Resilient Asynchronous State Estimation for Markovian Jump Neural Networks Subject to Stochastic Nonlinearities and Sensor Saturations
研究了一类离散时间马尔可夫跳跃神经网络在随机非线性、传感器饱和和参数不确定下的异步状态估计问题，利用隐马尔可夫模型和伯努利过程建模，保证了误差系统的随机稳定性和耗散性能。
This article studies the problem of dissipativity-based asynchronous state estimation for a class of discrete-time Markov jump neural networks subject to randomly occurring nonlinearities, sensor saturations, and stochastic parameter uncertainties. First, two stochastic nonlinearities occurring in the system are described by statistical means and obey two Bernoulli processes independently. Then, the hidden Markov model is used to characterize the real communication environment closely between the designed estimator and the system model due to the networked-induced phenomenons that also lead to randomly occurring parametric uncertainties of the estimator considered modeled by two Bernoulli processes. A new criterion is established to guarantee that the resulting error system is stochastically stable with predefined dissipativity performance. Finally, we provide a simulation example to validate the theoretical analysis.