New Brunswick, N.J. The National Science Foundation (NSF), in collaboration with the U.S. Department of Homeland Security (DHS), has awarded Rutgers University two grants to lead research on detecting smuggled nuclear materials. The work aims to thwart terrorists from using such materials to develop dirty bombs or environmental contaminants that could sicken large segments of the population and paralyze commerce.
The NSF has committed $631,509 to support the first years research on two projects, with $2.3 million in total funding anticipated over four years. The work will define mathematical principles for how to lay out sensor networks and accurately interpret the vast amounts of data they will capture.
While many vehicles already are being scanned for radiation at major ports, border crossings and urban bridges and tunnels, a detection system based on data from different types of sensors in widespread networks could boost security without hindering the flow of people and goods. Intelligent deployment and coordination of such systems could substantially reduce the costs of protecting persons and property.
Coordinating the research is the Rutgers Center for Dynamic Data Analysis (DyDAn), in collaboration with the School of Information and Library Studies (SCILS) and the Rutgers Center for Operations Research (RUTCOR). DyDAn is a DHS center of excellence, established last year to develop computing technologies that find patterns and relationships in massive amounts of public data. The grants are being funded by the DHS Domestic Nuclear Detection Office.
The goal is to immediately recognize and alert authorities to true threats, while weeding out false alarms, said Fred Roberts, director of DyDAn and its host institute, the Center for Discrete Mathematics and Theoretical Computer Science (DIMACS), based at Rutgers.
The first of the two projects, on sensor management for nuclear detection, will involve faculty from mathematics, computer science, statistics and engineering at Rutgers and two DyDAn partner institutions, Princeton University and Texas State University at San Marcos. The university researchers also will collaborate with scientists at three national research laboratories, Los Alamos, Pacific Northwest and Sandia. Roberts is the principal investigator. First-year funding is $485,767, with $2 million expected over four years.
Sensor management is challenging due to the variety of potential settings for sensor networks, such as seaports, land border crossings, major metropolitan traffic arteries and stadiums. Containers coming off ships and trucks crossing borders, for example, can be directed through scanning devices, but vehicles on bridges or in tunnels need to be examined while moving and tracked if suspicious. Container shipments can receive a quick initial scan, but the readings have to be analyzed quickly and reliably to determine if further examination is warranted. The alternative a detailed inspection of each shipment would quickly choke port traffic. At high-visibility events such as Olympic Games or football championships, temporary sensors could be deployed around stadiums and networked for the duration of the event.
Sensors in such environments also have to distinguish readings caused by legitimate nuclear sources, such as smoke detector shipments or people who have undergone medical tests or treatments, from readings caused by true threats. The second project aims to optimize the value of sensor information to clarify ambiguous or deceptive readings. It will involve faculty from Rutgers SCILS and RUTCOR. Paul Kantor, a professor in SCILS and director of the Center for Interdisciplinary Studies in Privacy and Security, is the principal investigator. First-year funding is $145,742, with $290,000 expected over two years.
Both projects will promote a well-planned defense against illicit nuclear materials with the least possible disruption of our day-to-day lives, while ensuring that our security dollars are spent as effectively as possible, Kantor said.
In addition to aiding research, the grant for sensor management will support workshops and seminars for the scientific community and students, including outreach to high schools and communities on mathematics for homeland security.
These projects are the latest in a series of Rutgers homeland security initiatives, which include privacy-preserving data analysis, epidemic detection and containment, worker and first responder training, and response to natural and man-made disasters.
|Contact: Carl Blesch|
Rutgers, the State University of New Jersey