Efficient Caching and Spreading of Traffic Videos by Optimized Topology in IoV

Zeyu Hu, Yong Li, Suhua Tang

Abstract: Traffic videos represent a crucial source of data for future vehicle technologies, such as autonomous driving, and proactive dissemination of such videos can reduce delay in the Internet of Vehicles (IoV). Nonetheless, the IoV presents challenges for data diffusion, including variable vehicle locations, timely video delivery, scarce resources, and complex link quality. This paper seeks to establish the underlying relationship between data diffusion efficiency and IoV topology, and explore topology optimization criteria to minimize total network delay. Specifically, we establish mathematical models for topology control, parallel transmission, and diffusion delay, respectively. We quantify the spreading index using the complex network spreading dynamics model to formulate accurate diffusion strategies. Simulation verifies the spreading index's effectiveness and limitations. Based on this index, we propose an efficient topology control strategy and hybrid diffusion strategy for delay reduction. Compared to traditional content-centric network (CCN) and proactive diffusion strategies, our approach is superior in optimizing total delay by leveraging the advantages of a more efficient topology. We verify the effectiveness of our approach through extensive simulation results and describe the applicable scenarios.

External IDs:dblp:journals/tvt/HuLT24