Markov Has Quantization And Packet Loss Transition Neural Network Feedback Control

The neural network, as an important large-scale complex system, exhibits rich and colorful dynamical behaviors. It has important and potential applications in solving associative memories, optimizing solvers, signal processing and so on. Since the limition of the neural network’s communication bandwidth, service capacity and carrying capacity, the performance of the NNCSs is suppressed. Recently, the stability of quantied NNCSs has attracted a large number of researchers, and a series of significant results have been achieved.This dissertation focuses on the out-feedback control design for markovian jumping neural network systems subject to quantization and dropout. Two well-known NNCSs models, i.e., the logarithmic quantize and Bernoulli model. Our purpose here is to develop several conditions such that the closed-loop systems are stable and satisfy corresponding performance indices. The main work of this dissertation is outlined as follows:1. The background and the research significance of Markovian jumping NNCSs are given. In addition, we summarize the basis problems of NCSs2. We establish the feedback control model for Markovian jumping NNCSs subject to quantization and dropout. By using the Lyapunov function, some sufficient conditions are given ensuring that the considered system are derived in terms of LMI. Then the controller design problem is also given.3. We study the observer-based output feedback control system. By using the Lyapunov function, some criteria for the robust exponentially stability of Markovian jumping neural networks with quantized are given. Based on these criteria, the corresponding controller design problem can be obtained.