Title :
A recurrent neural network for online design of robust optimal filters
Author :
Jiang, Danchi ; Wang, Jun
Author_Institution :
Chinese Univ. of Hong Kong, Shatin, China
fDate :
6/1/2000 12:00:00 AM
Abstract :
A recurrent neural network is developed for robust optimal filter design. The purpose is to fill the gap between the real-time computation requirement in practice and the computational complexity of the filter design in the case that the statistical properties of noise are unknown. First, an H∞ requirement and an L2 requirement of the filter design problem are formulated as a group of linear matrix inequalities. On this basis, an optimization problem is introduced to solve the robust optimal filter design problem. Then, a recurrent neural network is deliberately developed for solving the optimization problem in real time. The effectiveness and efficiency of the recurrent neural network is shown by use of theoretical and simulation results
Keywords :
H∞ optimisation; circuit CAD; computational complexity; filtering theory; linear systems; matrix algebra; recurrent neural nets; H∞ requirement; L2 requirement; computational complexity; linear matrix inequalities; online design; optimal filter design; optimization problem; real-time computation requirement; recurrent neural network; robust optimal filters; Computational complexity; Design methodology; Design optimization; Linear matrix inequalities; Neural networks; Noise robustness; Nonlinear filters; Recurrent neural networks; Riccati equations; Signal processing;
Journal_Title :
Circuits and Systems I: Fundamental Theory and Applications, IEEE Transactions on
