Analysis of the Performance of Decentralized Sensor Network with Correlated Observations

Abstract

In this paper, we study the performance of a decentralized sensor network in the presence of correlated additive Gaussian noise. We propose a parallel genetic algorithm approach to simultaneously optimize both the fusion rule and the local decision rules in the sense of minimizing the probability of error. Our results show that the algorithm converges to a majority-like fusion rule irrespective of the degree of correlation and that the local decision rules play a key role in determining the performance of the overall system in the case of correlated observations. We also show that the performance of the system degrades with increase in the correlation between the observations.