Geometric barycenters for covariance estimation in compound‐Gaussian clutter
- 1 March 2017
- journal article
- Published by Institution of Engineering and Technology (IET) in IET Radar, Sonar & Navigation
- Vol. 11 (3), 404-409
- https://doi.org/10.1049/iet-rsn.2016.0092
Abstract
The authors consider the problem of covariance matrix estimation in heterogeneous environments for radar signal processing applications, where the secondary data exhibit heterogeneity in local power and share the same covariance structure. Without resorting to the complete statistical characterisation of the sample support, a class of estimators, each of them defined as the geometric barycenter of a set of basic covariances estimates (obtained from the available secondary data) with a specific distance employed, is proposed. The basic estimates are obtained by exploiting the characteristics of positive-definite matrix space and a condition number upper bound constraint. Finally, they evaluate the detection capabilities of an adaptive normalised matched filter with the proposed estimators in the presence of compound-Gaussian disturbance comparing it with existing alternatives.Keywords
Funding Information
- National Natural Science Foundation of China (61201276, 61178068, 61301266)
This publication has 25 references indexed in Scilit:
- Optimality Claims for the FML Covariance Estimator with respect to Two Matrix NormsIEEE Transactions on Aerospace and Electronic Systems, 2013
- Covariance matrix estimation via geometric barycenters and its application to radar training data selectionIET Radar, Sonar & Navigation, 2013
- Information Geometry of Covariance Matrix: Cartan-Siegel Homogeneous Bounded Domains, Mostow/Berger Fibration and Fréchet MedianPublished by Springer Science and Business Media LLC ,2012
- Maximum Likelihood Estimation of a Structured Covariance Matrix With a Condition Number ConstraintIEEE Transactions on Signal Processing, 2012
- Barycenters in Alexandrov spaces of curvature bounded belowAdvances in Geometry, 2012
- Unified Framework to Regularized Covariance Estimation in Scaled Gaussian ModelsIEEE Transactions on Signal Processing, 2011
- Non-Euclidean statistics for covariance matrices, with applications to diffusion tensor imagingThe Annals of Applied Statistics, 2009
- Geometric Means in a Novel Vector Space Structure on Symmetric Positive‐Definite MatricesSIAM Journal on Matrix Analysis and Applications, 2007
- Fast converging adaptive processor or a structured covariance matrixIEEE Transactions on Aerospace and Electronic Systems, 2000
- Rapid Convergence Rate in Adaptive ArraysIEEE Transactions on Aerospace and Electronic Systems, 1974