Journal
CHINESE JOURNAL OF AERONAUTICS
Volume 30, Issue 6, Pages 1941-1957Publisher
ELSEVIER SCIENCE INC
DOI: 10.1016/j.cja.2017.08.010
Keywords
Deterministic bound; Grouping ability; Network calculus; Networking features; Switched networks
Categories
Funding
- National Natural Science Foundation of China [61301086, 71701020]
Ask authors/readers for more resources
The state of the art avionics system adopts switched networks for airborne communications. A major concern in the design of the networks is the end-to-end guarantee ability. Analytic methods have been developed to compute the worst-case delays according to the detailed configurations of flows and networks within avionics context, such as network calculus and trajectory approach. It still lacks a relevant method to make a rapid performance estimation according to some typically switched networking features, such as networking scale, bandwidth utilization and average flow rate. The goal of this paper is to establish a deterministic upper bound analysis method by using these networking features instead of the complete network configurations. Two deterministic upper bounds are proposed from network calculus perspective: one is for a basic estimation, and another just shows the benefits from grouping strategy. Besides, a mathematic expression for grouping ability is established based on the concept of network connecting degree, which illustrates the possibly minimal grouping benefit. For a fully connected network with 4 switches and 12 end systems, the grouping ability coming from grouping strategy is 15-20%, which just coincides with the statistical data (18-22%) from the actual grouping advantage. Compared with the complete network calculus analysis method for individual flows, the effectiveness of the two deterministic upper bounds is no less than 38% even with remarkably varied packet lengths. Finally, the paper illustrates the design process for an industrial Avionics Full DupleX switched Ethernet (AFDX) networking case according to the two deterministic upper bounds and shows that a better control for network connecting, when designing a switched network, can improve the worst-case delays dramatically. (C) 2017 Chinese Society of Aeronautics and Astronautics. Production and hosting by Elsevier Ltd.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available