Title :
Benefits of multi-weight optimization in OBE algorithms
Author :
Joachim, D. ; Deller, J.R., Jr.
Author_Institution :
Dept. of Electr. & Comput. Eng., Michigan State Univ., East Lansing, MI, USA
Abstract :
Optimal bounding ellipsoid identification algorithms feature a unique data selection process which recursively checks observations for innovation, then assigns weights in accordance with information content. Multi-weight optimization enhances this process by reoptimizing over a number of past weights. This technique offers faster unbiased convergence of the central estimator, improved tracking, increased selectivity, and can be used to decrease computational complexity. These properties potentially decrease idle time in multi-processor systems
Keywords :
computational complexity; convergence of numerical methods; optimisation; parameter estimation; OBE algorithms; computational complexity; data selection process; idle time; information content; multi-weight optimization; optimal bounding ellipsoid identification; selectivity; tracking; unbiased convergence; Approximation algorithms; Computational complexity; Convergence; Covariance matrix; Data engineering; Ellipsoids; Engineering education; Recursive estimation; Size measurement; Technological innovation;
Conference_Titel :
Circuits and Systems, 1998. Proceedings. 1998 Midwest Symposium on
Conference_Location :
Notre Dame, IN
Print_ISBN :
0-8186-8914-5
DOI :
10.1109/MWSCAS.1998.759480
