### Abstract

When a linear mixture of independent sources is contaminated by multiplicative noise, also called speckle noise, the statistic of the outputs of a linear transformation of the noise data is very different from the statistic that appears when the speckle noise is not present. Specifically, it is not possible find a linear transformation that provides independent outputs and it is necessary study the statistical structure that appears in this case. In this paper, a general approach to obtain the mixture when there exists speckle noise is developed. In order to do this, the linear transformation is searches as the one that reproduces the this theoretical statistic structure.

Original language | English |
---|---|

Title of host publication | Independent Component Analysis and Blind Signal Separation |

Subtitle of host publication | Fifth International Conference, ICA 2004, Granada, Spain, September 22-24, 2004. Proceedings |

Editors | Carlos G Puntonet, Alberto Prieto |

Publisher | Springer |

Pages | 248-255 |

Number of pages | 8 |

Volume | 3195 |

ISBN (Electronic) | 978-3-540-30110-3 |

ISBN (Print) | 978-3-540-23056-4 |

DOIs | |

Publication status | Published - 2004 |

Event | 5th International Conference on Independent Component Analysis and Blind Signal Separation - Granada, Spain Duration: 22 Sep 2004 → 24 Sep 2004 |

### Publication series

Name | Lecture Notes in Computer Science |
---|---|

Volume | 3195 |

ISSN (Print) | 0302-9743 |

### Conference

Conference | 5th International Conference on Independent Component Analysis and Blind Signal Separation |
---|---|

Abbreviated title | ICA 2004 |

Country | Spain |

City | Granada |

Period | 22/09/04 → 24/09/04 |

## Cite this

*Independent Component Analysis and Blind Signal Separation: Fifth International Conference, ICA 2004, Granada, Spain, September 22-24, 2004. Proceedings*(Vol. 3195, pp. 248-255). (Lecture Notes in Computer Science; Vol. 3195). Springer. https://doi.org/10.1007/978-3-540-30110-3_32