碩士 / 中原大學 / 資訊工程學系 / 87 / The recognition of underwater acoustic signal from ocean ambient noise is an important job of underwater signal processing. The propagation of underwater sound is affect due to multi-path, reverberation, and inhomogeneity. Besides, the random process and time vary characteristics of the signal cause it difficult to use an effective mathematics model to simulate the system. Hence, there are two emphases about this research. The one is underwater acoustic signal feature parameter extracted by using digital signal processing. The other is optimizing the fuzzy logic recognition system by applying genetic algorithm. Finally, the two parts are integrated to perform signal recognition task and prove that our research can be realized.
During the feature parameter extraction stage, signal characteristic analysis and feature selection is implemented. Upon careful analysis the feature parameters of the signal and practical data are obtained from calculation, the result shows the feature parameters obtaining from the Spectrum can distinguish ship signature effectively.
During the fuzzy logic recognition system modeling, each ship's feature parameters are utilized on to construct the preliminary fuzzy logic recognition system. In the systems, genetic algorithm is further used to fine-tune the membership functions of the if-then inference rules. Hence, the condition parts of the rules can be automatically adjusted to increase the recognition rate. Moreover, to make the recognition system intensively compact, applying genetic algorithm to filter out the insignificant or redundant rules. The pruned compact system not only needs less reasoning time, but also makes the reasoning process simpler and more explainable.
By using object-orient language, a modular recognition modulate system architecture was developed which shorten program develop time and improve system execute performance. An actually realized underwater acoustic signal recognition system was established, the performance of system is simple in operation and fast in response.
| Identifer | oai:union.ndltd.org:TW/087CYCU0392004 |
| Date | January 1999 |
| Creators | 林爭賢 |
| Contributors | 杜筑奎 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 60 |
Page generated in 0.0074 seconds