Three Dimentional Multiple Near-field Source Localization Based on MUSIC Algorithm to Increase the Localization Accuracy of Optimal Beamformer

Abstract

In this paper the MUSIC source localization algorithm is applied to the near-field narrowband optimal beamformer to increase its localization accuracy and resolution capability. Optimal beamformer cannot identify closely spaced multiple near-field signals. MUSIC algorithm is an Eigen-decomposition based source localization technique. A three dimensional MUSIC algorithm is used with near-field optimal beamformer to correctly localize the three parameters (range, elevation and azimuthal angle) of multiple sources. The robustness of this proposed beamformer against the white Gaussian noisy environment is also examined. The Root Mean Square Error (RMSE) to localize the multiple near-field targets is also studied. The simulation results show that the MUSIC based optimal beamformer can easily sense the multiple closely spaced sources in the noisy environment with sharper radiation lobe using minimum number of snapshots and sensors.