Optimal Control-based Computing Task Scheduling in Software-Defined Vehicular Edge Networks
Ershuai Peng, Zhiyuan Li · 2020
MEC (Mobile edge computing) is one of the key technologies to promote the development of 5G. Mobile edge computing server is the small base-station equipped with computation and storage capacity. It allows nearby-cars to access in a wireless way. Be different with cloud computing, Mobile edge computing can meet the strict requirements in network delay of the cars but the limitation computation capacity cannot meet all requirements of the cars. To solve this issue, This paper introduces OCTS(Optimal Control-based Computing Task Scheduling) algorithm. It can develop efficient resource allocation strategies and make full used of these resource. Software-Defined network is introduced to avoid obtaining local optimum and we use adjustable parameters to make the algorithm performance well in different circumstance. Experiments show that this method can effectively improve the utilization of resources and meet the requirements of computation delay and network delay for vehicle's tasks.