Performance evaluation using Markov model for a novel approach in Ethernet based embedded networked control communication

Mohamad K Ishak, Guido Herrmann, M Pearson

Research output: Chapter in Book/Report/Conference proceedingConference Contribution (Conference Proceeding)

2 Citations (Scopus)
266 Downloads (Pure)

Abstract

In earlier work, we suggested a low cost control communication scheme across half-duplex Ethernet allowing for reduced jitter and delay for an information sharing control communication bus. The principle advantage of half-duplex Ethernet is the reduced number of wires (two versus four wires) and the reduced cost of switch/router equipment. The basic principles for our communication has been the network wide introduction of a synchronisation signal and the modification of the back-off scheme for each node, by assigning to each a node specific minimal back-off time. In this paper, a Markov chain model for modeling network jitter and delays is developed for this control communication approach. For this, the events occurring after a synchronization signal are formally summarized within a state vector. A two node example is considered. The state vector describes the states of collision detection, back-off and transmission within the two node system. Relevant probabilities are assigned for transfer between different states. The process of calculating the overall Markov chain model and related parameters is described. This approach shows the advantage of our approach over a standard Carrier Sense Multiple Access/Collision Detection (CSMA/CD) setting.
Original languageEnglish
Title of host publication2016 Annual IEEE Systems Conference (SysCon 2016)
Subtitle of host publicationProceedings of a meeting held 18-21 April 2016, Orlando, Florida, USA
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Number of pages7
ISBN (Electronic)9781467395199
ISBN (Print)9781467395205
DOIs
Publication statusPublished - Jul 2016

Keywords

  • Backoff time
  • Delays
  • Ethernet Ethernet
  • Jitter
  • Markov Model
  • network model

Fingerprint

Dive into the research topics of 'Performance evaluation using Markov model for a novel approach in Ethernet based embedded networked control communication'. Together they form a unique fingerprint.

Cite this