Autonomous driving paper index
Analysis on Impact of Edge Computing on 5G
One-line summary
In this paper, the performance of an Edge-Computing-assisted 5G network is analyzed using the MATLAB 5G Toolbox in combination with the MATLAB 5G System Level and Link Level simulation framework.
Engineering notes
Key topics: autonomous driving, end-to-end. See the paper for implementation details and experimental results.
Chinese explanation / 中文解读
中文解读待补充：本站会优先为端到端自动驾驶、BEV感知、3D目标检测、轨迹预测、路径规划、LiDAR感知等高价值论文补充中文说明。
Original abstract
The fifth generation (5G) of mobile communication is expected to deliver ultra-low latency, massive connectivity, and multi-gigabit throughput, but these benefits are hard to achieve with computation being centralized in a remote cloud data center. To overcome this limitation, Multi-access Edge Computing (MEC) moves the computation, storage and application logic to the edge of the network near the location where the data is generated. In this paper, the performance of an Edge-Computing-assisted 5G network is analyzed using the MATLAB 5G Toolbox in combination with the MATLAB 5G System Level and Link Level simulation framework. The layered architecture is proposed and mathematically modeled with regard to queuing delay, task-offloading decisions and energy consumption, considering 5G New Radio (NR) access network, Multi-access Edge Computing (MEC) layer and centralized cloud layer. The simulation is designed as a discrete-event system-level simulation with realistic 5G NR numerology, a large number of user equipment (UEs) and varying edge-server capacities to measure latencies, throughput, packet delivery ratio (PDR), response time, energy consumption, bandwidth utilization and reliability. Edge-assisted offloading shows an average end-to-end delay of 46-58% lower than a traditional cloud-only architecture and a maximum of 61% delay reduction in response time compared to the conventional cloud-only architecture, while maintaining high packet delivery ratio and reliability under heavy user density, while consuming a small amount of extra energy at the edge servers. These results validate MEC as a key enabling complementary technology for 6G networks with its supporting use cases for Ultra-Reliable Low-Latency Communication (URLLC); and the paper also reveals open research challenges in adaptive resource allocation, security and AI-based orchestration for future edge architectures in 6G.
Links and sources
Need this topic turned into a technical roadmap?
Full Self Driving can prepare a custom autonomous driving literature review, code map, dataset map, and B2B technology assessment.
Request B2B research
Comments