Acoustic classification of submerged targets

Broadband signal processing techniques have introduced new possibilities in acoustic classification through detailed examination of the target´s form function (backscattered pressure vs frequency). The form function is unique for a body of given size, shape and composition and correlation with analytically or empirically obtained form function signatures can be accomplished with present computer capabilities. The relationship between the free body resonances of simple targets and thier signatures have been used to extract material constants. However, the most significant contribution of the resonance formalism, the separation of the signature of an elastic body into resonance and geometric contributions, has yet to be exploited. Simple targets, both hollow and solid, are used to illustrate an experimental technique for extracting an approximation of the geometric contribution from the backscattered echo These results are used in exploring possible classification schemes.