Title :
Fast identification of linear systems in the presence of input and output noises
Author_Institution :
Sch. of Quantitative Methods & Math. Sci., Univ. of Western Sydney, NSW, Australia
Abstract :
The paper proposes a fast identification method for linear systems in the presence of input and output noise. The proposed method entails a new scheme that makes use of added information regarding the identified system so that a better estimate of the unknown variances of the input and output noise can be attained in a quicker manner during the iterative estimation process. Simulations demonstrate that the proposed method can converge at a faster rate towards the true system parameters and the accuracy of the parameter estimates can also be raised.
Keywords :
control theory; convergence of numerical methods; iterative methods; linear systems; parameter estimation; random noise; signal processing; transfer functions; communications; control systems; input noise; iterative estimation; linear system identification method; output noise; parameter estimation; signal processing; transfer function; variances; Communication system control; Control system synthesis; Equations; Iterative methods; Linear systems; Noise cancellation; Parameter estimation; Process control; Transfer functions; White noise;
Conference_Titel :
Signal Processing, 2002 6th International Conference on
Print_ISBN :
0-7803-7488-6
DOI :
10.1109/ICOSP.2002.1180970
