Personalized Vehicle Task Offloading Based on the Recommender System

Mobile Edge Computing (MEC) provides a high-bandwidth, low-latency computing environment for a large number of computing tasks of intelligent and connected vehicles. It is a non-trivial task to select the personalized service nodes for each vehicle regarding task offloading. To address this challenge, a vehicle computing task offloading algorithm based on a recommender system is proposed. Specifically, the vehicle task offloading problem is formulated as an optimization problem that aims to maximize the utility, in which the utility function quantifies the personalized demand and satisfaction of the vehicles. Then, the optimization problem is solved by constructing a relevant dataset and a recommendation algorithm to acquire the personalized assignment of vehicle edge computing task offloading. The simulation results show that the proposed method can achieve higher system utility compared with traditional offloading schemes, indicating improved quality of service and user satisfaction.