A real-time sound field renderer based on digital Huygens' model

Modeling acoustical behavior in a room is complicated and computationally intense. Although many methods have been proposed for computer simulation and computers have become faster and faster, the procedure is deeply time-consuming as sound space increases. With the development of semiconductor technology, an alternative solution is to simulate and rebuild the sound field distribution using hardware. Compared with other methods, the method based on digital Huygens' model (OHM) is simple and easily implemented by hardware. In this paper, two schemes of OHM are derived to analyze sound propagation in a two-dimensional space, and are implemented by field programmable gate array (FPGA). In a two-dimensional space of size 2 m x 2 m, surrounded by rigid walls, the hardware simulation results of both schemes agree well with the software simulation result, and the calculation errors are relatively small. Compared with the software-based FDTD, original scheme, and updated scheme, a hardware system with 1024 nodes implemented by a Xilinx FPGA chip XC5VLX330T-FF1738 speeds up 7.1, 5.5, and 3.6 times, respectively. (C) 2011 Elsevier Ltd. All rights reserved.