Title :
Efficient implementation of SVD-updating algorithms for subspace tracking
Author :
Ali, Markus ; Götze, Jüryen
Author_Institution :
Inst. of Network Theory & Circuit Design, Tech. Univ. of Munich, Germany
Abstract :
An efficient parallel implementation of the SVD (singular value decomposition) updating algorithm is addressed. The modifications applied to the algorithm are: approximation of Kogbetliantz´s SVD algorithm, approximation of the SVD-rotation evaluations, and factorization of the rotations. This results in a square root and division free SVD-updating algorithm, which combines the speedups and hardware reductions obtained by the different modifications. The proposed SVD-based subspace tracking algorithm only requires the implementation of (+, *) operations in all processor cells
Keywords :
matrix decomposition; parallel algorithms; singular value decomposition; state-space methods; tracking filters; Kogbetliantz´s SVD algorithm; SVD-rotation evaluations; SVD-updating algorithms; factorization; parallel implementation; subspace tracking algorithm; Approximation algorithms; Circuit synthesis; Data mining; Hardware; Matrix decomposition; Multiple signal classification; Parallel processing; Singular value decomposition;
Conference_Titel :
Communications, Computers and Signal Processing, 1993., IEEE Pacific Rim Conference on
Conference_Location :
Victoria, BC
Print_ISBN :
0-7803-0971-5
DOI :
10.1109/PACRIM.1993.407345
