Event-triggered State Estimation And Filtering For Delayed Neural Networks With Quantization

Recently,neural networks have widely applied in various fields,such as signal processing,pattern recognition and combinatorial optimization.However,in networked application,under the factors such as time-delay,data packet dropouts,quantization occurs in the network environment,it will degrade the system performance and even may make the system unstable.So the studies of quantization and time-delay of neural networks have significant theoretic meaning and application value.Meanwhile,in order to save the limited network bandwidth resources and reduce the burden of the network,the research on how to design a suitable event-triggered scheme has become a hot issue.Nevertheless,the problem of event-triggered and quantization in neural networks system have not been fully discussed,there are still many research problems that need to be solved.In this paper,we investigates the state estimation and H_?filter design problem for the delayed neural network systems with event-triggered sheme and quantization,the main contents of this thesis are outlined as follows:(1)Considering a class of delayed neural network systems with event-triggered communication and quantization,by employing the Lyapunov-Krasovskii functional approach and linear matrix inequality technique,some sufficient conditions are obtained under which state estimator exists and the estimator error dynamics is asymptotically stable.Finally,a numerical example is provided to demonstrate the effectiveness of the proposed approach.(2)An event-based filter design model for neural network systems is constructed by taking the effect of the similar event-triggered scheme and the quantization into consideration.Besides,sufficient conditions are established in terms of linear matrix inequalities for the exponential stability of the filtering error dynamics and the explicit expression is given for the designed filter parameters.Finally,Matlab simulation example is given to show the feasibility of the conclusion.Finally,this paper summarizes the main results of the thesis and the future research direction.