Recursive Estimation of the False Alarm Probability using Approximations of Mixtures

S. Nakamori (Japan), R.Caballero-Águila, A. Hermoso-Carazo, J. Jiménez-López, and J. Linares-Pérez (Spain)


Identification, Estimation, Filtering.


The estimation problem of the false alarm probability from uncertain observations is addressed from a bayesian view point. Considering successive approximations of mixture distributions, approximations of the a posteriori densities of the unknown probability are obtained from an arbitrary a priori distribution. Then, by taking a Beta as a priori distribution, a recursive algorithm that provides approxi mations for the Bayes estimators of the false alarm proba bility is deduced. The estimators do not use the state-space model, but only covariance information about the processes involved in the observation equation, as well as the a priori distribution.

Important Links:

Go Back