Open Access

ARTICLE

5G-Aware Incremental Routing and Scheduling for Dynamic Time-Triggered Flow Admission in Time-Sensitive Networks

Zhihao Liu^1,2, Yi Zhang³, Wei Zhang^1,2, Jian Wang⁴, Huiling Shi^1,2, Xiaolong Wang^1,2,*

1 Key Laboratory of Computing Power Network and Information Security, Ministry of Education, Shandong Computer Science Center (National Supercomputer Center in Jinan), Qilu University of Technology (Shandong Academy of Sciences), Jinan, China
2 Shandong Provincial Key Laboratory of Computing Power Internet and Service Computing, Shandong Fundamental Research Center for Computer Science, Jinan, China
3 School of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu, China
4 College of Science, China University of Petroleum (East China), Qingdao, China

* Corresponding Author: Xiaolong Wang. Email:

Computers, Materials & Continua 2026, 89(2), 91 https://doi.org/10.32604/cmc.2026.088018

Received 26 June 2026; Accepted 21 August 2026; Issue published 15 September 2026

Abstract

Mobile edge services require deterministic communication across Time-Sensitive Networking (TSN) and 5G access, where the standardized integration architecture exposes the 5G System (5GS) to the TSN controller as a logical bridge. We study dynamic admission of time-triggered (TT) flows using reported 5GS bridge delay and TSN-to-5GS Quality of Service (QoS) mapping in route selection and Gate Control List (GCL) scheduling. Arrivals and departures can split available transmission time into noncontiguous windows. Online insertion preserves admitted schedules but may reduce subsequent schedulability, whereas full recomputation can restore schedulability but changes many routes and GCL entries, complicating coordinated activation. Coupling routing and GCL scheduling under timing, bridge-delay, QoS-mapping, and a bound on changes to admitted schedules yields an NP-hard problem. To address it, we propose a two-timescale scheduling mechanism. The fast timescale uses current 5GS bridge information to place arrivals without modifying admitted flows. Fragmented windows, repeated insertion failures, or changes in reported 5GS state invoke the slower timescale, which sequentially reschedules a bounded subset of admitted flows and commits only feasible improvements. At 0.95 offered load across A380, CEV, and Ring6, admission improves by 14.8–18.5 percentage points over online-only scheduling and remains within 1.7–2.2 points of full recomputation, while per-event runtime falls by over one order of magnitude with limited GCL changes.

---

Keywords

---