Paper No: 38
Probabilistic Geometric Approach to Blind Separation of Time-Varying Mixtures
Author(s): Ran Kaftory, Yehoshua Zeevi
Abstract
We consider the problem of blindly separating time-varying
instantaneous mixtures. It is assumed that the arbitrary
dependency of the mixing coefficient on time, is known up to a
finite number of parameters. Using sparse (or sparsified) sources,
we geometrically identify samples of the curves representing the
parametric model. The parameters are found using a probabilistic
approach of estimating the maximum likelihood of a curve, given
the data. After identifying the model parameters the mixing system
is inverted to estimate the sources. The new approach to blind
separation is demonstrated using both synthetic and real data. |
