Decentralized dynamic output feedback for globally asymptotic stabilization of a class of dynamic networks

The objective of this paper is to propose an approach to stabilization of a class of dynamic networks with each node being a non-linear system with multiple equilibria. The proposed algorithms, which are developed within the convex optimization framework, employ a decentralized dynamic output feedback structure. Furthermore, an interesting conclusion is reached, in which the stabilization problem for the whole Nn-dimensional dynamic networks can be converted into the simple n-dimensional space in terms of only three LMIs. An application of output stabilization of mutually coupled phase-locked loop networks is used to verify the effectiveness of the proposed methods.