A distributed quickest detection scheme for sensor networks

We consider a distributed quickest detection problem where local sensors provide information about the occurrence of a phenomenon to a fusion center. The objective of the fusion center is to detect the occurrence of the phenomenon as quickly as possible subject to a constraint on the mean time between false alarms. We propose a distributed scheme where each sensor performs a local Page's Test and sends its binary decision to the fusion center. The fusion center invokes a majority vote fusion rule to make the final decision. We compare our distributed scheme to the centralized Page's Test that has access to all sensor observations. Surprisingly, we show that in the low SNR regime, our distributed scheme significantly outperforms the centralized scheme in terms of quickest detection parameters.