9:00 am

G-1 Modern Development Tools for Dynamic Transducers

Martin Opitz and Richard Barnert, AKG Acoustics GmbH, Vienna, Austria

The growing use of multimedia applications in mobile equipment like notebooks, MP3-players or mobile phones causes the need for miniature speakers with optimized performance. In order to improve their acoustic performance the detailed behavior of transducers has to be studied. Modern tools like numeric simulation programs and laservibrometry give a deeper sight into the demanded characteristic features. In this paper the application of three important tools is described, examples are shown.

Convention Paper 5438

9:30 am

G-2 Controlling Loudspeaker Array Radiation in Three Dimensions

Justin Baird, John Meyer, and Perrin Meyer, Meyer Sound Laboratories, Berkeley, CA, USA
Peter Kassakian and David Wessel, University of California, Berkeley, Berkeley, CA, USA

Three different loudspeaker arrays - a straight line array, a curved line array, and a dodecahedron - each consisting of twelve four inch diameter transducers, were examined from the point of view of predicting and controlling their radiation patterns. These arrays are relatively small, allowing for three-dimensional rotation and measurement within an anechoic chamber. Comparisons are made between the actual results of three-dimensional measurement and classical array theory over all angles and operating frequencies. The primary goal of this research is to accurately predict a given array's response through the use of individual transducer measurements and a far field prediction model. This prediction tool affords the ability to prototype arbitrary array configurations and to visualize the resulting radiation patterns. Using the results of such predictions, we explore possible methods for obtaining consistent beam width across the operating frequency range for the line arrays. For the dodecahedron we present novel applications of dynamic radiation pattern control. Signal processing structures that allow real-time control of the radiated pattern are also demonstrated.

No Convention Paper Printed

10:00 am

G-3 Loudspeaker Complex Directional Response Characterization

William R. Hoy and Charles McGregor, Eastern Acoustic Works, Inc., Whitinsville, MA, USA

For complex directional response data to be useful, it must be gathered and deployed in a much more disciplined manner than has typically been applied to magnitude-only data. The loudspeaker under test and measurement microphone must be precisely positioned; geometrical errors must be corrected; and, temperature variations must be accounted for. An object oriented data structure is described which facilitates solutions to each of these challenges. Practical applications employing the new data structure are also presented.

Convention Paper 5439

10:30 am

G-4 Loudspeaker Transfer Function Averaging and Interpolation

David W. Gunness, Eastern Acoustic Works, Inc., Whitinsville, MA, USA

Transfer functions of acoustical systems usually include significant phase lag due to propagation delay. When this delay varies from one transfer function to another, basic mathematical operations such as averaging and interpolation produce unusable results. A calculation method is presented which produces much better results, using well-known mathematical operations. Applications of the technique include loudspeaker complex directional response characterization, complex averaging, and DSP filter design for loudspeaker steering.

Convention Paper 5440

11:00 am

G-5 New Results on Combined Audio Compression/Watermarking

Frank Siebenhaar, Christian Neubauer, Jürgen Herre and Ralph Kulessa, Fraunhofer Institute for Integrated Circuits, Erlangen, Germany

Today, music distribution over the Internet is an increasingly important business. In this context, watermarking can provide beneficial means to transmit rights information within the content. To convey the origin of such content, the combination of simultaneous low bitrate encoding and watermark embedding is a promising novel technique. This paper describes the basic concept of combined compression/watermarking for audio signals. In contrast to separate steps of encoding and watermarking the combined approach enables an optimal coordination between the quantization strategy in the audio encoder and the watermark embedding process. This allows adjustment of the system to specific needs in terms of audio quality and watermark robustness. Experimental results obtained from an extended MPEG-2/4 AAC encoder and a first implementation of an extended MPEG-1/2 Layer-3 encoder confirm the potential of the concept.

Convention Paper 5442