| bibtype |
J -
Journal Article
|
| ARLID |
0321670 |
| utime |
20240111140715.9 |
| mtime |
20090325235959.9 |
| WOS |
000263431900006 |
| DOI |
10.1109/TSP.2008.2009271 |
| title
(primary) (eng) |
Fast Approximate Joint Diagonalization Incorporating Weight Matrices |
| specification |
| page_count |
14 s. |
| media_type |
www |
|
| serial |
| ARLID |
cav_un_epca*0256727 |
| ISSN |
1053-587X |
| title
|
IEEE Transactions on Signal Processing |
| volume_id |
57 |
| volume |
3 (2009) |
| page_num |
878-891 |
|
| title
(cze) |
Rychlá přibližná diagonalizace s váhovými maticemi |
| keyword |
autoregressive processes |
| keyword |
blind source separation |
| keyword |
nonstationary random processes |
| 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*0213973 |
| name1 |
Yeredor |
| name2 |
A. |
| country |
IL |
|
| source |
|
| cas_special |
| project |
| project_id |
1M0572 |
| agency |
GA MŠk |
| ARLID |
cav_un_auth*0001814 |
|
| research |
CEZ:AV0Z10750506 |
| abstract
(eng) |
We propose a new low complexity Approximate Joint Diagonalization (AJD) algorithm, which incorporates nontrivial block-diagonal weight matrices into a Weighted Least-Squares (WLS) AJD criterion. We show how the new algorithm can be utilized in an iteratively-reweighted separation scheme, thereby giving rise to fast implementation of asymptotically optimal BSS algorithms in various scenarios. In particular, we consider three specific (yet common) scenarios, involving stationary or block-stationary Gaussian sources, for which the optimal weight matrices can be readily estimated from the sample covariance matrices (which are also the target-matrices for the AJD). Comparative simulation results demonstrate the advantages in both speed and accuracy, as well as compliance with the theoretically predicted asymptotic optimality of the resulting BSS algorithms based on the weighted AJD, both on large scale problems with matrices of the size 100 x 100. |
| abstract
(cze) |
V práci je navržena nová metoda přibližné vzájemné diagonalizace souboru matic, která obsahuje netriviální váhové matice, jimiž se nastavuje fungování algoritmu. Algoritmus má velmi nízkou výpočetní složitost. Je ukázáno iterativní použití algoritmu s adaptivním odhadováním váhových matic, a to při slepé separaci signálů ve třech různých variantách: separace nezávislých autoregresních procesů a separace po blocích stacionárních autoregresních procesů s řádem 1 nebo vyšším. Ve všech případech je tak možné získat asymptoticky eficientní odhady. Algoritmus lze použít na diagonalizaci velkých matic, např. velikosti 100x100. |
| reportyear |
2009 |
| RIV |
BB |
| permalink |
http://hdl.handle.net/11104/0170139 |
| mrcbT16-f |
2.954 |
| mrcbT16-g |
0.316 |
| mrcbT16-h |
7.2 |
| mrcbT16-i |
0.04685 |
| mrcbT16-j |
1.039 |
| mrcbT16-k |
15879 |
| mrcbT16-l |
431 |
| mrcbT16-q |
151 |
| mrcbT16-s |
3.059 |
| mrcbT16-y |
29.5 |
| mrcbT16-x |
3.31 |
| arlyear |
2009 |
| mrcbU34 |
000263431900006 WOS |
| mrcbU56 |
pdf |
| mrcbU63 |
cav_un_epca*0256727 IEEE Transactions on Signal Processing 1053-587X 1941-0476 Roč. 57 č. 3 2009 878 891 |
|