MODEL ORDER SELECTION OF DAMPED SINUSOIDS IN NOISE BY PREDICTIVE DENSITIES

We develop a procedure for the order selection of damped sinusoidal mo dels based on the maximum a posteriori (MAP) criterion. The proposed m ethod merges the concept of predictive densities with Bayesian inferen ce to arrive at a complex multidimensional integral whose solution is achieved by way of the efficient Monte Carlo importance sampling techn ique. The importance function, a multivariate Cauchy probability densi ty, is employed to produce stratified samples over the hypersurfaces s upport region. Centrality location parameters for the Cauchy are resol ved by exploiting the special structure of the compressed likelihood f unction (CLF) and applying the fast maximum likelihood (FML) procedure of Umesh and Tufts [38]. Simulation results allow for a comparison be tween our method and the singular value decomposition (SVD) based info rmation theoretic criteria in [28].