Showing papers on "Complex normal distribution published in 1992"
TL;DR: In this paper, the moduli of the eigenvalues of a complex Gaussian matrix in terms of x 2 distributions were characterized and shown to be stochastically smaller than the norm of a k × (k + 1) real Gaussian matrices.
116 citations
Metz1
TL;DR: In this paper, analogues of the normal distribution in Euclidean space for orientations represented by Rodrigues parameters are discussed, and it is emphasized that different characterizations of normal distribution can lead to different distributions in other spaces, none of which is mathematically better than any other one.
Abstract: Analogues of the normal distribution in Euclidean space for orientations represented by Rodrigues
parameters are discussed. It is emphasized that different characterizations of the normal distribution
in Euclidean space lead to different distributions in other spaces, none of which is mathematically
superior to any other one. Particular analogues of the normal distribution are the Bingham
distribution on S + 4 for the purposes of mathematical statistics, and the Brownian motion distribution
on S + 4 in terms of probability theory and stochastic processes. It is reminded of the fact that a simple
analogue of the central limit theorem in Euclidean space does not exist for the hyperspheres S P and
projective hyperplanes H P − 1 = S + 4 .
23 citations
TL;DR: In this paper, the authors compared the performance of the singular value decomposition (SVD) and the total least-squares (TLS) methods for spectral analysis of FIFO signals.
11 citations
TL;DR: In this article, two definitions of environmental factors for the log normal distribution are proposed and point estimates, upper and lower confidence limits are derived by classical and Bayesian methods, whose performances are discussed.
7 citations
26 Oct 1992
TL;DR: The problem of generating a positive definite maximum likelihood (ML) estimate of a complex Toeplitz covariance matrix given one N-length data vector is considered and it is shown that the measure of the set of such data vectors is small and converges to zero as the length of the data vector increases.
Abstract: The problem of generating a positive definite maximum likelihood (ML) estimate of a complex Toeplitz covariance matrix given one N-length data vector is considered. The data vector is assumed to be drawn from a complex Gaussian population with mean zero and covariance sigma /sup 2/I. An upper bound is derived on the measure of the set of such N-length data vectors such that, for one data vector, the ML estimation procedure yields a positive definite solution. These data vectors are denoted as those that do not satisfy the failure condition of the ML estimation procedure, and it is shown that the measure of the set of such data vectors is small and converges to zero as the length of the data vector increases. >
2 citations