|
Abstract:
|
Pervasive surveillance and monitoring provides an effective way to protect against terrorist attack by illicit nuclear materials . The advancement of technology makes nuclear detection devices both inexpensive and portable, while the GPS (Global Positioning System) is becoming commonly available. It is feasible for mass productions and installations of such devices on taxi cabs, police vehicles, fire trucks, and public transit systems in major U.S. cities. In this paper we propose nuclear signal models and sensor detection models, and develop a spatial multi-cluster detection method to analyze the signals from the proposed sensor network. The development is based on a latent modeling framework and the method can be used to detect multiple clusters of varying sizes simultaneously. For illustration, we conduct a simulation study for cluster detection based on several practical scenarios.
|
