An Energy-Efficient Framework in Radio Multi-Access Edge Cloud by Automatically Generating Network Slice

Abstract

Traditional radio multi-access edge computing (RMEC) mainly serves information transmission. However, networking devices tend to be powered by batteries, and the life time of the network is constrained by battery capacity. How to power networking devices in RMEC is a key challenge. Wireless energy transfer (WET), a technology that uses radio frequency signals to charge devices wirelessly, offers the potential to eliminate this performance bottleneck. In this article, a framework for automatically generating network slice (AGNS) is proposed, where “WET” is set as a special application to realize synchronous transmission of energy and information by network slices. First, we discuss the development trends of RMEC. Next, we embark on customized services and resource efficiency maximization to reduce device's energy consumption, and analyze technical challenges. AGNS provides a logical mapping method to bridge the qualitative demand of new applications and the quantitative supply of slices, and provides a slice orchestration strategy to automatically combine virtualized network functions and optimize virtualized network resources for virtual private network (VPN). AGNS also provides a physical mapping approach to establish the resource corresponding relation between the network slices and physical devices in dynamic environments. Simulation results show that AGNS is superior to existing methods which adopt the “best effort” mode with a “one size fits all” rule, in terms of acceptance rate, resource efficiency and energy consumption. The AGNS framework is measured by the data of 10T of China Telecom Sichuan Branch. The results are consistent with our theoretical analysis.