Title :
Inversion of a nonlinear communication channel by kriging
Author :
Costa, J.P. ; Pronzato, Luc ; Thierry, E.
Author_Institution :
Lab. I3S, CNRS-UNSA, Biot, France
Abstract :
We consider a problem of nonlinear system-inversion, with unknown underlying model structure. Classical parametric behavioural models (Volterra and NARMAX models) involve a lot of parameters, which are difficult to estimate from short training sequences. A similar difficulty is encountered when methods based on neural networks are used. We suggest in this paper to use a semi-parametric approach called kriging. We show on an example that good performances are obtained for short training sequences
Keywords :
covariance matrices; filtering theory; inverse problems; nonlinear systems; optimisation; prediction theory; statistical analysis; telecommunication channels; jamming; kriging; nonlinear communication channel; nonlinear system-inversion; optimisation; parametric behavioural models; radial basic function; semi-parametric approach; training sequences; Autoregressive processes; Communication channels; Covariance matrix; Neural networks; Nonlinear systems; Polynomials; Predictive models; Random processes; Signal processing; Vectors;
Conference_Titel :
Electronics, Circuits and Systems, 1999. Proceedings of ICECS '99. The 6th IEEE International Conference on
Conference_Location :
Pafos
Print_ISBN :
0-7803-5682-9
DOI :
10.1109/ICECS.1999.814475
