Abstract of Thesis presented at COPPE/UFRJ as a partial fulfillment of the requirements for the degree of Master of Science (M.Sc.)

Study of Tonals and Their Features to Use in a Classifier of Ships Based on Neural Networks

Marcos Damas Paula

September/2007

Advisor:

Luiz Pereira Calôba

Department:

Eletrical Engineering

      This work presents a study of extracted features of tonals aiming to identify classes of war ships. The tonals were obtained from hydroacustics signals radiated from four differents classes of Brazilian Navy ships. The features added to the rotation speed of the propeller shafts are the input variables for a neural network that performs the classification. It was used tonals which have demonstrated more relevance to identify a contact, being selected by a sonar expert. The results here obtained, point out to the good potencial of this classification using tonals.