A New Method for Topology Identification of Complex Dynamical Networks

Area of Science:

• Complex Systems
• Network Science
• Dynamical Systems Theory

Background:

• Topology identification of complex dynamical networks is crucial.
• Existing methods often rely on the linear independence condition (LIC), which has verification challenges and rigor issues.
• Network synchronization can lead to identification failures in current approaches.

Purpose of the Study:

• To propose a new, more robust method for complex dynamical network topology identification.
• To overcome the limitations and potential failures associated with the linear independence condition (LIC).
• To eliminate the need for verifying the linear independence condition.

Main Methods:

• Introducing a regulation mechanism to the original network.
• Constructing an auxiliary network composed of isolated nodes.
• Utilizing outer synchronization between the regulated network and the auxiliary network for identification.

Main Results:

• The proposed method successfully identifies the topology of complex dynamical networks.
• It effectively avoids identification failures caused by network synchronization.
• The method eliminates the necessity of checking the linear independence condition (LIC).

Conclusions:

• The novel method provides a reliable and high-performing solution for network topology identification.
• It offers a significant advancement over existing techniques by addressing critical limitations.
• The approach is demonstrated to be robust and effective through provided examples.