ADAPTIVE FUSION BY REINFORCEMENT LEARNING FOR DISTRIBUTED DETECTION SYSTEMS

Chair and Varshney have derived an optimal rule for fusing decisions b ased on the Bayesian criterion. To implement the rule, the probability of detection P-D and the probability of false alarm P-F for each dete ctor must be known, but this information is not always available in pr actice. An adaptive fusion model which estimates the P-D and P-F adapt ively by a simple counting process is presented, Since reference signa ls are not given, the decision of a local detector is arbitrated by th e fused decision of all the other local detectors, Furthermore, the fu sed results of the other local decisions are classified as ''reliable' ' and ''unreliable.'' Only reliable decisions are used to develop the rule, Analysis on classifying the fused decisions in term of reducing the estimation error is given and simulation results which conform to our analysis are presented.