A comprehensive approach for assessing the reliability of complex networks using OANN approach

Abstract

The optimistic model of network reliability assumes the nodes to be trustworthy while assigning a failure probability to the network's connections. For calculating a network's reliability, the persistent condition is used to guarantee that no two edges have the same failure probability. Accurately estimating the network's reliability across all of its endpoints is an NP-hard problem. The study of network reliability centers on problems arising from topological isolation of individual data nodes. Analysis of network reliability spans the phases of building, deploying, and testing a network. Issues with connection, capacity, and trip time are all metrics used in analyses of computer network reliability. In this paper, an Optimal ANN strategy is proposed as a means of assessing the reliability of the network. In this research, we analyze the methods used in and findings from recent studies on trustworthiness. The term "Optimized Artificial Neural Network" (OANN) is used to describe a strategy that takes into account aspects of both neural networks and another way for measuring trustworthiness. Results are well tested on network of 2^8nodes (mesh network) and hyper-tree network of n nodes. A network's performance may be evaluated with the help of the suggested method, which also contributes to the calculation of its cost and reliability metrics.