Ústav teorie informace a automatizace

Jste zde

Bibliografie

Journal Article

Cramer-Rao-Induced Bound for Blind Separation of Stationary Parametric Gaussian Sources

Doron E., Yeredor A., Tichavský Petr

: IEEE Signal Processing Letters vol.14, 6 (2007), p. 417-420

: CEZ:AV0Z10750506

: 1M0572, GA MŠk

: blind source separation, independent component analysis, autoregressive, ARMA, moving average stationary Gaussain random processes

(eng): The performance of blind source separation algorithms is commonly measured by the output interference to signal ratio (ISR). In this paper we derive an asymptotic bound on the attainable ISR for the case of Gaussian parametric (auto-regressive (AR), moving-average (MA) or ARMA) processes. Our bound is induced by the Cramer-Rao bound on estimation of the mixing matrix. We point out the relation to some previously obtained results, and provide a concise expression with some associated important insights. Using simulation, we demonstrate that the bound is attained asymptotically by some asymptotically efficient algorithms.

(cze): Kvalita algoritmu pro slepou separaci je měřena pomocí výstupního poměru interference vůči šumu (ISR). V tomto článku je odvozena asymptotická mez pro ISR pro Gaussovské parametrické (AR, ARMA, MA) stacionarní procesy. V simulacích je ukázáno, že existují asymptoticky eficientní algoritmy, které této hranice dosahují.

: BB

07.01.2019 - 08:39