Cognitive Vehicular Network Resource Allocation Algorithm Based on Network Slicing

To ensure that diverse applications in vehicular networks can achieve service quality (QoS) standards tailored to their requirements, this paper proposes a cognitive IoT resource allocation algorithm augmented by network slicing technology. The proposed algorithm leverages network slicing to provide differentiated QoS guarantees for heterogeneous vehicular services. Specifically, considering the reliability constraints of ultra-reliable low-latency (URLLC) V2V links, resource block (RB) allocation constraints, and network stability constraints, we establish a joint spectrum-power optimization model aimed at maximizing the satisfaction of V2I users. Within this framework, a Q-learning-based approach is designed to determine optimal strategies for resource allocation. Simulation outcomes demonstrate that our algorithm substantially enhances user satisfaction compared to existing schemes by achieving efficient spectrum utilization and adaptive resource management.