#include <PseudoTriCovarianceLS.hpp>
Inheritance diagram for PseudoTriCovarianceLS:
Purpose: - approximated prewhitening for FAST ICA - approximated sphering for multi dimensional gaussian classification Proto: Bernard De Cuyper 12/2003 proposal: Pseudo covariance Tri diagonal solver Used to approximate covariance with local elements U(unknown)= Cov-1(unknown) * W(known) is equivalent to Cov(known) * U(unknown) = W(known if covariance have only local correlations, a tridiagonal solver can give solutions in O(N) using no inversion. Basic Memory Usage: 9*N
@ Copyrights: Bernard De Cuyper & Eddy Fraiha 2003, Eggs & Pictures. MIT/Open BSD copyright model.
Public Methods | |
PseudoTriCovarianceLS (int asize) | |
virtual | ~PseudoTriCovarianceLS () |
virtual void | loadA (RTriDiagonalMatrix *m) |
virtual void | loadMatrix (RealVector *u1, RealVector *d1, RealVector *l1) |
virtual void | loadX (RealVector *data) |
virtual void | loadB (RealVector *data) |
virtual bool | solve () |
virtual void | output () |
virtual void | output (FILE *file) |
Protected Methods | |
void | computeLRdecomposition () |
Equivalent to filter parametrisation. | |
void | forwardSubstitution () |
Equivalent to filter causal part. | |
void | backwardSubstitution () |
Equivalent to filter non-causal part. | |
Private Attributes | |
RealVector * | uOut |
RealVector * | dOut |
RealVector * | lOut |
RealVector * | y |