QoS-aware Network Slicing and Resource Management for Internet of Vehicles in 5G networks

The development of Internet of Vehicles (IoV) technologies brings with it numerous challenges, such as heterogeneous Quality of Service (QoS) in 5G and beyond (B5G). These challenges go hand in hand with spectrum scarcity, one of the biggest challenges of future wireless technologies. This will be exacerbated especially in the era of Vehicle-to-Everything (V2X) communication due to the limitation of radio resources such as channels, bandwidth, and power. This challenge is further intensified by the increasing number of vehicles. Therefore, spectrum sharing initiatives have a significant impact on traffic safety and efficiency. In response to these challenges and to meet the diverse QoS requirements of vehicular applications within 5G/B5G, new paradigms such as New Radio Vehicle-to-Everything (NR-V2X) and Network Slicing (NS) have emerged as important solutions. Network Slicing efficiently divides the physical network into slices tailored to Ultra-Reliable Low Latency Communications (URLLC), Enhanced Mobile Broadband (eMBB) and massive Machine Type Communications (mMTC). In this paper, we investigate the IoV slicing problem with QoS support, focusing on partitioning the physical network into three different slices: URLLC, eMBB and mMTC. To ensure seamless communication in vehicular networks, our mixed method approach effectively incorporates handover mechanisms, emergency traffic prioritization, based services, and road network-specific parameters. In addition, we propose two resource allocation algorithms that enable efficient allocation of resources to vehicles and comply with the standardization principles of the 3rd Generation Partnership Project (3GPP). These algorithms aim to prioritize emergency traffic during incidents while maintaining an acceptable QoS for non-safety services in a resource-constrained environment, thus improving Key Performance Indicators (KPIs). Detailed simulations show the effectiveness of the proposed algorithms, and confirm their ability to improve the performance of emergency services in terms of end-to-end delay while ensuring acceptable reliability and throughput. Comparative evaluations further highlight the superiority of the proposed NS method over other existing approaches.