Parameter estimation of Oscillating Gaussian functions using the scaled reassigned spectrogram

Research output: Contribution to journalArticle

Abstract

In this paper we suggest an algorithm for estimation of the parameters detailing Oscillating Gaussian functions. The different components of the signal are first detected in the spectrogram. After this we exploit the fact that a Gaussian function may be perfectly reassigned into one single point given a correct scaling factor, where this scaling factor is a function of the unknown shape parameter of the Gaussian function. The scaled reassignment of the spectrogram is performed using a set of candidate scaling factors and the local Renyi entropy is used to measure the concentration of each component using every candidate scaling factor. The estimates are refined by using non-linear least squares. The algorithm is evaluated on both simulated and real data.

Details

Authors
Organisations
Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Probability Theory and Statistics

Keywords

  • Gabor atom, Gaussian functions, Logons, Parameter estimation, Reassigned spectrogram, Transients
Original languageEnglish
Pages (from-to)20-32
Number of pages13
JournalSignal Processing
Volume150
StatePublished - 2018 Sep 1
Peer-reviewedYes