Deep reinforcement learning-aided RAN slicing enforcement supporting latency sensitive services in B5G networks

Edge computing and artificial intelligence promise to turn future mobile networks into service- and radio-aware infrastructures, able to address the requirements of upcoming latency-sensitive applications. For instance, they can be used to dynamically and optimally manage the Radio Access Network Slicing. However, this is a challenging goal, due to the mostly unpredictably nature of the wireless channel. This paper presents a novel architecture using Deep Reinforcement Learning at the network edge for addressing Radio Access Network Slicing and Radio Resource Management. By considering the autonomous-driving use-case, computer simulations demonstrate the effectiveness of our proposal against baseline methodologies.