Re: [Help-gsl] Solving Ax=B where A is a complex, singular matrix

[Brian Gough]

At Tue, 28 Jun 2011 00:26:33 -0600, Benjamin Blumer wrote:
> My question: My program needs to solve the matrix equation Ax=B for
> a long list of complex, square matrices A.  Using the complex LU
> decomp/solve works great for most of them, but some of the A are
> singular.  It appears to me that GSL doesn't have an SVD or QR
> decomposition that works for complex matrices.  The matrix,
> unfortunately, isn't guaranteed to be positive definite so that
> eliminates the Cholesky decomp/solve pair (right?).  Does the
> library contain anything to solve the equation?

Thanks for your email.  I can confirm that the library is lacking the
complex SVD and QR unfortunately.  If anyone can write either I will
gladly include them.