DeepNC: A Fast GNN-based Pre-Verification Surrogate for TSN Configuration
2026-07-27 • Networking and Internet Architecture
Networking and Internet Architecture
AI summaryⓘ
The authors focus on improving how we check timing guarantees in safety-critical networks using Time-Sensitive Networking (TSN). They identify that the usual method, Network Calculus (NC), is very slow when testing many configurations. To speed this up, they created DeepNC, a tool that uses a special kind of neural network combined with NC knowledge to quickly filter out unlikely configurations before doing the slow formal checks. Their tests show DeepNC is much more accurate than previous learning methods and drastically cuts down the number of full checks needed, making the whole process much faster.
Time-Sensitive NetworkingNetwork Calculusschedulability analysisGraph Neural Networkformal verificationconfiguration-space explorationregression accuracyworst-case delaypre-verification filter
Authors
Jiayi Zhu, Jing Lin, Zelong Tian, Feng He, Luxi Zhao
Abstract
Time-Sensitive Networking (TSN) is critical to deterministic communication in safety-critical domains, with formal verification such as Network Calculus (NC) serving as the cornerstone for schedulability guarantees. However, during automated configuration-space exploration, repeated schedulability analysis consumes over 90% of the total configuration time, becoming the primary bottleneck for large-scale TSN configurations. To address this challenge, we propose DeepNC, a novel pre-verification surrogate module that pioneers the structural fusion of NC principles into a Graph Neural Network (GNN) for TSN configuration-space exploration. Rather than replacing formal verification, DeepNC acts as a high-speed pre-verification filter, reserving computationally expensive formal verification only for promising candidates. Extensive evaluations demonstrate that DeepNC significantly improves worst-case delay prediction accuracy over state-of-the-art learning-based methods, increasing the average $R^2$ by 55.8% and reducing the average MAPE by 65.3%. More importantly, its high-fidelity regression substantially reduces the number of formal verification calls during configuration-space exploration by 93.25%, while accelerating NC-based verification by more than two orders of magnitude.