michael J. Pont,et al.  Reducing message-length variations in resource-constrained embedded systems implemented using the Controller Area Network (CAN) protocol , 2009, J. Syst. Archit..

[8]  Robert I. Davis,et al.  Schedulability analysis of CAN with non-abortable transmission requests , 2011, ETFA2011.

[9]  Robert I. Davis,et al.  Controller Area Network (CAN) Schedulability Analysis with FIFO Queues , 2011, 2011 23rd Euromicro Conference on Real-Time Systems.

[10]  Alberto L. Sangiovanni-Vincentelli,et al.  Using Statistical Methods to Compute the Probability Distribution of Message Response Time in Controller Area Network , 2010, IEEE Transactions on Industrial Informatics.

[11]  Nicolas Navet,et al.  Impact of clock drifts on CAN frame response time distributions , 2011, ETFA2011.

[12]  Mathieu Grenier,et al.  Pushing the limits of CAN - scheduling frames with offsets provides a major performance boost , 2008 .

[13]  Alberto Sangiovanni-Vincentelli,et al.  Stochastic Analysis of CAN-Based Real-Time Automotive Systems , 2009, IEEE Transactions on Industrial Informatics.

[14]  Thomas Nolte,et al.  Probabilistic worst-case response-time analysis for the controller area network , 2003, The 9th IEEE Real-Time and Embedded Technology and Applications Symposium, 2003. Proceedings..

[15]  Yeqiong Song,et al.  Worst-case deadline failure probability in real-time applications distributed over controller area network , 2000, J. Syst. Archit..