Interactive fusion in distributed detection: Architecture and performance analysis

Abstract

Within the Neyman-Pearson framework we investigate the effect of feedback in two-sensor tandem fusion networks with conditionally independent observations. While there is noticeable improvement in performance of the fixed sample size Neyman-Pearson (NP) test, it is shown that feedback has no effect on the asymptotic performance characterized by the Kullback-Leibler (KL) distance. The result can be extended to an interactive fusion system where the fusion center and the sensor may undergo multiple steps of interactions.