Title :
Nonlinear filtering by kriging, with application to system inversion
Author :
Costa, J.-P. ; Pronzato, L. ; Thierry, E.
Author_Institution :
CNRS-UNSA, Biot, France
Abstract :
Prediction by kriging does not rely on any specific model structure, and is thus much more flexible than approaches based on parametric behavioural models. Since accurate predictions are obtained for extremely short training sequences, it generally performs better than prediction methods using parametric models. Application to nonlinear system inversion is considered
Keywords :
Gaussian processes; covariance matrices; filtering theory; inverse problems; maximum likelihood estimation; nonlinear filters; prediction theory; random processes; sequences; statistical analysis; Gaussian process; MLE; SISO nonlinear system; covariance matrix; kriging prediction; linear regression; nonlinear filtering; nonlinear system inversion; parametric models; random process; short training sequences; Autoregressive processes; Filtering; Linear regression; Nonlinear systems; Parametric statistics; Polynomials; Prediction methods; Predictive models; Random processes; Signal processing;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1999. Proceedings., 1999 IEEE International Conference on
Conference_Location :
Phoenix, AZ
Print_ISBN :
0-7803-5041-3
DOI :
10.1109/ICASSP.1999.756221
