A Novel Near-Field Localization Method Based on Second Order Statistics

The bearing and range estimation of near-field sources is studied and a novel method is presented. In our proposed algorithm the second-order statistics was used to estimate bearing of near-field sources. By exploiting the Toeplitz property of covariance matrix of far-field sources, we reconstruct a far-field approximate covariance matrix with respect to bearing only for near-field sources, which allows for the bearing estimation using the generalized far-field root MUSIC algorithm. The range estimation of all sources is consequently obtained in one time search by defining a novel 1-D MUSIC spectrum. The method has the merits of good estimation precision and its effectiveness is approved by the simulation results.