Impact of Local Decision Rules in Distributed Sensor Networks

Abstract

In this paper, we analyze the impact of the local decision rules on the performance of a distributed sensor network. We do this by fixing the fusion rule to be one of the three common binary decision rules: majority voting, AND and OR rules, and optimizing the local decision rules with respect to the probability of error at the fusion center. We assume that the sensor observations are correlated and we consider the general case of heterogeneous (non-identical) sensors. We show that systems having different fusion rules can all provide similar performance if the local decision rules are chosen appropriately.