Blind separation of audio signal mixtures
The separation of two unknown signals that mixed together in an unknown manner was shown to be possible. This process is called blind source separation. Two types of mixing process, which are instantaneous and convolutive mixing were simulated. It was found that by assuming that the two known signal...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Published: |
2008
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/4948 |
| _version_ | 1826125044146765824 |
|---|---|
| author | Ng, Hock Seng. |
| author2 | Wan, Chunru |
| author_facet | Wan, Chunru Ng, Hock Seng. |
| author_sort | Ng, Hock Seng. |
| collection | NTU |
| description | The separation of two unknown signals that mixed together in an unknown manner was shown to be possible. This process is called blind source separation. Two types of mixing process, which are instantaneous and convolutive mixing were simulated. It was found that by assuming that the two known signals and their signal mixtures were uncorrelated, the two original signals could be recovered. The signals undergoing instantaneous mixing can be easily separated by eigen-decomposition method processing. The signals undergoing convolutive mixing, however need a more robust separation algorithm. They still can be separated using gradient-based optimisation algorithm and simulated annealing algorithm, which generates the solution signals set that correspond to the least cross-correlation value. |
| first_indexed | 2024-10-01T06:30:18Z |
| format | Thesis |
| id | ntu-10356/4948 |
| institution | Nanyang Technological University |
| last_indexed | 2024-10-01T06:30:18Z |
| publishDate | 2008 |
| record_format | dspace |
| spelling | ntu-10356/49482023-07-04T15:49:28Z Blind separation of audio signal mixtures Ng, Hock Seng. Wan, Chunru School of Electrical and Electronic Engineering DRNTU::Engineering::Electrical and electronic engineering::Electronic systems::Signal processing The separation of two unknown signals that mixed together in an unknown manner was shown to be possible. This process is called blind source separation. Two types of mixing process, which are instantaneous and convolutive mixing were simulated. It was found that by assuming that the two known signals and their signal mixtures were uncorrelated, the two original signals could be recovered. The signals undergoing instantaneous mixing can be easily separated by eigen-decomposition method processing. The signals undergoing convolutive mixing, however need a more robust separation algorithm. They still can be separated using gradient-based optimisation algorithm and simulated annealing algorithm, which generates the solution signals set that correspond to the least cross-correlation value. Master of Science (Communication and Network Systems) 2008-09-17T10:02:02Z 2008-09-17T10:02:02Z 2000 2000 Thesis http://hdl.handle.net/10356/4948 Nanyang Technological University application/pdf |
| spellingShingle | DRNTU::Engineering::Electrical and electronic engineering::Electronic systems::Signal processing Ng, Hock Seng. Blind separation of audio signal mixtures |
| title | Blind separation of audio signal mixtures |
| title_full | Blind separation of audio signal mixtures |
| title_fullStr | Blind separation of audio signal mixtures |
| title_full_unstemmed | Blind separation of audio signal mixtures |
| title_short | Blind separation of audio signal mixtures |
| title_sort | blind separation of audio signal mixtures |
| topic | DRNTU::Engineering::Electrical and electronic engineering::Electronic systems::Signal processing |
| url | http://hdl.handle.net/10356/4948 |
| work_keys_str_mv | AT nghockseng blindseparationofaudiosignalmixtures |