Muhammad Sheraz Riaz

many areas of signal processing, e.g. telecommunication, chemistry,
biology and audio, the observed signals result from the combination of several sources. Source separation is the general problem of characterizing the
sources and estimating the source signals underlying a given mixture signal.
Early source separation techniques based on spatial filtering are now
established: beamforming and time frequency masking are employed in mobile phones and consumer audio systems to suppress environmental noise
and enhance spatial rendering [1, 2], while independent component analysis
ICA is used for the extraction of specific signals from electroencephalogram EEG, electrocardiogram ECG and functional magnetic resonance
imaging fMRI data [3, 4]. The emergence of more powerful source separation techniques in the last five years has led to a boom in the range
of applications. Data that were thought as too difficult to separate can
now be processed, as illustrated by companies providing com