Performance evaluation of censoring-enabled systems for sequential detection in large wireless sensor networks

In this paper, we consider a sequential binary hypothesis testing framework in wireless sensor networks. We study the effect of sensor censoring on network performance in terms of the average error probability and average number of observations required until a global decision is made. The detection process is mathematically modeled as a random walk process with two absorbing barriers. We resort to Chernoff bound in order to find upper bounds on the error probabilities and the average stopping time. The main contribution of this paper is to prove that in a sequential binary hypothesis network where sensors send their hard decisions to the fusion center, censoring can enhance the network performance in comparison to non-censoring networks in certain SNR regimes. Numerical evaluation is provided to illustrate the gains achieved through censoring.