Sequential decision fusion for abnormality detection via diffusive molecular communications

This paper considers the task of abnormality detection in a fluid medium, employing a molecular communications (MC) based network of nanoscale sensors. This task entails sensing, detection and reporting of abnormal changes in the environment that may characterize a disorder or an abnormal event. Such distributed detection (DD) problems are of paramount interest, especially in applications such as health monitoring, disease diagnosis, targeted drug delivery, environmental sensing and monitoring, contaminant detection and removal, and environmental remediation. This letter proposes, for the first time in the literature, to employ a sequential probability ratio test based approach to the decision fusion in diffusive MC based DD. The proposed approach leads to considerable gains in the average number of samples required for the decision compared to its fixed-sample size counterparts, resulting in a significant improvement in the average decision delay. In the investigated DD scenarios, we observe savings of up to 50% in the number of samples required for decision fusion.