Google
 

Trailing-Edge - PDP-10 Archives - decus_20tap2_198111 - decus/20-0026/eigen.doc
There are 2 other files named eigen.doc in the archive. Click here to see a list.
SUBROUTINE EIGEN

PURPOSE
   COMPUTE EIGENVALUES AND EIGENVECTORS OF A REAL SYMMETRIC
   MATRIX

USAGE
   CALL EIGEN(A,R,N,MV)

DESCRIPTION OF PARAMETERS
   A - ORIGINAL MATRIX (SYMMETRIC), DESTROYED IN COMPUTATION.
       RESULTANT EIGENVALUES ARE DEVELOPED IN DIAGONAL OF
       MATRIX A IN DESCENDING ORDER.
   R - RESULTANT MATRIX OF EIGENVECTORS (STORED COLUMNWISE,
       IN SAME SEQUENCE AS EIGENVALUES)
   N - ORDER OF MATRICES A AND R
   MV- INPUT CODE
	   0   COMPUTE EIGENVALUES AND EIGENVECTORS
	   1   COMPUTE EIGENVALUES ONLY (R NEED NOT BE
	       DIMENSIONED BUT MUST STILL APPEAR IN CALLING
	       SEQUENCE)

REMARKS
   ORIGINAL MATRIX A MUST BE REAL SYMMETRIC (STORAGE MODE=1)
   MATRIX A CANNOT BE IN THE SAME LOCATION AS MATRIX R

SUBROUTINES AND FUNCTION SUBPROGRAMS REQUIRED
   NONE

METHOD
   DIAGONALIZATION METHOD ORIGINATED BY JACOBI AND ADAPTED
   BY VON NEUMANN FOR LARGE COMPUTERS AS FOUND IN 'MATHEMATICAL
   METHODS FOR DIGITAL COMPUTERS', EDITED BY A. RALSTON AND
   H.S. WILF, JOHN WILEY AND SONS, NEW YORK, 1962, CHAPTER 7