Collaborative Task Offloading in Vehicular Edge Computing Networks

With the explosive increasing of various vehicle applications, vehicular edge computing (VEC) networks have emerged as a promising technology to extend the capability of the traditional vehicular networks. The computing tasks generated by the vehicles are mainly completed based on the cooperation of local and mobile edges in the past, which may not be able to meet the increasing computing resource requirements of the tasks. In this study, we propose a VEC architecture that groups vehicles with idle computing resources into a vehicular cloud (VC) to process tasks generated by vehicles. In order to reduce the delay and energy consumption of the VEC system, we formulate an optimization problem to find the optimal task assignment decision. Since the problem is NP-hard, two joint task offloading decision algorithms are proposed to solve the problem. Simulation results show that the algorithm is able to effectively reduce the latency and energy consumption of the system. Moreover, it is also demonstrated that the VC can be effectively constructed to further improve the utility of the system.