An improved method for passive localization of coexistent far-field and near-field sources

An improved algorithm based on the symmetric uniform linear array (ULA) and the second-order statistics (SOS) is proposed to cope with the mixed far-field and near-field sources localization problem. The direction-of-arrivals (DOAs) and powers of far-field sources can be firstly estimated from the MUSIC spectral function. Then, an oblique projection technique is adopted to eliminate the obtained far-field interference. Finally, both direction-of-arrival (DOA) and range estimations of near-field sources can be performed, exploiting the symmetry property of ULA geometry. Compared with the previous second-order MUSIC method, the resultant algorithm provides the improved accuracy, as well as alleviates the aperture loss. Computer simulations confirm the satisfactory performance of the proposed algorithm.