Distributed detection fusion with fuzzy priori probabilities and fuzzy cost functions

The optimal detection fusion in the sense of minimum Bayesian risk at the fusion center is considered when the a priori probabilities and cost functions are fuzzy. The fusion center receives decisions from various distributed sensors and the optimal detection fusion schemes at the fusion center are derived by using the URI (utility ranking index) and TDC (total distance criterion) fuzzy ordering criteria. It is discovered that the optimal detection fusion rule is a weighted sum of local decisions in this case, that the weight is dependent on the probability of detection P_Di and the probability of false alarm P_Fi of the local detector, and that the threshold depends not only on the fuzzy a priori probabilities and cost functions but also on the fuzzy ordering criterion