Showing papers by "Louis L. Scharf published in 1993"
TL;DR: This paper assumes that the parameters structure the mean value vector in a multivariate normal distribution and the Fisher matrix is a Gramian constructed from the sensitivity vectors that characterize the first-order variation in the mean with respect to the parameters.
111 citations
TL;DR: The accuracy with which deterministic modes can be identified from a finite record of noisy data is determined by computing the Cramer-Rao bound on the error covariance matrix of any unbiased estimator of mode parameters.
Abstract: The accuracy with which deterministic modes can be identified from a finite record of noisy data is determined by computing the Cramer-Rao bound on the error covariance matrix of any unbiased estimator of mode parameters. The bound is computed for many of the standard parametric descriptions of a mode, including autoregressive and moving-average parameters, poles and residues, and poles and zeros. Asymptotic, frequency-domain versions of the Cramer-Rao bound bring insight into the role played by poles and zeros. Application of the bound to second- and fourth-order systems illustrates the influence of mode locations on the ability to identify them. Application of the bound to the estimation of an energy spectrum illuminates the accuracy of estimators that presume to resolve spectral peaks. >
60 citations
01 Nov 1993
TL;DR: In this article, the generalized likelihood ratio (GLR) test is used to detect subspace signals in subspace interference and broadband noise, and it is shown that the GLR test is the uniformly most powerful invariant detector.
Abstract: Studies the problem of detecting subspace signals in subspace interference and broadband noise. The authors derive the generalized likelihood ratio (GLR) and establish its invariances. Then then prove that the GLR test (GLRT) is the uniformly most powerful invariant detector, establishing once and for all its optimality. >
41 citations
01 Nov 1993
TL;DR: In this paper, the properties of quadratic covariance bounds for parametric estimators are investigated and a projection operator and integral/kernel representation for this class of bounds are introduced.
Abstract: We investigate the properties of quadratic covariance bounds for parametric estimators. The Cramer-Rao, Bhattacharyya (1946), and Barankin (1949) bounds have this quadratic structure and the properties of these bounds are uniquely determined by their respective score functions. We enumerate some characteristics of score functions which generate tight bounds. We also introduce projection operator and integral/kernel representations for this class of quadratic covariance bounds. These representations are useful as analysis and synthesis tools. We also address the issue of efficiency for this class of bounds. >
16 citations
01 Nov 1993
TL;DR: In this paper, the authors extend the work of Tufts, Kot, and Vaccaro (1987, 1991) to improve the analytical characterization of threshold breakdown in SVD methods and lower bound the probability of a subspace swap in the SVD.
Abstract: We extend the work of Tufts, Kot, and Vaccaro (1987, 1991) to improve the analytical characterization of threshold breakdown in SVD methods. Our results sharpen the Tufts, Kot, and Vaccaro results and lower bound the probability of a subspace swap in the SVD. >
6 citations