Research on network time reliability evaluation method based on uncertainty theory

In view of the problem that existing network time reliability assessment methods often only consider the impact of inherent uncertainty, while ignoring the impact of epistemic uncertainty caused by insufficient fault information on reliability assessment results, a new method based on uncertainty theory is proposed. According to the node range of network reliability, two types of measurement parameters of single node pair and multi-node pair time reliability are designed. The extended uncertain network model is proposed to directly model the epistemic uncertainty characteristics on nodes and links, and further single node and multi-node pair time reliability algorithms based on the most reliable path and the most reliable extended uncertain subnetwork are designed. Finally, taking a six-node network and the backbone network of China Education and Research Network (CERNET) as examples, the method is applied to evaluate two time reliability indicators, verifying the correctness and effectiveness of the method.