Optimal state space partitioning
C. Olivier · IEEE Transactions on Systems Man and Cybernetics · 1994
The partitioning problem of a parameter state space /spl Omega/ into observation subsets is addressed. The initial knowledge about this parameter is a prior probability distribution over /spl Omega/. This distribution is recursively updated through parallel observation results, that are actually binary informations about the presence or the absence of the parameter inside subsets /spl omegasub i/ of /spl Omega/. Each subset is scanned with some errors, corresponding to misdetections and false alarms. It is shown how the partitioning of /spl Omega/ into the {/spl omegasub i/} may be optimized under different optimality criteria related to various measures of the "information" contained in the posterior probability density function. Simulations results are presented and computability issues are discussed.>