Publication: Blind source separation (bss) of underwater acoustic signal
| datacite.subject.fos | oecd::Engineering and technology::Electrical engineering, Electronic engineering, Information engineering | |
| dc.contributor.author | Azhar, Ahmad Saiful | |
| dc.date.accessioned | 2024-02-13T09:24:43Z | |
| dc.date.available | 2024-02-13T09:24:43Z | |
| dc.date.issued | 2020-08-01 | |
| dc.description.abstract | The study of underwater acoustic signal has become a great contribution to the new technology, especially for scientist, biologist and military. This study has a very big potential in which the signal separations could be used to spot the specific target under the sea that has the enormous size of dimensions such as identifying an underwater object or location of swarm of fish. However, two main problems that have been facing by the human when tracking the underwater object are incapability of human to keep monitor 24-hour and incapability of human to detecting something that too far or too deep from our capable senses. Thus, Blind Source Separation (BSS) method is proposed for the implementation of fish sound detection and ship sound detection. This method can help humans to solve their problems related to identifying underwater objects. The FastICA with kurtosis and FastICA with negentropy will be the key for separation of underwater acoustic signal. As the result, the Signal-to-Distortion ratio (SDR), Signal-to-Artifacts ratio (SAR) and Signal-to-Interference ratio (SIR) measurement would be used to compare the effectiveness between two approaches which is FastICA kurtosis or FastICA negentropy. In conclusion, the FastIca negentropy is more capable to separate the higher number of signals compare to FastIca kurtosis. This based on SIR measurement results shows that FastIca negentropy scored average 31.339 dB compare to FastIca kurtosis which scored average 26.148 dB only. Higher average scored shows that the separation is more success. | |
| dc.identifier.uri | https://erepo.usm.my/handle/123456789/18333 | |
| dc.language.iso | en | |
| dc.title | Blind source separation (bss) of underwater acoustic signal | |
| dc.type | Resource Types::text::report | |
| dspace.entity.type | Publication | |
| oairecerif.author.affiliation | Universiti Sains Malaysia |