A deconvolution approach to source localization

Because of its low resolution the conventional beamformer (CBF) has generally been neglected in the recent literature on array processing and considerable attention has been devoted to high-resolution (HR) techniques. A deconvolution approach was applied to the output of the CBF in order to obtain a source localization procedure whose performance is comparable with those of the most efficient HR techniques. Deconvolution is the process of recovering the input of a system from the output knowing the characteristics of the system. The scheme also furnishes estimates of the source powers and comprises a test that detects the number of sources present. It can be considered as a means to obtain maximum likelihood estimates (MLEs) when the CBF output is taken as the observed data.<<ETX>>