Small Cell Clustering for Efficient Distributed Fog Computing: A Multi-User Case

Abstract

Ultra-dense deployment of radio access points is a key enabler for future 5G networks. It allows the network to cope with the ever increasing mobile data traffic. In addition, these radio access points can serve as an infrastructure for a local mobile cloud computing platform referred to as fog computing. The fog is a capillary edge cloud that enables joint optimization of communication and computational resources for maintaining an efficient and scalable network design. In this paper, we address the problem of radio access points clustering for fog computing applications. We focus on the case where multiple users require fog computing services. We formulate the distributed clustering problem as a joint optimization of the computation and communication resources. We transform the non-convex original problem into an equivalent convex one. Our simulation results show that the clustering solution derived from this problem yields high users' satisfaction ratio while keeping low the communication power consumption of the computation cluster.