Complex matrices

Complex matrices

[Dr. David Kirkby]

Hi,
	I'm trying to solve an equation of the form:
[V]=[I].[Z], where V is a vector that is known, Z is square matrix that is known
and I is unknown. V, I and Z are all complex. 

I can't seem to find any routines in gsl for complex matrices - plenty for
complex numbers and plenty for matrices. Am I missing something?? It seems like
quite a standard thing. 

My square matrix Z is symmetrical (i.e. Zxy=Zyx). Can I use this to my
advantage? The matrix is not sparse. 
-- 
Dr. David Kirkby PhD,
email: drkirkby@ntlworld.com 
web page: http://www.david-kirkby.co.uk/
Amateur radio callsign: G8WRB

Complex matrices

[Dr. David Kirkby]

Hi,
	I'm trying to solve an equation of the form:
[V]=[I].[Z], where V is a vector that is known, Z is square matrix that is known
and I is unknown. V, I and Z are all complex. 

I can't seem to find any routines in gsl for complex matrices - plenty for
complex numbers and plenty for matrices. Am I missing something?? It seems like
quite a standard thing. 

My square matrix Z is symmetrical (i.e. Zxy=Zyx). Can I use this to my
advantage? The matrix is not sparse. 
-- 
Dr. David Kirkby PhD,
email: drkirkby@ntlworld.com 
web page: http://www.david-kirkby.co.uk/
Amateur radio callsign: G8WRB

Re: Complex matrices

[Eugene Eremin]

----- Original Message -----
From: "Dr. David Kirkby" <drkirkby@ntlworld.com>
To: "gsl discussion list" <gsl-discuss@sources.redhat.com>
Sent: Wednesday, March 06, 2002 1:08 AM
Subject: Complex matrices


> Hi,
> I'm trying to solve an equation of the form:
> [V]=[I].[Z], where V is a vector that is known, Z is square matrix that is
known
> and I is unknown. V, I and Z are all complex.
>
> I can't seem to find any routines in gsl for complex matrices - plenty for
> complex numbers and plenty for matrices. Am I missing something?? It seems
like
> quite a standard thing.
>
> My square matrix Z is symmetrical (i.e. Zxy=Zyx). Can I use this to my
> advantage? The matrix is not sparse.
> --
> Dr. David Kirkby PhD,
> email: drkirkby@ntlworld.com
> web page: http://www.david-kirkby.co.uk/
> Amateur radio callsign: G8WRB
>

Try this

Ax=b

//allocate memory for matrix

gsl_matrix_complex* A_matrix = gsl_matrix_complex_alloc(size,size);

gsl_vector_complex* B_vector = gsl_vector_complex_alloc(size);

gsl_vector_complex* X_vector = gsl_vector_complex_alloc(size);



//initilaze matrix A

gsl_matrix_complex_set(A_matrix,0,0 .............................



//solve system the LU methods

int s;

gsl_permutation * p = gsl_permutation_alloc (size);

gsl_linalg_complex_LU_decomp (A_matrix, p, &s);

gsl_linalg_complex_LU_solve (A_matrix, p, B_vector, X_vector);

//free memory after solve system

gsl_permutation_free(p);

Re: Complex matrices

[Eugene Eremin]

----- Original Message -----
From: "Dr. David Kirkby" <drkirkby@ntlworld.com>
To: "gsl discussion list" <gsl-discuss@sources.redhat.com>
Sent: Wednesday, March 06, 2002 1:08 AM
Subject: Complex matrices


> Hi,
> I'm trying to solve an equation of the form:
> [V]=[I].[Z], where V is a vector that is known, Z is square matrix that is
known
> and I is unknown. V, I and Z are all complex.
>
> I can't seem to find any routines in gsl for complex matrices - plenty for
> complex numbers and plenty for matrices. Am I missing something?? It seems
like
> quite a standard thing.
>
> My square matrix Z is symmetrical (i.e. Zxy=Zyx). Can I use this to my
> advantage? The matrix is not sparse.
> --
> Dr. David Kirkby PhD,
> email: drkirkby@ntlworld.com
> web page: http://www.david-kirkby.co.uk/
> Amateur radio callsign: G8WRB
>

Try this

Ax=b

//allocate memory for matrix

gsl_matrix_complex* A_matrix = gsl_matrix_complex_alloc(size,size);

gsl_vector_complex* B_vector = gsl_vector_complex_alloc(size);

gsl_vector_complex* X_vector = gsl_vector_complex_alloc(size);



//initilaze matrix A

gsl_matrix_complex_set(A_matrix,0,0 .............................



//solve system the LU methods

int s;

gsl_permutation * p = gsl_permutation_alloc (size);

gsl_linalg_complex_LU_decomp (A_matrix, p, &s);

gsl_linalg_complex_LU_solve (A_matrix, p, B_vector, X_vector);

//free memory after solve system

gsl_permutation_free(p);

complex matrices

[Travis R Miller]

Your manual covers none of the functions usable for complex matrices. 
Either how one sets their values or performs various operations on
them.  There is a hint of a function for performing LU decomposition on
such a matrix in the linear algebra chapter, but there is nothing more. 
Some help would be great.  Here is a bit of code that returns a
segmentation fault.  Perhaps you can explain what is wrong

#include<stdio.h>
#include<stdlib.h>
#include<math.h>
#include<gsl/gsl_rng.h>
#include<gsl/gsl_matrix_complex_double.h>

int main(void)
{
  const gsl_rng_type *T;
  gsl_rng *r;
  gsl_matrix_complex *m;
  gsl_complex x;
  int i, j;
  double u;

  gsl_matrix_complex_alloc(4,4);

  gsl_rng_env_setup();
  T = gsl_rng_default;
  r = gsl_rng_alloc(T);

  for(i=0; i<4; i++) {
    for(j=0; j<4; j++) {
      u = gsl_rng_uniform(r);
      printf("%.5f\n", u);
      x.dat[0] = u;
      u = gsl_rng_uniform(r);
      printf("%.5f\n", u);
      x.dat[1] = u;
      gsl_matrix_complex_set(m, i, j, x);
    }
  }

  gsl_rng_free(r);

  return 1;
}



Thank you for your time,
Travis

Re: complex matrices

[Brian Gough]

Travis R Miller writes:
 > Your manual covers none of the functions usable for complex matrices. 
 > Either how one sets their values or performs various operations on
 > them.  There is a hint of a function for performing LU decomposition on
 > such a matrix in the linear algebra chapter, but there is nothing more. 
 > Some help would be great.  Here is a bit of code that returns a
 > segmentation fault.  Perhaps you can explain what is wrong

Hi,

The chapter "Debugging Numerical Programs" in the GSL manual should
help you here.

regards
Brian Gough