Anti-periodic Solutions For Two Class Neutral-type Neural Networks With Distributed Delays

Cellular neural networks have attracted considerable attention because of pattern recognition,optimization,model identification,signal processing,potential applicatio-ns in associative memory,etc.In the last few years,many authors have study many p-roblems to neural networks,such as,the existence and uniqueness?asymptotic beha-aviour of pseudo almost periodic solutions?asymptotically almost periodic solutions for neural networks,and so on.However,very few results are available on the existe-nce and exponential stability of anti-periodic solutions for neural networks,particulary,neutral-type neural networks.We mainly discussed the existence and stability of anti-periodic solution for the neural networks,in the second chapter,we study the following discrete time neutral type neutral-type neural networks with time-varying delays:where Ai is a difference operator defined by(Aixi)(n)=xi(n)-ci(n)xi(n-?),i=1,2,...,m,(4)Where y is a positive integer.In the third chapter,we considered the following with impulsive discrete-time neutral-type inhibitory cellular neural networks with distrib-uted delays:where Aij is a difference operator defined by(Aijxij)(n)= xij(n)-bij(n)xij(n-?),i=1,2,...,m,j=1,2,...,p.? is a positive integer.These two kinds of neural networks are both neutral-type neural networks,thus,the second kind of neural networks is neutral-type neural networks witih neutral type.By using the method of coincidence degree theory and constructing suitable Laypunov functional,transformation and some analysis methods such as inequality sufficient co-nditions are given to guarantee global exponential stability of anti-periodic solution an-d through strict proof.Finally,a numerical example is given to show the effectivenes of the results in this paper.