Scilab Website | Contribute with GitLab | Mailing list archives | ATOMS toolboxes
Scilab Online Help
2025.0.0 - English


polar

polar decomposition of a square matrix

Syntax

[Rho, Theta] = polar(A)

Arguments

A

a real or complex square matrix.

Rho

a real matrix.

Theta

a real or complex matrix.

Description

[Rho, Theta] = polar(A) returns the polar form of A i.e. A = Rho * expm(%i*Theta) with Rho symmetric ≥ 0 and Theta hermitian ≥ 0.

Examples

// Build an input matrix A:
Theta_0 = toeplitz(1.5:-0.5:0)
Rho_0 = toeplitz(1:4)
A = Rho_0 * expm(%i*Theta_0)

// Compute its polar matricial components:
[Rho, Theta] = polar(A);
clean(Rho)    // is real symetric
clean(Theta)  // is hermitian

// Check that the computed decomposition builds A as well
norm(A - Rho*expm(%i*Theta), 1)
--> Theta_0 = toeplitz(1.5:-0.5:0)
 Theta_0  =
   1.5   1.    0.5   0.
   1.    1.5   1.    0.5
   0.5   1.    1.5   1.
   0.    0.5   1.    1.5

--> Rho_0 = toeplitz(1:4)
 Rho_0  =
   1.   2.   3.   4.
   2.   1.   2.   3.
   3.   2.   1.   2.
   4.   3.   2.   1.

--> A = Rho_0 * expm(%i*Theta_0)
 A  =
  -1.2699509 - 2.1374364i  -2.6833779 - 1.9074687i  -3.0456968 - 0.7713555i  -1.5838266 + 0.7745137i
  -0.3831402 - 0.8547475i  -2.7866925 - 1.5182796i  -2.3759306 - 0.8785558i  -0.745459  + 0.2813657i
  -0.745459  + 0.2813657i  -2.3759306 - 0.8785558i  -2.7866925 - 1.5182796i  -0.3831402 - 0.8547475i
  -1.5838266 + 0.7745137i  -3.0456968 - 0.7713555i  -2.6833779 - 1.9074687i  -1.2699509 - 2.1374364i


--> // Compute its polar matricial components:
--> [Rho, Theta] = polar(A);

--> clean(Rho)    // is real symetric
 ans  =
   4.441742  + 0.i   2.4611614 + 0.i   1.4611614 + 0.i   1.441742  + 0.i
   2.4611614 + 0.i   2.6572775 + 0.i   1.6572775 + 0.i   1.4611614 + 0.i
   1.4611614 + 0.i   1.6572775 + 0.i   2.6572775 + 0.i   2.4611614 + 0.i
   1.441742  + 0.i   1.4611614 + 0.i   2.4611614 + 0.i   4.441742  + 0.i

--> clean(Theta)  // is hermitian
 ans  =
  -0.6115539 + 0.i          0.2232029 + 1.3108917i  -0.2767971 + 1.3108917i   1.0300387 + 0.i
   0.2232029 - 1.3108917i  -1.1008351 + 0.i          1.5407576 + 0.i         -0.2767971 - 1.3108917i
  -0.2767971 - 1.3108917i   1.5407576 + 0.i         -1.1008351 + 0.i          0.2232029 - 1.3108917i
   1.0300387 + 0.i         -0.2767971 + 1.3108917i   0.2232029 + 1.3108917i  -0.6115539 + 0.i

--> // Check that the computed decomposition builds A as well
--> norm(A - Rho*expm(%i*Theta), 1)
 ans  =
   1.435D-14

See also

  • expm — square matrix exponential
  • svd — singular value decomposition
Report an issue
<< lu Factorization qr >>

Copyright (c) 2022-2024 (Dassault Systèmes)
Copyright (c) 2017-2022 (ESI Group)
Copyright (c) 2011-2017 (Scilab Enterprises)
Copyright (c) 1989-2012 (INRIA)
Copyright (c) 1989-2007 (ENPC)
with contributors
Last updated:
Thu Oct 24 11:13:08 CEST 2024