Date: Thu, November 9, 17:15-19:15
Place: Room Dw601, D Block, IIS, The University of Tokyo
Invited Speaker: Prof. Danilo P. Mandic (Imperial College London)

Title: Online Methods for Tracking the Modality Within Real World Signals
 
Abstract:
Methods for signal modality characterisation in an off-line manner
already exist and have found their application in biomedicine and
engineering. It is however equally important to track the changes in
signal modality online, since for instance the change in the nature of
heart rates from stochastic to deterministic may indicate heart
hazard. To that cause we present our recent results on on-line signal
modality characterisation, based on the use of hybrid adaptive
filters. This framework is supported by simulation results on real-life
EEG signals describing epileptic seizure. We next extend these results
to the complex domain in order to track some intricate changes in the
nature of complex signals, such as their split-complex and fully-complex
natures. This is supported by examples on real world maritime radar
data.