You are required to read and agree to the below before accessing a full-text version of an article in the IDE article repository.

The full-text document you are about to access is subject to national and international copyright laws. In most cases (but not necessarily all) the consequence is that personal use is allowed given that the copyright owner is duly acknowledged and respected. All other use (typically) require an explicit permission (often in writing) by the copyright owner.

For the reports in this repository we specifically note that

  • the use of articles under IEEE copyright is governed by the IEEE copyright policy (available at http://www.ieee.org/web/publications/rights/copyrightpolicy.html)
  • the use of articles under ACM copyright is governed by the ACM copyright policy (available at http://www.acm.org/pubs/copyright_policy/)
  • technical reports and other articles issued by M‰lardalen University is free for personal use. For other use, the explicit consent of the authors is required
  • in other cases, please contact the copyright owner for detailed information

By accepting I agree to acknowledge and respect the rights of the copyright owner of the document I am about to access.

If you are in doubt, feel free to contact webmaster@ide.mdh.se

Reducing pessimism in CAN response time analysis

Authors:


Publication Type:

Student Thesis

Publisher:

MRTC


Abstract

This thesis investigates the level of pessimism in the traditional schedulability analysis for the Controller Area Network (CAN). Specifically, we investigate the effects of considering bit-stuffing distributions instead of worst case bit-stuffing. This allows us to obtain bus utilisation values more close to reality. On the other hand, since our analysis is based on assumptions concerning distributions of stuff-bits, our response times will only be met with some probability.We introduce a model and some methods, that relax the pessimism of the worst case analysis, and we show the effect of our methods by considering both an artificial traffic model and samples of real CAN traffic. Also, we propose a simple coding scheme that substantially reduces the number of stuff-bits in the considered real traffic.Delay variations (jitter) in computations and communications cause degradation of performance in control applications. There are many sources of jitter, including variations in execution time and bus contention. By introducing some restrictions when using CAN, such as a small reduction of available frame priorities, we are able to reduce the number of stuff-bits in the worst case. We also combine this with the methods mentioned above that reduces the number of stuff-bits in the data part of the frame. We show the actual penalty introduced by forbidding priorities and we show the overall improvement by using these techniques together in a small case study.------------------------------------------------------------ Svenska ------------------------------------------------------------Examensarbetet utreder graden av pessimism i traditionell schemaläggningsanalys för CAN-bussen (även kallad Controller Area Network). Arbetet fokuserar på användandet av bit-stuffing-distributioner istället för traditionella värsta falls scenarion. Genom att använda distributioner kan vi erhålla en analytisk utnyttjandegrad närmare den verkliga, dvs. då pessimismen i analysen är lägre återspeglar de analytiska värdena mer de verkliga. Å andra sidan, då vår analys är baserad på antaganden om distributionernas utseenden, kommer de erhållna bara att hållas med en viss grad av sannolikhet.I examensarbetet introducerar vi en modell samt några metoder som lättar på pessimismen i den traditionella analysen. Vi påvisar effekten av våra metoder genom att använda både modellerad trafik samt ”riktig” trafik uppsamlad hos en av våra industriella partners. Vi föreslår en metod för kodning av meddelanden som avsevärt minskar antalet stuff-bits.Variationer (jitter) i beräkningstider samt kommunikation orsakar en degradering av prestanda i reglersystem. Det finns många orsaker till jitter, till exempel variationer i exekveringstider samt trafik på meddelandebussen. Genom att införa små begränsningar vid användandet av CAN, som till exempel förbjudandet av vissa utvalda meddelandeidentifierare, kan vi minska antalet stuff-bitar avsevärt. I rapporten visar vi även kostnaden av att förbjuda dessa meddelandeidentifierare. Vi visar även den totala prestandaökningen vid användandet av alla dessa metoder i ett exempel.

Bibtex

@mastersthesis{Nolte383,
author = {Thomas Nolte},
title = {Reducing pessimism in CAN response time analysis},
month = {November},
year = {2002},
publisher = {MRTC},
url = {http://www.es.mdu.se/publications/383-}
}