Localization and tracking of multiple moving sources based on local polynomial approximation of DOA

The windowed linear local polynomial approximation (LPA) of the time-varying direction-of-arrival (DOA) is developed for nonparametric multiple high-resolution estimation and tracking q rapidly moving sources. The method gives estimates of instantaneous values of the directions as well as their first derivatives. The asymptotic variance and bias of these estimates are derived and used for the window size optimization. Marginal beamformers, 2D functions in the /spl theta/ - /spl theta//sup (1)/ space, are proposed in order to be used for estimation and moving sources visualization. These marginal beamformers are able to localize and track every source individually nulling signals from all other moving sources.