QML Blind Deconvolution: Asymptotic Analysis
نویسندگان
چکیده
Blind deconvolution is considered as a problem of quasi maximum likelihood (QML) estimation of the restoration kernel. Simple closed-form expressions for the asymptotic estimation error are derived. The asymptotic performance bounds coincide with the Cramér-Rao bounds, when the true ML estimator is used. Conditions for asymptotic stability of the QML estimator are derived. Special cases when the estimator is super-efficient are discussed.
منابع مشابه
Quasi maximum likelihood blind deconvolution: super- and sub-Gaussianity vs. asymptotic stability
In this note we consider the problem of quasi maximum likelihood (QML) blind deconvolution. We examine two classes of estimators, which are commonly believed to be suitable for superand sub-Gaussian sources. We state the asymptotic stability conditions and demonstrate a distribution, for which the studied estimators result unsuitable, in the sense that they are asymptotically unstable.
متن کاملProbabilistic Formulation of Independent Vector Analysis Using Complex Gaussian Scale Mixtures
We propose a probabilistic model for the Independent Vector Analysis approach to blind deconvolution and derive an asymptotic Newton method to estimate the model by Maximum Likelihood.
متن کاملAsymptotic Diversity Analysis of Alamouti Transmit Diversity with Quasi-ML Decoding Algorithm in Time-Selective Fading Channels
This paper addresses the asymptotic diversity performance of Alamouti transmit diversity technique especially in time-selective fading channels. In particular, a linear quasi-maximum-likelihood (QML) decoding method is employed for the Alamouti transmit diversity system to solve the error floor problem induced by the conventional linear ML decoding method. By judiciously utilizing the derived a...
متن کاملQML Estimators in Linear Regression Models with Functional Coefficient Autoregressive Processes
This paper studies a linear regression model, whose errors are functional coefficient autoregressive processes. Firstly, the quasi-maximum likelihood QML estimators of some unknown parameters are given. Secondly, under general conditions, the asymptotic properties existence, consistency, and asymptotic distributions of the QML estimators are investigated. These results extend those of Maller 20...
متن کاملGeneralized mutual information approach to multichannel blind deconvolution
In this paper, we propose an approach to the multichannel blind deconvolution problem, based on the mutual information criterion and more generally on an appropriate system of estimating equations. Formulas for the quasiNewton algorithm and the asymptotic covariance matrix of the estimator are provided. More interesting results have been obtained in the pure deconvolution case. By a clever para...
متن کامل