QoS-Aware Network Slicing for Integrated 5G Terrestrial and LEO Non-Terrestrial Networks in Vehicular Communication

Abstract

The growing demand for high-throughput, low-latency, and high-reliability in the Internet of Vehicles (IoV) applications highlights the limitations of standalone terrestrial or non-terrestrial networks, especially in the environment of high mobility and heterogeneity of traffic conditions. To overcome these issues, we propose a Quality of Service (QoS) aware network slicing architecture for an integrated 5G terrestrial and Low Earth Orbit (LEO) non-terrestrial network for IoVs. The architecture allocates heterogeneous vehicular services such as ultra-reliable low-latency communications (URLLC), massive machine-type communications (mMTC), navigation, and infotainment traffic to appropriate network slices and access technologies. To ensure service continuity and performance isolation, a slice-aware traffic management mechanism is enabled between 5G and LEO links. We implement this framework in the simulation environment NS-3, integrating the 5G-LENA New Radio (NR) module, the realistic orbital model of LEO satellite (SNS3), based on the Simplified General Perturbations 4 (SGP4) propagation model, and vehicular traffic mobility traces from the Simulation of Urban Mobility Model (SUMO). The simulation results demonstrate that the proposed hybrid slicing method is highly effective compared with 5G-only, LEO-only and Round Robin, significantly reducing end-to-end delays, enhancing throughput stability, and reducing packet loss. Prioritised terrestrial access is especially beneficial for latency-sensitive services, while bandwidth-intensive applications efficiently utilise satellite resources without affecting vital traffic. The findings validate that QoS-aware slicing in integrated 5G-LEO networks is an effective solution for vehicular communications of the next generation, providing valuable insights into the convergence of non-terrestrial and terrestrial networks.