Fine-Grained Service Scheduling Scheme Based on Application-Aware Network for Internet of Vehicles

Le résumé fourni par la source

A novel fine-grained service scheduling scheme based on application-aware network (APN) for Internet of Vehicles (IoV) is proposed to achieve efficient and differentiated network resource orchestration. Application and service flow information is translated to application identifier (ID) and Sub-service ID encapsulated in IPv6 header. In this way, network can perceive various applications and provide suitable network channel resource for them. The APN-based information identifying method for application does not require unpacking of data packets and does not overly rely on network controllers, which can improve service efficiency. The experimental verification has been conducted on IoV experimental network in Xiong'an, China. In the experiment, the end-to-end delay of remote driving of APN autonomous vehicle is decreased by 67.5% compared to that of normal autonomous vehicle. The reason is that the network can identify the critical service flow sent by the APN vehicle and guide it to the dedicated channel to avoid competing for bandwidth resources in the shared channel which has 10% packet loss rate. In addition, the packet loss rate thresholds for all remote driving service flows are obtained through experiment, so that the optimal flow orchestration can be used to realize efficient utilization of network resource while ensuring smooth remote control.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.