| bibtype |
I -
Internal Report
|
| ARLID |
0411386 |
| utime |
20240103182323.7 |
| mtime |
20060210235959.9 |
| title
(primary) (eng) |
Performance Analysis of the FastICA Algorithm and Cramé-Rao Bounds for Linear Independent Component Analysis |
| publisher |
| place |
Praha |
| name |
ÚTIA AV ČR |
| pub_time |
2005 |
|
| specification |
|
| edition |
| part_name |
Interní publikace DAR - ÚTIA. |
| volume_id |
2005/6 |
|
| keyword |
blind source separation |
| keyword |
independent component analysis |
| keyword |
Cramér-Rao lower bound |
| author
(primary) |
| ARLID |
cav_un_auth*0101212 |
| name1 |
Tichavský |
| name2 |
Petr |
| institution |
UTIA-B |
| full_dept |
Department of Stochastic Informatics |
| fullinstit |
Ústav teorie informace a automatizace AV ČR, v. v. i. |
|
| author
|
| ARLID |
cav_un_auth*0108100 |
| name1 |
Koldovský |
| name2 |
Zbyněk |
| institution |
UTIA-B |
| full_dept |
Department of Stochastic Informatics |
| fullinstit |
Ústav teorie informace a automatizace AV ČR, v. v. i. |
|
| author
|
| ARLID |
cav_un_auth*0213223 |
| name1 |
Oja |
| name2 |
E. |
| country |
FI |
|
| COSATI |
12B |
| cas_special |
| project |
| project_id |
1M0572 |
| agency |
GA MŠk |
| ARLID |
cav_un_auth*0001814 |
|
| research |
CEZ:AV0Z10750506 |
| abstract
(eng) |
The FastICA or fixed point algorithm is one of the most successful algorithms for linear independent component analysis (ICA) in terms of accuracy and computational complexity. The main result of this paper are (1) analytic closed form expressions that characterize the separating ability of the algorithm in a local sense, assuming a good initialization and (2) derivation ofthe Cramer-Rao bound for the estimation problem, used as algorithm independent bound on the separation quality. |
| abstract
(cze) |
Algoritmus FastICA je jeden z dosud nejvíce používaných algoritmů pro analýzu nezávislých komponent. V článku jsou odvozeny teoretické výrazy charakterizující schopnost separace nezávislých komponent tímto algoritmem v závislosti na použité variantě algoritmu, na použité volitelné nelineární funkci a na pravděpodobnostním rozložení odhadovaných komponent. Analýza umožnuje optimálně kombinovat obě varianty algoritmu a je ověřena na simulacích. Dále je v práci odvozena Rao-Cramerova mez pro kvalitu separace. |
| RIV |
BB |
| reportyear |
2006 |
| department |
SI |
| permalink |
http://hdl.handle.net/11104/0131468 |
| ID_orig |
UTIA-B 20050116 |
| arlyear |
2005 |
| mrcbU10 |
2005 |
| mrcbU10 |
Praha ÚTIA AV ČR |
|