Distributed Binary Detection over Fading Channels: Cooperative and Parallel Architectures

Abstract

This paper considers the problem of binary distributed detection of a known signal in correlated Gaussian sensing noise in a wireless sensor network, where the sensors are restricted to use likelihood ratio test (LRT), and communicate with the fusion center (FC) over bandwidth-constrained channels that are subject to fading and noise. To mitigate the deteriorating effect of fading encountered in the conventional parallel fusion architecture, in which the sensors directly communicate with the FC, we propose new fusion architectures that enhance the detection performance, via harvesting cooperative gain (so-called decision diversity gain). In particular, we propose: (i) cooperative fusion architecture with Alamouti's space-time coding (STC) scheme at sensors, (ii) cooperative fusion architecture with signal fusion at sensors, and (iii) parallel fusion architecture with local threshold changing at sensors. For these schemes, we derive the LRT and majority fusion rules at the FC, and provide upper bounds on the average error probabilities for homogeneous sensors, subject to uncorrelated Gaussian sensing noise, in terms of signal-to-noise ratio (SNR) of communication and sensing channels. Our simulation results indicate that, when the FC employs the LRT rule, unless for low communication SNR and moderate/high sensing SNR, performance improvement is feasible with the new fusion architectures. When the FC utilizes the majority rule, such improvement is possible, unless for high sensing SNR.