Accession Number : ADA302269

Title :   Modifying Two-Sided Orthogonal Decompositions: Algorithms Implementation, and Applications.

Descriptive Note : Technical rept. Dec 93-Oct 95,

Corporate Author : PENNSYLVANIA STATE UNIV UNIVERSITY PARK APPLIED RESEARCH LAB

Personal Author(s) : Yoon, Peter A.

PDF Url : ADA302269

Report Date : OCT 1995

Pagination or Media Count : 196

Abstract : In this thesis we propose several algorithms for rank-one updates and downdates to these decompositions with strong stability properties and efficient implementations on high-performance computers. We seek algorithms which only require O(n2) operations per update or downdate unlike recomputing the two-sided orthogonal decomposition (TSOD) in O(n3). We also desire highly regular data movement inherited in these algorithms in order to implement these efficiently on the distributed memory MIMD multiprocessors. The algorithms are based upon 'chasing' strategies for updating and downdating procedures for orthogonal decompositions. (AN)

Descriptors :   *ALGORITHMS, *MATHEMATICAL PROGRAMMING, MATHEMATICAL MODELS, SIGNAL PROCESSING, IMAGE PROCESSING, STABILITY, COMPUTATIONS, DATA MANAGEMENT, DISTRIBUTED DATA PROCESSING, PERFORMANCE(ENGINEERING), MATRICES(MATHEMATICS), ACCURACY, EFFICIENCY, PARALLEL PROCESSING, TRANSFORMATIONS(MATHEMATICS), ERROR ANALYSIS, LEAST SQUARES METHOD, PERTURBATION THEORY, MULTIPROCESSORS, PATTERN RECOGNITION, SUBROUTINES, LINEAR REGRESSION ANALYSIS, DECOMPOSITION, ORTHOGONALITY.

Subject Categories : Operations Research

Distribution Statement : APPROVED FOR PUBLIC RELEASE