Synchronization for Markov jump neural networks with generally uncertain transition rates
Rui-ping XU,Cun-chen GAO,Yong-gui KAO
DOI: https://doi.org/10.7641/CTA.2015.41107
2015-01-01
Abstract:Synchronization for Markov jump neural networks with generally uncertain transition rates XU Rui-ping1,2†, GAO Cun-chen3 , KAO Yong-gui4 (1. College of Information Science and Engineering, Ocean University of China, Qingdao Shandong, 266100, China; 2. College of Mathematics, Qingdao University, Qingdao Shandong, 266071, China; 3. School of Mathematical Sciences, Ocean University of China, Qingdao Shandong, 266100, China; 4. Department of Mathematics, Harbin Institute of Technology, Weihai Shandong, 264209, China) Abstract: This paper investigates the asymptotical synchronization problem for a class of Markov jump neural networks (MJNNs) with generally uncertain transition rates (GUTRs). In the GUTR neural network model, each transition rate may be completely unknown or only its estimate value is known. This new uncertain model could be applied in many practical cases. Based on the Lyapunov-Krasovskii function method, a sufficient condition for the asymptotical synchronization in mean square is derived in terms of linear matrix inequalities (LMIs). Finally, a numerical example is presented to illustrate the effectiveness of the developed method.