A Novel Internet of Vehicles’s Task Offloading Decision Optimization Scheme for Intelligent Transportation System

Abstract

In the future intelligent transportation system (ITSs), there will be a lot of negotiation work between vehicle and vehicle (V2V) and between vehicle and infrastructure (V2I), so it is very necessary to design efficient and energy-saving offloading strategy. Aiming at the three conflicting optimization objectives of offloading delay, energy consumption and load balancing, an efficient and energy-saving offloading decision scheme in the scenario of Internet of vehicles was proposed in this paper. Firstly, the task segmentation model, offloading delay model, energy consumption model, load balancing model and multi-objective optimization model were constructed. Then, based on the comprehensive consideration of data offloading delay, energy consumption and load balance, a task offloading scheme based on MOEA/D was proposed. Finally, the proposed scheme was compared with NSGA-II-based scheme, NSGA-III-based scheme,PESA-II-based scheme and SPEA-II-based scheme. The simulation results show that a task offloading scheme based on MOEA/D is obviously superior to the above schemes in terms of offloading delay, energy consumption and load balancing, and can provide efficient and energy-saving offloading service.