Detecting the number of signals for an undamped exponential model using cross-validation approach

Abstract Detecting the number of signals and estimating the parameters of the signals are important problems in statistical signal processing. Quite a number of papers appeared in the last 20 years in estimating the parameters of an exponential signal quite efficiently but not that much of attention has been paid in estimating the number of signals of an exponential signal model. Recently, it is observed that different information theoretic criteria can be used to estimate the number of signals in this situation. But it is also observed that the choice of the penalty function is very important particularly for small sample sizes. In this paper we suggest to use the cross-validation technique on estimating the number of signals and give its practical implementation procedures. Numerical experiments reveal that the new procedure performs quite comparable to the best performed information theoretic criteria at least for small sample sizes and it has certain desirable properties also.

[1]  J. Rissanen.  A UNIVERSAL PRIOR FOR INTEGERS AND ESTIMATION BY MINIMUM DESCRIPTION LENGTH , 1983 .

[2]  Debasis Kundu.  Detecting the number of signals for undamped exponential models using information theoretic criteria , 1992 .

[3]  P. Laguna,et al.  Signal Processing , 2002, Yearbook of Medical Informatics.

[4]  Debasis Kundu,et al.  Consistent method of estimating superimposed exponential signals , 1995 .

[5]  C. R. Rao,et al.  Some Recent Results in Signal Detection. , 1986 .

[6]  J. Rissanen,et al.  Modeling By Shortest Data Description* , 1978, Autom..

[7]  M. Stone.  Asymptotics for and against cross-validation , 1977 .

[8]  H. Akaike.  A new look at the statistical model identification , 1974 .

[9]  S. Gupta,et al.  Statistical decision theory and related topics IV , 1988 .

[10]  Ramdas Kumaresan,et al.  ESTIMATING THE PARAMETERS OF EXPONENTIALLY DAMPED OR UNDAMPED SINUSOIDAL SIGNALS IN NOISE , 1982 .

[11]  Yoram Bresler,et al.  Exact maximum likelihood parameter estimation of superimposed exponential signals in noise , 1986, IEEE Trans. Acoust. Speech Signal Process..

[12]  M. Stone.  Cross‐Validatory Choice and Assessment of Statistical Predictions , 1976 .

[13]  Calyampudi R. Rao,et al.  Linear statistical inference and its applications , 1965 .

[14]  R. Kumaresan,et al.  Estimation of frequencies of multiple sinusoids: Making linear prediction perform like maximum likelihood , 1982, Proceedings of the IEEE.

[15]  Debasis Kundu,et al.  Estimating the Parameters of Undamped Exponential Signals , 1993 .

[16]  Z. Bai,et al.  On detection of the number of signals in presence of white noise , 1985 .

[17]  M. Bartlett.  A Note on the Multiplying Factors for Various χ2 Approximations , 1954 .

[18]  D. Lawley.  TESTS OF SIGNIFICANCE FOR THE LATENT ROOTS OF COVARIANCE AND CORRELATION MATRICES , 1956 .

[19]  J. Varah.  On Fitting Exponentials by Nonlinear Least Squares , 1982 .

[20]  Petre Stoica.  List of references on spectral line analysis , 1993, Signal Process..

[21]  Debasis Kundu,et al.  On asymptotic behavior of least squares estimators and the confidence intervals of the superimposed exponential signals , 1999, Signal Process..

[22]  M. Stone.  An Asymptotic Equivalence of Choice of Model by Cross‐Validation and Akaike's Criterion , 1977 .

[23]  D. Kundu.  Shorter Communication: Estimating the Parameters of Undamped Exponential Signals , 1993 .

[24]  Peter A. Lachenbruch,et al.  Some unsolved practical problems in discriminant analysis , 1975 .

[25]  H. Akaike,et al.  Information Theory and an Extension of the Maximum Likelihood Principle , 1973 .

[26]  G. Schwarz.  Estimating the Dimension of a Model , 1978 .