Novel multi-dimensional task offloading techniques for vehicular edge computing networks

The Internet of Things (IoT) has experienced rapid expansion, which has led to the creation of several safety services and applications for the global transportation network. These vehicular applications provide resource-intensive tasks that must be processed instantly. Tasks generated by the time-sensitive applications are offloaded to the prospective Vehicular Edge Computing (VEC) network for processing in order to meet the computation demands of these applications. The high mobility of the vehicles and the time-sensitive nature of vehicular applications are the main issues in VEC that make effective offloading and processing strategies necessary. Based on the literature, several task offloading schemes have been proposed to achieve optimal offloading that meets the application computation demands and enhances the performance of the VEC network. In this study, a comprehensive review of novel Multi-dimensional Task Offloading Techniques for Vehicular Edge Computing Networks is presented. Task offloading schemes based on metaheuristic algorithms, machine learning algorithms, deep learning algorithms, game theory, and multi-criteria decision-making algorithms were considered and analysed. Finally, we present various research opportunities based on the inferences drawn from the state-of-the-art techniques in the literature.