The concept of sound zones has recently gained
much attention in the acoustics community. The
idea of sound zones is to separate a space into
multiple zones for which the audio content can
be individually controlled. This is done by use of
an array of loudspeakers. The most commonly
studied objective is to play sound in one part
of a space while simultaneously attempting to
keep another part of the space quiet. There are
several known methods for steering an array in
order to obtain this result. An issue is that these
methods typically assume the input to be sinu-
soidal with fixed frequency. This is not a realis-
tic assumption for common audio signals such
as speech or music. The problem of creating
sound zones for such signals has not been ex-
tensively studied in the literature. In this the-
sis we investigate some possibilities for creating
sound zones for wideband signals of this type. To
facilitate the investigation of sound zone meth-
ods, a software framework for room simulation
with the mirror image model has been devel-
oped. Among the discussed methods are both
methods which have previously been discussed
in the literature and methods which have not.
We show that the task of extending sound zone
methods to wideband signals introduces compli-
cations which are not seen for simple sinusoidal
signals. We also discuss means of overcoming
these complications. The main outcome of the
presented investigations is a sound zone method
which performs well for wideband signals.