Open AccessExponential H ∞ filtering for switched neural networks with mixed delays
Author(s) -
Su Ziyi,
Wang Hongxia,
Yu Li,
Zhang Dan
Publication year - 2014
Publication title -
iet control theory and applications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.059
H-Index - 108
eISSN - 1751-8652
pISSN - 1751-8644
DOI - 10.1049/iet-cta.2013.0879
Subject(s) - dwell time , filter (signal processing) , control theory (sociology) , artificial neural network , computer science , exponential stability , attenuation , exponential function , noise (video) , activation function , function (biology) , convex combination , stability (learning theory) , regular polygon , topology (electrical circuits) , mathematics , algorithm , convex optimization , artificial intelligence , image (mathematics) , mathematical analysis , optics , biology , geometry , control (management) , quantum mechanics , evolutionary biology , machine learning , computer vision , medicine , clinical psychology , physics , nonlinear system , combinatorics
The study focuses on the exponential H ∞ filtering problem of biological neural nets (BNNs). By considering some realistic factors including delays, disturbance and topology changes, the well‐known leaky integrate‐and‐fire model is modified as a switched neural network so that function of a single neuron is identified via the H ∞ filtering instead of biological experimental methods. With the aid of average dwell time method, we provide a delay‐dependent sufficient condition, under which the designed filter for the function of every individual neuron in BNNs satisfies H ∞ noise attenuation and exponential stability. Moreover, the design of such a filter is converted into a convex optimisation problem, which can be easily solved by using standard numerical software. Finally, two examples are given to show the effectiveness of the proposed method.