State Estimation for Nonuniformly Sampled Neural Networks With Hidden Information
针对采样间隔对估计器不可知的非均匀采样神经网络，利用隐藏马尔可夫链描述量化模型，设计了基于观测模式的估计器，并通过线性矩阵不等式求解增益。
This study addresses estimator design for a class of nonuniformly sampled neural networks under the scenario of the sampling interval being inaccessible to the estimator. A new quantization model is described by a hidden Markov chain, where the emission probability depends on the network status and sampling interval. Two variables called hidden mode and observed mode are defined based on the assumption that the data receiver can recognize the quantization density instead of the sampling interval, and the associated observed mode-dependent estimator is designed. An augmented estimation error system is obtained, and the strict <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$(\mathcal {Q},\mathcal {S},\mathcal {R})-\gamma -$ </tex-math></inline-formula> dissipativity for the nonuniformly sampled neural networks is investigated. Then the estimator gain is calculated by solving a set of linear matrix inequalities. Finally, the effectiveness of the proposed approach is demonstrated via an example.