Scalable Modulation based Computation Offloading in Vehicular Edge Computing System

Abstract

Vehicular Edge Computing (VEC) is becoming popular due to the high offloading demand of computation-intensive vehicular network applications. Although the mobility of vehicles poses great challenges to the computation offloading as the fast time-varying fading channel, the moving vehicles can be served as mobile relays to achieve opportunistic transmission for improving offloading efficiency. In this paper, a scalable modulation (S-Mod) based task offloading (SMO) strategy is proposed, in which the vehicles can be selected as mobile relay to implement cooperative transmission based on S-Mod scheme. To reduce the energy consumption of vehicles, the energy consumption minimization offloading decision problem with delay and transmission quality requirement constraints is formulated. A hybrid offloading algorithm with S-Mod scheme design and relay selection is proposed. Our simulation results show that the proposed SMO strategy can outperform the existing strategy up to by 72.68% on energy consumption and 14.15% on execution time.