An Analytical Latency Model for AVB Traffic in TSN Considering Time-Triggered Traffic

Time-Sensitive Networking (TSN) is an emerging network technique used in real-time communication. There are three typical traffic types in TSN: Time-Triggered (TT) traffic with strict time stability requirements, Audio-Video-Bridging (AVB) traffic with large bandwidth and bounded latency requirements, and Best-Effort (BE) traffic. This paper presents an analytical latency model for AVB traffic in TSN considering TT traffic. According to the time-slots of TT traffic, an accurate single-hop service curve for AVB traffic is established based on Network Calculus, from which a more precise queueing delay through one hop can be calculated. Moreover, a strict worst-case end-to-end latency of AVB traffic can be obtained by the calculated queueing delays. We evaluate the proposed model in a real network environment, and the results show that the proposed model is more accurate and more accordance with actual circumstance compared with other methods.