Multizone wideband sound field reproduction

This thesis deals with the problem of multizone wideband sound field reproduction using an array of loudspeakers. A pressure matching approach is researched to control the sound field within the zones through the calculation of loudspeaker weights. The loudspeaker weights are computed first using a regularized least-squares (LS) approach and then a least-absolute shrinkage and selection operator (Lasso). It is demonstrated that the single-stage LS technique outperforms the single-stage Lasso in multizone wideband sound field reproduction, while the single-stage Lasso enables the judicious placement of loudspeakers. To improve the multizone sound reproduction performance of wideband sources using a limited number of loudspeakers, it is assumed that the virtual sources are fixed in positions. A new two-stage Lasso-LS pressure matching approach is then proposed to optimize both the loudspeaker locations and weights. In the first stage, a Lasso algorithm is used to select the loudspeakers' positions for all sources and frequency bands. A second stage then optimizes reproduction using all selected loudspeakers on the basis of a regularized LS algorithm. The results show that a horizontal array of limited number of loudspeakers (e.g. 52) can be used to effectively create personal audio spaces for multiple users of variable heights. The proposed method is then extended to a nested Lasso-LS method which employs harmonic nested arrays in the first stage Lasso to reduce the computational complexity. Effectively, the nested arrays provide a priori knowledge of prospective loudspeaker locations based on the frequency bands of interest. The final loudspeaker locations and weightings are then estimated during the two-stage Lasso-LS optimization.