Abstract: As a new generation of Ethernet technology, time-sensitive networking (TSN) can effectively meet the low-latency and high-rate requirements of information transmission in vehicular networks, and has attracted extensive attention from automotive network engineers. This paper establishes a TSN network simulation model for battery electric vehicles, takes the autonomous driving function as the research scenario, and uses a combination of simulation and experiment to systematically verify the effects of strict priority (SP), credit-based shaper (CBS), and time-aware shaper (TAS) algorithms on the transmission delay of time-sensitive parameters. The results show that under various load conditions, the combination of CBS and TAS can control the transmission delay of time-sensitive parameters within 100 μs, and the performance is significantly better than that without quality of service (QoS) mechanism and SP mechanism. The research provides necessary theoretical support and practical reference for the design of vehicular TSN networks.

Key words: time-sensitive networking; credit-based shaper; time-aware shaper; transmission delay

摘要： 时间敏感型网络（TSN）作为新一代以太网技术，可有效满足车载网络信息传输的低 延迟、高速率需求，已引起车载网络工程师的广泛关注。文章搭建纯电动汽车 TSN 网络仿真 模型，以自动驾驶功能为研究场景，采用仿真与实验相结合的方法，系统验证了严格优先级 （SP），基于信用的整形机制（CBS）算法、时间感知调度（TAS）算法对时间敏感参数传输 延时的影响。研究结果表明，在多种负载工况下，采用 CBS+TAS 组合机制可将时间敏感参 数的传输延时控制在 100 μs 以内，效果明显优于无服务质量（QoS）机制和 SP 机制。该研究 结果为车载 TSN 网络设计提供了必要的理论支撑与实践参考。

关键词: 时间敏感型网络；基于信用的整形机制；时间感知调度；传输延时