5G-RAN Slicing by utilizing RL-based Approaches for V2X Service

초록

main challenge on designing 5G networks is how to address different types of users with diverse QoS requirements on a shared physical network infrastructure. Radio Access Network (RAN) slicing has been introduced as a solution to these challenges. The RAN slicing have a limitation to provide two common services for 5G, namely eMBB and Cellular V2X. We propose an RAN slicing strategy based on reinforcement learning (RL) to allocate radio resources for different slices while considering the resource utilization to maximize the efficiency.