Multi Agent Reinforcement Learning for Handover and Offlaoding in Vehicular Environment

This dataset contains simulation data generated to support the study on multi-agent reinforcement learning for joint base station handover and priority-aware task offloading in vehicular networks. The data were produced using the extended DriveNetSim simulator and capture vehicular mobility dynamics, wireless channel conditions, computational resource utilization, task characteristics, and offloading decisions across vehicular, edge (MEC), and cloud computing tiers. The dataset includes performance metrics such as end-to-end delay, processing and transmission latency, CPU and memory utilization, handover events, and task priority distributions. These data enable reproducibility and further analysis of mobility-aware and priority-driven offloading strategies in dynamic vehicular environments.