This article deals with spatial analysis of real 3D sound fields, and presents an optimal beamforming algorithm minimizing the power of side lobes, used to improve the spatial representation of sound fields. Optimal tap vectors are computed for a set of incidence directions mapping the whole sphere. The robustness of the algorithm to sensorpositioning errors and to sensor noise is tested with real data, acquired using microphone and loudspeaker arrays. It is shown that there is good agreement between the real and the corresponding simulated data regarding the incidence direction. Moreover, early reflections, providing new peaks in the representation depending on the location of secondary sources are also clearly visible.
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.