Composite wavefront decomposition via multidimensional digital filtering of array data

A heuristic solution is obtained for the problem of estimating the number, vector velocity, and wave shape of overlapping plane waves in the presence of interfering plane waves and channel noise. The solution is shown to be optimum if the number of waves and vector velocities are known. Wave detection and estimation of wave parameters is accomplished by digital filtering of the frequency-wavenumber power spectrum of the array data, which is computed by an efficient estimator. The multiwave estimation problem is approached by reducing it to a succession of single-wave problems using specially developed frequency-wavenumber filters. The results are placed in perspective by showing how important approaches to the processing of array data are related by spatio-temporal filtering of the frequency-wavenumber spectrum. Examples of processed array data are presented and analyzed which demonstrate the effectiveness of spatio-temporal spectral estimation and filtering.