An Enhanced Routing and Scheduling Mechanism for Time-Triggered Traffic with Large Period Differences in Time-Sensitive Networking

In the field of the automotive area as well as industrial control, real-time communication requires deterministic delivery with low delay and bounded jitter. Real-time communication in these networks requires transmission schedule and routing, which is an NP-hard problem. In this paper, we present a...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Applied sciences Ročník 12; číslo 9; s. 4448
Hlavní autoři: Nie, Hongrui, Li, Shaosheng, Liu, Yong
Médium: Journal Article
Jazyk:angličtina
Vydáno: Basel MDPI AG 01.05.2022
Témata:
Communication
Ethernet
flow classification
Genetic algorithms
Heuristic
integer linear programming (ILP)
Linear programming
Mathematical programming
routing and scheduling algorithm
Schedules
Scheduling
Success
time-sensitive network (TSN)
ISSN:2076-3417, 2076-3417
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:In the field of the automotive area as well as industrial control, real-time communication requires deterministic delivery with low delay and bounded jitter. Real-time communication in these networks requires transmission schedule and routing, which is an NP-hard problem. In this paper, we present an offline routing and scheduling method based on integer linear programming (ILP), with a flow preprocessing step to explore the period correlation of time-triggered (TT) traffic in time-sensitive networking (TSN). First, a multiperiod flow routing and scheduling algorithm based on flow classification is proposed to improve the scheduling success rate and reduce execution time. The flow classification technique obtained a more fine-grained TT traffic classification, which can be superimposed on any routing and scheduling algorithms. Second, an adaptive period compensation scheduling algorithm based on flow classification is proposed in simple network architecture conditions. The evaluations demonstrate that the proposed algorithms improve scheduling success rate and reduce execution time compared with baseline methods in all test cases. In addition, we can adapt our different proposed algorithms in different network architecture conditions to schedule various flows with different periods and sizes.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2076-3417
2076-3417
DOI:10.3390/app12094448