Hearing Protection for High-Noise Environments

We propose to develop high fidelity software tools which would allow identification and understanding of the relevant bioacoustic and psychacoustic mechanisms responsible for the transition of acoustic energy through non-airborne pathways to the cochlea and significantly reduce the cost of design of noise suppression devices. The principal numerical tool we propose to use is the fast integral equation solver for elasto-acoustic media, development of which we initiated during Phase I of this project. The following features make the choice of our solver uniquely suited to the proposed numerical simulations: - the non-lossy compression technique, which allows to reduce, without compromising accuracy of the solution, the memory and computer time requirements from O(N^3) to O(Nlog N), where N is the number of unknowns (we note that in the envisioned numerical simulations of the acoustic wave propagating through the human head, the expected number of unknowns can be of the order of one to ten millions; thus the compression technique we propose to use would reduce the memory and computational complexity by a gigantic factor of 10^{12} - 10^{14}, and hence make the realistic numerical simulations tractable), - numerical accuracy over a sufficient dynamic range to describe the alternative sound propagation paths (empirically, it is known that the intensity of the bone-conducted sound is about 30 - 40 dB lower than that of the air-conduction mechanism, - ability to simulate regions/surfaces of different material properties, - applicability of the compression technique to the entire interval of acoustic frequencies in which the problem size ranges from sub-wavelength (fraction of a wavelength) to moderate (several wavelengths), - parallel, scalable implementation of the algorithm which reduces the wall clock time of the simulations.