Biological Agent Detection Network

Physical Sciences Inc. in cooperation with University of Notre Dame proposes to develop a method for the persistent surveillance and detection of aerosolized biological warfare agents using a distributed point bioaerosol detector network, local meteorological sensors, existing infrared surveillance cameras, and crowdsourced georeferenced network traffic and phrase monitoring. The proposed technology will utilize a propriety technique for optimal sensor placement, and an urban plume dispersion model to predict concentration profile measurements and sensor intercept times, resulting in networked threat confidence greater than the individual sensor level. A Bayesian classifier will be used to track possible threats and assess the probability that they contain a biological agent. The proposed system is designed as a wide-area surveillance system, where the ultimate goal is timely treatment and post event triage. The specific capability to be developed under this proposed effort is the wide-area point sensor system and accompanying algorithms. In the Phase I effort PSI will design a point sensor network, and develop sensor fusion and biological agent detection algorithms. These algorithms will be validated using a synthetic dataset based chemical agent release measurements, acquired at Dugway Proving Grounds.