FiWi-Enhanced Vehicular Edge Computing Networks: Collaborative Task Offloading

Abstract

Intelligent connected vehicles (ICVs) equipped with onboard sensors, communication, and computation units, are expected to be the promising paradigm of future vehicles. With the development of ICVs, the conflict between resource-hungry ICV applications and resource-constrained ICVs becomes prominent, and mobile edge computing provides a promising solution to address it. However, facing diverse computation tasks with different granularities and quality-of-service (QoS) requirements, it is obviously not enough to solely rely on the lightweight edge servers placed at the roadside units (RSUs). Making the most of the computational resources at the edge servers and the cloud center appears to be particularly important. In addition, considering the diversity of vehicle-to-everything (V2X) communications and their QoS requirements, the coexistence of multiple wireless communication technologies will persist for a long time, and designing a networking technology to integrate them is necessary and urgent. Toward this end, we introduce fiber-wireless (FiWi) integration to enhance vehicular edge computing networks to provide support for the coexistence of remote cloud and edge servers and propose two collaborative task offloading schemes in FiWienhanced vehicular edge computing networks, where both vehicle-to-RSU and vehicle-to-vehicle (V2V) communications are considered. Extensive numerical results corroborate the superior performance of our schemes on reducing the tasks' processing delay.