Spectral analysis of nouniformly sampled data: a new approach versus the periodogram

Authors:
Petre Stoica;Jian Li;Hao He
Affiliations:
Department of Information Technology, Uppsala University, Uppsala, Sweden;Department of Electrical and Computer Engineering, University of Florida, Gainesville, FL;Department of Electrical and Computer Engineering, University of Florida, Gainesville, FL
Venue:
IEEE Transactions on Signal Processing
Year:
2009

Citing 6
Cited 6

The Regular Fourier Matrices and Nonuniform Fast Fourier Transforms

SIAM Journal on Scientific Computing
Convex Optimization

Convex Optimization
Spectral analysis of irregularly-sampled data: Paralleling the regularly-sampled data approaches

Digital Signal Processing
An adaptive filtering approach to spectral estimation and SARimaging

IEEE Transactions on Signal Processing
Efficient mixed-spectrum estimation with applications to targetfeature extraction

IEEE Transactions on Signal Processing
Nonuniform fast Fourier transforms using min-max interpolation

IEEE Transactions on Signal Processing

Spectral analysis of nonuniformly sampled data -- a review

Digital Signal Processing
Reliable wideband multichannel spectrum sensing using randomized sampling schemes

Signal Processing
Consistent estimation of non-bandlimited spectral density from uniformly spaced samples

IEEE Transactions on Information Theory
Wideband spectrum sensing technique based on random sampling on grid: Achieving lower sampling rates

Digital Signal Processing
Evaluation of Fourier transform estimation schemes of multidimensional signals using random sampling

Signal Processing
SAR imaging via efficient implementations of sparse ML approaches

Signal Processing

Quantified Score

Hi-index	35.74

Visualization

Abstract

We begin by revisiting the periodogram to explain why arguably the plain least-squares periodogram (LSP) is preferable to the "classical" Fourier periodogram, from a data-fitting viewpoint, as well as to the frequently-used form of LSP due to Lomb and Scargle, from a computational standpoint. Then we go on to introduce a new enhanced method for spectral analysis of nonuniformly sampled data sequences. The new method can be interpreted as an iteratively weighted LSP that makes use of a data-dependent weighting matrix built from the most recent spectral estimate. Because this method is derived for the case of real-valued data (which is typically more complicated to deal with in spectral analysis than the complex-valued data case), it is iterative and it makes use of an adaptive (i.e., data-dependent) weighting, we refer to it as the real-valued iterative adaptive approach (RIAA). LSP and RIAA are nonparametric methods that can be used for the spectral analysis of general data sequences with both continuous and discrete spectra. However, they are most suitable for data sequences with discrete spectra (i.e., sinusoidal data), which is the case we emphasize in this paper.