Although sound-field reproduction methods have been extensively studied over the past decades, there is no single approach that dominates. This paper describes a method called Co-Variance Method (CVM) that reproduces a spatial covariance matrix of the original sound field without knowledge of sound source locations. Conventional methods based on physical theory require an impractically large number of transducers. In contrast, CVM is based on the statistical properties of a sound field. Simulation experiments evaluated the relation between reproduction errors and the number of microphone-loudspeaker pairs when reconstructing the wave fronts of sinusoidal, impulse, and saw-tooth-like impulsive signals. With CVM, a wave front resembling that of the original field was obtained in the listening area.
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.