MAID: Mobility-aware information dissemination in mobile IoT using temporal point processes

Yongqing Cai, Dianjie Lu, Jing Chen, Guijuan Zhang

Abstract: The mobile Internet of Things (IoT) integrates mobile communication technology with IoT to connect various physical devices (e.g., sensors, smart devices, vehicles, and home appliances) for data collection, processing, and distribution. However, current research on mobile IoT information dissemination overlooks the stochastic nature of device mobility, leading to inaccurate predictions of dissemination scale. To address this, we propose a mobility-aware information dissemination model (MAID) using the temporal point process (TPP) to investigate the stochastic dynamics introduced by device mobility. First, we develop a TPP-based model to describe random events, such as movement, linking, unlinking, and information dissemination. We propose a mobility-aware intensity prediction method to calculate event intensities within the TPP framework. Finally, we predict the scale of information dissemination on the basis of the calculated intensity and develop an event-driven simulation system to model network structure changes and information dissemination within the mobile IoT. The simulation results indicate that device mobility accelerates network structure changes, thereby increasing the scope and scale of information dissemination. This dynamic has a two-sided effect on dissemination efficiency, depending on the initial network sparsity. Extensive experiments on synthetic datasets show that our method improves the accuracy of dissemination scale prediction by over 92% compared to four baseline methods.