Latency minimization for the MEC-aided vehicular network under Doppler Effect

Abstract

In this paper, we consider a heterogeneous Mobile Edge Computing (MEC)-aided multi-layer vehicular network consisting of multiple vehicles as Edge Device (EDs), a road side unit (RSU), and a cloud center (CC). The raw data produced by the vehicles can be processed at the three nodes separately. EDs transmit the data to the RSU via wireless communication link, and a wired link is considered between the RSU and the CC. Taking the mobility of vehicles and the limited amount of resources into account, we jointly discuss the task offloading and resource allocation in this three-layer MEC-aided vehicular network. Specifically, the interactions of Doppler Effect and the bandwidth provided by RSU to different vehicles are investigated. Aiming at minimizing the latency while increasing the utilization of resource, we design an iterative task assignment and resource allocation (ITR) algorithm. The simulation results indicate that: 1) The average latency of all tasks is significantly decreased, and the optimal bandwidth allocation scheme to the EDs is achieved; 2) Due to the Doppler Effect of vehicles, it is important to consider both the mobility of vehicles and the length of communication links when allocating resource.