DocumentCode :
1402863
Title :
Split versions of the Levinson-like and Schur-like fast algorithms for solving block-slanted-Toeplitz systems of equations
Author :
Joshi, Rajashri R. ; Yagle, Andrew E.
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., Michigan Univ., Ann Arbor, MI, USA
Volume :
46
Issue :
7
fYear :
1998
fDate :
7/1/1998 12:00:00 AM
Firstpage :
2027
Lastpage :
2030
Abstract :
In Joshi and Yagle (1998) the Fredholm equations of one-dimensional (1-D) inverse scattering and LLS estimation were transformed via the orthonormal wavelet transform into a series of symmetric “block-slanted-Toeplitz” (BST) systems of equations. Levinson-like and Schur-like fast algorithms were presented for solving the BST systems. Here, we present split versions of the Levinson-like and Schur-like fast algorithms. The significance of these split algorithms is as follows. Although the Levinson-like and Schur-like fast algorithms reduce the complexity of solving the BST systems from O(n3) to O(n2), there still exists an inherent redundancy in these algorithms in the case where the BST system matrices have centrosymmetric blocks. This situation arises when a symmetric wavelet basis function (like the Littlewood-Paley) is used in the problem transformation. This redundancy is exploited here to derive the split Levinson-like and split Schur-like fast algorithms. These split algorithms reduce the number of multiplications required at each iteration by a factor of two, as compared with the Levinson-like and Schur-like algorithms
Keywords :
Toeplitz matrices; computational complexity; integral equations; iterative methods; matrix multiplication; redundancy; wavelet transforms; BST systems; Fredholm equations; LLS estimation; Levinson-like fast algorithms; Littlewood-Paley wavelet basis function; Schur-like fast algorithms; block-slanted-Toeplitz systems of equations; centrosymmetric blocks; complexity; iteration; matrices; multiplications; one-dimensional inverse scattering; orthonormal wavelet transform; problem transformation; redundancy; split algorithms; symmetric wavelet basis function; Binary search trees; Computer science; Integral equations; Inverse problems; Kernel; Moment methods; Random processes; Signal processing algorithms; Symmetric matrices; Wavelet transforms;
fLanguage :
English
Journal_Title :
Signal Processing, IEEE Transactions on
Publisher :
ieee
ISSN :
1053-587X
Type :
jour
DOI :
10.1109/78.700975
Filename :
700975
Link To Document :
بازگشت