ARMA parameter estimation using only output cumulants
Published 6 January 2003
Ananthram Swami, Jerry M. Mendel
Citations105
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
A fundamental result is presented pertaining to the identifiability of AR parameters, based on the Yule-Walker equations drawn from a (specific) set of (p+1) 1-D slices of the kth (k>2) order output cumulant.
Abstract
The authors develop two algorithms to estimate the parameters of a nonminimum phase autoregressive moving average (ARMA) system which is excited by an unobservable i.i.d. nonGaussian process. AR parameters are obtained in the usual way-via the higher-order Yule-Walker equations based either on the autocorrelations or the cumulants. The first method uses q (where q is the MA order) one-dimensional slices of the output cumulant, whereas the second method uses only two such slices to compute the MA parameters. Neither method involves computation of the residual (i.e., AR compensated) time series of polynomial factorization. Extensions to 2D and multichannel systems, as well as to time-recursive versions, are also developed. Both methods handle Gaussian noise of unknown power spectral densities, at the input as well as the output.>
Keywords
Computer ScienceEngineering
IEEE Transactions on Circuits and SystemsKronecker products and matrix calculus in system theory
1,896 Citations1978John Brewer
The calculus of matrix valued functions of matrices is used to develop an interesting new method for the identifiication of parameters of lnear time-invariant system models.
Proceedings of the IEEEBispectrum estimation: A digital signal processing framework
1,368 Citations1987C.L. Nikias, M.R. Raghuveer
It is the purpose of this tutorial paper to aid engineers in grasping the utility of the available bispectrum estimation techniques, to discuss application problems that can directly benefit from the use of the bispectrums, and to motivate research in this area.
Applied OpticsSpeckle masking in astronomy: triple correlation theory and applications
581 Citations1983Adolf W. Lohmann, G. Weigelt +1 more
The theory of speckle masking is presented; it makes use of triple correlations and their Fourier counterparts, the bispectra, and shows algorithms for the recovery of the object from genuine astronomical Bispectra data.
Proceedings of the IEEESpectral estimation: An overdetermined rational model equation approach
493 Citations1982J.A. Cadzow
IEEE Transactions on Acoustics Speech and Signal ProcessingIdentification of nonminimum phase systems using higher order statistics
483 Citations1989Georgios B. Giannakis, Jerry M. Mendel
The authors model a finite-dimensional system as an ARMA (autoregressive moving-average) rational function of known orders, but the special cases of AR, MA, and all-pass models are also considered.
The Annals of StatisticsDeconvolution and Estimation of Transfer Function Phase and Coefficients for Nongaussian Linear Processes
318 Citations1982K. S. Lii, M. Rosenblatt
Deconvolution and Estimation of Transfer Function Phase and Coefficients for Nongaussian Linear Processes
303 Citations2011Richard Davis, Keh‐Shin Lii +1 more
IEEE Transactions on Acoustics Speech and Signal ProcessingBispectrum estimation: A parametric approach
267 Citations1985M.R. Raghuveer, C.L. Nikias
It is shown that the proposed parametric approach to bispectrum estimation based on a non-Gaussian white noise driven autoregressive (AR) model provides bispectral estimates that are far superior to the conventional estimates in terms of bispectrals fidelity and that in the case of detecting phase coupling among sinusoids, the method provides significantly better resolution.
IEEE Transactions on Information TheoryIdentification of linear stochastic systems via second- and fourth-order cumulant matching
221 Citations1987J.K. Tugnait
The identification problem for time-invariant single-input single-output linear stochastic systems driven by non-Gaussian white noise is considered and a least-squares criterion that involves matching the second- and the fourth-order cumulant functions of the noisy observations is proposed.
GeophysicsRecursion filters for digital processing
218 Citations1967John L. Shanks
Digital filters are most often applied to discrete time series by convolving the time series with the unit impulse response, or weighting function of the filter, and the resulting recursion formulas will be quite simple, providing efficient and rapid digital processing.
Proceedings of the IEEECumulants: A powerful tool in signal processing
174 Citations1987Georgios B. Giannakis
This expression permits the use of one-dimensional processing of the output cumulants of a linear, time-invariant system for identification of non-minimum-phase systems, and opens new directions in other signal processing applications.
IEEE Transactions on Acoustics Speech and Signal ProcessingCumulant-based order determination of non-Gaussian ARMA models
169 Citations1990Georgios B. Giannakis, Jerry M. Mendel
The objective is the order determination of non-Gaussian and nonminimum phase autoregressive moving average (ARMA) models, using higher-order cumulant statistics, and the two methods developed assume knowledge of upper bounds on the ARMA orders.
IEEE Transactions on Acoustics Speech and Signal ProcessingOn estimating noncausal nonminimum phase ARMA models of non-Gaussian processes
95 Citations1990Georgios B. Giannakis, A. Swami
The authors address the problem of estimating the parameters of non-Gaussian ARMA (autoregressive moving-average) processes using only the cumulants of the noisy observation using the least-squares solution to a quadratic model fitting of a sampled cumulant sequence.
AutomaticaIdentification of non-minimum phase linear stochastic systems
77 Citations1986J.K. Tugnait
Strong consistency of the proposed estimator is proved under certain sufficient conditions and simulation results are presented in support of the theory.
Constrained total least squares
67 Citations2005Theagenis J. Abatzoglou, Jerry M. Mendel
This work obtains a generalization of the TLS criterion called the Constrained Total Least Squares (CTLS) method by taking into account the linear dependence of the noise terms in both A andb by taking the largest eigenvalue and corresponding eigenvector of a certain matrix.
Adaptive system identification using cumulants
48 Citations2003Ananthram Swami, Jerry M. Mendel
A lattice version of the recursive instrumental variable method for adaptive parameter identification of ARMA (autoregressive moving-average) processes is developed, and adaptive versions of cumulant-based MA parameter identification algorithms are directly applicable.
IEEE Transactions on Automatic ControlTime and lag recursive computation of cumulants from a state-space model
42 Citations1990Ananthram Swami, Jerry M. Mendel
Journal of Time Series AnalysisESTIMATION OF COEFFICIENTS OF AN AUTOREGRESSIVE PROCESS BY USING A HIGHER ORDER MOMENT
39 Citations1981Mituaki Huzii
Identification of nonminimum phase linear stochastic systems
38 Citations1984J.K. Tugnait
This work considers the identification of time-invariant, nonminimum phase, stochastic systems driven by non-Gaussian white noise, given only the noisy observations of the system output.
AR Identifiability Using Cumulants
34 Citations2005Ananthram Swami, Jerry M. Mendel
Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIEUse Of Higher-Order Statistics In Signal Processing And System Theory: An Update
32 Citations1988Jerry M. Mendel
This paper provides a brief overview of much of the work that has occurred when parametric models are used in conjunction with higher-order statistics in a wide range of signal processing and system theory problems.
Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIENew Results On State-Space And Input-Output Identification Of Non-Gaussian Processes Using Cumulants
26 Citations1988Georgios B. Giannakis, A. Swami
ARMA Modeling using cumulant and autocorrelation statistics
22 Citations2005Georgios B. Giannakis, Jerry M. Mendel +1 more
An overdetermined system of equations whose least-squares solution yields the coefficients of an ARMA model is presented, which makes the solution practical for moderate output data lengths, when additive white Gaussian noise is considered.
On selection of maximum cumulant lags for noncausal autoregressive model fitting
21 Citations2003J.K. Tugnait
The author gives specific values for the maximum cumulant lags, and proves strong consistency of the two estimators of the AR parameters under the specified choice of lags.
Cumulant based parameter estimation of multichannel moving-average processes
17 Citations2003Yasushi Inouye, Georgios B. Giannakis +1 more
Given finite samples of a stationary, perhaps noisy, nonGaussian r-variate moving-average, MA(q) process, the authors study cumulant based identifiability conditions, under which the MA coefficient matrices, the input statistics, and the order q, can be uniquely determined.
Approximate realization and model reduction of non-minimum phase stochastic systems
17 Citations1986Georgios B. Giannakis, Jerry M. Mendel
Cross-bispectrum computation for multichannel quadratic phase coupling estimation
17 Citations1992M.R. Raghuveer, Sohail A. Dianat
A two-channel, quadratic, nonlinear process driven by sinusoidal signals is considered and the derivation is based on the bispectrum and cross-power spectrum of the two channels.
On estimating noncasual ARMA nonGaussian processes
11 Citations2003Georgios B. Giannakis, Ananthram Swami
Optimal estimates of MA and ARMA parameters of non-Gaussian processes from high-order cumulants
11 Citations2003B. Porat, B. Friedlander
The authors describe an asymptotically minimum-variance algorithm for estimating the moving average and autoregressive moving average parameters of nonGaussian processes from sample second- and third-order moments.
Adaptive Cumulant-Based Estimation of ARMA Parameters
10 Citations1988Ananthram Swami, Jerry M. Mendel
Time-recursive lattice algorithms for the estimation of the AR parameters of an ARMA process, using a 1-D slice of the k-th order cumulant, shown to be equivalent to linear prediction with non-conventional orthogonality conditions.
The Asymptotic Second Order Moments of the Estimated Third Order Cumulants of Stationary Processes
2 Citations1988Boaz Port, B. Friedlander