School of Information Science and Engineering, Northeastern University, Shenyang 110819, China.
Neural Netw. 2016 Dec;84:47-56. doi: 10.1016/j.neunet.2016.08.002. Epub 2016 Aug 30.
This paper investigates the H state estimation problem for a class of discrete-time memristive neural networks (DMNNs) with time-varying delays. For the sake of coping with the switched weight matrices, the DMNNs are recast into a tractable model by defining a series of state-dependent switched signals. Based on the tractable model, the robust analysis method and Lyapunov stability theory are developed to devise a sufficient condition which ensures the global asymptotical stability of the estimation error system with a prescribed H performance. The desired state estimator gain matrix and optimal performance index can be accomplished via solving a convex optimization problem subject to several linear matrix inequalities (LMIs). Finally, one numerical example is presented to check the effectiveness of the theoretical results.
本文研究了一类具有时变时滞的离散时间忆阻神经网络（DMNNs）的 H 估计问题。为了应对切换权矩阵，通过定义一系列状态相关的切换信号，将 DMNNs 转换为一个可处理的模型。基于可处理的模型，利用鲁棒分析方法和 Lyapunov 稳定性理论，设计了一个充分条件，该条件确保了具有给定 H 性能的估计误差系统的全局渐近稳定性。通过求解几个线性矩阵不等式（LMIs）下的凸优化问题，可以得到期望的状态估计增益矩阵和最优性能指标。最后，通过一个数值实例验证了理论结果的有效性。
