Mobility-Aware Computation Offloading for Cloud-Assisted Mobile Edge Computing in Vehicular Networks

Abstract

Mobile Edge Computing (MEC) is a promising solution for vehicular task offloading to meet the stringent delay and high-reliability requirements. In addition, cloud-assisted MEC can further improve resource utilization of MEC and cloud computing server. However, the current cloud-assisted MEC offloading schemes have high algorithm complexity and a single optimization target (e.g., latency). At the same time, the mobility of vehicles will also lead to varying influences on communication links. Frequent disconnection and reconnection of communication links will easily lead to failure of computing task transmission. In this paper, we present a Collaborative Computation Offloading and System Energy optimization (CCOSEO) scheme. First, the computation complexity of the scheme is reduced by a computation task classifying algorithm. Then, combined with the mobility of vehicles, we formulate the offloading decision as a joint optimization problem of latency and energy consumption. Finally, a Game Theory-based Offloading Strategy optimization (GTOSO) algorithm is proposed to achieve the optimal solution. The simulation results show that the proposed scheme can greatly enhance system energy efficiency and reduce the total delay by 20% in the dense-vehicle scenario.