Sweet-spot-independent binaural reproduction with a listener-adaptive loudspeaker array
Marcos Simón Gálvez, Filippo Maria Fazi · 2016
Adaptive cross-talk cancellation systems require a smooth and fast update of the control filters for different listening positions, which is often achieved by simultaneous filtering and cross-fading. An alternative method is presented here, in which the control filters are calculated analytically in real-time, assuming each acoustical source is represented by a point-monopole radiator. This allows for a time-domain formulation of each filter as a sum of gain-delay elements, rather than as a single finite impulse response (FIR) filter. This leads to a very simple signal processing scheme, wherein the gain-delay elements are varied to adapt the response towards the listener. The performance of the proposed system is evaluated by means of numerical simulations with a geometry based on a 28 point-source loudspeaker array.