Decentralized binary detection with noisy communication links

This correspondence presents a Bayesian framework for distributed detection in sensor networks with noisy communication links between the sensors and the fusion center (or access point (AP)). Noisy links are modeled as binary symmetric channels (BSCs), but the proposed framework can be extended to other communication link models. To improve the system robustness against observation and communication noises, we propose schemes with 1) multiple observations and a single AP and 2) single observations and multiple APs. By using the De Moivre-Laplace approximation, we derive simple and accurate expressions for the probability of decision error in scenarios with a large number of nodes.