Imperial College London > Talks@ee.imperial > COMMSP Seminar > Resonance-Based Signal Analysis

Resonance-Based Signal Analysis

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Danilo Mandic.

This talk describes a nonlinear signal analysis method based not on frequency or scale, as provided by the Fourier and wavelet transforms, but on ‘resonance’. This decomposition is motivated by the fact that many signals arising from physiological and physical processes, in addition to being non-stationary, are moreover a mixture of sustained oscillations (constituting a high-resonance component) and non-oscillatory transients (constituting a low- resonance component) that are difficult to disentangle by linear methods. Examples of such signals include speech, biomedical, and geophysical signals; for example, EEG signals contain rhythmic oscillations (alpha and beta waves, etc) but they also contain transients due to measurement artifacts and non-rhythmic brain activity. While frequency components are straightforwardly defined and can be obtained by linear filtering, resonance components are more difficult to define and procedures to obtain resonance components are necessarily nonlinear, as will be shown. The resonance-based signal decomposition algorithm presented in this talk utilizes several recent developments in signal processing, including sparse signal representations, morphological component analysis, constant-Q (wavelet) transforms with adjustable Q-factor, and fast algorithms for sparsity-regularized linear inverse problems.

This talk is part of the COMMSP Seminar series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


Changes to Talks@imperial | Privacy and Publicity