Computational and Location Aware Middleware to Enable Edge Computing in Mobile Devices

Abstract

Edge computing in 5G and Beyond 5G (B5G) networks is an essential infrastructure enabler for future technology and business developments, also known as Industry 4.0, which requires high computational power and fast service deliveries. There are many proprietary Multi-access Edge Computing (MEC) architectures defined by the service providers and the standard bodies. However, all of them require changes not only in the existing end devices, but also in the existing applications to connect with the MEC platform for edge server discovery. Furthermore, these client applications are unaware of the list of server applications running on the edge servers. It also requires installation of device application in mobile devices to subscribe and authorize with the MEC Platform, to communicate with edge servers and to find the available MEC services, which cause additional overhead to the device. In this paper, we propose a novel solution named Computational and Location-Aware Middleware (CLM), to compute and locate the nearest server for service discovery and computational offload for the mobile device. The proposed solution does not require any changes in existing applications for edge server discovery, and it has been successfully prototyped in Samsung devices with Android Pie for evaluation. The results show a reduction in edge server discovery overheads to one Round Trip Time and remarkable performance gain of up to 85% for service discovery.