MiFo: A novel edge network integration framework for fog computing

Abstract

Fog computing has been recently proposed to move certain cloud computing services down to edge networks especially for mobile applications, aiming at improving the throughput and delay through the collaboration of mobile devices at edge wireless networks. Yet, the current edge wireless networks are the bottleneck that prohibits mobile devices to efficiently share their resources, due to the lack of an efficient mechanism of coordinating heterogeneous wireless networks (e.g., LTE, WiFi, and WiMax) to better utilize the radio resources. In this paper, we propose MiFo, a novel hierarchical dual-layer edge network integration framework for fog computing. At the lower level, the Mist layer manages the baseband resources of homogeneous networks in a centralized way, which can enhance the transmission performance through physical layer cooperation among base stations. At the upper level, the Fog layer coordinates the heterogeneous network access and the resource scheduling. In MiFo, we also propose a multi-stream concurrent (MSC) transmission protocol, which can make full use of diverse wireless network resources and significantly improve the data transmission rate of multi-mode mobile devices. The proposed MSC transmission protocol can also achieve smart handover and improve the energy efficiency as well.