IEEE Trans Neural Netw Learn Syst. 2018 May;29(5):1888-1899. doi: 10.1109/TNNLS.2017.2688582. Epub 2017 Apr 11.
The problem of dissipativity-based resilient filtering for discrete-time periodic Markov jump neural networks in the presence of quantized measurements is investigated in this paper. Due to the limited capacities of network medium, a logarithmic quantizer is applied to the underlying systems. Considering the fact that the filter is realized through a network, randomly occurring parameter uncertainties of the filter are modeled by two mode-dependent Bernoulli processes. By establishing the mode-dependent periodic Lyapunov function, sufficient conditions are given to ensure the stability and dissipativity of the filtering error system. The filter parameters are derived via solving a set of linear matrix inequalities. The merits and validity of the proposed design techniques are verified by a simulation example.
本文研究了存在量化测量的离散时间周期马尔可夫跳跃神经网络基于耗散的弹性滤波问题。由于网络介质的容量有限，对底层系统应用了对数量化器。考虑到滤波器是通过网络实现的，滤波器的随机发生的参数不确定性通过两个模式相关的伯努利过程来建模。通过建立模式相关的周期 Lyapunov 函数，给出了确保滤波误差系统稳定性和耗散性的充分条件。通过求解一组线性矩阵不等式来推导滤波器参数。通过仿真示例验证了所提出设计技术的有效性和有效性。
