Dual hidden Markov model for characterizing wavelet coefficients from multi-aspect scattering data

Angle-dependent scattering (electromagnetic or acoustic) is considered from a general target, for which the scattered signal is a non-stationary funct ion of the target-sensor orientation. A statistical model is presented for the wavelet coefficients of such a signal, in which the angular non-station arity is characterized by an "outer" hidden Markov model (HMM,). The statis tics of the wavelet coefficients, within a state of the outer HMM, are char acterized by a second, "inner" HMMi, exploiting the tree structure of the w avelet decomposition. This dual-HMM construct is demonstrated by considerin g multi-aspect target identification using measured acoustic scattering dat a. (C) 2001 Elsevier Science B.V. All rights reserved.