Underwater signal classification has been an area of considerable importance due to its applications in multidimensional fields. The selection of the source specific features in a classifier is very significant, as it determines the efficiency and performance of the classifier. Discrete sine transform (DST)-based features possesses the essential traits suitable for the design of statistical models in underwater signal classifiers. These when incorporated in hidden markov models (HMMs), can provide an effective architecture which can be utilized in the classification of underwater noise sources. The design and performance analysis of a 12-state HMM-based classifier for underwater signals in Rayleigh fading channel conditions are presented in this paper. The HMMs utilizing the DST features are found to perform efficiently in underwater signal classification scenario, compared to existing cepstral feature-based classifiers. The fading channel estimation has been carried out and the classifier performance has been improved by providing Rayleigh fading compensation. The success rates of the classifier has been estimated under different operating conditions. The system performance has been analyzed in MATLABTM platform for the class of underwater signals, which include actual field collected data and the results have been presented in this paper.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.