An Algorithm for 4-D Parameters Jointly Estimating of Near-Field Sources

Source localization is an important problem in array signal processing and it is used to estimate the direction of arrival (DOA) of multiple sources. In this paper, we present a four-order cumulant based ESPRIT like method for passive localization of near-field narrowband non-Gaussian sources using a centrosymmetric cross array. The proposed method uses the multiple-rotational invariance among certain cumulant-domain signal subspaces for 4D parameter (carrier frequency, elevation, bearing and range) estimation of multiple sources. The method does not require any spectral peak search, only needs 1D parameter pairing, and can be applied to additive Gaussian noise environment. The performance of the proposed method is verified by computer simulations.