Optimal source placement for sound zone reproduction with first order reflections

Abstract

The problem of delivering personal audio content to listeners sharing the same acoustic space has recently attracted attention. It has been shown that a perceptually acceptable level of acoustic separation between the listening zones is difficult to achieve with active control in non-anechoic conditions. A common problem of strong first order reflections has not been examined in detail for systems with practical constraints. Acoustic contrast maximization combined with optimization of source positions is identified as a potentially effective control strategy when strong individual reflections occur. An analytic study is carried out to describe the relationship between the performance of a 2×2 (two sources and two control sensors) system and its geometry in a single-reflection scenario. The expression for acoustic contrast is used to formulate guidelines for optimizing source positions, based on three distinct techniques: Null-Split, Far-Align, and Near-Align.The applicability of the techniques to larger systems with up to two reflections is demonstrated using numerical optimization. Simulation results show that optimized systems produce higher acoustic contrast than non-optimized source arrangements and an alternative method for reducing the impact of reflections (sound power minimization).

Combining Artificial and Natural Background Noise in Personal Audio Systems

“…This line of enquiry, i.e. consideration of the perceptual relevance of sound leaking from one zone into another, has been taken by contributors to the POSZ Project [7] to inform loudspeaker positioning, [8], [9] and the choice of sound zoning methods [10]- [12]. A recent article by Donley et.…”