Computation Offloading- Based Task Scheduling in the Vehicular Communication Environment for Computation-Intensive Vehicular Tasks

With the advent of the cellular-V2X (C-V2X) of 5G, autonomous vehicles are allowed to offload their computation-intensive tasks to other near-by vehicles for the expedited processing of enormous data while driving on the road. This enables vehicles to overcome their own computational resource limitations and to run various real-time applications with satisfying constrained timing requirements. In this paper, we propose to use computation offloading among vehicles so that vehicles can cooperate in performing their tasks. Using this technique, a vehicle, called a task vehicle, can request the computation of some subtasks of its task to a subset of its neighboring vehicles, called service vehicles, for the efficient and fast execution of the task. In order to investigate the effectiveness of computation offloading in the vehicular communication environment, three task scheduling algorithms, the Min-min, the Max-min and the HEFT algorithms, are adopt to this environment. Throughout simulations, we evaluate the performance of the three algorithms with computation offloading.¹