Dynamic Trust Modeling In Siov: A Hybrid Multilayer Framework Of Fuzzy Logic, Temporal Dynamics, And Shapley-Based Cooperative Game Theory Under Adversarial Conditions

Abstract

In the complex and dynamic environment of the "Social Internet of Vehicles" (SIoV), recognized as a transformative advancement within the Internet of Things domain focusing on human-machine interactions, the concept of trust emerges as a multidimensional, context-dependent, and behaviorally sensitive variable. It plays a fundamental role in ensuring communication security, facilitating intelligent cooperation, and enhancing decision-making processes. This research presents the design and evaluation of a multilayer framework for modeling trust within the intricate and evolving ecosystem of the Social Internet of Vehicles (SIoV). The proposed framework integrates fuzzy logic, fuzzy temporal value analysis, and cooperative game theory to examine trust’s behavioral dimensions—honesty, sincerity, privacy, and connectivity—through dynamic temporal modeling and strategic analysis of agent interactions. Leveraging fuzzy approximate inference and simulation within the Veins platform, the model achieved trust estimation accuracy exceeding 89% and identified the IC strategy as the most stable behavioral pattern. Additionally, risk and utility analyses improved decision-making accuracy. Comparative assessments with baseline models demonstrated the proposed framework’s performance improvement of 5 to 11 percent. By offering a coherent and generalizable approach, this study significantly contributes to advancing intelligent social systems within the SIoV environment.