Proposal of Maximum Delay Estimation Method for Unidentified Networks by Generating Pseudo-Network

Abstract

The maximum network delay needs to be estimated for maintaining the stability of delay-sensitive systems, such as remote-control applications. Conventional estimation approaches, including network digital twin technologies, typically rely on complete network configuration information. However, when such information is incomplete or unidentified, these approaches cannot be applied. In this study, we propose a method to estimate the maximum delay by generating a pseudo-network whose delay characteristics closely resemble those of the target network. Candidate networks with different configurations are created, and the pseudo-network is selected by comparing their delay distributions with that of the target network. Experimental results show that a pseudo-network with similar delay characteristics to the target network can be found, even when its configuration differs from that of the target network. This result holds for multiple target networks. These findings support the potential applicability of our pseudo-network-based approach to unidentified networks.