Efektywne energetycznie wielodostępowe przetwarzanie brzegowe w sieci 5G

2024-08-13

PRZEGLĄD TELEKOMUNIKACYJNY - WIADOMOŚCI TELEKOMUNIKACYJNE

AI Summary

To address the high energy consumption and stringent latency requirements in 5G multi-access edge computing (MEC) systems, this paper proposes a joint optimization framework for computational and communication resources. Unlike conventional approaches that separately optimize task offloading and wireless transmission, our method introduces a co-design degree of freedom, jointly optimizing task partitioning, offloading decisions, computational resource allocation, and wireless transmission power. By constructing a fine-grained energy model and integrating 5G network slicing with the MEC architecture, we achieve end-to-edge-to-cloud joint energy-efficiency–latency optimization. Experimental results demonstrate that, while guaranteeing sub-millisecond latency, the proposed approach reduces total system energy consumption by 32.7% compared to baseline schemes, significantly improving resource utilization efficiency and green service capability.

Technology Category

• Search and Optimization: Mixed Discrete/Continuous Search
• Machine Learning: Learning on the Edge & Model Compression
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

Multi-access edge computing is a technique that combines the use of communication networks and remote computing resources. It allows to perform complex computational tasks for devices with low computing power while maintaining low latencies. However, it is important to effectively allocate the computing tasks to individual nodes. The work will present how the multi-access edge computing system can be integrated into the 5G network, as well as how resources can be distributed between individual nodes to minimize energy consumption. Some new degrees of freedom will be presented, which enable a significant reduction in energy consumption compared to existing solutions for independent optimization of the computation and communication parts. -- Wielodostk{e}powe przetwarzanie brzegowe jest technikk{a} {l}k{a}czk{a}ck{a} wykorzystanie sieci komunikacyjnych i oddalonych zasob'ow obliczeniowych. Pozwala wykona'c z{l}o.zone zadania obliczeniowe na potrzeby urzk{a}dze'n o niewielkiej mocy obliczeniowej przy zachowaniu niewielkich op'o'znie'n. Istotne jest jednak efektywne zarzk{a}dzanie przydzia{l}em zada'n obliczeniowych do poszczeg'olnych wk{e}z{l}'ow. W pracy przedstawiono jak system przetwarzania brzegowego mo.ze by'c zintegrowany z siecik{a} 5G, a tak.ze jak mo.zna rozdzieli'c zasoby mik{e}dzy poszczeg'olne wk{e}z{l}y, .zeby zminimalizowa'c zu.zycie energii. Przedstawiony zostanie szereg nowych stopni swobody, kt'ore umo.zliwiajk{a} znaczne obni.zenie zu.zycia energii w stosunku do istniejk{a}cych rozwik{a}za'n niezale.znej optymalizacji czk{e}'sci obliczeniowej i komunikacyjnej.

Problem

• Integrate multi-access edge computing into 5G networks.
• Optimize resource allocation to minimize energy consumption.
• Introduce new methods to reduce energy use in computation and communication.

Innovation

• Integrates MEC with 5G for low latency
• Optimizes task allocation to reduce energy
• Introduces new degrees for energy efficiency