Power-Efficient and Trustworthy Data Dissemination for Social Vehicle Associations in the Internet of Vehicles

Abstract

<div>In modern era, with the global spread of massive devices, connecting, controlling, and managing a significant amount of data in the IoT environment, especially in the Internet of vehicles (IoV) is a great challenge. There is a big problem of high-energy consumption due to overhead-unwanted data communication to the non-participatory vehicles, at high enduring connection rate. Therefore, this article proposed a social vehicle association-based data dissemination approach, which was segregated into three parts: <i>First</i>, develop an improved power evaluation approach for discovering power-efficient vehicles. <i>Second</i>, using the Fokker–Planck equation, the connection likelihood of these vehicles is calculated in the second phase to find trustworthy and steady connections. <i>Last</i>, develop an evaluation approach for vehicles community association using convolutional neural network (CNN). It filtered most likely vehicles to form a community for data dissemination by considering temporal, spatial, and social attributes of vehicles. The proposed approach has evaluated using widespread simulation tests in a highway environment. It verified the efficacy of proposed approach regarding power, linking, and community score of vehicles. The finding of experiment shows that, with advancement of power, connectivity, and community score of vehicles, data dissemination also enhanced. Furthermore; it guarantees that data will be shared efficiently with great reliability.</div>