Parallel To Limit The Global Stability Of Cellular Neural Networks

Cellular neural networks (CNNs) were introduced by Chua and Yang in 1988. Since then, many authors have paid much attention to the research on the theory and application of the CNNs. A new class of CNNs, namely the shunting inhibitory cellular neutral networks (SICNNs) have been introduced in 1991 by Bouzerdoum and Pinter. Now SICNNs have been extensively applied in psychophysics, speech, perception, adaptive pattern recognition and emage processing.In this paper, the global stability for inhibitory cellular neural networks (SICNNs) are considered. There are some sufficient conditions guaranteeing the global robust stability of for shunting inhibitory CNNs with delays, but when the delay is very small, the notion of delay-independent stability may be overly restrictive. So we give a delay-dependent stability criteria, which give information on the delay-dependence property. But in practice, absolute constant delay for process of dynamics change is almost not existent, constant delay is only an ideal approach of time-varying delay. Therefore, the studies of model with time-varying delays have more important significance than the ones of model with constant delay. So we consider the time-varying delays SICNNs. We give some delay-independent or delay-dependent stability criteria for global stability of shunting inhibitory CNNs with time-varying delays.