共 47 条
State Estimation for Periodic Neural Networks With Uncertain Weight Matrices and Markovian Jump Channel States
被引:59
作者:
Xu, Yong
[1
,2
]
Wang, Zhuo
[3
]
Yao, Deyin
[1
,2
]
Lu, Renquan
[1
,2
]
Su, Chun-Yi
[1
,2
]
机构:
[1] Guangdong Univ Technol, Sch Automat, Guangzhou 510006, Guangdong, Peoples R China
[2] Guangdong Key Lab IoT Informat Technol, Guangzhou, Guangdong, Peoples R China
[3] Beihang Univ, Sch Instrumentat Sci & Optoelect Engn, Beijing 100191, Peoples R China
来源:
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
|
2018年
/
48卷
/
11期
基金:
中国国家自然科学基金;
关键词:
Markov chain;
neural networks (NNs);
packet dropouts;
state estimator;
stochastic parameter;
GENETIC REGULATORY NETWORKS;
DISSIPATIVITY ANALYSIS;
SENSOR NETWORKS;
DISCRETE;
SYSTEMS;
STABILITY;
STABILIZATION;
PARAMETERS;
D O I:
10.1109/TSMC.2017.2708700
中图分类号:
TP [自动化技术、计算机技术];
学科分类号:
0812 ;
摘要:
This paper studies the state estimator design for periodic neural networks, where stochastic weight matrices B(k) and packet dropouts are considered. The stochastic variables, which may influence each other, are introduced to describe uncertainties of weight matrices. In order to model the time-varying conditions of the communication channel, a Markov chain is employed to study the jumping cases of the stochastic properties of the packet dropouts (i.e., Bernoulli process with jumping means and variances being used to handle the packet dropouts). A state estimator is constructed such that the augmented system is stochastically stable and satisfies the H-infinity performance. The estimator parameters are derived by means of the linear matrix inequalities method. Finally, a numerical example is provided to illustrate the effectiveness of the proposed results.
引用
收藏
页码:1841 / 1850
页数:10
相关论文