Loading…

Computation of tight bounds for the worst-case end-to-end delay on Avionics Full-Duplex Switched Ethernet

Avionics Full-Duplex Switched Ethernet (AFDX) is a fault-tolerant real-time communication bus for safety–critical applications in aircraft. AFDX configures communication channels, denoted as virtual links (VLs), ensuring bounded message delays through traffic shaping at both end-systems and switches...

Full description

Saved in:
Bibliographic Details
Published in:Journal of systems architecture 2024-11, Vol.156, p.103278, Article 103278
Main Authors: Ayhan Kilinc, Zeynep, Schmidt, Ece Guran, Schmidt, Klaus Werner
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Avionics Full-Duplex Switched Ethernet (AFDX) is a fault-tolerant real-time communication bus for safety–critical applications in aircraft. AFDX configures communication channels, denoted as virtual links (VLs), ensuring bounded message delays through traffic shaping at both end-systems and switches. Effective AFDX network design necessitates computing the worst-case end-to-end delay of time-critical VLs to meet specified message deadlines. This paper presents a new method for calculating tight bounds on the worst-case end-to-end delay for each VL in an AFDX network. We introduce the new notion of an extended uninterrupted transmission interval, which is the prerequisite for computing the worst-case queuing delay at switches. Adding up these queuing delays along the path of each VL between end-systems yields a tight upper bound on the worst-case end-to-end delay. The correctness of our results is formally proved, and comprehensive simulation experiments on different example networks confirm the tightness of our bound. These simulations also demonstrate the superior performance of our method compared to existing approaches that offer more pessimistic as well as optimistic results.
ISSN:1383-7621
DOI:10.1016/j.sysarc.2024.103278