MINIMAX ROBUST DECENTRALIZED DETECTION
Venugopal V. Veeravalli
Tamer Basar
H. Vincent Poor
Abstract
Decentralized detection problems are studied where the sensor
distributions are not specified completely. The sensor distributions
are assumed to belong to known uncertainty classes. It is shown for a
broad class of such problems that a set of least favorable
distributions exists for minimax robust testing between the
hypotheses. It is hence established that the corresponding minimax
robust tests are solutions to simple decentralized detection problems
for which the sensor distributions are specified to be the least
favorable distributions.