Recursive least-squares quadratic filtering and fixed-point smoothing algorithms for signal estimation from uncertain observations are derived when the uncertainty is modeled by not necessarily independent variables and the observations contain white plus coloured noise. The proposed estimators do not require the knowledge of the state-space of the model generating the signal, but only the moments, up to the fourth one, of the processes involved, along with the probability that the signal exists in the obervations and the (2,2)-element of the conditional probability matrix of the sequence describing the uncertainty.
@article{urn:eudml:doc:38754,
title = {Quadratic estimation from non-independent uncertain observations with coloured noise.},
journal = {Extracta Mathematicae},
volume = {19},
year = {2004},
pages = {399-413},
zbl = {1118.62099},
mrnumber = {MR2135832},
language = {en},
url = {http://dml.mathdoc.fr/item/urn:eudml:doc:38754}
}
Nakamori, S.; Caballero, R.; Hermoso, A.; Jiménez, J.; Linares, J. Quadratic estimation from non-independent uncertain observations with coloured noise.. Extracta Mathematicae, Tome 19 (2004) pp. 399-413. http://gdmltest.u-ga.fr/item/urn:eudml:doc:38754/