An SVD-like matrix decomposition and its applications (Q1399229)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | An SVD-like matrix decomposition and its applications |
scientific article |
Statements
An SVD-like matrix decomposition and its applications (English)
0 references
30 July 2003
0 references
A matrix \(S\in {\mathbb C}^{2m\times 2m}\) is symplectic if \(SJS^\star=J\), where \(J=\left( \begin{smallmatrix} 0&I_m\\ -I_m&0\end{smallmatrix}\right)\) and \({\mathbb C}\) denotes the complex numbers. The author obtains (i) and (ii) below and other similar results. (i) If \(B\in{\mathbb R}^{n\times 2m}\), then \(B=QDS^{-1}\), where \(Q\) is real orthogonal, \(S\) is real symplectic, \(D\) is permuted diagonal, and \({\mathbb R}\) denotes the real numbers. (ii) If \(S\in{\mathbb C}^{2m\times 2m}\), then \(S=UDV^\star\), where \(U,V\) are unitary and symplectic, respectively, \(D=\text{diag} (\Omega,\Omega^{-1})\), and \(\Omega\) is positive diagonal. It is known that every real skew-symmetric matrix \(K\in{\mathbb R}^{n\times n}\) has a factorization \(BJB^T\) for some real matrix \(B\). The author applies his results to the \(BJB^T\) factorization and studies numerical applications.
0 references
skew-symmetric matrix
0 references
symplectic matrix
0 references
orthogonal (unitary) symplectic matrix
0 references
Hamiltonian matrix
0 references
eigenvalue problem
0 references
singular value decomposition
0 references
factorization
0 references
Schur form
0 references
Jordan canonical form
0 references