Optimal Soundfield Generation for Sound Reinforcement in Time and Frequency Domain
Large-scale sound reinforcement for speech and music is nowadays typically realised with line array systems. Their drive has to be optimised with respect to several degrees of freedom, e.g.arrangement, geometry, construction of the boxes and different boundary constraints, e.g. geometry of the auditory, areas that have to be avoided, available acoustic power. It is an ill-posed inverse problem in mathematical terms. The number of sources is less than the number of receivers. There are no scientifically reported solutions and no approved theoretical methods for this application. The project therefore aims at implementing an adjoint-based approach in the time domain which has proven its capabilities in fluid dynamics. The driving functions optimised by numerical simulations in the time and frequency domains and the predicted sound fields are to be verified by an experimental setup in an anechoic room. By means of the adjoint method the influence of abase flow on the generated sound field is taken into account for the first time, which can be of great practical relevance if it comes to the influence of wind flows on speech intelligibility. Due to different methodological challenges, the project goals can only be reached by a cooperation of departments with expertise in acoustics and audio technology (with specific experience in the field of sound field synthesis) and numerical fluid dynamics (with specific experience in the adjoint method) as it is to be implemented in this project.