Abstract | ||
---|---|---|
Blind deconvolution is considered as a problem of quasi maximum likelihood (QML) estimation of the restoration kernel. Simple closed-form ex- pressions for the asymptotic estimation error are derived. The asymptotic perfor- mance bounds coincide with the Cram´ er-Rao bounds, when the true ML estima- tor is used. Conditions for asymptotic stability of the QML estimator are derived. Special cases when the estimator is super-efficient are discussed. |
Year | DOI | Venue |
---|---|---|
2004 | 10.1007/978-3-540-30110-3_86 | ICA |
Keywords | Field | DocType |
asymptotic analysis,asymptotic stability,quasi maximum likelihood,blind deconvolution | Kernel (linear algebra),Discrete mathematics,Applied mathematics,Mathematical optimization,Blind deconvolution,Exponential stability,Independent component analysis,Numerical analysis,Blind signal separation,Asymptotic analysis,Mathematics,Estimator | Conference |
Citations | PageRank | References |
1 | 0.52 | 8 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Alexander M. Bronstein | 1 | 2978 | 143.17 |
Michael M. Bronstein | 2 | 4032 | 167.52 |
Michael Zibulevsky | 3 | 1087 | 124.28 |
Yehoshua Y. Zeevi | 4 | 610 | 248.69 |