In Audio Signal Processing for Next-Generation Multimedia Communication Systems, pp. 323-342, Y.A. Huang, J. Benesty, Kluwer, Jan. 2004 Chapter 13 SOUND FIELD SYNTHESIS Sascha Spors, Heinz Teutsch, Achim Kuntz, Rudolf Rabenstein Telecommunications Laboratory Multimedia Communications and Signal Processing University of Erlangen–Nuremberg spors@LNT.de, teutsch@LNT.de, kuntz@LNT.de, rabe@LNT.de Abstract Conventional multichannel audio reproduction systems for entertain- ment or communication are not capable of immersing a large number of listeners in a well defined sound field. A novel technique for this pur- pose is the so-called wave field synthesis. It is based on the principles of wave physics and suitable for an implementation with current mul- tichannel audio hard- and software components. A multiple number of fixed or moving sound sources from a real or virtual acoustic scene is reproduced in a listening area of arbitrary size. The listeners are not restricted in number, position, or activity and are not required to wear headphones. A successful implementation of wave field synthesis sys- tems requires to address also spatial aliasing and the compensation of non-ideal properties of loudspeakers and of listening rooms. Keywords: wave field synthesis, multichannel audio, loudspeaker array, loudspeaker compensation, room compensation 1. INTRODUCTION State-of-the-art systems for the reproduction of spatial audio suffer from a serious problem: The spatial properties of the reproduced sound can only be perceived correctly in a small part of the listening area, the so-called sweet spot. This restriction occurs because conventional reproduction of spatial audio is based on psychoacoustics, i. e. mainly intensity panning techniques. A solution to this problem calls for a new reproduction technique which allows the synthesis of physically correct 323