Sequential MCMC estimation of nonlinear instantaneous frequency

Y. Li, D. Simon, Antonia Papandreou-Suppappola, Darryl Morrell, R. L. Murray

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

Instantaneous frequency (IF) estimation of signals with nonlinear phase is challenging, especially for online processing. In this paper, we propose IF estimation using sequential Bayesian techniques, by combining the particle filtering method with the Markov chain Monte Carlo (MCMC) method. Using this approach, a nonlinear IF of unknown closed form is approximated as a linear combination of the IFs of non-overlapping waveforms with polynomial phase. Simultaneously applying parameter estimation and model selection, the new technique is extended to the IF estimation of multicomponent signals. Using simulations, the performance of this sequential MCMC approach is demonstrated and compared with an existing IF estimation technique using the Wigner distribution.

Original languageEnglish (US)
Title of host publication2007 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '07
PagesIII1177-III1180
DOIs
StatePublished - 2007
Event2007 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '07 - Honolulu, HI, United States
Duration: Apr 15 2007Apr 20 2007

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume3
ISSN (Print)1520-6149

Other

Other2007 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '07
Country/TerritoryUnited States
CityHonolulu, HI
Period4/15/074/20/07

Keywords

  • Bayes theorem
  • Frequency estimation
  • Markov chain Monte Carlo
  • Particle filter

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Sequential MCMC estimation of nonlinear instantaneous frequency'. Together they form a unique fingerprint.

Cite this