Deep-VFog: When Artificial Intelligence Meets Fog Computing in V2X

【Author】 Rihan, Mohamed; Elwekeil, Mohamed; Yang, Yatao; Huang, Lei; Xu, Chen; Selim, Mahmoud M.

【Source】IEEE SYSTEMS JOURNAL

【影响因子】4.802

【Abstract】Next-generation vehicle-to-everything (V2X) networks are envisioned as communication platforms, which not only aim to improve road safety and drastically increase traffic efficiency but also introduce a plethora of innovative applications that mostly require ultrareliable low-latency communications. Motivated by such new applications and its accompanying services, future vehicular networks can easily switch from current connected vehicles, mostly vehicle-to-vehicle, to fully heterogeneous V2X networks that support autonomous driving. Toward this target, the deep integration between artificial intelligence (AI) and fog computing (FC) technologies within the future V2X networks is a step whose time has come. In this survey, we address a novel and practical AI-enabled vehicular network architecture with FC in its core. Moreover, we present AI-enabled, fog-assisted V2X use cases that accommodate important FC capabilities and exploit the power of AI for enabling the desired evolution of vehicular networks. Finally, some potential issues of AI in FC-based V2X networks for future work are discussed.

【Keywords】Tools; Edge computing; Reinforcement learning; Computer architecture; Resource management; Autonomous vehicles; Artificial intelligence (AI); deep learning (DL); deep reinforcement learning (DRL); federated learning (FL); fifth generation (5G); fog computing (FC); vehicular communications (VCs)

【发表时间】2021 SEP

【收录时间】2022-01-02

【文献类型】

【主题类别】

--