Abstract
The rapid growth of the Internet of Things (IoT) offers significant opportunities for network optimisation, particularly with the increasing use of connected vehicles and IoT devices in enterprise environments. In these areas, service vehicles can serve as mobile data collectors for battery-powered IoT devices, enhancing cloud communication, extending battery life, and reducing operational costs. However, current mobility models used in VANET research primarily address urban or rural roads and fail to capture the specific movement patterns of vehicles in confined domains. The limitation in mobility models emphasises the need for specialised models for confined industrial areas. This paper proposes a synthetic mobility model for vehicles in enterprise environments derived from thousands of real-world traces. Our contributions include detailed insights into mobility patterns, probability distributions, and their corresponding best-fit mathematical models. The synthetic mobility is based on two Markovian models that can predict future speed and direction based on the previous speed. Additionally, we have developed highly accurate mathematical models for each distribution. The novelty of this work, compared to existing models, lies in our new dataset and the model's precision. It delivers a match of 86% to 98% with our collected mobility data.