Very fast blind source separation by signal to noise ratio based stopping threshold for the SHIBBS/SJAD algorithm
This paper works on joint approximate diagonalization of simplified fourth order cumulant matrices for very fast and large scale blind separation of instantaneous mixing model sources. The JADE algorithm is widely accepted but only limited to small scale separation tasks. The SHIBBS algorithm calcul...
| Main Authors: | , , |
|---|---|
| Format: | Journal Article |
| Published: |
ELSEVIER
2010
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/20.500.11937/17428 |
| _version_ | 1848749464018223104 |
|---|---|
| author | Liu, Xianhua Cardoso, J. Randall, R. |
| author_facet | Liu, Xianhua Cardoso, J. Randall, R. |
| author_sort | Liu, Xianhua |
| building | Curtin Institutional Repository |
| collection | Online Access |
| description | This paper works on joint approximate diagonalization of simplified fourth order cumulant matrices for very fast and large scale blind separation of instantaneous mixing model sources. The JADE algorithm is widely accepted but only limited to small scale separation tasks. The SHIBBS algorithm calculates a fraction of the fourth order cumulant set and avoids eigenmatrix decomposition to reduce calculation cost. However, it was seen to be slower than JADE at the time of its first publication and is hence less known. On the other hand, the SJAD algorithm using the same approach is shown to be very fast. This paper studies the iteration convergence criterion and proposes to use a signal to noise ratio based iteration stopping threshold approach. The improved SHIBBS/SJAD algorithm is very fast, and capable of large scale separation. Experimental separation comparisons between the SHIBBS/SJAD and FastICA are presented. |
| first_indexed | 2025-11-14T07:21:21Z |
| format | Journal Article |
| id | curtin-20.500.11937-17428 |
| institution | Curtin University Malaysia |
| institution_category | Local University |
| last_indexed | 2025-11-14T07:21:21Z |
| publishDate | 2010 |
| publisher | ELSEVIER |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | curtin-20.500.11937-174282017-09-13T15:45:17Z Very fast blind source separation by signal to noise ratio based stopping threshold for the SHIBBS/SJAD algorithm Liu, Xianhua Cardoso, J. Randall, R. Independent component analysis Blind source separation Signal to noise ratio Joint approximate diagonalization Fourth order cumulant matrices This paper works on joint approximate diagonalization of simplified fourth order cumulant matrices for very fast and large scale blind separation of instantaneous mixing model sources. The JADE algorithm is widely accepted but only limited to small scale separation tasks. The SHIBBS algorithm calculates a fraction of the fourth order cumulant set and avoids eigenmatrix decomposition to reduce calculation cost. However, it was seen to be slower than JADE at the time of its first publication and is hence less known. On the other hand, the SJAD algorithm using the same approach is shown to be very fast. This paper studies the iteration convergence criterion and proposes to use a signal to noise ratio based iteration stopping threshold approach. The improved SHIBBS/SJAD algorithm is very fast, and capable of large scale separation. Experimental separation comparisons between the SHIBBS/SJAD and FastICA are presented. 2010 Journal Article http://hdl.handle.net/20.500.11937/17428 10.1016/j.ymssp.2010.03.009 ELSEVIER restricted |
| spellingShingle | Independent component analysis Blind source separation Signal to noise ratio Joint approximate diagonalization Fourth order cumulant matrices Liu, Xianhua Cardoso, J. Randall, R. Very fast blind source separation by signal to noise ratio based stopping threshold for the SHIBBS/SJAD algorithm |
| title | Very fast blind source separation by signal to noise ratio based stopping threshold for the SHIBBS/SJAD algorithm |
| title_full | Very fast blind source separation by signal to noise ratio based stopping threshold for the SHIBBS/SJAD algorithm |
| title_fullStr | Very fast blind source separation by signal to noise ratio based stopping threshold for the SHIBBS/SJAD algorithm |
| title_full_unstemmed | Very fast blind source separation by signal to noise ratio based stopping threshold for the SHIBBS/SJAD algorithm |
| title_short | Very fast blind source separation by signal to noise ratio based stopping threshold for the SHIBBS/SJAD algorithm |
| title_sort | very fast blind source separation by signal to noise ratio based stopping threshold for the shibbs/sjad algorithm |
| topic | Independent component analysis Blind source separation Signal to noise ratio Joint approximate diagonalization Fourth order cumulant matrices |
| url | http://hdl.handle.net/20.500.11937/17428 |