Application of the conjugate gradient and steepest descent for computing the eigenvalues of an operator

Tapan K. Sarkar, Xiaopu Yang

Research output: Contribution to journalArticle

20 Scopus citations


This paper describes the method of steepest descent and the method of conjugate gradient for iteratively finding the first few large/small eigenvalues and eigenvectors of a Hermitian operator. Both the methods have been applied for the computation of the prolate spheroidal functions. Since the methods are iterative, it is expected to yield accurate solutions for the first few large/small eigenvalues particularly when the condition number (ratio of the largest to the smallest eigenvalue) is large.

Original languageEnglish (US)
Pages (from-to)31-38
Number of pages8
JournalSignal Processing
Issue number1
StatePublished - May 1989



  • Hermetian operators
  • conjugate gradient
  • steepest descent

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering

Cite this