Divide and conquer algorithm for computer simulation and application in the matrix eigenvalue problem

Author

Cui, Yuhuan ; Qu, Jingguo ; Chen, Weili ; Yang, Aimin

Author_Institution

Coll. of Light Ind., Hebei Polytech. Univ., Tangshan, China

Volume

1

fYear

2009

fDate

5-6 Dec. 2009

Firstpage

319

Lastpage

322

Abstract

DC is a divide and conquers method for short. Its basic idea is to divide the given real symmetric tridiagonal matrix A into two lower-order submatrix A1 , A2 first, then glue the eigenvalues of the two submatrixes together and find the eigenvalue and eigenvector of the original symmetric triangular matrix by means of some calculations. It is that we solve the submatrix eigenvalue problem, and then pose a question easy to answer than the original eigenvalue problem. Finally we uses computer to simulate, using mathematical software programming to solve the eigenvalues of each submatrix separately.

Keywords

digital simulation; divide and conquer methods; eigenvalues and eigenfunctions; mathematics computing; matrix algebra; parallel algorithms; computer simulation; divide and conquer algorithm; mathematical software programming; submatrix eigenvalue problem; symmetric tridiagonal matrix; Application software; Computer aided instruction; Computer networks; Computer simulation; Concurrent computing; Distributed computing; Eigenvalues and eigenfunctions; Jacobian matrices; Matrix decomposition; Symmetric matrices; divide and conquer; eigenvalue; parallel algorithm; the divide and conquer method based on dichotomy method;

fLanguage

English

Publisher

ieee

Conference_Titel

Test and Measurement, 2009. ICTM '09. International Conference on

Conference_Location

Hong Kong

Print_ISBN

978-1-4244-4699-5

Type

conf

DOI

10.1109/ICTM.2009.5412930

Filename

5412930