A new near-field source localization algorithm using Second-Order Statistics

In this paper, based on Second-Order Statistics (SOS), a new algorithm is proposed for jointly estimating ranges, frequencies and Direction-Of-Arrival (DOA) of multiple near-field narrowband sources. The proposed method constructs three matrices using the second-order statistics of the properly chosen sensor outputs, and forms two new ones using the three matrices, then estimates the source parameters from the new ones' eigenvalues and the corresponding eigenvectors. In comparison with other methods based on SOS, the proposed method can save one sensor, and only uses three SOS matrices; in addition, the estimated parameters are paired automatically. Finally, the simulation results are presented to validate the performance of the proposed method.