Sound field reproduction based on Pressure Matching with transfer functions modeled by equivalent sources and image sources

Abstract

Pressure matching (PM) method is an effective method to control a sound field using many loudspeakers. It is difficult to implement PM method because it requires multi-point room impulse response (RIR) measurements. Multi-point measurements are challenging because they require a large number of microphones and a narrow sampling interval when dealing with high frequencies. Therefore, to synthesize a sound field in a local region, we proposed a PM method using estimated RIRs from only a small number of measurements based on the sparse Equivalent Source Method (ESM). In the previous method, we modeled the transfer functions considering only the direct sound from the loudspeaker. However, when the sound field is synthesized in the actual environment, the reflected sounds must be considered. In this study, we propose a PM method using the estimated transfer functions by sparse ESM based on the image source method considering the direct sound and early reflected sounds. In the experiment, we evaluated the accuracies of synthesized sound field by the proposed method compared to the methods only considering the direct sound.