Spatial soundfield recording using compressed sensing techniques

We present a new method for spatial soundfield recording based on the application of compressed sensing theory. The major problem in spatial soundfield recording system is how to record the higher order harmonic components of a given soundfield using as few as possible of microphones. The proposed method is under the assumption that the soundfield is sparse in source domain. More specifically, based on the general model of soundfield, we apply sparse decomposition to the harmonics, which are calculated according to the sound pressure received by a circular microphone array, and then upscale the harmonics to higher order. Simulation results indicate that our proposed method can drastically reduce the required number of microphones for capturing large region soundfield compared to the current methods, especially in high frequencies.