Sound field control for multiple listener virtual imaging.

Advances in technology have made it possible to use loudspeakers to generate the perception in a listener of a "virtual" image of a source of sound that appears to be at a position other than that of the loudspeakers. This paper deals with the challenge of simultaneously generating the perception of the same virtual sound image at the ears of multiple listeners. The basis of the analysis is the so-called Optimal Source Distribution (OSD) that enables well-conditioned crosstalk cancellation at the ears of a single listener. This consists of a hypothetically continuous distribution of monopole acoustic strength which has a remarkable frequency independent radiation pattern. This directivity pattern enables crosstalk cancellation at a series of defined listener positions in the far field. However, in any practical application using current technology, the sources will be discrete loudspeaker elements. Previous work already shows the promise of this technique and the work described here aims to evaluate a number of methods for determining the strength of such discrete sources while still ensuring crosstalk cancellation at a number of listener positions. • The Optimal Source Distribution (OSD) delivers multiple listener crosstalk cancellation. • Solutions are presented for a discrete approximation to the continuous OSD. • Minimisation of the L 1 source strength norm has advantages over L 2 norm minimisation. • Minimisation of the deviation from the OSD sound field also yields viable solutions. • L 1 norm minimisation for a single listener gives equivalent results to the OSD. [ABSTRACT FROM AUTHOR]