Application Specific Mobile Edge Computing through Network Softwarization

Mobile Edge Computing (MEC) has been recognizedas one of the key emerging technologies in the evolution towards5G cellular wireless networks. In this paper, we present anapplication-specific MEC architecture that applies the concept ofsoftware-defined data plane to Mobile Virtual Network Operators(MVNOs) network. We first characterize the flow properties ofpopular mobile applications with data trace collected from theMVNO network connecting customized smartphones that enableus to identify applications and devices with 100% accuracy. Wethen develop an application-specific MEC processing to optimizeMVNO network through applying different MEC services todifferent mobile applications. As a case study of applicationspecificMEC optimization, we present a theoretical analysisof the optimal number of TCP connections per mobile userand develop a mobile-friendly optimization mechanism to achievefairness between tethering and smartphone users by regulatingthe concurrent TCP connections.