Xhaul Latency Dimensioning of 5G Drone Control
2022 International Conference on Unmanned Aircraft Systems (ICUAS)(2022)
摘要
This work considers the latency boundaries for Ultra Reliable Low Latency (URLLC) applications in 5G mobile networks using a Cloud-Radio Access Network (C-RAN) architecture. We present a use case for drone control, where the 5G network provides connectivity between the drone and the Ground Control Station (GCS). The 5G C-RAN architecture under investigation is the virtualised URLLC scenario proposed by Open-RAN (O-RAN) Alliance where radio processing is divided into three units, connected by an xhaul network. Hence, this work analyses opportunities for using the O-RAN virtualised URLLC scenario for drone control in two states; autonomous and under manual real-time control. The 5G network consists of different segments, each with different latency requirements, and this work outlines the deployment and dimensioning options in each segment. This work explores an Ethernet xhaul network, and thus the impact of queuing latency since the traffic link can be shared by different sources. The segments in the xhaul network must transport data that has not yet been through all radio processing layers, and thus, it is interesting to investigate how this affects the use case. Results illustrate how the maximum distance between the drone and GCS is limited by various xhaul parameters, while the maximum number of supported drones is not. This distance depends on both the xhaul network topology and link rates. This work shows the impact different message sizes has on the distance between drone and GCS, for example an Ethernet xhaul network with 4 hops requires at least 10 Gbps xhaul link rate to reach an xhaul length of 10 km. Hence, over-provisioning of xhaul bandwidth is a good solution seen from a latency perspective.
更多查看译文
关键词
crosshaul,delay,functional split,latency,O-RAN,xhaul,drones,UAV
AI 理解论文
溯源树
样例
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要