Sapienza University of Rome, DIET, Via Eudossiana 18, Rome, Italy, CEA, LETI, Minatec Campus, 17 rue des Martyrs, Grenoble, France

Mobile Edge Computing (MEC) plays a key role in the 5G roadmap, as a way to bring information technology (IT) services closer to the mobile users by empowering radio access points with additional functionalities, like caching or computation offloading capabilities. At the physical layer, some of the key technologies enabling very low latency mobile services are dense deployment of radio access points, massive MIMO and millimeter-wave (mmW) communications for radio access as well as for radio fronthaul/backhaul. In this paper, we merge computation offloading techniques for mobile edge computing with mmW communications and show how the joint optimization of computation/communication resources is crucial to design an energy efficient mobile edge computing system. In particular, we tackle the intermittency of mmW links by considering multi-link communications and show how to take advantage of preliminary estimation of blocking probabilities. © 2017 IEEE.

Computation offloading, Millimeter-wave communications, Mobile edge computing

Blocking probability, Energy efficiency, Millimeter waves, Network layers, Radio, Computation offloading, Edge computing, Energy efficient, Joint optimization, Key technologies, Millimeter wave (MMW), Millimeter-wave communication, Radio access points, Radio communication