State Transition Matrix for an HMM based underwater target classifier

The problem of identification of noise sources in the ocean is of prime importance because of its diverse practical applications. Hidden Markov Models provide an effective architecture for the classification of underwater noise sources. A technique for the estimation of State Transition Matrix for a twenty state Hidden Markov Model for the classification of noise sources in the ocean is presented in this paper. The results of classification studies using the noise data waveforms of certain marine species, utilising the state transition matrix approach is also presented.