-
Covariance matrix estimation in the singular case using regularized Cholesky factor
Authors:
Olivier Besson
Abstract:
We consider estimating the population covariance matrix when the number of available samples is less than the size of the observations. The sample covariance matrix (SCM) being singular, regularization is mandatory in this case. For this purpose we consider minimizing Stein's loss function and we investigate a method based on augmenting the partial Cholesky decomposition of the SCM. We first deriv…
▽ More
We consider estimating the population covariance matrix when the number of available samples is less than the size of the observations. The sample covariance matrix (SCM) being singular, regularization is mandatory in this case. For this purpose we consider minimizing Stein's loss function and we investigate a method based on augmenting the partial Cholesky decomposition of the SCM. We first derive the finite sample optimum estimator which minimizes the loss for each data realization, then the Oracle estimator which minimizes the risk, i.e., the average value of the loss. Finally a practical scheme is presented where the missing part of the Cholesky decomposition is filled. We conduct a numerical performance study of the proposed method and compare it with available related methods. In particular we investigate the influence of the condition number of the covariance matrix as well as of the shape of its spectrum.
△ Less
Submitted 13 June, 2025; v1 submitted 22 May, 2025;
originally announced May 2025.
-
On the distributions of some statistics related to adaptive filters trained with $t$-distributed samples
Authors:
Olivier Besson
Abstract:
In this paper we analyse the behaviour of adaptive filters or detectors when they are trained with $t$-distributed samples rather than Gaussian distributed samples. More precisely we investigate the impact on the distribution of some relevant statistics including the signal to noise ratio loss and the Gaussian generalized likelihood ratio test. Some properties of partitioned complex $F$ distribute…
▽ More
In this paper we analyse the behaviour of adaptive filters or detectors when they are trained with $t$-distributed samples rather than Gaussian distributed samples. More precisely we investigate the impact on the distribution of some relevant statistics including the signal to noise ratio loss and the Gaussian generalized likelihood ratio test. Some properties of partitioned complex $F$ distributed matrices are derived which enable to obtain statistical representations in terms of independent chi-square distributed random variables. These representations are compared with their Gaussian counterparts and numerical simulations illustrate and quantify the induced degradation.
△ Less
Submitted 3 March, 2021; v1 submitted 26 January, 2021;
originally announced January 2021.
-
Impact of covariance mismatched training samples on constant false alarm rate detectors
Authors:
Olivier Besson
Abstract:
The framework of this paper is that of adaptive detection in Gaussian noise with unknown covariance matrix when the training samples do not share the same covariance matrix as the vector under test. We consider a class of constant false alarm rate detectors which depend on two statistics $(β,\ttilde)$ whose distribution is parameter-free in the case of no mismatch and we analyze the impact of cova…
▽ More
The framework of this paper is that of adaptive detection in Gaussian noise with unknown covariance matrix when the training samples do not share the same covariance matrix as the vector under test. We consider a class of constant false alarm rate detectors which depend on two statistics $(β,\ttilde)$ whose distribution is parameter-free in the case of no mismatch and we analyze the impact of covariance mismatched training samples. More precisely, we provide a statistical representation of these two variables for an arbitrary mismatch. We show that covariance mismatch induces significant variations of the probability of false alarm and we investigate a way to mitigate this effect.
△ Less
Submitted 2 December, 2020;
originally announced December 2020.
-
Wind driven 3D Navier-Stokes circulation in the Atlantic
Authors:
Olivier Besson,
Julien Straubhaar
Abstract:
A finite element method for the numerical solution of the anisotropic Navier-Stokes equations in shallow domain is presented. This method take into account aspect ratio in the hydrostatic approximation of the Navier-Stokes equations \cite{beslay,azthese,azgui}. A projection method \cite{guermond,shen} is used for the time discretization. The linear systems are solved via a some preconditioned conj…
▽ More
A finite element method for the numerical solution of the anisotropic Navier-Stokes equations in shallow domain is presented. This method take into account aspect ratio in the hydrostatic approximation of the Navier-Stokes equations \cite{beslay,azthese,azgui}. A projection method \cite{guermond,shen} is used for the time discretization. The linear systems are solved via a some preconditioned conjugate algorithm, well adapted to massively parallel computers \cite{julien_these,julien_precond,julien_precond_paral}. Some results are presented for the wind driven water circulation in the North Atlantic.
△ Less
Submitted 2 July, 2012;
originally announced July 2012.
-
A Transport for imaging process
Authors:
Olivier Besson,
Martine Picq,
Jérôme Pousin
Abstract:
This work originates from a heart's images tracking which is to generate an apparent continuous motion, observable through intensity variation from one starting image to an ending one both supposed segmented. Given two images $ρ_0$ and $ρ_1$, we calculate an evolution process $ρ(t,\cdot)$ which transports $ρ_0$ to $ρ_1$ by using the optical flow. In this paper we propose an algorithm based on a fi…
▽ More
This work originates from a heart's images tracking which is to generate an apparent continuous motion, observable through intensity variation from one starting image to an ending one both supposed segmented. Given two images $ρ_0$ and $ρ_1$, we calculate an evolution process $ρ(t,\cdot)$ which transports $ρ_0$ to $ρ_1$ by using the optical flow. In this paper we propose an algorithm based on a fixed point formulation and a space-time least squares formulation of the transport equation for computing a transport problem. Existence results are given for a transport problem with a minimum divergence for a dual norm or a weighted $H^1_0$-semi norm, for the velocity. The proposed transport is compare with the transport introduced by Dacorogna-Moser. The strategy is implemented in a 2D case and numerical results are presented with a first order Lagrange finite element, showing the efficiency of the proposed strategy.
△ Less
Submitted 23 June, 2012;
originally announced June 2012.
-
Computing the time-continuous Optimal Mass Transport Problem without Lagrangian techniques
Authors:
Olivier Besson,
Martine Picq,
Jérôme Pousin
Abstract:
This work originates from a heart's images tracking which is to generate an apparent continuous motion, observable through intensity variation from one starting image to an ending one both supposed segmented. Given two images p0 and p1, we calculate an evolution process p(t, \cdot) which transports p0 to p1 by using the optimal extended optical flow. In this paper we propose an algorithm based on…
▽ More
This work originates from a heart's images tracking which is to generate an apparent continuous motion, observable through intensity variation from one starting image to an ending one both supposed segmented. Given two images p0 and p1, we calculate an evolution process p(t, \cdot) which transports p0 to p1 by using the optimal extended optical flow. In this paper we propose an algorithm based on a fixed point formulation and a time-space least squares formulation of the mass conservation equation for computing the optimal mass transport problem. The strategy is implemented in a 2D case and numerical results are presented with a first order Lagrange finite element, showing the efficiency of the proposed strategy.
△ Less
Submitted 17 September, 2010;
originally announced September 2010.
