The role of functional data analysis for instantaneous frequency estimation

Authors:
Minjeong Park;Sinsup Cho;Hee-Seok Oh
Affiliations:
Statistical Research Institute, Statistics Korea, Daejeon, Korea 302-280;Department of Statistics, Seoul National University, Seoul, Korea 151-747;Department of Statistics, Seoul National University, Seoul, Korea 151-747
Venue:
Computational Statistics
Year:
2013

Citing 4
Cited 0

Estimating the frequency of a noisy sinusoid by linear regression

IEEE Transactions on Information Theory
Multicomponent polynomial phase signal analysis using a trackingalgorithm

IEEE Transactions on Signal Processing
Characterization of signals by the ridges of their wavelettransforms

IEEE Transactions on Signal Processing
One or Two Frequencies? The Empirical Mode Decomposition Answers

IEEE Transactions on Signal Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes a method for estimating the instantaneous frequency of a nonstationary signal; this method is based on a combination of empirical mode decomposition and functional data analysis. The proposed method incorporates a basis expansion technique for a functional data into time-varying phase derived by empirical mode decomposition and Hilbert transform, which provides a stable instantaneous frequency function. The superiority of the proposed method for instantaneous frequency estimation is demonstrated by various simulation studies. The analysis of multicomponent signals by the proposed method is also discussed. Furthermore, it is shown that the proposed method is highly effective for identifying groups (clusters) of nonstationary signals on the basis of the instantaneous frequency information.