Task-Specific Trust Modeling and Resource Matching Enabled by Hypergraph for Efficient Task Completion in Collaborative IoT Systems

Rapid advancements in collaborative computing have enabled the exponential expansion of Internet of Things (IoT) systems, supporting pervasive connectivity and intelligent task execution across heterogeneous devices. However, achieving reliable and efficient task completion in IoT systems remains challenging due to dynamic network conditions and diverse physical attributes of computing resources and tasks. To overcome these challenges, this thesis introduces a novel task-specific trust modeling and resource matching enabled by hypergraph to achieve efficient task completion in IoT networks. Firstly, a task-oriented trusted collaboration enabled by hypergraph (TTC-hypergraph) is modeled to assign tasks to suitable collaborators based on dynamic trust evaluation that integrate task-specific requirements, and resource availability. Secondly, a trusted task-resource matching (TTRM) framework is constructed by leveraging historical collaboration-based hypergraph and task hypergraph to capture device historical performances and task requirements respectively. Then, task-resource matching is performed to identify the most suitable collaborators through a novel trust-driven reweighted random walk mechanism. Simulation results demonstrate that the proposed solutions achieve higher task execution efficiency, providing a robust foundation for trusted collaboration in IoT systems. The proposed frameworks hold significant practical implications, particularly in large-scale, complex IoT applications such as autonomous vehicular networks, industrial automation, and real-time healthcare systems, where efficient and reliable collaboration among devices is <a href="http://essential.

Virtual:” target=”_blank” title=”essential.

Virtual:”>essential.

Virtual: https://events.vtools.ieee.org/m/496565