Spatial PCM Sampling: A New Method for Sound Recording and Playback

This paper presents the mathematical and physical framework of a new technology, named SPS (Spatial PCM Sampling): it is the equivalent, in a two-dimensional spherical-coordinate space, of the traditional PCM representation of a waveform (in the one-dimensional time domain). It is nowadays possible to record an SPS multichannel stream (also called P-format) by processing the signals coming from massive microphone arrays, now widely employed in the broadcasting industry and in research labs. Some types of sound processing are easy when operating on P-format signals; some, indeed, require more work. At playback, it is possible to drive loudspeaker arrays of arbitrary shape and complexity, providing in general better spatial accuracy than competing well known methods, such as Ambisonics or WFS.