Topological structure identification method for complex networks with anomalous edges

Key Points

• The aim is to improve the identification of complex network topologies that include anomalous edges.
• Introduced a sparse Bayesian learning algorithm (S-Block-SBL) for network topology identification.
• Implemented a data purification strategy to enhance identification accuracy.
• Conducted experiments on benchmark networks to validate the proposed algorithm.
• Demonstrated the convergence and non-increasing property of the S-Block-SBL loss function.
• Validated the effectiveness and superiority of the algorithm over existing methods through experimental results.

Abstract