Sequential Fusion for Distributed Detection Over BSC Channels in an Inhomogeneous Sensing Environment

In this work, we propose a sequential fusion scheme for distributed detection in an inhomogeneous sensing environment. More specifically, the transmissions of sensors are ordered according to the observation quality and the communication channel quality to reduce energy consumption in wireless sensor networks. We devise the corresponding stopping and decision rules, as well as two transmission ordering rules, based on the log-likelihood ratio and mutual information, respectively. The simulation results demonstrate that the proposed scheme can reduce the required number of transmissions substantially, while achieving the same detection performance as the conventional maximum a posteriori probability fusion scheme.