Research on sound field reproduction technology aims to achieve a high sense of realism. However, conventional methods have the problem of difficulty in separating direct sound and reflected sounds, resulting in a lack of realism and accuracy. We propose a method using data obtained by the closely located four point microphone method to generate sound fields more accurate and freer. We use the four point method to determine the arrival sound source distribution and identify the direction of the reflected sounds. Furthermore, since virtual sound sources exist in a very wide space, it can be assumed that the spherical waves emitted from them arrive as plane waves in the listening area. We have previously proposed a method for sound field generation by adding together plane waves from all angles. Based on the three-dimensional wave equation as the ideal sound field, we compared the impulse response near the origin of the sound field generated by simulation using a spherical loudspeaker array and the error distribution. As a result, the impulse responses were very similar in shape, confirming sufficient reproducibility. Looking at the error distribution, it was found that the error increased as time passed.
Eriguchi et al. (Wed,) studied this question.