scientific article; zbMATH DE number 1140234

MaRDI QIDQ4385253zbMATH OpenFDO

Authors Dipa Choudhury, Roger A. Horn

Publication date 26 November 1998

zbMATH Keywords

complex bilinear formscomplex orthogonal matricesGram-Schmidt algorithmHermitian formisotropic vectorrectanormal set

Mathematics Subject Classification ID

Canonical forms, reductions, classification (15A21) Quadratic and bilinear forms, inner products (15A63) General theory of numerical methods in complex analysis (potential theory, etc.) (65E05) Orthogonalization in numerical linear algebra (65F25)

Review text

A set of vectors \(\{x_i\} \in\mathbb{C}^n\) is rectangular if \(x^T_i x_j=0\) whenever \(i \neq j\); it is rectanormal if it is rectangular and \(x^T_i x_i=1\) for all \(i\). Here, the superscript \(T\) denotes the transpose. In fact, these notions are analogues of orthogonal and orthonormal sets. In this paper, analogues of the Gram-Schmidt algorithm and some of their consequences are presented. Some interesting results involving complex orthogonal matrices are proved. As there are fundamental differences between the bilinear form and the Hermitian form, further investigations of rectangular sets are valuable.

Recommendations

• Birkhoff-James \(\varepsilon\)-orthogonality sets of vectors and vector-valued polynomials
• Iterative algorithms for Gram-Schmidt orthogonalization
• Birkhoff-James -orthogonality sets in normed linear spaces
• The Gram-Schmidt algorithm for a general angle
• Extension of an approximate orthogonalization algorithm to arbitrary rectangular matrices

Cited in

(1)

• Anti-symmetrical linear change and Gram-Schmidt orthogonalization process

This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4385253)