0411034 20240103182257.020060210235959.9 15 s. cav_un_epca*02567720890-6327171 (2003)51-65Wiley mixture models estimation Bayesian approach cav_un_auth*0101167 Nagy Ivan UTIA-B Department of Signal Processing Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0101124 Kárný Miroslav UTIA-B Department of Adaptive Systems Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0101168 Nedoma Petr UTIA-B Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0213007 Voráčová Š. CZ http://library.utia.cas.cz/prace/20030021.ps 09I GA102/03/0049 GA ČR cav_un_auth*0001805 IBS1075351 GA AV ČR cav_un_auth*0001804 CEZ:AV0Z1075907 The paper deals with modelling and estimation by a model described as a mixture of distributions. In this case, the exact application of the Bayes theory adopted is not feasible and its approximation is used. The general algorithm is specified for mixtures with components from exponential family. The theory is demonstrated on estimation of the basic relation between density and intensity of traffic flow in a single point of a vehicular communication and it can provides us with state classification. BB AS http://hdl.handle.net/11104/0131121 UTIA-B 20030021 2003 cav_un_epca*0256772 International Journal of Adaptive Control and Signal Processing 0890-6327 1099-1115 Roč. 17 č. 1 2003 51 65 Wiley