Asynchronous filtering for discrete-time Markov jump neural networks

摘要：

　　In this paper, the asynchronous H∞ filtering problem for discrete-time Markov jump neural networks is investigated.Since the information on the jump mode of the neural networks is not always available, the mode of the filter often doesn”t correspond to that of the discrete-time Markov jump neural networks.To overcome this kind of asynchronous phenomenon, a novel filtering design method is proposed.Here the mode of the neural networks and the mode of the filter are subject to two different Markov chains.By introducing a unified Lyapunov functional, we derive a sufficient condition in terms of linear matrix inequality(LMI) such that the resultant filtering error system is stochastically stable.Finally a numerical example is given to demonstrate the effectiveness of the proposed theoretical results.

关键词： Markov jump neural networks asynchronous filer linear matrix inequalities(LMI)

作者: Zhaowen Xu Hongye Su Zheng-Guang Wu

作者单位: National Laboratory of Industrial Control Technology, Institute of Cyber-Systems and Control, Zhejiang University, Yuquan Campus, Hangzhou Zhejiang, 310027, PR China

会议类型: 国内会议

会议名称: 西南大学2014年全国博士生学术论坛(电子技术与信息科学领域)

会议地点: 重庆

会议语种:英文

页码: 52-60

在线出版日期: 2014-12-01（万方平台首次上网日期，不代表论文的发表时间）