Dynamic Trust Evaluation and Resilience Assessment in Edge Computing Networks

Cloud computing has become the cornerstone of low-latency and resource-efficient processing in distributed systems, particularly for applications such as the Internet of Things (IoT) and autonomous systems. However, these edge networks present significant challenges in trust and security management due to their inherently decentralized nature. This paper addresses this challenge by presenting a novel dynamic trust evaluation framework. The proposed framework models the spatial and temporal evolution of edge networks over time. It incorporates attack-specific impact analysis and introduces new trust propagation mechanisms that account for the cascading effects of security events. Additionally, a comprehensive set of metrics is developed to evaluate detection rates, average trust levels, and network resilience, assessing performance against various attack scenarios. The framework is designed with a modular architecture, and its implementation has been tested in simulated environments. Results demonstrate that the proposed framework can maintain high detection accuracy with minimal trust degradation, even in the presence of severe attacks occurring at high frequencies. It outperforms existing state-of-the-art methods in terms of adaptability and the fine-grained modeling of trust dynamics.