Timing analysis of AVB traffic in TSN networks using network calculus

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Time-Sensitive Networking (TSN) is a collection of standards that extend Ethernet to support safety-critical and real-time applications. TSN integrates multiple traffic types, i.e., Time-Triggered (TT) traffic scheduled based on Gate-Control-Lists (GCLs), Audio-Video-Bridging (AVB) traffic that requires bounded latencies, and Best-Effort (BE) traffic, for which no guarantees are provided. This paper proposes a Network Calculus-based approach to determine the worst-case end-to-end delays of AVB traffic in a TSN network with both non-preemption and preemption modes. We consider the effects of TT traffic due to GCLs, 'guard bands', i.e., time windows that block other traffic from transmitting, and preemption overhead on the service for AVB traffic. We provide a proof of non-overflow condition for AVB credit, which is used to control the AVB traffic transmission. The analysis method is evaluated on realistic test cases, and compared to related work.

Original languageEnglish
Title of host publicationProceedings - 24th IEEE Real-Time and Embedded Technology and Applications Symposium, RTAS 2018
EditorsRodolfo Pellizzoni
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages25-36
Number of pages12
ISBN (Electronic)9781538652954
DOIs
StatePublished - 8 Aug 2018
Event24th IEEE Real-Time and Embedded Technology and Applications Symposium, RTAS 2018 - Porto, Portugal
Duration: 11 Apr 201813 Apr 2018

Publication series

NameProceedings of the IEEE Real-Time and Embedded Technology and Applications Symposium, RTAS
ISSN (Print)1545-3421

Conference

Conference24th IEEE Real-Time and Embedded Technology and Applications Symposium, RTAS 2018
Country/TerritoryPortugal
CityPorto
Period11/04/1813/04/18

Keywords

  • Delay
  • Real time network
  • TSN
  • Timing analysis
  • Upper bound

Fingerprint

Dive into the research topics of 'Timing analysis of AVB traffic in TSN networks using network calculus'. Together they form a unique fingerprint.

Cite this