Network Slicing with MEC and Deep Reinforcement Learning for the Internet of Vehicles

The interconnection of vehicles in the future fifth generation (5G) wireless ecosystem forms so-called Internet Vehicles (IoV). IoV offers new kinds applications requiring delay-sensitive, compute-intensive, and bandwidth-hungry services. Mobile edge computing (MEC) network slicing are two key enabler technologies 5G networks that can be used to optimize allocation resources guarantee diverse requirements applications. As traditional model-based optimization techniques generally end up with NP-hard strongly non-convex nonlinear mathematical programming formulations, this article, we introduce a model-free approach based on deep reinforcement learning (DRL) solve resource problem MEC-enabled slicing. Furthermore, solution uses non-orthogonal multiple access (NOMA) enable better exploitation scarce channel resources. considered addresses jointly power allocation, slice selection, vehicle selection (vehicle grouping). We model as single-agent Markov decision process. Then it using DRL well-known Q (DQL) algorithm. show our is robust effective under different conditions compared benchmark solutions.