智能网联汽车的智能驾驶与智能座舱业务推动车载网络向以太网跨域融合架构演进。本文围绕IEEE 802.1Q 标准，研究车载以太网服务质量（Quality of Service，QoS）的关键技术，梳理了优先级的流量分类方式，以及严格优先级（SP）、加权轮询（WRR）及基于信用的整形器（CBS）三类核心调度机制。结合带宽与缓冲区资源分配规则，搭建精细化时延分析模型，以量化不同优先级数据流的端到端时延指标。试验结果表明，所提混合调度 QoS 方案可有效保障高优先级车控流量的传输质量。即便在 90% 高网络负载工况下，关键高优先级业务仍可实现毫秒级低时延传输，能够满足自动驾驶、车身控制等车载实时业务的通信需求，为跨域架构下车载以太网通信提供可行的 QoS 落地方案。
Abstract
The intelligent driving and intelligent cabin services of intelligent connected vehicles drive the vehicle network to evolve towards the cross-domain integration architecture of Ethernet. This paper focuses on the IEEE 802.1Q standard and studies the key technologies of vehicle Ethernet Quality of Service (QoS). It summarizes three core scheduling mechanisms: priority-based traffic classification, Strict Priority (SP), Weighted Round Robin (WRR), and Credit-Based Shaper (CBS). Combined with bandwidth and buffer resource allocation rules, a refined delay analysis model is established to quantify the end-to-end delay indicators of different priority data streams. The experimental results show that the proposed hybrid scheduling QoS scheme can effectively guarantee the transmission quality of high-priority vehicle control traffic. Even under 90% high network load conditions, the key high-priority services can still achieve millisecond-level low delay transmission, meeting the communication requirements of real-time vehicle-based services such as autonomous driving and vehicle body control, providing a feasible QoS solution for cross-domain architecture vehicle Ethernet communication.
关键词
智能网联汽车 /
通信系统 /
服务质量
Key words
intelligent connected vehicles /
communication system /
QoS
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 呼布钦，秦贵和，刘颖，等 . 下一代汽车网络：车载以太网技术现状与发展 [J]. 计算机工程与应用，2016，52（24）：29-36.
[2] 晏欣炜，周奎，朱政泽，等 . 新能源汽车车内实时以太网技术发展综述 [J]. 软件导刊，2018，17（1）：4-7.
[3] IEEE Std 802.3br—2016, IEEE Standard for Ethernet Amendment: Specification and Management Parameters for Interspersing Express Traffic[S].
[4] IEEE Std 802.1Qcc—2018, IEEE Standard for Local and Metropolitan Area Networks-Bridges and Bridged Networks-Amendment 31: Stream Reservation Protocol (SRP)Enhancements and Performance Improvements[S].
[5] IEEE Std 802.1Qav—2009, IEEE Standard for Local and Metropolitan Area Networks——Virtual Bridged Local Area Networks Amendment 12: Forwarding and Queuing Enhancements for Time-Sensitive Streams[S].
[6] Deng Libing, Xie Guoqi, Liu Hong, et al. A Survey of RealTime Ethernet Modeling and Design Methodologies: From AVB to TSN[J]. ACM Computing Surveys, 2022,55(2): 1-36.
[7] 吴海林，王明 . 基于车载以太网的智能汽车通信系统设计 [J]. 通信电源技术，2024，41（20）：32-34.
[8] Norman Finn. P802.1Qat Delay and Bandwidth Parameterization [R]. New York: IEEE 802.1 Working Group, 2008.
[9] Geoffrey M. Garner. Reconciliation of formulas for maxBurst in 802.1Qav [R]. New York: IEEE 802.1 AVB TG, 2008.
[10] Lin Zhao, Feng He, Ershuai Li, et al. Improving Worst-Case Delay Analysis for Traffic of Additional Stream Reservation Class in Ethernet-AVB Network[J]. Sensors, 2018, 18 (11)：3849.
PDF(1696 KB)
