Abstract
From the most known Gaussian mixture to the cutting-edge multi-Bernoulli
mixture of various forms, mixture offers a fundamental means to deal
with uncertainties, which has led to a variety of appealing applications
in the state estimation realm based on a single sensor or a sensor
network. Like noise is often used to model unknown system input, one may
use various hypotheses to deal with the uncertain state space model or
data association. Meanwhile, consensus may be sought over the
cross-correlated sensors. These all drive a need for representing the
probability distribution by a mixture of properly weighted component
distributions, which fuse the information gained from different
models/hypotheses or from different sensors. This technical note
presents information-theoretical results which answer how the
averaging/mixture approach makes sense and how the fusing weights should
be designed.