We propose an approximate maximum likelihood parameter estimation algorithm
, combined with a model order estimator, for superimposed undamped exponent
ials in noise. The algorithm combines the robustness of Fourier-based estim
ators and the high-resolution capabilities of parametric methods. We use a
combination of a Wald statistic and a MAP test for order selection and init
ialize an iterative maximum likelihood descent algorithm recursively based
on estimates at higher candidate model orders. Experiments using simulated
data and synthetic radar data demonstrate improved performance over MDL, MA
P, and AIC in cases of practical interest.