Title :
Multiscale realization and estimation for space and space-time problems
Author :
Ho, Terrence T. ; Frakt, Austin B. ; Willsky, Alan S.
Author_Institution :
MIT, Cambridge, MA, USA
Abstract :
Although there have been significant advances in realization techniques for a class of multiscale stochastic models introduced by Willsky, Benveniste, Chou, et al. (1994), there remain a number of important open problems. We address two of them in this paper. The first is a new static realization method that is two orders of magnitude more efficient than methods previously available, but no less general. The second is the extension of this multiscale methodology to modeling dynamic systems
Keywords :
parameter estimation; parameter space methods; dynamic systems modeling; multisacle model parameters; multiscale estimation; multiscale realization; multiscale stochastic models; open problems; space problems; space-time problems; static realization method; Ear; Error analysis; Error correction; Estimation error; Reduced order systems; Smoothing methods; Statistics; Stochastic processes; Tree data structures; White noise;
Conference_Titel :
Information Theory, 1998. Proceedings. 1998 IEEE International Symposium on
Conference_Location :
Cambridge, MA
Print_ISBN :
0-7803-5000-6
DOI :
10.1109/ISIT.1998.708600
