Synchronization of discrete-time neural networks with time delays subject to missing data

• Published in: Neurocomputing (Amsterdam) Vol. 122; pp. 418 - 424
• Main Authors: Wu, Zheng-Guang, Park, Ju H.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 25.12.2013; Elsevier
• Subjects: Applied sciences; Artificial intelligence; Computer science; control theory; systems; Computer systems and distributed systems. User interface; Connectionism. Neural networks; Control system analysis; Control system synthesis; Control theory. Systems; Controllers; Criteria; Exact sciences and technology; Gain; Linear matrix inequality (LMI); Missing data; Neural networks; Software; Synchronism; Synchronization; Time delay; Missing data; Time-delay; Synchronization; Linear matrix inequality (LMI); Neural networks; Probabilistic approach; Transit time; Lyapunov method; Feedback regulation; Control synthesis; Lossy channel; Functional analysis; Discrete data; Neural network; Delay dependent system; Modeling; Markov chain; Efficiency; Discrete time; Linear matrix inequality; Delay time; Incomplete information; Transmission loss; Lyapunov function
• ISSN: 0925-2312; 1872-8286
• DOI: 10.1016/j.neucom.2013.06.011

This paper is concerned with the problem of synchronization of discrete-time neural networks with time-delays under unreliable communication links, which are modeled as stochastic dropouts. The process of missing data satisfies a discrete-time Markov chain with two state components. By using Lyapunov functional approach, some delay-dependent synchronization criteria are first obtained and formulated in the form of linear matrix inequalities (LMIs). Then, sufficient conditions on the existence of feedback controllers are derived by employing these newly obtained synchronization criteria. The controller gains can be achieved by solving a set of LMIs. Finally, a numerical example is given to illustrate the effectiveness of the design method. •This paper is concerned with the problem of synchronization of discrete-time neural networks with time-delays under unreliable communication links.•Some delay-dependent synchronization criteria are obtained and formulated in the form of linear matrix inequalities.•A feedback controller has been designed such that the master system and the slave system are globally exponentially synchronized in the mean square.