Determining class of underwater vehicles in passive sonar using hidden Markov model with Hausdorff similarity measure

The main purpose of this paper is about detection and classification of underwater vehicles (UVs) using features extracted from their acoustic signals. We have proposed an algorithm for the above purpose based on hidden Markov model (HMM) with Hausdorff similarity measure (HSM). The HMM is a proper stochastic model for speech recognition and classification of spoken words. We used this model to recognize UVs acoustic signal among other environmental noise and therefore would be a good candidate to classify some UVs. We considered three classes and simulate their sounds, and constructed three trained HMM with optimal number of states. Measurement of similarity is done by HSM instead Euclidean measure, in train and test section. Robustness and better performance are two promising results. The new method only demands more computation and therefore faster processors.