O-RAN Slicing for Multi-Service Resource Allocation in Vehicular Networks
Yaping Cui, Xisheng Yang, Peng He, Dapeng Wu, Ruyan Wang
Abstract
The automobile industries are undergoing a technological revolution, which has facilitated the emergence of multiple Vehicle-to-Everything (V2X) applications with diverse Quality of Service (QoS) requirements. The conventional resource allocation methods are no longer applicable in V2X due to the diverse vehicular services with different QoS requirements in various practical scenarios. Therefore, this paper proposes a multi-service resource allocation strategy for V2X. Specifically, the problem of allocating resources for multiple V2X services is decoupled into two sub-problems, i.e., resource allocation between slices, and resource scheduling within slice. Then, a heuristic resource allocation algorithm is adopted to solve the resource allocation between slices sub-problem. For Real-Time (RT) and Non-Real Time (NRT) services in the vehicular network, specialized scheduling has been designed for the near-real-time Radio Access Network (RAN) Intelligent Controller and non-real-time RAN Intelligent Controller in the Open Radio Access Network (O-RAN) architecture, to meet the heterogeneous QoS service requirements of vehicles. Simulation results demonstrate that compared to the modified largest weighted delay first (M-LWDF) algorithm, the proposed resource allocation strategy can enhance the AoI of the proportional fairness (PF) services by 15%, and provide 7.39% transmission rate gain for NRT services.