Statistical Non-Gaussian Model of Sea Surface with Anisotropic Spectrum for Wave Scattering Theory. Part I

Tatarskiǐ, V. I.; Tatarskii, V. V.

doi:10.2528/pier98120230

Cited by 6 publications

(3 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With such an interpretation, weights {w n } have the meaning of the probabilities that a realization belongs to a certain nor mal set. Such an approach obviously corresponds to nonergodic poly Gaussian random processes [28] and cannot be used for constructing the models describing uniform roughness profiles.…”

Section: Mixtures Of Probability Distributionsmentioning

confidence: 99%

“…The repre sentation of random processes by a combination of Gaussian distributions was analyzed theoretically in [25][26][27], where the completeness of the Gaussian functions in L 2 (-∞, ∞) was proved and the possibility of an arbitrarily exact description of an arbitrary com plex valued function by a finite number of Gaussian components was demonstrated. Poly Gaussian super positions have found wide application in describing random non Gaussian signals and noise for solving problems of laser ranging and approximation of distri butions of random quantities encountered in technical applications [28,29].…”

Section: Mixtures Of Probability Distributionsmentioning

confidence: 99%

See 1 more Smart Citation

Poly-Gaussian models of a non-Gaussian randomly rough surface

Litvak

Malyugin

2012

Tech. Phys.

View full text Add to dashboard Cite

Poly Gaussian models are developed for non Gaussian random processes, which make it possible to describe and imitate rough surfaces with various densities of roughness height distribution and correlation properties; the algorithms for numerical and analytical calculations of statistical characteristics of non Gaus sian reliefs are also worked out. Examples are given for the application of the integral poly Gaussian model.

show abstract

Section: Mixtures Of Probability Distributionsmentioning

confidence: 99%

Section: Mixtures Of Probability Distributionsmentioning

confidence: 99%

Poly-Gaussian models of a non-Gaussian randomly rough surface

Litvak

Malyugin

2012

Tech. Phys.

View full text Add to dashboard Cite

show abstract

“…Existing works on modeling and fitting clutter amplitude distribution include Gaussian based Rayleigh distribution and Compound-Gaussian distributions such as Weibull, Log-normal, and K, etc. [1][2][3][4][5]. Based on these studies, clutter amplitude distribution is further incorporated into design and analysis of constant false alarm ratio (CFAR) system [6,7], small target detection [8], target recognition [9], and so on.…”

Section: Introductionmentioning

confidence: 99%

A Modified Goodness-of-Fit Measurement for Radar Clutter Amplitude Statistics

Liang¹,

Kong²,

Yang³

2012

PIER M

View full text Add to dashboard Cite

This paper addresses the application of measurement on goodness-of-fit (GoF) for amplitudes of radar clutter sample data against reference/theoretic parameterized probability density function (PDF). In general, various existing methods for this problem highly depend on empirical PDF parameters. This makes GoF assessments with these methods less perceivable and their accuracies are hard to control. A new method based on chi-squared type of measurement is proposed to overcome these difficulties. This method evaluates GoF by estimating the distance between the true PDF of the clutter data amplitude and the reference PDF. Hence the distance is statistically approximately independent with empirical PDF parameters. The new method has higher accuracy and symmetric property. It is especially useful for GoF comparison over multiple radar clutter data sets.

show abstract