Modularity-Driven Group Scheduling for Time-Sensitive Networks

Time-Sensitive Networks (TSN) are designed to ensure real-time and reliability performance in industrial applications. While Integer Linear Programming (ILP) based scheduling methods can achieve optimal solutions for network design, they face scalability issues and are primarily suitable for small-scale problems. This paper presents a group scheduling strategy targeting for large-scale TSN applications. The traffic dependency is modeled as an undirected graph to reveal the naturally existing intrinsic community structure among flows. Critical flows and intra-group flows are explicitly identified, allowing the scheduling problem to be decomposed into several subset scheduling problems. The scheduling process prioritizes critical flows, while intra-group flows are scheduled in parallel. Experimental evaluations confirm that the proposed method significantly reduces flowspan and computational overhead, offering an efficient and scalable solution for large scale industrial TSN applications.