Probabilistic PCR6 fusion rule

This chapter defines and implements a non-Bayesian fusion rule for combining densities of probabilities, derived from imprecise knowledge. This rule is the restriction to a strict probabilistic paradigm of the Proportional Conflict Redistribution rule no 6 (PCR6) developed in the DSmT framework for fusing basic belief assignments. A sampling method for probabilistic PCR6 (p-PCR6) is defined. It is shown that p-PCR6 allows to keep the modes of local densities and preserve as much as possible the whole information inherent to each densities to combine. In particular, p-PCR6 is able of maintaining multiple hypotheses/modes, when they are too distant for fusion, contrariwise to classical technique. The question of sequential filtering by p-PCR6 is addressed, thus implying the necessity to handle the redundancy of the information.