Robustly adaptive synchronization for stochastic Markovian neural networks of neutral type with mixed mode-dependent delays

Cheng-De Zheng,Ziping Wei,Zhanshan Wang
DOI: https://doi.org/10.1016/j.neucom.2015.07.066
IF: 6
2016-01-01
Neurocomputing
Abstract:A robust adaptive feedback controller is designed to achieve complete synchronization of a class of neutral-type stochastic Markovian jump neural networks with mode-dependent and mixed delays. The mixed delays consist of time-varying discrete delays and continuously distributed delays. By employing Wirtinger-based integral inequality, a reciprocally convex combination technique and a generalized LaSalle-type invariance principle for stochastic Markovian differential delay equations, this paper establishes two globally almost surely asymptotical stability conditions of the error dynamical system in the mean-square sense for all admissible uncertainties. An example and its numerical simulation are given to demonstrate the effectiveness of the theoretical results. HighlightsFirst LMI-based synchronization of neutral-type stochastic Markovian jump NNs with mode-dependent.Effective combination of Wirtinger-based inequality and reciprocally convex combination.Application of LMI and generalized LaSalle-type invariance principle.Effective and less conservative delay-dependent sufficient conditions.
What problem does this paper attempt to address?