Task Offloading for MEC-V2X Assisted Autonomous Driving

Mobile edge computing (MEC) and vehicle-to-everything (V2X) communications are two promising technics to support delay-sensitive applications for autonomous driving. In this paper, road side unit (RSU) and assistant vehicle offloading are jointly considered to minimize latency for vehicular tasks in an MEC-V2X network. The impact of Doppler spread caused by high moving speed is also considered. The offloading decision, transmit power, bandwidth, and computation resource allocations are solved by formulating them as a joint optimization problem. The simulation results show that the proposed scheme significantly reduced the task latency, compared to three traditional methods, where task either run locally, or at MEC server, or adaptively between MEC server and local.