Bayes-based identification and localization of faulty links in LEO/MEO satellite communication networks

For LEO satellite systems employing intersatellite links (ISL´s), we present a Bayes-based method to identify and locate faulty links, which applies to a satellite network assuming a connection-oriented network structure, e.g., ATM or ATM-type switches on-board satellites. In the method a management satellite can autonomously and real-time identify a ranked list of the most probable failed, network links through building discrete-time dynamic virtual topology graph (DT-DVTG), then testing can be used to quickly pinpoint the actual faulty links. The performance of the method is evaluated through simulation.