Maximum likelihood estimation of structural wave components from noisy data.
In this paper, a general framework is developed for determining the underlying parameters of general signal models through the application of maximum likelihood estimation theory for functions whose variables separate. This method extends previous work in sinusoidal and exponential estimation to include models with other functional bases, such as exponential functions with nonconstant amplitudes and Bessel functions. Nonuniform spatial sampling is also possible with this technique. The maximum likelihood method is applied to the identification of wave components along one-dimensional structural elements. Results are given which demonstrate the viability and accuracy of the technique estimating exponential and Bessel function model parameters from noisy simulation data.