| Summary: | 碩士 === 中原大學 === 資訊工程學系 === 88 === During propagation of the underwater acoustic signal is affected by ocean interference and environmental noise disturbance, in order to distinguish the weaken signal caused by long distance propagation loss the received signal must be processed properly. There are two emphases about this research. The one is underwater acoustic signal feature parameter extraction by using wavelet packet decomposition. The other is the signal pattern recognition by using of Hidden Markov Model. Finally, combine the two procedures and establish a practical recognition system.
During the feature parameter extraction stage, signal characteristic analysis and feature selection is discussed. It has been proved that using the wavelet packet decomposition method for feature parameter selective can obtain multi-resolution characteristics. Therefore, the feature parameters obtain by above maintain method can distinguish the different characteristics of ships. Besides, use vector quantization to clustering data, and find the characteristics of data gathering can get representative pattern feature parameters of each sample classification individually.
During the recognition system modeling, use Hidden Markov Model theory to establish recognition system. The organization of the system includes two parts, in first part the stochastic probability process is used for statistical modeling of the underwater acoustic signal, in second part, the Viterbi algorithm is used to find the best recognition result.
|
|---|
